Morbidity after neck dissection in head and neck1 Morbidity after neck dissection in head and neck...

University of Groningen

Morbidity after neck dissection in head and neck cancer patientsWilgen, Cornelis Paul van

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2004

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Wilgen, C. P. V. (2004). Morbidity after neck dissection in head and neck cancer patients: a studydescribing shoulder and neck complaints, and quality of life. s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 23-02-2020

https://www.rug.nl/research/portal/en/publications/morbidity-after-neck-dissection-in-head-and-neck-cancer-patients(92f7282a-cfda-4607-a337-72b1fec6e7cf).html



1

Morbidity after neck dissection in head and neck

cancer patients; a study describing shoulder and neck

complaints, and quality of life.

2

Cover: Steffen van Bergenhenegouwen, Benjamin R. Hol

The publication of this thesis was supported by:Nucletron B.V., ABBOTT B.V., Stichting Beatrixoord Noord-Nederland

Wilgen, Cornelis Paul

Morbidity after neck dissection in head and neck cancer patients;

a study describing shoulder and neck complaints, and quality of life.

3

RIJKSUNIVERSITEIT GRONINGEN

Morbidity after neck dissection in head and neckcancer patients; a study describing shoulder and neck

complaints, and quality of life

Proefschrift

ter verkrijging van het doctoraat in de

Medische Wetenschappen

aan de Rijksuniversiteit Groningen

op gezag van de

Rector Magnificus, dr. F. Zwarts,

in het openbaar te verdedigen op

woensdag 9 juni 2004

om 16.15 uur

door

Cornelis Paul van Wilgen

geboren op 11 oktober 1967

te Groningen

4

Promotor: Prof. dr. J.L.N. Roodenburg

Copromotor Dr. P.U. Dijkstra

Beoordelingscommissie: Prof. dr. J.H.B. Geertzen

Prof. dr. G.J. Hordijk

Prof. dr. J.M.K.H. Wierda

ISBN: 90.367.202.14

5

Contents

Chapter 1 Pagina 7Introduction

Chapter 2 Pagina 19Shoulder pain and disability in daily life, following supraomohyoid neck dissection: a pilot study.Journal of Cranio-Maxillofacial Surgery (2003) 31:183-186

Chapter 3 Pagina 29Incidence of shoulder pain after neck dissection: a clinical explorative study for risk factors.Head and Neck (2001) Nov.:947-953

Chapter 4 Pagina 45Shoulder complaints after nerve sparring neck dissection.Int. Journal of Oral and Maxillofacial Surgery (2004),33(3):253-257

Chapter 5 Pagina 59Shoulder complaints after neck dissection; is the spinal accessory nerve involved?British journal of oral and maxillofacial surgery (2003),41:7-11

Chapter 6 Pagina 73Morbidity of the neck after head and neck cancer therapy.Accepted Head and Neck (december 2003)

Contents

6

Chapter 7 Pagina 91Shoulder and neck morbidity in quality of life after surgery for head and neck cancer. Accepted Head and Neck (februari 2004)

Chapter 8 Pagina 105Measuring somatic symptoms with the CES-D to assess depression in cancer patients after treatment, valid or not?(Comparison between patients with head and neck, gynaecological, colo-rectal, and breast cancer) Submitted

Chapter 9: General Discussion Pagina 119

Chapter 10: Summary Pagina 125Samenvattingen Pagina 132Dankwoord Pagina 139Appendix 1 and 2 Pagina 142List of publications Pagina 145

Contents

7

CHAPTER 1

INTRODUCTION

8

A century after Crile (1906)

It is almost a century since Crile1 described the radical neck dissection forpatients with head and neck cancer. In that paper Crile complained about thelack of attention of other physicians concerning the surgical progress in thetreatment of head and neck cancer patients. Now, almost one hundred yearslater, many things have changed.In his historical paper Crile described, as one of the first, the importance ofremoving lymphatic structures of the neck (the complete lymphatic block orradical neck dissection) in head and neck cancer patients. In great detail hedescribed the anaesthesiology procedures, the surgical techniques, and thesacrificed structures, all illustrated with anatomical sketches. Additionally hedescribed complications like: infections, hemorrhage, shock and collapse. Inthis same paper an evaluation study was presented in which he comparedpatients operated with a radical neck dissection (n = 12) to patients in whichthe lymphatic structures were not removed (n = 48). Patients operated with aradical neck dissection had a survival rate that was four times higher thanpatients without a radical neck dissection. Crile’s paper is one of thekeystones of head and neck surgery, the radical neck dissection is still thesurgical standard against which various modifications are compared. Crile did not describe morbidity after surgery with radical neck dissection.Sacrificing the accessory nerve was not even mentioned in the text but onlyexplained in one of the sketches as one of the structures to be sacrificed.Almost 50 years after the paper of Crile, Ewing (1952) was one of the firstwho described morbidity after radical neck dissection.2 Several types ofmorbidity were mentioned: disfigurement, pain in the shoulder region, lossof strength, reduced range of motion, loss of function of the shoulder,sensory disturbance, and restrictions in daily activities. Up to 62 % of thepatients, operated with a radical neck dissection, had disfigurement of theshoulder and 46 % experienced shoulder pain. Other authors describedhigher incidences of shoulder complaints after radical neck dissection, evenup to 100 %.3 Shoulder complaints after radical neck dissection arecompletely attributed to sacrificing of the spinal accessory nerve. The high morbidity rates, and the gained insight in the biological behaviourof various cancer types, to estimate if a tumour has metastasised, led tomodifications on the classical radical neck dissection. Bocca (1980)

Chapter 1

9

described the functional neck dissection4, later to be called the modifiedradical neck dissection.5 In this modified radical neck dissection one or more of thefollowing non-lymphatic structures are preserved: spinal accessory nerve,sternocleidomastoideus muscle, and internal jugular vene. The spinalaccessory nerve seemed an important structure to preserve, in order toprevent shoulder morbidity. After modified radical neck dissections, withpreservation of the spinal accessory nerve, the prevalence of shouldercomplaints decreased, the prevalence rates were 18% to 61%.6,7 Furthermodifications in neckdissections were developed by removing only certainlevels of lymphatic structures, that are prone to metastasis, instead ofremoving all levels. Four selective neck dissections (figure 1) are described:supraomohyoid neckdissection (levels I, II, and III), lateral neckdissection(levels II,III, and IV) posterolateral neckdissection (levels II,III,IV,and V),and anterior neckdissection (level VI). 5 In these selective neckdissections allnon-lymphatic structures (spinal accessory nerve, sternocleidomastoideusmuscle, and internal jugular vene) are preserved. Supraomohyoid neckdissections are often performed in case of oropharyngeal squamous cellcarcinoma, with a N0 tumour status and possible sufficient in the treatmentof a selected group of patients with positive nodes at level I. 8,9

Figure 1 Classification of levels used for neck dissections (1a). Anatomical structures that cross the cervical lymphatic structures and lymph nodes (1b). Situation after a radical neck dissection (1c).

1 a 1 b 1c(pictures from Mondziekten & Kaakchirurgie, B. Stegenga, A. Vissink, LGM de Bont)

Chapter 1

10

Recently the classification of Robbins of 1991 has been updated. In this newclassification the levels have been further modified. The levels I, II and Vhave been split into two levels, and are to be called level I a/b, level II a/band level V a/b.10 (figure 2)

Figure 2 New classification described by Robbins et al. 2002 10

Morbidity after selective neck dissections has been described scarcely, andthe studies included relatively small samples.3,11,12 Pinsolle described a groupof 41 patients after supraomohyoid neckdissection of which 32% had minorproblems, 5 % moderate and 3 % severe shoulder problems. 13 Probablyshoulder morbidity rates have decreased further as a result of selectiveneckdissections, but the evidence is still limited.

Shoulder complains after neck dissection

Sacrificing the spinal accessory nerve

The spinal accessory nerve is a motor nerve, innervating thesternocleidomastoideus muscle and the trapezius muscle. Resection of thespinal accessory nerve leads to denervation of the trapezius muscle. Thetrapezius muscle exists of an upper, middle and lower part and has twomajor functions, shrugging the shoulder and stabilising the scapula on thethorax. Paralysis of the trapezius muscle will lead to a lateral gliding of thescapula and a lateral rotation.14 (figure 3)

Chapter 1

11

Figure 3 Lateral rotation of the scapula and changed shoulder joint position

As a consequence of the changed scapula position its possibility to moveduring shoulder movements decreases. As a consequence a reduced range ofmotion of the shoulder, abduction and forward flexion, appears. (figure 4) Areduced range of motion of the shoulder can cause dysfunction in activitiesin which the shoulder is needed like lifting heavy objects or reaching aboveshoulder level. 6,15 Several patients also complain about pain after head andneck cancer treatment. The exact cause of this pain is not properlyinvestigated. Several possible causes have been described like frozenshoulder16, sternoclavicular joint hypertrophy17, or myofascial pain as aconsequence of stretching of muscles due to the changed scapula position.18

But these hypothesised causes for shoulder pain have never beeninvestigated properly. Is some cases, after resection of the spinal accessory nerve, the trapeziusmuscle function remains (partly) intact. This is due to a double innervationby the cervical plexus of the musculus trapezius in about 18% of thepatients.14

Preservation of the spinal accessory nerve

Several patients after neck dissection with preservation of the spinalaccessory nerve still suffer from shoulder complaints. The mechanisms ofthe shoulder complaints after modified or selective procedures with sparringof the spinal accessory nerve are unclear.

Chapter 1

View from above

View from behind

Figure 4 Scapula alata as a consequence of trapezius atrophy (4a) and a reduced abduction as a consequence of a changed scapula position (4b).

During neck disssurrounding structmay result in a neuresults in a neurotpresumed preservaconsequence shouland trapezius muscinterpreted as neuresearch has been dclinical importance

More morbidity aBeside shoulder morbidity. Little iscancer treatment.sensation, pain, oliterature only a feperformed on smConsequences of n

Chapter 1

4a

ection the spinal accessory nerveures, and its supplying blood vesselsropraxia or the nerve may be dissectemesis (permanent deficit). In these ction leads to spinal accessory nerve der complaints. But with an intact sple function still shoulder complaints ropathic pain or myofascial pain.19

one to substantiate these hypothesis for treatment possibilities.

fter head and neck cancer treatmencomplaints patients may suffer fr known about morbidity of the neck

Clinically several patients compr a reduced range of motion of thw papers describe this morbidity.19,20

all samples, often without a pheckdissection with or without radiati

4b

12

is lifted from its are dissected whichd accidentally whichases preservation ordysfunction and as ainal accessory nervemay arise which are Unfortunately littlewhereas it is of great

tom other types of after head and necklain about loss ofe cervical spine. In,21 These studies areysical examination.on therapy on range

13

of motion of the cervical spine are only described in studies that usedquestionnaires without actually measuring range of motion.22 Loss ofsensation is thoroughly described in only one study by Saffold,23 althoughthat study was primarily aiming at the results of the preservation of cervicalroot branches in selective neck dissections. The influence of radiation therapy on loss of sensation, and range of motionis scarcely described, although clinically patients seem to suffer fromfibrosis of soft tissues which decreases range of motion. Pain is probably the most important type of morbidity of patients after headand neck cancer treatment. As mentioned before some hypotheses exist ashow to explain post cancer treatment pain. Pain can be explained asneuropathic pain, pain in the clavicular or acromio-clavicular joint (probablynociceptive) and myofacial pain. The causes of pain are still hardly studiedbut clinically important for treatment modalities. Treatment modalities mightbe medication, physical therapy, and informing patients.

Psychological consequences after head and neck cancer therapyAfter head and neck cancer treatment also psychological consequences aredescribed. Beside the physical morbidity, psychological as well as socialproblems have their effect on well being and quality of life. 24 Psychologicalconsequences may be depression,25 distress,26 fear of recurrence,27

inadequate coping strategies28 problems in reintegration in work and lack ofsocial support.29 After cancer treatment depression is, with a prevalence of24 % (range 1.5 % - 50 %), an important psychological morbidity.30

Depression has a construct in which physical and psychological domains arecombined. Depression effects survival, recovery, treatment compliance, painand quality of life.31,32 Therefor depression should be assessed adequately inthe post treatment phase.33 Most studies describe psychological morbiditywithout taking physical morbidity into account. Especially in head and neckcancer patients both factors seem to be related. Therefor more studies areneeded in which psychological as well as physical morbidity are analysed. Head and neck cancer and rehabilitationIn the Netherlands 60.000 new patients are yearly diagnosed with cancer.34

About 10 % of these patients, mostly man, are diagnosed with head and neckcancer.35Although mortality rates for head and neck cancer patients are still

Chapter 1

14

high, through early diagnosis and better treatment possibilities more patientswill survive cancer.36 In the last decade attention has been given to postcancer treatment rehabilitation.37,38 The last year’s treatments are more oftenmultidimensional, aiming on physical and psychological problems. Theseprograms are executed in groups to gain from peer support of fellow cancerpatients.39

Specific physical therapy programs for patients with shoulder complaintsafter neck dissections are proposed since 1976. (appendix 2) Only one studyis performed as a controlled trial.40 Physical therapy programs differ incontent and aims. Most programs aim at gaining a full, active and passive,range of motion, strengthening of shoulder muscles to stabilise the shoulder,and training shoulder function in activities in daily life. Besides exercises aphysical therapy progam may contain, relaxation or massage. Most programsare developed because of assumed spinal accessory nerve dysfunction.Several authors claim good results, but only the study of Salerno used acontrol group.40 She showed that physical therapy can be helpful in therehabilitation after neck dissection for shoulder complaints. Neverthelessbefore setting up a program more insight in the exact morbidity after headand neck cancer treatment should be achieved to known where the programshould focus on.

Aim of our studyIn this thesis we are aiming at more insight in morbidity after head and neckcancer treatment (resection of the primary tumour, neck dissection and pre-or post-operative radiation therapy). We will study shoulder complaints andthe role of the spinal accessory nerve, pain and the underlying painmechanisms, range of motion, and loss of sensation. The consequences ofdysfunction such as shoulder disability and activities of daily living, but alsopsychological problems (depression) and quality of life will be analysed.This thesis aims at a better understanding of consequences of type of neckdissection, especially the supraomohyoid neck dissection and the influenceof radiation therapy. A better understanding of morbidity after cancertreatment must lead to specific rehabilitation treatment options in the nearfuture.

Chapter 1

15

References1. Crile G. Excision of cancer of the head and neck. JAMA 1906;47:1780-

1786.2. Ewing MRMH. Disability following radical neck dissection. Cancer

1952;5:873-883.3 Cheng PT, Hao SP, Lin YH, Yeh AR. Objective comparison of shoulder

dysfunction after three neck dissection techniques. Ann. Otol. Rhinol.Laryngol. 2000;109:761-766.

4. Bocca E, Pignataro O, Sasaki CT. Functional neck dissection. A descriptionof operative technique. Arch.Otolaryngol. 1980;106:524-527.

5. Robbins KT, Medina JE, Wolfe GT, Levine PA, Sessions RB, Pruet CW.Standardizing neck dissection terminology. Official report of theAcademy's Committee for Head and Neck Surgery and Oncology.Arch.Otolaryngol.Head Neck Surg. 1991;117:601-605.

6. Short SO, Kaplan JN, Laramore GE, Cummings CW. Shoulder pain andfunction after neck dissection with or without preservation of the spinalaccessory nerve. Am.J.Surg. 1984;148:478-482.

7. Carenfelt C, Eliasson K. Occurrence, duration and prognosis of unexpectedaccessory nerve paresis in radical neck dissection. Acta Otolaryngol.1980;90:470-473.

8. Medina JE, Byers RM. Supraomohyoid neck dissection: rationale,indications, and surgical technique. Head Neck 1989;11:111-122.

9. Kowalski LP, Magrin J, Waksman G, Santo GF, Lopes ME, de Paula RP,Pereira RN, Torloni H. Supraomohyoid neck dissection in the treatment ofhead and neck tumors. Survival results in 212 cases.Arch.Otolaryngol.Head Neck Surg. 1993;119:958-963.

10. Robbins KT, Clayman G, Levine PA, Medina J, Sessions R, Shaha A, SomP, Wolf GT. Neck dissection classification update: revisions proposed bythe American Head and Neck Society and the American Academy ofOtolaryngology-Head and Neck Surgery. Arch.Otolaryngol.Head NeckSurg. 2002;128:751-758.

11. Sobol S, Jensen C, Sawyer W, Costiloe P, Thong N. Objective comparisonof physical dysfunction after neck dissection. Am.J.Surg. 1985;150:503-509.

12. Leipzig B, Suen JY, English JL, Barnes J, Hooper M. Functional evaluationof the spinal accessory nerve after neck dissection. Am.J.Surg.1983;146:526-530.

Chapter 1

16

13. Pinsolle V, Michelet V, Majoufre C, Caix P, Siberchicot F, Pinsolle J.[Spinal accessory nerve and lymphatic neck dissection.Rev.Stomatol.Chir.Maxillofac. 1997;98:138-142.

14. Krause HR. Shoulder-arm-syndrome after radical neck dissection: itsrelation with the innervation of the trapezius muscle. Int.J.OralMaxillofac.Surg. 1992;21:276-279.

15. Shone GR, Yardley MP. An audit into the incidence of handicap afterunilateral radical neck dissection. J.Laryngol.Otol. 1991;105:760-762.

16. Patten C, Hillel AD. The 11th nerve syndrome. Accessory nerve palsy oradhesive capsulitis? Arch.Otolaryngol.Head Neck Surg. 1993;119:215-220.

17. Cantlon GE, Gluckman JL. Sternoclavicular joint hypertrophy followingradical neck dissection. Head Neck Surg. 1983;5:218-221.

18. Nori S, Soo KC, Green RF, Strong EW, Miodownik S. Utilization ofintraoperative electroneurography to understand the innervation of thetrapezius muscle. Muscle Nerve 1997;20:279-285.

19. Sist T, Miner M, Lema M. Characteristics of postradical neck painsyndrome: a report of 25 cases. J.Pain Symptom.Manage. 1999;18:95-102.

20. Chaplin JM, Morton RP. A prospective, longitudinal study of pain in headand neck cancer patients. Head Neck 1999;21:531-537.

21. Shah S, Har-El G, Rosenfeld RM. Short-term and long-term quality of lifeafter neck dissection. Head Neck 2001;23:954-961.

22. Schuller DE, Reiches NA, Hamaker RC, Lingeman RE, Weisberger EC,Suen JY, Conley JJ, Kelly DR, Miglets AW. Analysis of disability resultingfrom treatment including radical neck dissection or modified neckdissection. Head Neck Surg. 1983;6:551-558.

23. Saffold SH, Wax MK, Nguyen A, Caro JE, Andersen PE, Everts EC, CohenJI. Sensory changes associated with selective neck dissection.Arch.Otolaryngol.Head Neck Surg. 2000;126:425-428.

24. White CA, Macleod U. Cancer. BMJ 2002;325:377-380.25. de Leeuw JR, de Graeff A, Ros WJ, Blijham GH, Hordijk GJ, Winnubst

JA. Prediction of depressive symptomatology after treatment of head andneck cancer: the influence of pre-treatment physical and depressivesymptoms, coping, and social support. Head Neck 2000;22:799-807.

26. Zaza C, Baine N. Cancer pain and psychosocial factors. A critical review ofthe literature. J.Pain Symptom Manage. 2002;24:526-542.

27. Lee-Jones C, Humphris G, Dixon R, Hatcher MB. Fear of cancerrecurrence--a literature review and proposed cognitive formulation toexplain exacerbation of recurrence fears. Psychooncology. 1997;6:95-105.

Chapter 1

17

28. Petticrew M, Bell R, Hunter D. Influence of psychological coping onsurvival and recurrence in people with cancer; systematic review. BMJ2002;325:1066-1076.

29. Hassanein KA, Musgrove BT, Bradbury E. Functional status of patientswith oral cancer and its relation to style of coping, social support andpsychological status. Br.J.Oral Maxillofac.Surg. 2001;39:340-345.

30. Hann D, Winter K, Jacobsen P. Measurement of depressive symptoms incancer patients: evaluation of the Center for Epidemiological StudiesDepression Scale (CES-D). J.Psychosom.Res. 1999;46:437-443.

31. Hjerl K, Andersen EW, Keiding N, Mouridsen HT, Mortensen PB,Jorgensen T. Depression as a prognostic factor for breast cancer mortality.Psychosomatics 2003;44:24-30.

32. Krishnan KR, Delong M, Kraemer H, Carney R, Spiegel D, Gordon C,McDonald WI, Dew MA, Alexopoulos G, Buckwalter K, Cohen PD, EvansD, Kaufmann PG, Olin J, Otey E, Wainscott C. Comorbidity of depressionwith other medical diseases in the elderly. Biological Psychiatry 2002;Vol52:559-588.

33. Bottomley A. Depression in cancer patients: a literature review.Eur.J.Cancer Care 1998;7:181-191.

34. Visser O, Coebergh JWW, Dijck van JAAM, Siesling S. Incidence ofcancer in the Netherlands in 1998. Tenth report of the Netherlands CancerRegistry. Utrecht 2002.

35. Netherlands Cancer Registry. Head and Neck tumours in the Netherlands1989-1995. Lulof Almelo, 1998.

36. Brenner H. Long-term survival rates of cancer patients achieved by the endof the 20th century: a period analysis. The Lancet 2002;360:1131-1135.

37. DeLisa JA. A history of cancer rehabilitation. Cancer 2001;92:970-974.38. Ronson A, Body JJ. Psychosocial rehabilitation of cancer patients after

curative therapy. Support.Care Cancer 2002;10:281-291.39. Weert van E, Hoeksta-Weebers JEHM, Grol BMF, Otter R, Arendzen JH,

Postema K, Schans van der CP. Physical functioning and quality of lifeafter cancer rehabilitation Int. J. Rehab.Res. 2004;27: 27-35.

40. Salerno G, Cavaliere M, Foglia A, Pellicoro DP, Mottola G, Nardone M,Galli V. The 11th nerve syndrome in functional neck dissection.Laryngoscope 2002;112:1299-1307.

Chapter 1

18

Chapter 1

19

CHAPTER 2

SHOULDER PAIN AND DISABILITY IN DAILY LIFE,

FOLLOWING SUPRAOMOHYOID NECK DISSECTION:

A PILOT STUDY.

C. Paul van Wilgen, 1,2, Pieter U. Dijkstra, 1,2, Jan M. Nauta, 2

Albert Vermey,3 Jan L.N. Roodenburg, 21. Department of Rehabilitation, 2. Department of Oral and MaxillofacialSurgery, 3. Department of Surgery,Head and Neck Surgery

Journal of Cranio-Maxillofacial Surgery (2003) 31, 183-186

20

SummaryIntroduction: the purpose of this pilot study was to assess shouldermorbidity; i.e. pain and disability in daily activities, at least one year afterunilateral or bilateral supraomohyoid neck dissection. Patients and methods: 52 patients having been subjected to a supraomohyoidneck dissection filled in a questionnaire assessing pain and daily activities.Results: 14 (28%) patients complained of ipsilateral shoulder pain followingsupraomohyoid neck dissection. The disability perceived during daily life,because of shoulder complaints, was minor. The pain and disabilityexperienced during daily activities led to dependency upon other people intwo patients. This dependency only existed during heavy householdactivities.Conclusion: Despite the fact that this type of neck dissection was developedto reduce shoulder morbidity, 28% of the patients experienced shoulder painfollowing supraomohyoid neck dissection. The degree of disability due toshoulder complaints, however, was minor.

IntroductionNeck dissection is performed in the treatment of carcinoma of head andneck. In radical neck dissection (Crile, 1906) all lymphnodes at one side ofthe neck are resected, plus the spinal accessory nerve, the internal jugularvein, and the sternocleidomastoid muscle. Morbidity following radical neckdissection includes, disfigurement, sensory changes, shoulder pain, reducedstrength of the trapezius muscle, reduced range of motion of the shoulder,disability in activities of daily life, and even loss of work (Ewing, 1952).Shoulder morbidity is probably the result of sacrificing the spinal accessorynerve, resulting in a paralysis of the trapezius muscle (Remmler et al., 1986).The latter results in a reduction of active abduction, forward flexion of thearm, shoulder pain, and disability in daily activities. The incidence ofmorbidity after radical neck dissection varies between 60 % and 100 %(Ewing, 1952, Short et al., 1984, Brown et al., 1988, Shone and Yardley,1991). This is why Bocca et al. (1980) introduced the “functional” or“modified radical” neck dissection in which at least one of the non-lymphatic structures is spared; the spinal accessory nerve, internal jugularvein, or sternocleidomastoid muscle. In most cases the spinal accessory nerve is spared to prevent shoulder morbidity. However the incidence of

Chapter 2

21

shoulder pain and disability following modified radical neck dissection werestill high, varying between 36 % and 77 % (Leipzig et al., 1983, Schuller etal., 1983, Pinsolle et al., 1997). Increased insight in the biological behaviour of head and neck squamous cellcarcinoma, and improved staging, led to the development of four types ofselective neck dissection (Figure 1): supraomohyoid neck dissection (levelsI, II, and III), lateral neck dissection (levels II, III, and IV) posterolateralneck dissection (levels II, III, IV,and V), and anterior neck dissection (levelVI) (Robbins et al., 1991). With these selective neck dissections all non-lymphatic structures mentioned are spared.Supraomohyoid neck dissection is often performed for oral or oropharyngealsquamous cell carcinoma, in patients with N0 tumours and is possiblysufficient in the treatment of a selected group of patients with positive nodesat level I (Medina and Byers,1989, Kowalski et al.,1993, Kligerman et al.,1994, Spiro et al., 1996). Although supraomohyoid neck dissections arefrequently performed little is published about shoulder morbidity, and theconsequences for daily activities. Pinsolle et al. (1997) described 41 patientsfollowing supraomohyoid neck dissection, with 32% having minor, 5 %moderate, and 2.5% severe shoulder problems (Pinsolle et al.,1997). Allpatients could manage their activities of daily living independently.

Figure 1 Classification of selective neck dissection

In other studies, of supraomohyoid neck dissections, the sample size wassmall (n = 7 to 36) (Leipzig et al., 1983, Sobol et al., 1985, Cheng et al., Chapter 2

Supraomohyoid (levels I,II,III)

Lateral (levels II,III,IV)

Postero-lateral (levels II,III,IV,V)

Anterior (levels VI)

Reprinted from Robbins et al. 1991

22

2000), the follow-up was short (6 weeks to 6 months) (Leipzig et al., 1983,Sobol et al., 1985, Cheng et al., 2000), or the impact of shoulder morbidityon daily activities was not well described (Leipzig et al., 1983, Sobol et al.,1985, Cheng et al., 2000). Remmler et al., (1986) described 103 patients, ofwhich 76 underwent supraomohyoid neck dissection. In their study the dropout after a year was 73 %, and of the remaining 25 patients it was notreported how many had been subjected to a supraomohyoid neck dissection,and hindrance during daily activities was not described (Remmler etal.,1986).The purpose of this pilot study was to analyse the prevalence of shoulderpain and disability, following supraomohyoid neck dissection, and todetermine which daily activities were disturbed.

Material and methodsPatients in the care of the Department of Oral and Maxillofacial Surgery(University Hospital Groningen) who had undergone supraomohyoid neckdissection, for squamous cell carcinoma of the oral cavity or oropharynx, atleast 1 year previously, were asked to participate in this study. Patients whomatched our criteria were asked, during a regular follow-up appointment, toparticipate in the study. A standardized questionnaire was used to assess painand disability. Patients operated upon bilaterally were asked to refer to thepainful shoulder only. The questionnaire was a combination of two valid andreliable questionnaires: the shoulder disability questionnaire (SDQ) (van-derHeijden et al., 2000) and the Groningen activity restriction scale (GARS)(Suurmeijer et al.1994). From these two questionnaires only questionsassessing typical shoulder functions were selected. Patients who perceivedshoulder pain completed the whole questionnaire while patients without painfilled out if they have perceived temporarily shoulder pain. The frequency ofshoulder pain was assessed on a four point Likert scale (shoulder pain wasperceived constantly, often, occasionally or never in the last month). Furthermore the questionnaire assessed: waking up because of shoulderpain, pain when lying on the affected shoulder, pain when moving theshoulder, pain when leaning on the arm or elbow, pain when reaching aboveshoulder level, pain when carrying heavy or light objects, and pain whenreaching for the neck. Disability perceived during daily activities was alsoassessed on a four point Likert scale: patients filled out the level of effort

Chapter 2

23

during daily activities or whether they were dependent on others for theirdaily activities (fully independently without difficulty, some difficulty, greatdifficulty, only able to do it with someone’s help). The following dailyactivities were assessed: dressing, washing, washing hair, light and heavyhousehold activities, facial care and toilet use. In addition informationconcerning physiotherapy, history of shoulder pain, type of surgery wereasked for or collected from the medical records. Data were analysed usingSPSS 10.0: chi-square analysis with a continuity correction, and t-test forindependent samples

ResultsFifty-six patients were asked to fill in the questionnaire. Of these, 52returned the questionnaire of which two were excluded from analysisbecause of missing data. Thus 50 questionnaires (27 females 23 males, mean63 years, SD: 9) could be analysed. The mean follow-up was 2.3 years (SD:1.3). Forty patients were operated upon unilaterally, and 10 bilaterally. Alloperations were between 1995 and 2000. Fourteen patients (28 %, 95%Confidence Interval: 18% - 42%) perceived shoulder pain, of which fourexperienced this constantly, three often, and seven experienced shoulder painoccasionally. Twenty-seven patients had received radiotherapy of which 10(37 %) complained of shoulder pain (Table 1).

Table 1 Shoulder complaints and radiation therapy, gender, follow-up and age

Shouldercomplaints

No shouldercomplaints

p

Radiation Therapy Yes 10 17 No 4 19 0.22 *Gender Male 5 18 Female 9 18 0.55 *Follow-up (mean; yrs) 2.2 (SD 2.2) 2.4 (SD 1.2) 0.71 **

Age (mean; yrs) 63 (SD 8.8) 62 (SD 9.0) 0.79 *** Results of Chi-square with continuity correction,**Results of t-test for independent sample analyses. SD = standard deviation

Chapter 2

24

No significant relation was found between shoulder complaints and gender,follow-up interval, or age (Table 1). Of the 14 patients with shouldercomplaints 8 had an were operation on their dominant side. One of thepatients with shoulder pain reported shoulder pain before surgery.Activities typically provoking shoulder pain were: moving arm or shoulder,reaching above shoulder level, reaching for the neck, carrying heavy objects,and laying on the shoulder (Figure 2). Due to the shoulder complaints twopatients needed help from others during heavy house-hold activities, such aswashing windows, or cleaning the floor. Eight patients had difficulties withdaily activities but did not require help, and of these eight some perceiveddifficulties with: washing their hair (five patients), washing and dryingthemselves (five), light household activities (five), and facial care (four).

Figure 2 Shoulder pain during daily activities in 14 out of 50 patients following supraomohyoid neck dissection

Chapter 2

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

0% 20% 40% 60% 80% 100%

waking up

lying on shoulder

moving shoulder/arm

leaning on arm/elbow

reaching above shoulder level

carrying heavy objects

carrying light objects

reaching for the neck

frequency of pain

�� never

�� sometimes

�� often

�� always

25

The correlation between the frequency of shoulder pain and the amount ofperceived disability during daily activities was r =.89 (95 % ConfidenceInterval: .69 - .96)Only six out of 14 patients with shoulder complaints had been or were stillbeing treated by a physiotherapist. Of the 36 patients (72%) without shouldercomplaints, four patients had experienced temporary shoulder complaintsfollowing neck dissection.

DiscussionFourteen of the 50 patients (28%) experienced shoulder problems, despitethe preservation of the spinal accessory nerve. As the nerve is partiallystripped of its vascular supply during surgery a temporary non-functionalnerve or only partially functional nerve may be the result (Soo et al.,1990).Consequently the trapezius muscle may be too weak to stabilize the scapulasufficiently.Although the group interviewed was only 50 patients, 28% of theseperceiving shoulder pain is a significant proportion following this selectivetype of neck dissection. The exact source of the post-operative shoulder pain is unknown. Manysuggestions o f the possible cause have been made: secondary frozenshoulder (Patten and Hillel, 1993), hypertrophic sternoclavicular joint(Cantlon and Gluckman, 1983), and excessive stretching of the rhomboidand levator scapulae muscle (Nori, et al., 1997). Post-operative shoulder painis not always caused by spinal accessory nerve dysfunction, Saunders etal.(1985) found a weak relationship between trapezius muscle dysfunctionand subjective symptoms of shoulder pain (Saunders et al., 1985). Cutting ofcutaneous sensory nerves, causing neuropathic pain, or neuromata may alsocause shoulder pain (Brown et al., 1988). Most of the authors agree that animportant mechanism behind the shoulder pain is the overload of theshoulder girdle as a result of the inability of the trapezius muscle to stabilizethe scapula.Although supraomohyoid neck dissections seem to create less disabilitywhen compared with radical- or modified radical neck dissections, thepercentage of patients (28%) with shoulder pain after a supraomohyoid neckdissection is relatively high, but considerably lower when compared withradical neck dissections (60%-100%)% (Ewing, 1952, Short et al., 1984,

Chapter 2

26

Brown et al., 1988, Shone and Yardley, 1991) and modified radical neckdissections (36%-77%) (Leipzig et al., 1983, Schuller et al., 1983, Pinsolle etal., 1997). Despite having some difficulties in heavy household activities,most patients could perform their daily activities without the help of others.In this study radiation therapy was not significantly associated with shoulderpain or disability, confirming the statement of Chaplin and Morton (1999)that radiation therapy has no effect on shoulder pain. However others statethat radiation therapy may add significantly to permanent post-operativedisability (Schuller et al., 1983). These conflicting findings indicate thateffects of radiation therapy on shoulder complaints need furtherinvestigation. In this study only six of 14 patients with shoulder complaints were treatedwith physiotherapy. Physiotherapy is often prescribed to help patients withshoulder complaints following neck dissection the aim being to reduce orprevent shoulder pain by reducing shoulder load, and to increase the strengthof other scapula stabilizing muscles to compensate for loss of function of thetrapezius muscle. Patients seem to benefit from these physical therapyprogrammes. (Saunders and Johnson, 1975, Gluckman et al., 1983, Fialkaand Vinzenz, 1989, Salerno et al., 2002).

ConclusionDespite the fact that the supraomohyoid neck dissection was developed toreduce shoulder morbidity, 28 % of the patients experienced shoulder painand disability in daily activities. As this study group was relatively small, thecause and extent of morbidity requires further research.

AcknowledgementThe authors thank C. Bron, J.L.M. Franssen and B.G.M. de Valk for their support indeveloping the questionnaire for this study

Chapter 2

27

References1. Bocca E, Pignataro O, Sasaki CT. Functional neck dissection. A description

of operative technique. Arch Otolaryngol 1980; 106:524-527.2. Brown H, Burns S, Kaiser CW. The spinal accessory nerve plexus, the

trapezius muscle, and shoulder stabilization after radical neck cancersurgery. Ann Surg 1988; 208:654-661.

3. Cantlon GE, Gluckman JL. Sternoclavicular joint hypertrophy followingradical neck dissection. Head Neck Surg 1983; 5:218-221.

4. Chaplin JM, Morton RP. A prospective, longitudinal study of pain in headand neck cancer patients. Head Neck 1999; 21:531-537.

5. Cheng PT, Hao SP, Lin YH, Yeh AR. Objective comparison of shoulderdysfunction after three neck dissection techniques.Ann.Otol.Rhinol.Laryngol 2000;109:761-766.

6. Ewing MRMH. Disability following radical neck dissection. Cancer 1952;5:873-883.

7. Fialka V, Vinzenz K. Zur physikalischen therapie und diagnostik derpostoperativ geschädigten schulter nach radikaler neck-dissection. Dtsch ZMund Kiefer Gesichtschir 1989; 13:220-225.

8. Gluckman JL, Myer CM, Aseff JN, Donegan JO. Rehabilitation followingradical neck dissection. Laryngoscope 1983; 93:1083-1085.

9. van-der Heijden G, Leffers P, Bouter LM. Shoulder disability questionnairedesign and responsiveness of a functional status measure. J Clin Epidemiol2000; 53:29-38.

10. Kligerman J, Lima RA, Soares JR, Prado L, Dias FL, Freitas EQ, OlivattoLO. Supraomohyoid neck dissection in the treatment of T1/T2 squamouscell carcinoma of oral cavity. Am J Surg 1994; 168:391-394.

11. Kowalski LP, Magrin J, Waksman G, Santo GF, Lopes ME, de Paula RP,Pereira RN, Torloni H. Supraomohyoid neck dissection in the treatment ofhead and neck tumors. Survival results in 212 cases. Arch OtolaryngolHead Neck Surg 1993; 119:958-963.

12. Leipzig B, Suen JY, English JL, Barnes J, Hooper M. Functional evaluationof the spinal accessory nerve after neck dissection. Am J Surg 1983;146:526-530.

13. Medina JE, Byers RM. Supraomohyoid neck dissection: rationale,indications, and surgical technique. Head Neck 1989; 11:111-122.

14. Nori S, Soo KC, Green RF, Strong EW, Miodownik S. Utilization ofintraoperative electroneurography to understand the innervation of thetrapezius mu scle. Muscle Nerve 1997; 20:279-285.

Chapter 2

28

15. Patten C, Hillel AD. The 11th nerve syndrome. Accessory nerve palsy oradhesive capsulitis? Arch Otolaryngol Head Neck Surg 1993; 119:215-220.

16. Pinsolle V, Michelet V, Majoufre C, Caix P, Siberchicot F, Pinsolle J.Branche externe du nerf spinal et évidements ganglionnaires cervicaux. RevStomatol Chir Maxillofac 1997; 98:138-142.

17. Remmler D, Byers R, Scheetz J, Shell B, White G, Zimmerman S, GoepfertH. A prospective study of shoulder disability resulting from radical andmodified neck dissections. Head Neck Surg 1986; 8:280-286.

18. Robbins KT, Medina JE, Wolfe GT, Levine PA, Sessions RB, Pruet CW.Standardizing neck dissection terminology. Official report of theAcademy's Committee for Head and Neck Surgery and Oncology. ArchOtolaryngol Head Neck Surg 1991; 117:601-605.

19. Salerno G, Cavaliere M, Foglia A, Pellicoro DP, Mottola G, Nardone M,Galli V. The 11th nerve syndrome in functional neck dissection. TheLaryngoscope 2002; 112:1299-1307.

20. Saunders WH, Johnson EW. Rehabilitation of the shoulder after radicalneck dissection. Ann Otol Rhinol Laryngol 1975; 84:812-816.

21. Schuller DE, Reiches NA, Hamaker RC, Lingeman RE, Weisberger EC,Suen JY, Conley JJ, Kelly DR, Miglets AW. Analysis of disability resultingfrom treatment including radical neck dissection or modified neckdissection. Head Neck Surg 1983; 6:551-558.

22. Shone GR, Yardley MP. An audit into the incidence of handicap afterunilateral radical neck dissection. J Laryngol Otol 1991; 105:760-762.

23. Short SO, Kaplan JN, Laramore GE, Cummings CW. Shoulder pain andfunction after neck dissection with or without preservation of the spinalaccessory nerve. Am J Surg 1984; 148:478-482.

24. Sobol S., Jensen C., Sawyer W., Costiloe P., Thong N. Objectivecomparison of physical dysfunction after neck dissection. Am J Surg 1985;150:503-509.

25. Soo KC, Guiloff RJ, Oh A, Della RG, Westbury G. Innervation of thetrapezius muscle: a study in patients undergoing neck dissections. HeadNeck 1990; 12:488-495.

26. Spiro RH, Morgan GJ, Strong EW, Shah JP. Supraomohyoid neckdissection. Am J Surg 1996; 172:650-653.

27. Suurmeijer TP, Doeglas DM, Moum T, Briancon S, Krol B, Sanderman R,Guillemin F, Bjelle A, van den Heuvel WJ. The Groningen ActivityRestriction Scale for measuring disability: its utility in internationalcomparisons. Am J Public Health 1994; 84:1270-1273.

Chapter 2

29

Chapter 3INCIDENCE OF SHOULDER PAIN AFTER NECK

DISSECTION: A CLINICAL EXPLORATIVE STUDY FOR

RISK FACTORS

Pieter U Dijkstra, PhD1,2,3 C. Paul van Wilgen, BSc1,2,3 , Ron PBuijs, BSc4, Wim Brendeke, BSc5, Cornelis JT de Goede,BSc6, Ad Kerst, BSc7, Muriel Koolstra, BSc6, Johan MarinusMsc8, Elisabeth M Schoppink, MSc8, Martijn M Stuiver, BSc9,Caroline F van de Velde, MSc10 , Jan LN Roodenburg, PhD1.1: Dept of Oral and Maxillofacial Surgery, University Hospital Groningen,2: Pain Center, University Hospital Groningen, 3: Dept of Rehabilitation,University Hospital Groningen, 4: University Hospital Rotterdam / Daniel,PO Box 5201, 3075 EA, Rotterdam, 5: Rijnstate Hospital Arnhem, PO Box9555, 6800 TA, Arnhem, 6: University Hospital Vrije UniversiteitAmsterdam, PO Box 7057, 1007 MB, Amsterdam, 7: University HospitalUtrecht, Heidelberglaan 100, 3584 CX, Utrecht, 8: Haaglanden MedicalCenter, PO Box 432, 2501 CK, Den Haag, 9: Netherlands CancerInstitute/Antoni van Leeuwenhoek Hospital, Plesmanlaan 121, 1066 CX,Amsterdam, 10: University Hospital Maastricht, PO Box 5800, 6202 AZ,Maastricht.

Head and Neck (2001) Nov. 947-953

30

AbstractBackgroundIt is the purpose of this study to determine the incidence of shoulder pain andrestricted range of motion of the shoulder after neck dissection, and toidentify risk factors for the development of shoulder pain and restrictedrange of motion.MethodsClinical patients who underwent a neck dissection completed a questionnaireassessing shoulder pain. The intensity of pain was assessed using a visualanalog scale (100 mm). Range of motion of the shoulder was measured.Information about reconstructive surgery, and side and type of neckdissection was retrieved from the medical records.ResultsOf the patients (n=177, mean age 60.3 years (SD: 11.9)) 70% experiencedpain in the shoulder. Forward flexion and abduction of the operated side wasseverely reduced compared to the non-operated side, 21° and 47°respectively. Non-selective neck dissection was a risk factor for thedevelopment of shoulder pain (9.6 mm) and a restricted shoulder abduction(55°). Reconstruction was risk factor for a restricted forward flexion of theshoulder (24.5°). ConclusionsShoulder pain after neck dissection is clinically present in 70% of thepatients. Non-selective neck dissection is a risk factor for shoulder pain anda restricted abduction. Reconstruction is a risk factor for a restricted forwardflexion of the shoulder.

Chapter 3

31

IntroductionThe presence of shoulder complaints is a common problem following radicalneck dissection. Accordingly, Krause1 found that 72% of the patientssuffered from shoulder complaints, whereas 44% of these patients weredisabled due to the severity of these shoulder complaints. Other authorsfound prevalences of shoulder complaints after radical neck dissectionvarying from 50 to 100%.2,3 In general shoulder complaints after radicalneck dissection consist of pain in the neck-shoulder region and a restrictedactive range of motion of the shoulder girdle. It is assumed that thesecomplaints are based on sacrificing the accessory nerve during the neckdissection, which results in most patients in paralysis of the descending andtransverse part of the trapezius muscle.4 Due to loss of strength of thetrapezius muscle the scapula shifts downward and the inferior angle rotatesmedially, resulting in a downward facing of the glenoid fossa.5 Due to thisshift and downward facing, the active range of motion of forward flexionand abduction of the shoulder girdle is restricted. Additionally, duringactivities of the arm the scapula can not be adequately stabilized to thethorax as a result of insufficient muscle strength. The change in position ofthe scapula and the inadequate stabilization may lead to a mechanicaloverload of the shoulder causing pain. Several structures of the shouldergirdle, including the gleno-humeral joint, acromio-clavicular joint, sterno-clavicular joint, levator scapula and rhomboid muscles have been heldresponsible for shoulder pain.6,7 Even frozen shoulder and lesions of thebrachial plexus have been reported after neck dissection.8,9 However, if the accessory nerve is not sacrificed, as in functional ormodified neck dissections, shoulder complaints are still reported by 31% to60 % of the subjects.2,3,10,11 Even after selective neck dissections shouldercomplaints have been reported in approximately 29% to 39% of thepatients.12,13

Most studies identifying shoulder complaints after neck dissection have beenperformed after discharge from the hospital, at least a month after surgery,or have been performed retrospectively. Only two studies included patientsbefore discharge from the hospital.14,15 However, in the study of Leipzig etal.14 ,no information is provided about shoulder complaints during thehospital stay. In the study of Nowak et al.15 , only the range of motion of theneck and shoulder was assessed, but shoulder pain was not investigated.

Chapter 3

32

Therefore, little is known as to whether complaints are present immediatelyafter the operation or whether they develop after discharge from thehospital? Another problem is that subjects with shoulder complaints beforethe neck dissection are not excluded in most studies, except in the study ofCarentfelt et al.16 Therefore, most studies cannot be conclusive with regardto the relationship between cause and effect, i.e., neck dissection resulting inshoulder complaints, because the shoulder pain may already be presentbefore the dissection.17 The aims of this explorative study were to determine the incidence ofshoulder pain and restricted range of motion of the shoulder after neckdissection the day before discharge from the hospital, to analyse the effect ofshoulder pain on daily activities in the clinical phase, and to identify riskfactors for the development of shoulder pain and restricted range of motionof the shoulder.

Material and methodsIn this multicentre study seven Dutch hospitals participated: NetherlandsCancer Institute/Antoni van Leeuwenhoek Hospital Amsterdam, UniversityHospital Groningen, University Hospital Maastricht, University HospitalRotterdam/ Daniel, University Hospital Vrije Universiteit Amsterdam,Haaglanden Medical Center Den Haag, and Rijnstate Hospital Arnhem.Clinical patients who underwent a neck dissection because of a tumor in thehead and neck area were assessed by a physical therapist the day beforedischarge from the hospital. The assessment consisted of a standardized questionnaire, developed for thisstudy, which was completed by the patient. The questionnaire addressed thefollowing information: hand dominance, the presence of preoperativecomplaints of the shoulders (operated and/or non-operated side) threemonths prior to the neck dissection and the presence of shoulder pain at theside of the neck dissection under various circumstances. Thesecircumstances included rest, movements of the shoulder, lying on theshoulder, walking with the arm unsupported, dressing, and washing. Inaddition, the questionnaire assessed problems while washing, dressing, andreaching forward and the reasons for these problems (i.e., shoulder pain,stiffness, and/or loss of strength). Finally, the amount of pain in the shoulderof the operated side was assessed on a visual analog scale (100 mm).

Chapter 3

33

The following information was retrieved from the medical records: tumortype, localization, staging, type and extent of surgery, type of reconstructivesurgery, side and type of neck dissection (radical, modified, withpreservation of the accessory nerve, or selective), preoperative radiotherapy,and pain medication (non opiate, mild opiates and opiates). When the studyproved to be feasible, range of motion of the shoulder (forward flexion,abduction external rotation) was measured by the physical therapist using aninclinometer according to a standardized protocol. The questionnaire and therange of motion measurements became part of the standard dischargeprocedure. Inclusion criteria for the study were: a neck dissection for acarcinoma of the head and neck region in one of the participating hospitals,good understanding of the Dutch language and age 18 years or older.Patients with shoulder complaints (at the side of the operation) within 3months before the neck dissection were excluded as were patients with anhistory of mental illness. In this way patients at risk for developing shouldercomplaints as a result of the neck dissection were selected.Of the patients with a bilateral dissection the most painful side was enteredin the database. The database was checked for missing data and theparticipating institutes were requested to provide the missing data, ifavailable.

Data analysisData analysis in SPSS version 9 and CIA version 2 comprised descriptivestatistics, 95% confidence interval calculation, t-tests for paired data, chisquare tests, and product moment correlation (Pearson’s r). In the univariateanalyses risk factors for shoulder pain, ristrictions in forward flexion andabduction were identified. In the multivariate analyses, linear regressions(method stepwise forward), the extent of shoulder pain and range ofabduction and forward flexion of the operated side were predicted on thebasis of the risk factors identified in the univariate analyses. The risk factorswere dissection type, preservation of cervical plexus, reconstruction, genderand age.For the analysis of differences in range of motion between the operated andthe non-operated side, patients with bilateral dissections as well as patientswith shoulder complaints of the non-operated side in the 3 months before theneck dissection, were excluded.

Chapter 3

34

Table 1 Descriptive statistics of the population under study, type and stage of tumor, side and type of neck dissection, radiotherapy, reconstructive therapy, medication and days after surgery

Variables (number of valid observations)* Frequency %Gender (n=171)-Women 68 40%-Men 103 60%Age in years (sd) 60.3 (12)Tumor type (n=141)-Squamous cell carcinoma 109 77%-Salivary gland tumor 12 9%-Melanoma 5 4%-Other types of cancer 15 11%#

Tumor stage (n=145)-T0 6 4%-T1 32 22%-T2 49 34%-T3 29 20%-T4 24 17%-Tx 5 3%Side of dissection (n=169)-Dominant side 81 49%-Non dominant 84 48%-Bilateral dissection 4 1%Type of dissection (n=169)-Radical neck dissection 42 25%-Modified neck dissection (preserving NXI) 95 56%-Selective neck dissection 32 19%Preservation of cervical branches (n=68)-Yes 44 65%-Partial 3 4%-No 21 31%*Because of missing data the totals of the analyses do not always add up to 177. # Due to rounding of, the sum of the percentages exceeds 100.

Chapter 3

35

Table 1 (continued)

Variables (number of valid observations)* Frequency %Preoperative radiotherapy (n=170)-Yes 20 12%-No 150 88%Reconstructive surgery (n=171)-Pectoral cutaneous flap 20 12%-Radial cutaneous flap 9 5%-Other 25 15%-No 117 68%Medication (n=167)-No medication 87 52%-Non opiates (NSAIDs) 59 35%-Mild opiates 5 4%-Opiates 2 1%-Unknown 14 8%Days after surgery (sd) 13.2 (10) Median [interquartile range] 10 [7-16]*Because of missing data the totals of the analyses do not always add up to 177.

ResultsThe initial database consisted of 75 women (39%) and 119 men (61%). Themean age of the total group was 60.5 (SD: 12.1) years. After excluding thepatients with shoulder complaints before the neck dissection, a cohort of 177patients remained, consisting of 68 women (40%) and 103 men (60%) with amean age of 60.3 (sd: 11.9) years. Gender was not recorded in 6 patients.Data of range of motion of the shoulder was available of 100 patients, afterexcluding patients with shoulder complaints before the operation (operatedand/or non-operated side) and patients with a bilateral dissection.Descriptive statistics of the research population, type of tumor, tumor stage,type of dissection, preservation of N. XI and cervical branches, radiotherapy,reconstructive surgery, medication use and days after surgery aresummarized in Table 1. Shoulder pain was experienced by 70% of thepatients (mean intensity 14 mm , sd 16) (Table 2). The number of situationsin which the patient experienced pain was significantly related to theintensity of pain (Pearson’s r = .73; 95% CI: 0.63 to 0.80). Pain intensity wasnot significantly related (r = .032) to the number of days after surgery. Pain Chapter 3

36

medication was used by 60% of the patients experiencing shoulder painwhereas 43% of the patients without shoulder pain used pain medication.This difference in percentage was not significant (chi square = 2.77; p =.096).

Table 2 Frequency and intensity of shoulder pain after neck dissection

Variables (number of valid observation)

Frequency(%)

(95% CI)

Shoulder pain present* (n=128): -Yes 89 (70%) (62% to 78%)-Intensity of shoulder pain (sd: range) 14.0 (16: 0-66)-Median [inter quartile range] 8 [0-23]Pain in the shoulder (n=177)-During rest 30 (17%) (11% to 22%)-Moving the shoulder 54 (31%) (24% to 38%)-Lying on the shoulder 49 (30%) (23% to 37%)-Walking (arm not supported) 20 (11%) (7% to 16%)-Washing the opposite arm 15 (9%) (4% to 13)-Dressing 17 (10%) (5% to 14%)Difficulties, due to pain, stiffness or weakness of the shoulder, with (n=177)-Washing 54 (31%) (24% to 38%)-Dressing 52 (30%) (13% to 37%)Number of situations in which the patients experience shoulder pain (n=177)-0 99 (56%) (49% to 63%)-1 30 (17%) (11% to 22%)-2 21 (12%) (7% to 17%)-3 9 (5%) (3% to 9%)-4 9 (5%) (3% to 9%)-5 4 (2%) (1% to 6%)-6 5 (3%) (1% to 6%)* Pain was assessed on a 100 mm Visual Analoge scale (VAS): Presence of painindicates pain > 0 on the VAS. Because of missing data the totals of the analyses do not always add up to 177.

Chapter 3

37

In women as well as in men, 70% had shoulder pain. Moving the shoulderand lying on the shoulder were most frequently reported to be painful, 31%and 30% respectively. Forward flexion and abduction of the operated sidewas significantly reduced compared with the non-operated side, 21° and 47°respectively (Table 3).Risk factors for shoulder pain were: sacrificing the cervical plexus versuspreservation of this plexus (relative risk (RR) = 1.7), radical dissectionversus selective dissection (RR = 1.5) and modified radical dissection versusselective dissection (RR = 1.4). Risk factors for a restricted abduction were:sacrificing the cervical plexus versus preservation of this plexus (RR = 1.7),radical dissection versus selective dissection (RR = 2.3) and modified radicaldissection versus selective dissection (RR = 1.8). Risk factors for a restrictedforward flexion were: sacrificing the cervical plexus versus preservation ofthis plexus (RR = 1.8) and reconstructive surgery versus no reconstruction(RR = 1.8) (Table 4). In the linear regression analysis (multivariate analysis), a non selectivedissections was a risk factor for the development of shoulder pain andrestricted shoulder abduction. Reconstruction was a risk factor for restrictedforward flexion of the shoulder (Table 5).

Tabel 3 Differences in range of motion (in degrees) between operated side and non-operated side and the 95% confidence interval of the differences. (n=100)

Operated side Non-operated sideMean SD Mean SD difference (95% CI)

Forward flexion 138.4 26.3 159.1 24.8 20.7 (14.7 to 26.6)Abduction 99.2 46.6 145.7 35.4 46.5 (37.5 to 55.6)External rotation 59.5 17.8 65.1 17.7 5.6 (3.6 to 7.6)

Chapter 3

38

Table 4 Risk factors for shoulder pain, restricted abduction and restricted forward flexion

Shoulder pain Restrictedabduction

Restrictedforward flexion

Cervical plexus preserved-Yes 57%* 53%* 37%*-No 95% 90% 65%Reconstructed-Yes 68% 73% 67%*-No 69% 55% 37%Radiotherapy-Yes 88% 63% 50%-No 66% 60% 46%Dissection-Radical 79% 84% 58%-Modified radical 75% 66% 49%Dissection-Radical 79%* 84%* 58%-Selective 52% 36% 32%Dissection-Modified radical 75%* 66%* 49%-Selective 52% 36% 32%Restricted abduction was defined as abduction (non-operated side) - abduction(operated side) ≥ 20°. Restricted forward flexion was defined as forward flexion (non-operated side) -forward flexion (operated side) ≥20°. Percentages differ significantly from each other (results of chi square tests). Note: during the analyses preservation of some branches of the cervical plexus wasleft out because of the small number of subjects.

Chapter 3

39

Table 5 Results of the linear regression analysis to predict shoulder pain, and range of motion of abduction and forward flexion of the operated side

Variable β 95% CI of β R squareShoulder pain operated side-Selective dissection -9.6 (-19.1 to -0.2)-Constant 20.1 (14.4 to 25.8) .06Abduction operated side-Selective dissection 55.0 (35.0 to 75.1)-Constant 76.1 (64.3 to 87.9) .35Forward flexion operated side-Reconstruction -24.5 (-35.5 to -13.4)-Constant 148.4 (141.8 to 155.1) .26In the regression analyses the following variable were entered step wise forward:Selective dissection (yes/no), preservation of the cervical plexus (yes/no)reconstruction (yes/no), gender (male /female) and age in years.

DiscussionFollowing neck dissection, 70% of the patients reported some form ofshoulder pain in the clinical phase. The intensity of the pain was notexcessive judged from the mean, the median and inter-quartile range of thepain score. The intensity of pain was 23 or less in 75% of the patients on a100 mm-VAS. It was striking that only 44% of the patients claimed to havepain during one or more provoking situations/activities while 70% scoredshoulder pain more than zero on a VAS. This discrepancy in reporting painmight be explained that patients experience a continuous pain, resulting in ascore above 0 on the VAS, which is not aggravated by the activitiesassessed. Our clinical finding that 79% of the patients who had a radical neckdissection reported shoulder pain is similar to the percentage reported byKrause 1 in his retrospective study in the post-clinical phase (after dischargefrom the hospital) (Table 4). It is possible that the shoulder is alreadyoverloaded with relative non-strenuous activities in the clinical phase. Of thepatients who had been operated with preservation of the N. XI (modified

Chapter 3

40

radical) still 65% reported pain (Table 4). This percentage is somewhathigher compared to other studies performed post-clinically.2,3

Intensity of shoulder pain was not significantly correlated with the numberof post operative days (r = .032). Intensity of the shoulder pain wassignificantly related (r = .73) to the number of activities in which patientsexperienced shoulder pain, indicating that the shoulder pain influences ADLduring hospital stay considerably (53% explained variance). Range of motion was significantly and considerably affected by neckdissection. The mean difference between the operated and non operated sidewas 21° for forward flexion and 47° for abduction. The mean difference inexternal rotation was significant but small. This indicates that neckdissection has the greatest impact on shoulder abduction. The difference inimpact can be explained by the fact that the trapezius muscle is active duringabduction while during forward flexion the serratus anterior muscle is active.The risk factors for shoulder pain, restricted abduction and restricted forwardflexion were entered in regression analyses. In the regression analysis, aselective dissection was the only variable contributing significantly to theprediction of shoulder pain. The mean difference in shoulder pain between apatient with a selective dissection and a patient with a non-selectivedissection (modified or radical) was 9.6 mm on a 100 mm VAS. Thus,selective dissection is a protective factor for shoulder pain compared to non-selective dissections. Although significant, it must be noted that the strengthof the protection is weak. Predicting the abduction of the operated side,again selective dissection was the only variable contributing to the equation.The mean difference in abduction between patients with a selectivedissection compared to patients with non selective dissection (modified orradical) is 55°. Our findings that selective dissection provides protectionagainst shoulder pain and a restricted abduction are in agreement with thefindings of post-clinical studies.2,3,18,19 In the prediction of the forward flexion of the operated side, reconstructioncontributed significantly to the regression equation. Clinically this indicatesthat a reconstruction reduces forward flexion with approximately 25° on theaverage, compared to non reconstructed patients. Probably the extent ofsurgery, the tunnelling of the pectoralis muscle on the side of the surgery, orpain due to the radialis flap reduces forward flexion. This finding is inagreement of Nowak et al.17 who found that reconstruction using a pectoral

Chapter 3

41

myo-cutaneous flap reduced range of motion of the cervical spine andforward flexion of the shoulder. Because of the hospital setting of this study our results can only begeneralized to the post-clinical phase to a limited extent. Although thedissection was performed preserving the N. XI, in many subjects, still theonly procedure that had a protective effect on shoulder pain was a selectivedissection (Table 5). It is possible that during non-selective procedures, butwith preservation of the N. XI, the nerve loses its conductive functiontemporarily due to stripping of the nerve from its surrounding tissuesresulting in a neurapraxia. This neurapraxia may recover in the post clinicalphase.18,20

The type and extent of dissection is dictated by the tumor site, size and stage.However, when possible, surgery should be as selective as possible to reduceshoulder pain and restriction in abduction. Additionally a modified neckdissection preserving the N. XI in a clinical positive neck does not adverselyaffect survival and neck control.21

A weakness of this study is the considerable amount of missing data, whichin part can be attributed to incompleteness of the medical files. For instance,quite often it could not be found in the surgery reports whether the cervicalplexus was preserved or not. Even the tumor type was not available in 36cases. Strength of this study was that subjects with shoulder complaints priorto the dissection were excluded from the analysis. In a post hoc analyses, theintensity of pain (VAS score) in the group with complaints before thedissection was 39.1 mm and in the group without complaints 14.0 mm (95%CI of the difference: 10.8 to 39.3). This illustrates the impact on the resultsof the patients with complaints before the dissection, if they are notexcluded. In conclusion, pain after neck dissection is clinically present in 70% of thepatients. A risk factor for development of shoulder pain is a non-selectivedissection. The pain has a considerable impact on activities of clinical dailyliving.

AcknowledgementWe would like to thank dr B Stegenga for his critical comments on previousversions of this manuscript.

Chapter 3

42

References1. Krause HR. Shoulder-arm-syndrome after radical neck dissection: its

relation with the innervation of the trapezius muscle. Int J Oral MaxillofacSurg 1992;21:276-279.

2. Short SO, Kaplan JN, Laramore GE, Cummings CW. Shoulder pain andfunction after neck dissection with or without preservation of the spinalaccessory nerve. Am J Surg 1984;148:478-482.

3. Saunders-JR J, Hirata RM, Jaques DA. Considering the spinal accessorynerve in head and neck surgery. Am J Surg 1985;150:491-494.

4. Krause HR, Bremerich A, Herrmann M. The innervation of the trapeziusmuscle in connection with radical neck-dissection. An anatomical study. JCraniomaxillofac Surg 1991;19:87-89.

5. Remmler D, Scheetz J, Byers R, et al. Morbidity of modified neckdissection. In: Larson DL, Ballantyne AJ, Guillamondegul OM, editors.Cancer in the neck. New York: Macmillan Inc.; 1986. p 141-152.

6. Saunders WH, Johnson EW. Rehabilitation of the shoulder after radicalneck dissection. Ann Otol Rhinol Laryngol 1975;84:812-816.

7. Cantlon GE, Gluckman JL. Sternoclavicular joint hypertrophy followingradical neck dissection. Head Neck Surg 1983;5:218-221.

8. Pfeifle K, Koch H. [Pain syndromes as late sequelae of neck dissection]Schmerzsyndrome als Spatfolge nach "Neck dissection". Dtsch Zahnarztl Z1973;28:968-972.

9. Patten C, Hillel AD. The 11th nerve syndrome. Accessory nerve palsy oradhesive capsulitis? Arch Otolaryngol Head Neck Surg 1993;119:215-220.

10. Bocca E, Pignataro O, Sasaki CT. Functional neck dissection. A descriptionof operative technique. Arch Otolaryngol 1980;106:524-527.

11. Robbins KT, Medina JE, Wolfe GT, Levine PA, Sessions RB, Pruet CW.Standardizing neck dissection terminology. Official report of theAcademy's Committee for Head and Neck Surgery and Oncology. ArchOtolaryngol Head Neck Surg 1991;117:601-605.

12. Pinsolle V, Michelet V, Majoufre C, Caix P, Siberchicot F, Pinsolle J.[Spinal accessory nerve and lymphatic neck dissection] Branche externe dunerf spinal et evidements ganglionnaires cervicaux. Rev Stomatol ChirMaxillofac 1997;98:138-142.

13. Cheng PT, Hao SP, Lin YH, Yeh AR. Objective comparison of shoulderdysfunction after three neck dissection techniques. Ann Otol RhinolLaryngol 2000;109:761-766.

Chapter 3

43

14. Leipzig B, Suen JY, English JL, Barnes J, Hooper M. Functional evaluationof the spinal accessory nerve after neck dissection. Am J Surg1983;146:526-530.

15. Nowak P, Parzuchowski J, Jacobs JR. Effects of combined modalitytherapy of head and neck carcinoma on shoulder and head mobility. J SurgOncol 1989;41:143-147.

16. Carenfelt C, Eliasson K. Radical neck dissection and permanent sequaleassociated with spinal accesorry nerve injury. Acta Otolaryngol1981;91:155-160.

17. Chaplin JM, Morton RP. A prospective, longitudinal study of pain in headand neck cancer patients. Head Neck 1999;21:531-537.

18. Kuntz AL, Weymuller EA. Impact of neck dissection on quality of life.Laryngoscope 1999;109:1334-1338.

19. Terrell JE, Welsh DE, Bradford CR, et al. Pain, quality of life, and spinalaccessory nerve status after neck dissection. Laryngoscope 2000;110:620-626.

20. Remmler D, Byers R, Scheetz J, et al. A prospective study of shoulderdisability resulting from radical and modified neck dissections. Head NeckSurg 1986;8:280-286.

21. Andersen PE, Shah JP, Cambronero E, Spiro RH. The role ofcomprehensive neck dissection with preservation of the spinal accessorynerve in the clinically positive neck. Am J Surg 1994;168:499-502.

Chapter 3

44

Chapter 3

45

CHAPTER 4SHOULDER COMPLAINTS AFTER NERVE

SPARING NECK DISSECTION

C.Paul van Wilgen,1,2,5 ,Pieter U. Dijkstra,1,2,5 , Bernhard F.A.M. van der Laan, 3 , John Th.M. Plukker,4 ,Jan L.N. Roodenburg,11. Department of Oral and Maxillofacial Surgery, 2. Department ofRehabilitation 3. Department of Otorhinolaryngology Head & NeckSurgery, 4. Department of Surgical Oncology 5. Pain Expertise Center

International Journal of Oral and Maxillofacial Surgery (2004), 33 (3),253-257.

46

AbstractThe purpose of the study was to analyse the prevalence of shouldercomplaints after nerve sparing neck dissection at least 1 year after surgery,and to analyse the influence of radiation therapy on shoulder complaints.Patients were interviewed for shoulder complaints, and patients filled out theshoulder disability questionnaire to evaluate shoulder disability in dailyactivities.In total 137 patients; 51 after modified radical neck dissection (MRND), 21after postero-lateral neck dissection (PLND), and 65 after supraomohyoidneck dissection (SOHND) were analysed. After MRND 33.3% of thepatients experienced shoulder complaints, after PLND 66.7 %, and afterSOHND 20 % of the patients experienced shoulder complaints. Type of neckdissection was significantly (p < 0.001) related to shoulder complaints.Outcome on the shoulder disability questionnaire also showed a significant(p< 0.01) difference in outcome for type of neck dissection. The prevalenceof shoulder complaints after SOHND is low, and reduces disability in dailyactivities. Radiation therapy does not have a significant effect on shouldercomplaints and disability.

Chapter 4

47

IntroductionNeck dissections are performed as an elective or therapeutic procedure in thetreatment of carcinoma of head and neck especially in squamous cellcarcinoma. There are three types of neck dissections: radical neck dissection(RND), modified radical neck dissection (MRND), and selective neckdissection (SND).18

During RND all levels of lymph nodes on one side of the neck, and severalimportant surrounding non-lymphatic structures are resected, including thespinal accessory nerve, internal jugular vein, and sternocleidomastoidmuscle. In the MRND all levels are resected, but one or more of the non-lymphatic structures, are spared, usually the spinal accessory nerve. In theSND all non-lymphatic structures are spared. Which type of SND will beperformed depends on the location and size of the primary tumour and therisk factors or presence of pathological lymph nodes in the neck.10,11 Fourtypes of SND are described: supraomohyoid neck dissection (SOHND; levelI,II, and III), posterolateral neck dissection (PLND; level II,III,IV, and V)anterior neck dissection (AND; level VI) and lateral neck dissection (LND;level II,III, and IV).18 Shoulder complaints can be a direct cause of neck dissections and can bepresented as; pain, reduced range of motion of the neck and shoulder, loss ofsensitivity and loss of shoulder function. Shoulder complaints may have animportant influence on quality of life.21 Prevalence of shoulder complaints after RND range from 47% to 100 %, andfrom 18% to 77% after MRND (Figure 1).1,2,6,9,12,19,22,23 Prevalence ofshoulder complaints after SND range from 31 % to 40 %. (Table 1) Kuntzconcluded that all type of neck dissections affect patients quality of lifedifferently, RND patients had a worse quality of life compared to SND.13 But the number of studies addressing shoulder complains after SND is small.Additionally the number of patients included in those studies is limited.3,14,17

Chapter 4

48

Figure 1 Prevalence of perceived shoulder pain after radical neck dissection and modified radical neck dissection.

The influence of radiation therapy is not often described in literature. Theavailable literature suggests radiation therapy to be of importance forshoulder complaints. Nowak described a twenty percent reduction in activerange of motion of the shoulder as a result of radiation therapy16, andaccording to Schuller radiation therapy adds significantly to permanentdisability.20 The aim of this study was to analyse shoulder complaints after nerve sparingneck dissection, and it’s impact on daily activities, and to analyse theinfluence of radiation therapy on shoulder complaints.

Chapter 4

47 50

65 6772

7682

100

18

31

5361

0

10

20

30

40

50

60

70

80

90

100

Ewin

g (19

52) R

Saun

ders

(198

5) R

Brow

n (19

88) R

Care

nfelt

(198

0) R

Krau

se (1

992)

RSh

one (

1991

) RHi

llel (

1989

) RSh

ort (

1984

) R

Care

nfeld

(198

0) M

od

Saun

ders

(198

5) M

odKr

ause

(199

4) R

R S

hort

(198

4) M

od

Authors

Perc

enta

ges Lower limit

Point estimation

Upper limit

R= Radical neck dissectionMod= Modified radical neck dissectionRR= Reconstruction of spinal accessory nerve

49

Material and methodsPatients who underwent a neck dissection, in the University Hospital inGroningen, between 1994 and 2000, by either the Department of Oral andMaxillofacial Surgery, Department of Otorhinolaryngology Head & Neck surgery or the Department of Surgical Oncology participated in this study.Patients were informed about this study, by means of an information letter, aweek before a standard appointment in our hospital. During this appointmentpatients were asked to participate in the study. Excluded from the study werepatients with recurrence of the tumour at the time of the study and inabilityto understand the Dutch language. Data concerning the surgery (levelsdissected, structures spared, reconstructions), and radiation therapy werecollected from a computerized medical record and surgery reports. Shouldercomplaints (pain, temporary complaints, reduced range of motion, loss offunction) before and after surgery were registered by means of a structuredinterview.All patients filled out the Shoulder Disability Questionnaire (SDQ). TheSDQ evaluates functional status limitation in patient with shouldercomplaints, and covers 16 items including: complaints when lying on theaffected shoulder, complaints at movements of the shoulder, leaning on thearm, reaching above shoulder level, carrying objects, and complaints whenreaching for/to neck or back. The SDQ ratings range from 0 (minimum) to100 (maximum). A high score indicates more shoulder disability. The SDQis a reliable and valid questionnaire for Dutch patients with shouldercomplaints.23

Statistical analyses are performed in SPSS 10.0 included: Chi-squareanalysis t-test for independent samples, linear regression and logisticregression. Confidence intervals were calculated in statistics with confidence(2nd edition).

Chapter 4

50

Table 1 Percentages and number of patients with shoulder complaints after selective neck dissection compared to modified radical neck dissection and radical neck dissection

Follow-up SND MRND RNDLeipzig (1983) n=99 6 months 31% (11/36) 36% (10/28) 60% (21/35)

Pinsolle (1997) n=127 1 year 39% (16/41) 77% (36/47) 92% (36/39)

Cheng (2000) n=21 6 months 29% (2/5) 56% (5/9) 100% (5/5)SND: selective neck dissection; MRND: modified radical neck dissection; RND:radical neck dissection.

Results Of the 220 patients who were invited, 154 (70 %) participated in the study(103 men, 51 women), mean age of 61 years (min: 13, max: 88, SD:11.9).The following types of neck dissections were performed: 5 RND, 54 MRND,22 PLND, 72 SOHND, and 1 Lateral Neck Dissection (LND). Excluded from the statistical analyses were patients after RND (n=5), andLND (n=1) because of the small number of patients. Additionally ninepatients who had shoulder complaints before surgery (3 MRND, 5 SOHND,1 PLND), and two patients who could not remember whether they had pre-operative shoulder complaints were excluded (2 SOHND). In the statistical analyses 137 patients were included (89 men, 48 women)mean age 61 years (SD: 12.1). Location of the primary tumour and type ofneck dissection are summarized in Table 2. Spinal accessory nerve wassacrificed in one case in the MRND group, and in 1 case in the PLND group.In all SOHND the spinal accessory nerve was spared. The followingMRNDs were performed: type 1 (preservation of the spinal accessorynerve); 7, type 2 (preservation of the spinal accessory nerve and internaljugular vein); 5 and type 3 (preservation of the spinal accessory nerve andinternal jugular vein and sternocleidomastoid muscle); 38.

Chapter 4

51

After MRND 33.3% of the patients perceived shoulder complaints,after PLND 66.7 %, and after SOHND 20 % of the patients perceivedshoulder complaints. Type of neck dissection was significantly (p<0.001) associated with postoperative shoulder complaints. Follow-up, age, and gender were not significantly related to shouldercomplaints (Table 3,4).

Table 2 Primary tumour with type of neck dissection and reconstruction’s performed

Tumor location Type of neck dissection ReconstructionMRND PLND SOHND PM Rad. Fibula Naso.

Larynx (n=17) 5 11 1 1 - - -Hypopharynx(n=1)

- 1 - - 1 - -

Lip (n=4) 2 - 2 1 - - -Oral Cavity,tongue (n=68)

16 - 52 3 8 6 9

Base of tongue(n=5)

4 - 1 1 - - -

Oropharynx(n=16)

12 2 2 2 8 - -

Thyroid (n=6) 3 3 - - - - -Unknownprimary (n=5)

3 2 - - - - -

Others (n=15) 6 2 7 1 - - -Sum (n=137) 51 21 65 9 17 6 9MRND: modified radical neck dissection, PLND: posterolateral neck dissection,SOHND: supraomohyoid neck dissection. PM: pectoralis myocutaneous flap, Rad: free radialis forearm flap, Fibula: free fibula bone and Naso: nasolabial flap

After radiation therapy patients had significantly (p < 0.05) more shouldercomplaints. (Table 3). Radiation therapy was given significantly morefrequent after PLND (91 %) and MRND (80 %) compared to SOHND (54%). In logistic regression analyses radiation therapy did not contribute Chapter 4

52

significantly to the prediction of shoulder complaints if type of surgery wasentered before radiation therapy in the equation. PLND and MRND were performed more frequently on T3-T4 tumors(PLND: 68.8 % (n = 11), MRND 61.9 % (n = 26)), compared to SOHND 34% (n = 21). Side of neck dissection or bilateral neck dissection was notsignificantly related to shoulder complaints (Table 3). No significantrelationship was found between shoulder complaints, operation side andhand dominance.

Table 3 Shoulder complaints after neck dissection: percentages and population analyses

Shoulder complaints

Gender-Male (n=89) 27.0 % (n=24)-Female (n=48) 41.7 % (n=20)Type of neck dissection (n=137)-MRND (n=51) 33.3 % (n=17)** -PLND (n=21) 66.7 % (n=14)** -SOHND (n=65) 20.0 % (n=13)** Radiation Therapy-RTX yes (n=95) 37.9 % (n=36)*-RTX no (n=42) 19.0 % (n=8) *Side of dissection-Left (n=52) 38.5 % (n=20) -Right (n=55) 25.5 % (n=14)-Bilateral (n=30) 33.3 % (n=10)MRND: modified radical neck dissection, PLND: posterolateral neck dissection, SOHND: supraomohyoid neck dissection* significant p < 0.05**significant p < 0.001 (result of Chi-square analyses)

In this population 4 types of reconstructions were applied: nasolabial flap (n=9), pectoralis myocutaneous flap (n=9), free radialis forearm flap (n=17)and free fibula bone (n=6). (Table 2) No significant differences in shouldercomplaints were found between these types of reconstructions.

Chapter 4

53

In our population 21 patients had temporarily shoulder complaintsafter neck dissection, which resolved in time, 12 after MRND, 3 afterPLND and 6 after SOHND.The mean scores of the SDQ were: 48.6 (SD:35.1) for the PLND, 22.2 (SD:28.6) for the MRND, and 11.6 (SD: 26.1) for the SOHND, this differencewas significant (ANOVA p < 0.01). The mean score on the SDQ weresignificantly higher for the patients with radiation therapy, compared to thepatients without radiation therapy: mean difference 12.2 (95% CI: 2.1 to22.3). But in the regression analyses radiation therapy did not contributesignificantly to the prediction of the SDQ score. Items most frequentlymentioned were: reaching above shoulder level (38 %), rubbing the shouldermore than ones a day (34 %), activities in and around the house (31%),lifting an object (31 %), reaching for the neck (29%) and lying on theaffected shoulder (29%).

Table 4 Means of follow up and age in patients with and without shoulder complaints after neck dissection

Shouldercomplaints

No shouldercomplaints

Mean (SD) Mean (SD) 95 % CI of thedifferences

Follow up (years) 3.2 (1.8) 2.9 (1.6) -0.8 to 0.3

Age (years) 61.1 (9.9) 60.9 (12.9) -4.3 to 4.1

DiscussionIncidence of shoulder complaints after SOHND are relatively low (20%).Prevalence of shoulder complaints in western population varies from 1.9 %to 26% for subjects between 30-65 years and from 5% to 34% for peopleabove 65 years.25 This indicates the low prevalence of shoulder complaintsin the group after SOHND. It can even be discussed whether these shouldercomplaints are the result of neck dissections or just the natural prevalence ofshoulder complaints in the general population. In literature the reported prevalence of shoulder complaints after RND ishigh (47%-100%). Because of this high prevalence and the extendedknowledge of biological behavior of tumours in the head and neck, RND are Chapter 4

54

hardly performed anymore in the University Hospital Groningen. In addition SOHND is an adequate procedure in N0 and selected N1 squamous cellcarcinomas of the oral and oropharyngeal cavity. 11,15

The prevalence of shoulder complaints after PLND in this study was high.Reasons for this high prevalence may be that a PLND is more extensive(level II,III,IV,V) compared to SOHND (level I,II,III) and the spinalaccessory nerve and cervical plexus are manipulated more extensively,especially in level V. It is unclear why the prevalence of shoulder complaintsafter PLND is much higher compared to MRND. An explanation might bethat PLND are more frequently performed on large (T3-T4) tumours of thelarynx. Another factor of importance might be the preservation of branchesof the cervical plexus. Unfortunately in 61 % of the surgery reportsinformation about preservation or sacrificing of branches of the cervicalplexus was missing.In our study radiation therapy had no significant influence on shoulderdisability, in contrast to the study of Schuller20 in which radiation therapyadded significantly to shoulder disability. An explanation for this finding isthat radiation therapy is significantly associated with type of neck dissection.Neck dissection type is therefore a confounder for radiation therapy. We distinguished selective neck dissections according to Robbins18 andshowed differences in prevalence of shoulder complaints between thesedissection types. Several authors did not distinguish the different types ofSND. Because of the significant differences in outcome after different kindof SND, SND should be distinguished according to the levels resectedduring surgery. Furthermore the resected non-lymphatic structures are ofimportance. Although in selective neck dissections, according to Robbins, allnon-lymphatic structures are supposed to be spared, in our population thespinal accessory nerve was sacrificed in one case in a PLND. Shoulder complaints were operationalized with a shoulder disabilityquestionnaire (SDQ). A specific questionnaire to evaluate shouldercomplaints is useful to give an indication of shoulder complaints. In ouropinion this questionnaire gives more specific information in shoulderdysfunction compared to other often in head and neck cancer research usedquestionnaires, and can be used in the evaluation of rehabilitation outcome.We did not correct our data for effects of reconstruction’s on shouldercomplaints. The effects of reconstructions on shoulder complaints are

Chapter 4

55

described in several studies, mostly containing consequences of pectoralmyocutaneous flaps to mobility of the shoulder. Nowak et al. concludedpectoral myocutaneous flaps to have a negative influence on shoulder andneck mobility. Patients were measured pre-treatment, 10-days after surgery,before and upon completion of radiation therapy. Shoulder abduction was5%-10% lower in the pectoral myocutaneous flap reconstruction group. 16

Haribhakti et al. concluded in a study of patients 14 months after RND, thespinal accessory nerve dysfunction as most important factor on shoulderdisability. In addition, the pectoral myocutaneous flap seems to have anegative influence on shoulder abduction. 7 Dijkstra et al. described, in astudy in the immediate post-operative clinical phase, a decrease in shoulderforward flexion of 24.5 % in patients with different reconstructions.5 In ourpopulation 29 % of the patients had one of 4 types of reconstruction. Nosignificant differences for shoulder complaints where found within thesesmall groups.Retrospective studies generally do not correct data of post-operativeshoulder complaints for pre-operative shoulder complaints. In our study allpatient with pre-operative shoulder complaints, and those who did notremember if they had complaints were excluded from this study. Eventhough recall bias may be of influence on this selection.Patients were invited for the study by means of a letter in which the purposeof this study was explained. This way of invitation may have stimulatedselection of patients with shoulder complaints to participate in the study. Asa result of this the percentages of shoulder complaints in our study would bean overestimation, and the actual prevalence might be somewhat lower. Theinvitation was however the same for all types of neck dissection, differencesin the prevalence of shoulder complaints between the types of neckdissection seem not to be influenced by this selection bias.This study contains a retrospective study based on one observation after aminimum of 1 year after neck dissection. A prospective study design wouldhave given more information about the process of shoulder complaints in thefirst postoperative year. We were interested in long-term outcomes.

Chapter 4

56

In the first year after surgery distress and risk of tumour recurrence are highand will certainly influence the perception of shoulder pain. Also the chanceof recovery of shoulder complaints is higher in the first postoperative year.In our population 14 % of the patients had temporarily shoulder complaints,mostly after MRND, in the first postoperative year. If shoulder complaintsare the result of neurapraxis of the spinal accessory nerve, recovery of thisneurapraxis might occur even long after surgery. Improved shoulder functionis mainly described in the MRND group.13 Beside regeneration of the SAN,effective coping strategies (doing exercises, reducing heavy physicalactivities), physical therapy 8, and changing of psychological factors 4 mayhave a positive effect on shoulder complaints. This is why we chose afollow-up for a minimum of 1 year, in contrast to Cheng et al. and Leipzig etal. who used a follow up of 6 months.In conclusion, the prevalence of shoulder complaints after SOHND are lowand reduce disability in daily life compared to other neck dissection types.The prevalence of shoulder complaints and disability rates after PLND arehigh and needs further investigation. Radiation therapy does not have asignificant effect on shoulder complaints and disability.

Chapter 4

57

References1. Brown H, Burns S, Kaiser CW. The spinal accessory nerve plexus, the

trapezius muscle, and shoulder stabilization after radical neck cancersurgery. Ann Surg 1988:208: 654-661.

2. Carenfelt C, Eliasson K. Occurrence, duration and prognosis of unexpectedaccessory nerve paresis in radical neck dissection. Acta Otolaryngol1980:90: 470-473.

3. Cheng PT, Hao SP, Lin YH, Yeh AR. Objective comparison of shoulderdysfunction after three neck dissection techniques. Ann Otol RhinolLaryngol 2000:109: 761-766.

4. de Graeff A, de Leeuw J, Ros WJ, Hordijk GJ, Blijham GH, Winnubst JA.A prospective study on quality of life of patients with cancer of the oralcavity or oropharynx treated with surgery with or without radiotherapy.Oral Oncol 1999:35: 27-32.

5. Dijkstra PU, van Wilgen PC, Buijs RP, Brendeke W, de Goede CJ, Kerst A,Koolstra M, Marinus J, Schoppink EM, Stuiver MM, de Velde CF,Roodenburg JL. Incidence of shoulder pain after neck dissection: A clinicalexplorative study for risk factors. Head and Neck 2001:23: 947-953.

6. Ewing MRMH. Disability following radical neck dissection. Cancer1952:5: 873-883.

7. Haribhakti VV, Kavarana NM, Tibrewala AN. Oral cavity reconstruction:an objective assessment of function. Head Neck 1993:15: 119-124.

8. Herring D, King AI, Connelly M. New rehabilitation concepts inmanagement of radical neck dissection syndrome. A clinical report. PhysTher 1987:67: 1095-1099.

9. Hillel AD, Kroll H, Dorman J, Medieros J. Radical neck dissection: asubjective and objective evaluation of postoperative disability. JOtolaryngol 1989:18: 53-61.

10. Kligerman J, Lima RA, Soares JR, Prado L, Dias FL, Freitas EQ, OlivattoLO. Supraomohyoid neck dissection in the treatment of T1/T2 squamouscell carcinoma of oral cavity. Am J Surg 1994:168: 391-394.

11. Kowalski LP, Magrin J, Waksman G, Santo GF, Lopes ME, de Paula RP,Pereira RN, Torloni H. Supraomohyoid neck dissection in the treatment ofhead and neck tumors. Survival results in 212 cases. Arch OtolaryngolHead Neck Surg 1993:119: 958-963.

12. Krause HR. Reinnervation of the trapezius muscle after radical neckdissection. J Craniomaxillofac Surg 1994:22: 323-329.

13. Kuntz AL, Weymuller-EA J. Impact of neck dissection on quality of life.Laryngoscope 1999:109: 1334-1338.

Chapter 4

58

14. Leipzig B, Suen JY, English JL, Barnes J, Hooper M. Functional evaluationof the spinal accessory nerve after neck dissection. Am J Surg 1983:146:526-530.

15. Medina JE, Byers RM. Supraomohyoid neck dissection: rationale,indications, and surgical technique. Head and Neck 1989:11: 111-122.

16. Nowak P, Parzuchowski J, Jacobs JR. Effects of combined modalitytherapy of head and neck carcinoma on shoulder and head mobility. J SurgOncol 1989:41: 143-147.

17. Pinsolle V, Michelet V, Majoufre C, Caix P, Siberchicot F, Pinsolle J.Spinal accessory nerve and lymphatic neck dissection. Rev Stomatol ChirMaxillofac 1997:98: 138-142.

18. Robbins KT, Medina JE, Wolfe GT, Levine PA, Sessions RB, Pruet CW.Standardizing neck dissection terminology. Official report of theAcademy's Committee for Head and Neck Surgery and Oncology. ArchOtolaryngol Head Neck Surg 1991:117: 601-605.

19. Saunders JR, Hirata RM, Jaques DA. Considering the spinal accessorynerve in head and neck surgery. Am J Surg 1985:150: 491-494.

20. Schuller DE, Reiches NA, Hamaker RC, Lingeman RE, Weisberger EC,Suen JY, Conley JJ, Kelly DR, Miglets AW. Analysis of disability resultingfrom treatment including radical neck dissection or modified neckdissection. Head Neck Surg 1983:6: 551-558.

21. Shah S, Har-El G, Rosenfeld RM. Short term and long term quality of lifeafter neck dissection. Head and Neck 2001:nov:954-961.

22. Shone GR, Yardley MP. An audit into the incidence of handicap afterunilateral radical neck dissection. J Laryngol Otol 1991:105: 760-762.

23. Short SO, Kaplan JN, Laramore GE, Cummings CW. Shoulder pain andfunction after neck dissection with or without preservation of the spinalaccessory nerve. Am J Surg 1984:148: 478-482.

24. van-der Heijden G, Leffers P, Bouter LM. Shoulder disability questionnairedesign and responsiveness of a functional status measure. J Clin Epidemiol2000:53: 29-38.

25. van der Windt DAWM, Croft PR. Shoulder pain. In: Crombie IK, ed.:Epidemiology of pain. Seattle: IASP press 1999: 257.

Chapter 4

59

CHAPTER 5SHOULDER COMPLAINTS AFTER NECK DISSECTION;

IS THE SPINAL ACCESSORY NERVE INVOLVED?

C.P. van Wilgen, PT 1,2 , P.U. Dijkstra, PT, MT, PhD 1,2 ,B.F.A.M. van der Laan, MD, PhD 3, J.Th. Plukker, MD, PhD, 4

J.L.N. Roodenburg, DDS, MD, PhD 51. Department of Oral and Maxillofacial Surgery, 2. Department ofRehabilitation, 3. Department of Otorhinolaryngology, Head & NeckSurgery, 4. Department of Surgical Oncology, University HospitalGroningen.

British journal of oral and maxillofacial surgery (2003) 41,7-11.

60

Summary

The purpose of the current study was to investigate the relation betweenshoulder morbidity (pain and range of motion), and the function of the spinalaccessory nerve after neck dissection. Identifying dysfunction of the nervegives insight in the mechanisms of post-operative shoulder complaints. Intotal 112 patients after neck dissection (73 male/ 39 female), mean (SD) age61 (13) years, participated in the study. The mean duration of follow up was3 (2) years. Five patients had radical, 43 modified radical, 48supraomohyoid, and 16 posterolateral neck dissection. Thirty-ninecomplained of shoulder pain of whom 20 (51%) had dysfunction of thespinal accessory nerve, and 19 (49%) did not. In total 29 patients (26%) haddysfunction of the spinal accessory nerve of whom 20 (69%) had shoulderpain. Shoulder pain was significantly related to dysfunction of the nerve (p <0.001). Twenty-three patients had a difference in active range of motion inshoulder abduction of ≥ 40° , of whom 22 (96%) had dysfunction of thenerve. A difference in active shoulder abduction of ≥ 40° was significantlyrelated to loss of function of the spinal accessory nerve (p < 0.001).Conclusion: Shoulder pain after neck dissection can only be attributed todysfunction of the spinal accessory nerve in about 50%. If patientsexperience shoulder pain after neck dissection examination of the trapeziusmuscle and active bilateral abduction of the shoulder should be made to findout if the spinal accessory nerve is involved.

Chapter 5

61

IntroductionNeck dissections are either elective or therapeutic procedures in thetreatment of cancer of head and neck. Ewing was one of the first to describeshoulder complaints after radical neck dissection.1 These complaintsconsisted of reduced range of motion, reduced strength in the trapeziusmuscle, pain, disfigurement, and disability in daily activities. In that study of100 patients, 47% developed shoulder complaints after radical neckdissection. These were attributed to resection of the spinal accessory nerve.Other authors have described higher incidences after radical neck dissection,ranging from 47 % to 100%.1-3 Resection of the nerve during radical neckdissection usually leads to loss of function of the trapezius muscle, but insome cases the muscle will function normally, because of the innervation bybranches from the cervical plexus. Innervation from the cervical plexus maybe through connections with the spinal accessory nerve, or through anindependent double motor supply directly to the trapezius muscle. Anindependent double innervation by means of the nerve, and the cervicalplexus is present in about 18 % of patients.4 The trapezius muscle isinnervated solely through the cervical plexus in 6 % of patients.5,6 Krausestated that in about 25 %, radical neck dissection will not lead to loss offunction of the trapezius muscle if enough cervical branches are preserved.5

Because of the high incidence of shoulder complaints after radical neckdissection, modified and selective neck dissections with preservation of thespinal accessory nerve were developed.7 However, even with preservation ofthe nerve, shoulder complaints developed in 18% to 77% after modifiedradical neck dissection,2,8,12 and in 29% to 39% after selectivedissection.2,11,12

Shoulder complaints after nerve-preserving procedures were still attributedto dysfunction of the spinal accessory nerve. However, Cheng et al. in asmall study described shoulder complaints after neck dissections with nodysfunction of the nerve. He described 5 patients after radical neckdissection who complained of shoulder pain, but only four had signs of lossof function of the trapezius muscle. Additionally in a group of 7 patientsafter selective neck dissection, 2 had pain with no signs of loss of function ofthe trapezius muscle.2 On the other hand, loss of function of the trapezius muscle does not alwaysaffect the shoulder. Saunders et al. showed in a study of 100 patients after

Chapter 5

62

radical and modified radical neck dissections that there was a weak relationbetween shoulder complaints (pain, ache, numbness, and weakness), andphysical signs of loss function of the trapezius muscle (atrophy, shoulderdrop, winging of the scapula, and reduced abduction).9 The purpose of the current study was to investigate the function of the spinalaccessory nerve after neck dissections, and the relation between it’s function,shoulder pain, and range of motion of the shoulder.

Materials and methodsPatients who had a neck dissection done by the multidisciplinary Head andNeck Oncology Group of the University Hospital Groningen, during theperiod 1994 to 2000, were invited to participate in the study. A week beforethey visited the hospital for a regular follow-up appointment, all patientswere sent a letter telling them about the study. During the appointment theywere asked by the physician to participate in the study. After given writteninformed consent they were included in the study. Patients with bilateralneck dissection, recurrence of the tumour, or who were unable to understandDutch were excluded. All patients had a follow up of at least 1 year afterneck dissection. From the medical record, the following data were retrieved:date of operation, type of resection, type of neck dissection, whether thespinal accessory nerve or the cervical plexus or both were preserved, thetype of reconstruction, stage and whether they had preoperative orpostoperative radiotherapy. Neck dissections were classified as described byRobbins et al.7 In this classification lymph nodes of the neck are divided into6 anatomical levels, and types of neck dissection are divided into: radical,modified radical, and four types of selective dissections. We also looked forextended posterolateral neck dissections in which parts of the trapeziusmuscle and the splenius muscle are sacrificed.13

Function of the spinal accessory nerve was assessed by examining thetrapezius muscle.5,9,14 This examination included: visual and palpable signsof atrophy of the trapezius pars descendens muscle during shrugging of theshoulders; visual signs of shoulder drop; and quantifying scapula posture bymeasuring the distance of the superior angle of the scapula to the spine. Theside of the neck dissection was compared to the non-dissected side, and ifthere was a difference of 2 cm or more a change in scapula posture wasrecorded.

Chapter 5

63

If a patient had two or three signs of loss of function of the trapezius musclewe assumed that the spinal accessory nerve was malfunctioning as aconsequence of the neck dissection. This may be caused by neurapraxia orneurotmesis.Shoulder pain was evaluated by means of an interview. Beside pain in theshoulder on the dissected side we also recorded pain on the non-dissectedside, and whether the patient had shoulder pain preoperatively. Shoulderpain was measured with a numbered visual analogue scale VAS from 0 to10. They were asked to report their mean painscore over the last week.Active abduction of the shoulder of the dissected and non-dissected sideswere measured with an inclinometer. The patient stood with the back, heels,and buttocks against the wall, and were asked to abduct both arms to themaximum of their ability. Reduced abduction was assumed if the differencebetween the dissected and non-dissected side was ≥ 40°. All patients were measured by the same observer. Statistical analyses weremade using the statistical package for the social sciences 10.0 (SPSSInc.,Chicago) and Chi-square analysis, and Student’s t-test for independentsamples. Probabilities of less than 0.05 were accepted as significant.

Results In total 122 patients participated in the study, (41 female/ 81 male) mean age61 (13) years, and mean follow-up of 3 (1-7) years. All patients withshoulder complaints before operation (n=7), and patients who could notremember whether they did or did not (n=2) were excluded from furtheranalyses. Of the remaining 113; 5 underwent radical, 43 modified radical, 48supraomohyoid, 16 posterolateral, and 1 lateral neck dissection. Beforestatistical analyses the patient with the lateral neck dissection was excluded.The spinal accessory nerve was dissected in all radical neck dissections, andin one modified radical, and one posterolateral neck dissection.

Chapter 5

64

Table 1 Characteristics of the 112 patients.

Gender:-Female -Male

39 (35)73 (65)

Age (years) Mean (SD) 61 (13)Type of neck dissection:-Radical -Modified radical -Supraomohyoid -Posterolateral

5 (4)43 (38)48 (43)16 (14)

Radiotherapy: -Yes-No

75 (67)37 (33)

Tumour stage:-T1-T2-T3-T4-Unknown

17 (15)31 (28)15 (13)24 (21)25 (22)

Reconstructions: -No-Nasolabial-Pectoral cutaneuos flap-Radial cutaneous flap-Fibula

84 (75)5 (4)8 (7)

13 (12)2 (2)

Data are number (%) expect were otherwise stated

In total the records of 112 patients (73 male/ 39 female) mean age 61 (13)years were analysed. The mean (SD) follow-up was 3 (2) years. Theircharacteristics are summarised in Table 1. Table 2 shows the relation between the type of neck dissection and the signsof dysfunction of the trapezius muscle. Thirty-nine patients (35%)complained of shoulder pain of whom 19 (49%) had no dysfunction of thenerve, and 20 (51%) did. Dysfunction was present in 29 patients, of whom20 (69%) had shoulder pain. Shoulder pain was significantly associated with Chapter 5

65

dysfunction of the nerve (Table 3). Sixteen (14 %) patients reported temporary shoulder complaints after neck dissection, that were no longerpresent at the time of the study. The cervical plexus was preserved in 30patients (27%), sacrificed in 6 (5 %), partially sacrificed in 5 (4 %) and itwas unknown in 71 (63 %) of the patients. From these data we could notanalyse the contribution of the cervical plexus to the function of thetrapezius muscle.

Table 2 Assessment of trapezius muscle and scapula as signs of dysfunction of the spinal accessory nerve, in patients after radical, modified radical, posterolateral, and supraomohyoid neck dissection.

Type of neckdissection

Numberof

patients

Atrophyof

trapezius

Shoulderdrop

Scapuladistance > 2 cm

Spinal accessorynerve

dysfunction*Radical 5 5 5 5 5 (100)Modified radical 43 12 13 10 12 (28)Postero lateral 16 9 9 8 9 (56)Supraomohyoid 48 3 6 5 3 (6)Total 112 29 (26) 33 (29) 28 (25) 29 (26 )Data are number of patients having each type of neck dissection who were affected. Percentages of totals are in parentheses.*Two or three signs of loss of function of the trapezius muscle function present

The mean (SD) VAS for patients with shoulder pain was 4.2 (2.3). Ninepatients perceived shoulder pain on the non-dissected side (VAS 4.2 (2.6)). Active range of motion (abduction) was measured in 111 patients, and themean (SD) active abduction on the operated side 146° (42°), wassignificantly less than on the non-operated side 162° (26°). There was adifference in active range of motion of ≥ 40° in 23 patients, which wassignificantly associated with dysfunction of the spinal accessory nerve(Table 3).

Chapter 5

66

Table 3 Dysfunction of the spinal accessory nerve in relation to shoulder pain, and active range of motion (abduction) after neck dissection

Number ofpatients

Dysfunction No dysfunction P values

Shoulder pain 39 20 (51) 19 (49)No shoulder pain 73 9 (12) 64 (88)Total 112 29 (26) 83 (74) 0.001

Difference ≥ 40°* 23 22 (96) 1 (4)Difference < 40° 88 7 (8) 81 (92)Total 111# 29(26) 82 (74) 0.001Data are number (%) of patients. Percentages are row percentages.* Difference in abduction between the dissected side and the non-dissected side # One patient was not physically examined

DiscussionDysfunction of the spinal accessory nerve occurs in all cases after neckdissection with resection of the nerve and in about 22 % when it ispreserved. It may cause shoulder pain but such pain may also be present in49 % of the cases without signs of dysfunction. Shoulder pain can beattributed to dysfunction of the spinal accessory nerve in only 51% ofpatients. As well as by a physical examination the function of the nerve can also beinvestigated by an electromyography (EMG), which provides informationabout the extent of denervation. However there is a strong relation betweenEMG findings and the findings of physical examination of the shouldergirdle.15,16

The active range of motion of the shoulder girdle decreases after neckdissection, particularly if the nerve has been resected.17 A common way ofevaluating the descending trapezius muscle is to elevate the shoulder girdle,but because the levator scapulae also elevates the shoulder girdle, this is nota valid method. A clinical interview, and simple physical examination several weeks after neck dissection is a useful way of accessing the functionof the spinal accessory nerve and complaints about the shoulder. Thephysical examination after neck dissection should include bilateral active

Chapter 5

67

abduction and inspection of the shoulder girdle, looking for atrophy of thetrapezius muscle, changed posture of the scapulae, and shoulder drop. We assumed that the nerve was not functioning if two of three physical signsof dysfunction of the trapezius were present. Only one physical sign mightbe caused by postoperative immobilisation by pain, neck dissection, or theprimary resection. We therefore arbitrarily choose two physical signs out ofthree. All patients who were known to have had the nerve resected had atleast two signs of dysfunction. Shoulder pain has been claimed to be a consequence of dysfunction of theaccessory nerve, and although there is a significant relationship, only 51 %of the patients with shoulder pain had a dysfunctional nerve. Differenttissues have been suggested to be responsible for shoulder pain, in patientswith dysfunction of the spinal accessory nerve: a secondary frozenshoulder,18 a hypertrophic sternoclavicular joint,19 and over-stretching of therhomboid and levator scapulae muscle.20 As well as these tissues, damage toor cutting of cutaneous sensory nerves causing deafferentation pain, orneuromas, may also cause shoulder pain.21 Why many patients have shoulderpain after neck dissection with a normal nerve function is unknown.Deafferentation pain, myofascial pain or neuromas may be the cause.20

Of the patients with a dysfunctional spinal accessory nerve 9 of 29 (31%)had no shoulder pain, and 7 had no major change in abduction. Thesefindings of dysfunction without shoulder complaints are similar to thefindings of Saunders et al.9 Patients in that study had atrophy of thetrapezius, shoulder drop, and a changed position of the scapula but did notdevelop pain or a big reduction in the range of motion. Probably thesepatients managed to cope with the dysfunction, which makes them aninteresting group for further investigation.After neck dissection with preservation of the nerve, neurapraxia may resultin a loss of function of the trapezius muscle. Several authors havehypothesised about the causes of this: traction during the operation,microtraumata or devascularisation of the nerve during, or as a consequenceof the operation.22,23 The chance for microtraumata may be more likelybecause of the anatomical variations in the course of the nerve, particularlyin the passage of the sternocleidomastoid muscle, which may lead to moreextensive damage.12,22 Looking into the course of the nerve, the mostimportant levels are II and V. At level V the C3 and C4 branches can be

Chapter 5

68

damaged by the surgeon, and preservation of level V is probably the mainreason why supraomohyoid neck dissection cause less morbidity of thenerve. The prevalence of dysfunction in posterolateral neck dissectionsshould be, according to the dissected level II and V, comparable to that inmodified radical neck dissections, but in our study the incidence isconsiderably higher. A reason for this might be that posterolateral neckdissections are more likely to be done in combination with removal of largetumours in the larynx. In this study no extended posterolateral neckdissections, in which the trapezius muscle is partly dissected, were found. The preservation of the cervical plexus may decrease the incidence ofshoulder pain in 25 %. 5 In our group all seven patients in whom the spinalaccessory nerve was sacrificed lost function in the trapezius muscle. Of theseseven, in one the cervical plexus was sacrificed, one was partly sacrificed,and in the others cases it was unknown. So no conclusions can be drawnabout the function of the cervical plexus after neck dissection. It seemsworthwhile to detect, and to preserve the branches of the cervical plexus, andto try to spare or damage these branches as little as possible. This sparingmainly consists the preparation of level V in which the branches of C3 andC4 are located. Our retrospective study was based on observations at least a year after neckdissection. Because of the long follow up (mean 3 years), recall of pre-operative shoulder problems may have been biased, and missing databecause of incomplete medical records resulted in little information aboutpreservation of branches of the cervical plexus.We were interested in long term outcome and did not include patients withina year after neck dissection. In the first year after operation psychologicaldistress and risk of recurrence of the tumour are high, and may influenceperception of shoulder pain. Shoulder pain may also recover during the firstpostoperative year. Fourteen percent of the patients in our group hadtemporary shoulder complaints, mostly after modified radical neckdissection, in the first year. If shoulder complaints are the result ofneurapraxia of the spinal accessory nerve, regeneration might occur evenafter the first postoperative year. Regeneration has mainly been describedafter modified radical neck dissection.24 As well as regeneration, effectivecoping strategies, reducing heavy physical activities, physiotherapy,25 andchanges in psychological factors 26 may reduce pain and disability.

Chapter 5

69

In this group nine patients (8%) had shoulder complaints on the unoperatedside. According to epidemiological research in a western population this is alow percentage.27

Based on the results of this study we conclude that shoulder pain after neckdissection can only partly (51%) be attributed to neurotmesis or neurapraxiaof the spinal accessory nerve. Further investigations into the causes ofshoulder pain, with and without dysfunction of the nerve, is required, and isimportant for postoperative rehabilitation.

Chapter 5

70

References

1. Ewing MRMH. Disability following radical neck dissection. Cancer 1952;5: 873-883.

2. Cheng PT, Hao SP, Lin YH, Yeh AR. Objective comparison of shoulderdysfunction after three neck dissection techniques. Ann Otol RhinolLaryngol 2000; 109: 761-766.

3. Remmler D, Byers R, Scheetz J, et al. A prospective study of shoulderdisability resulting from radical and modified neck dissections. Head NeckSurg 1986; 8: 280-286.

4. Krause HR. Reinnervation of the trapezius muscle after radical neckdissection. J Craniomaxillofac Surg 1994; 22: 323-329.

5. Krause HR. Shoulder-arm-syndrome after radical neck dissection: itsrelation with the innervation of the trapezius muscle. Int J Oral MaxillofacSurg 1992; 21: 276-279.

6. Stacey RJ, O'Leary ST, Hamlyn PJ. The innervation of the trapeziusmuscle: a cervical motor supply. J Craniomaxillofac Surg 1995; 23: 250-251.

7. Robbins KT, Medina JE, Wolfe GT, Levine PA, Sessions RB, Pruet CW.Standardizing neck dissection terminology. Official report of theAcademy's Committee for Head and Neck Surgery and Oncology. ArchOtolaryngol Head Neck Surg 1991; 117: 601-605.

8. Carenfelt C, Eliasson K. Occurrence, duration and prognosis of unexpectedaccessory nerve paresis in radical neck dissection. Acta Otolaryngol 1980;90: 470-473.

9. Saunders JR, Hirata RM, Jaques DA. Considering the spinal accessorynerve in head and neck surgery. Am J Surg 1985; 150: 491-494.

10. Short SO, Kaplan JN, Laramore GE, Cummings CW. Shoulder pain andfunction after neck dissection with or without preservation of the spinalaccessory nerve. Am J Surg 1984; 148: 478-482.

11. Leipzig B, Suen JY, English JL, Barnes J, Hooper M. Functional evaluationof the spinal accessory nerve after neck dissection. Am J Surg 1983; 146:526-530.

12. Pinsolle V, Michelet V, Majoufre C, Caix P, Siberchicot F, Pinsolle J.[Spinal accessory nerve and lymphatic neck dissection]. Rev Stomatol ChirMaxillofac 1997; 98: 138-142.

13. Plukker JT, Vermey A, Roodenburg JL, Oldhoff J. Posterolateral neck dissection: technique and results. Br J Surg 1993; 80: 1127-1129.

Chapter 5

71

14. Fialka V, Vinzenz K. [Physiotherapy and diagnosis of shoulder lesions afterradical neck dissection]. Dtsch Z Mund Kiefer Gesichtschir 1989; 13: 220-225.

15. Sobol S, Jensen C, Sawyer W, Costiloe P, Thong N. Objective comparisonof physical dysfunction after neck dissection. Am J Surg 1985; 150: 503-509.

16. Zibordi F, Baiocco F, Bascelli C, Bini A, Canepa A. Spinal accessory nervefunction following neck dissection. Ann Otol Rhinol Laryngol 1988; 97:83-86.

17. Dijkstra PU, van Wilgen PC, Buijs RP et al. Incidence of shoulder painafter neck dissection: A clinical explorative study for risk factors. HeadNeck 2001; 23: 947-953.

18. Patten C, Hillel AD. The 11th nerve syndrome. Accessory nerve palsy oradhesive capsulitis? Arch Otolaryngol Head Neck Surg 1993; 119: 215-220.

19. Cantlon GE, Gluckman JL. Sternoclavicular joint hypertrophy followingradical neck dissection. Head Neck Surg 1983; 5: 218-221.

20. Nori S, Soo KC, Green RF, Strong EW, Miodownik S. Utilization ofintraoperative electroneurography to understand the innervation of thetrapezius muscle. Muscle Nerve 1997; 20: 279-285.

21. Brown H, Burns S, Kaiser CW. The spinal accessory nerve plexus, thetrapezius muscle, and shoulder stabilization after radical neck cancersurgery. Ann Surg 1988; 208: 654-661.

22. Soo KC, Guiloff RJ, Oh A, Della RG, Westbury G. Innervation of thetrapezius muscle: a study in patients undergoing neck dissections. HeadNeck 1990; 12: 488-495.

23. Shankar K, Means KM. Accessory nerve conduction in neck dissectionsubjects. Arch Phys Med Rehabil 1990; 71: 403-405.

24. Kuntz AL, Weymuller-EA J. Impact of neck dissection on quality of life.Laryngoscope 1999; 109: 1334-1338.

25. Herring D, King AI, Connelly M. New rehabilitation concepts inmanagement of radical neck dissection syndrome. A clinical report. PhysTher 1987; 67: 1095-1099.

26. de Graeff A, de Leeuw J, Ros WJ, Hordijk GJ, Blijham GH, Winnubst JA.A prospective study on quality of life of patients with cancer of the oralcavity or oropharynx treated with surgery with or without radiotherapy.Oral Oncol 1999; 35: 27-32.

Chapter 5

72

27. van der Windt DAWM, Croft PR. Shoulder pain. In: Crombie IK, Croft PR,Linton SJ, Leresche L, Korff von M, eds. Epidemiology of pain. Seattle:IASP Press, 1999: 257-281.

Chapter 5

73

CHAPTER 6MORBIDITY OF THE NECK AFTER HEAD

AND NECK CANCER THERAPY

C. Paul van Wilgen, PT1,2 ,Pieter U. Dijkstra, PhD1,2 , Berend F.A.M. van der Laan, PhD3,John T. Plukker, PhD4,Jan L.N. Roodenburg, PhD 11.Department of Oral and Maxillofacial Surgery, 2. Department ofRehabilitation, 3. Department. of Otorhinolaryngology, Head & NeckSurgery, 4. Department of Surgical Oncology. University HospitalGroningen.

Accepted Head and Neck (december 2003)

74

AbstractBackground Studies on morbidity of the neck after head and neck cancer therapy aredescribed scarcely.MethodsPatients who underwent surgery, including neck dissection, with and withoutradiation therapy, at least 1 year before the study were asked to participate.We assessed neck pain, loss of sensation, range of motion of the cervicalspine, and shoulder pain.ResultsOf the 220 patients who were invited 153 (70 %) participated in the study.Neck pain was present in 33% of the patients (n=51), and shoulder pain in37% of the patients (n=57). Neuropathic pain of the neck was present in 32%(n=49), myofascial pain was present in 46% (n=70), and joint pain in 24%(n=37). Loss of sensation of the neck was present in 65% (n=99) and wasrelated to type of neck dissection and radiation therapy. Range of motion ofthe neck was significantly decreased, because of the neck dissection and orradiation therapy in lateral flexion away from the operated side. Conclusion The occurrences of morbidity of the neck after cancer therapy wereconsiderable and consisted of neck pain, loss of sensation, and decreasedrange of motion.

Chapter 6

75

Introduction Head and neck tumours account for 10 % of all new cancer diagnoses in TheNetherlands. 1 The treatment of head and neck tumours consists of surgery,radiation therapy or both. During surgery, often an elective or therapeuticneck dissection is performed. As a result of head and neck cancer therapy,morbidity of neck and shoulder region may occur. This morbidity manifestsitself through pain, loss of sensation, disfigurement, reduced range of motionof the shoulder, and changes in quality of life.2,3,4,5

Morbidity of the neck itself has only been described in small number ofstudies. Neck tightness was reported in 71% of the cases, together withshoulder discomfort in 53% of the cases, having a substantial negative effecton quality of life. This was reported by Shah et al.6 in a retrospective studyof 51 patients after different types of neck dissection. Head and neck painwas reported by Chaplin et al.2 in 25% of 93 patients after neck dissection,radiation therapy, or both. In a study of 25 patients with persistent neck painafter neck dissection, Sist et al.7 described two types of neck pain:neuropathic pain (100%) and myofascial pain (72%). In contrast to thesestudies Talmi et al.8 described three groups of patients (n=88) after radicaland modified radical neck dissection and claimed that pain in the neck afterneck dissection was uncommon. The results of previous studies regarding the presence of neck pain afterhead and neck cancer therapy are conflicting. It is unclear whether the typeof neck dissection has any relationship with the occurrence of neck pain. Theexact cause of neck pain after neck dissection was unclear. Besidesneuropathic pain, other causes have been described: sternoclavicular jointpain caused by subluxation9 or hypertrophy10 and myofascial pain in headand neck muscles.11 Loss of sensation after selective neck dissection related to sacrificing thesensory cervical root branches. For the analyses of loss of sensation, Saffolddivided the neck in eight regions. After neck dissection in which the sensorycervical root branches were sacrificed, significantly more sensory deficitswere present than in neck dissections in which the sensory cervical rootbranches were preserved.12 Loss of sensation after neck dissection, includingdissection of the sensory nerve branches, was reported most frequently in theregions D, E, and F (Figure 1). However, loss of sensation of the neck could

Chapter 6

76

also be related to radiation therapy. 13 To the best of our knowledge, no otherstudies have evaluated the loss of sensation after neck dissection adequately.

Figure 1 Regions in the neck for the assessment of loss of sensation after Saffold

Haribhakti et al. described impairment of neck movement after neckdissection and additional reconstruction. They found no differences inpatients with or without a pectoralis major myocutaneous flap.14 BothSchuller et al. and Haribhakti et al. used questionnaires, without a physicalexamination or measuring the range of motion. Studies measuring the rangeof motion of the cervical spine after neck dissection were not available.The aim of our study was to analyze neck morbidity, including neck painand loss of sensation and range of motion of the neck after head and neckcancer therapy, and to study the relationship between morbidity and type ofneck dissection, number of dissected levels, radiation therapy, and shoulderpain.

Materials and MethodsPatients who un derwent a neck dissection with or without radiation therapy,performed by the multidisciplinary Head and Neck Oncology Group of theUniversity Hospital Groningen in the period 1994 to 2000, were asked toparticipate. They were informed about the study by a personal letter, whichwas send 1 week before they visited the hospital for an appointment as part

Chapter 6

Region A: represents the lower half of the external ear, extending fromthe root of the helix to the tip of the lobule.Region B: is the midface and includes the face above a line drawnbetween the oral commissure and the angle of the mandibule.Region C: the lower face, extends from below this line (region b) to theinferior border of the mandible. The neck is divided into upper and lowerportions based on a horizontal line at the level of the thyroid prominence.Region D: is the upper posterior neck behind the anterior border of thesternocleidomastoideus.Region E: is the upper anterolateral neck, extending from the anteriorborder of the sternocleidomastoideus to a vertical line drawn from thefacial notch of the mandible.Region F: is the lower posterior neck behind the anterior border of thesternocleidomastoideus

77

of the standard postoperative care. Patients who had a recurrence of thetumor, or who could not understand Dutch were excluded. All patients wereat least 1 year after surgery. From the medical records, we retrieved the dateof surgery, type of surgery, type of neck dissection, preserved and/orsacrificed structures, preoperative or post operative radiation therapy, andtype of reconstruction. Types of neck dissections were recorded in accordance with theclassification of Robbins et al.15: radical neck dissection, modified radicalneck dissection, and four types of selective neck dissections. In thisclassification, lymph nodes in the neck are divided into six anatomic levels.Sensory cervical root branches were not preserved during surgery. Neckdissections were performed by use of the upper McFee incision for thedissection of level I, II, and III, and the lower McFee incision for thedissection of level IV, V, and VI. For the analyses of pain and sensation, wecategorized the patients in two groups: patients who onderwent dissection ofthe upper areas I,II, and III (all supraomohyoid neck dissections) andpatients who underwent dissection of the upper and lower area’s I to VI (allother neck dissections). A lower McFee incision was always performed incombination with an upper McFee incision.14 We asked patients with neck pain for their average pain intensity over theprevious week. Pain was measured by a numbered (0 – 10) visual analogscale. Allodynia was defined as an abnormal evoked pain tested with anonpainful stimulus. 16 Allodynia was elicited by touching the neck andcheek gently with a fingertip several times. When a patient reported pain,allodynia was assumed to be present. Hyperpathia was defined as anabnormally painful reaction to a painful stimulus compared with the reactionon an unoperated or radiated body part.16 Hyperpathia was tested with a pinprick. To get patients acquainted with the procedure, the nonirradiated cheekor upper-arm was tested first. Hyperpathia was present if pain was felt moreintensely in the neck, or cheek, at the side of the neck dissection compared tothe control region. Myofascial pain was tested by palpating the trapeziusmuscle, the levator scapula, rhomboid muscles, and the pectoral muscle.17

Myofascial pain was present if the same spot was reported as painful at leasttwo times during palpation. If muscles could not be palpated properly (forinstance, in case of severe trapezius muscle atrophy) the muscle wasreported as not painful. Myofascial pain could be distinguished from

Chapter 6

78

allodynia or hyperpathia. Allodynia and hyperpathia were sensations testedby stimulating the skin, whereas myofascial pain was investigated bysystematically palpating the muscles and looking for taught bands,triggerpoints and pain.17 Furthermore, the sterno-clavicular joint andacromio-clavicular joint were tested for pain by means of a “joint play”test.18 Because it was difficult in some cases to distinguish between neck pain andshoulder pain, we also assessed shoulder pain. Patients were asked whetherthey had shoulder pain, and their average pain intensity over the previousweek was assessed with a numbered VAS (0 – 10). Patients who hadundergone bilateral surgery, were assessed on the (most) painful side, or ifpain was not present, on the dominant side. Loss of sensation was assessedin six distinct regions (Figure 1) related to the anatomical regions of thesensory cervical root branches.12 Saffold described eight regions; we did notuse the two central regions over the larynx. Patients were examined for senseof touch with a wisp of cotton or pin prick. Patients were first acquaintedwith the nature of the stimulus by applying it to a unoperated, andnonirradiated body part as mentioned previously. Patients were asked torespond each time they felt the wisp of cotton or pin. All regions weretouched at least five times, when more than 50 % of the attempts were notfelt or reported incorrectly, loss of sensation was assumed. To analyze lossof sensation, we calculated the number of anatomic regions with loss ofsensation. Range of motion was measured by means of a cervicalmeasurement system. This system consisted of a helmet with twoinclinometers and a compass.19 All motions of the cervical spine (rotation toboth sides, lateral flexion to both sides, extension and forward flexion) weremeasured actively, while the patient was seated. For statistical analyses we used SPSS 10.0. Descriptive statistics, chi- squaretest, 95% confidence intervals calculation, and linear regression (methodstepwise forward) were performed. ResultsIn total 220 patients were invited, of which 153 patients (70 %) participatedin this study. The study population had 102 male, and 51 female patients(mean age 61.3 years; SD 11.9). The mean follow up was 3.0 years (SD 1.7).Unilateral neck dissection was performed in 121 patients, and bilateral neck

Chapter 6

79

dissection was performed in 32 patients. The following type of neckdissections were performed: 72 supraomohyoid neck dissections, 22posterolateral neck dissections, 54 modified radical neck dissections, and 5radical neck dissections. In seven patients, the spinal accessory nerve wassacrificed. In all supraomohyoid neck dissections, the upper McFee incisionwas used; for all others, the upper and lower McFee incisions were used.Radiation therapy was given to 107 patients (90 postoperative, 21preoperative and 4 patients preoperative and post operative). Fifty-onepatients (33%) experienced neck pain, of which 20 (39 %) had allodynia inthe neck, and 49 (96%) experienced hyperpathia; the mean intensity of neckpain was 3.5 (SD 2.3). Of the patients who underwent unilateral surgery four(3 %) had neck pain on the non surgical side; the mean pain intensity was 2(SD 2.1). Neck pain was significantly related to radiation therapy but not tothe number of levels dissected (Table 1). Shoulder pain, on the surgical side,was present in 57 patients (37 %); the mean pain intensity was 3.7 (SD 2.3).Shoulder pain was significantly related to neck pain and allodynia (p < .01)(Table 1). Myofascial pain in the levator scapulae muscle and trapezius parsdescendens muscle was found most frequently (Table 2). Myofascial painwas present significantly more on the surgical side (p < .05). Of the patientwho underwent unilateral surgery, in the nonsurgical side myofascial painwas present in 10 patients (8%) in the levator scapulae, 11 (9%) in thetrapezius muscle, and in 6 (5%) in the rhomboid muscles. Myofascial painwas significantly related to the number of levels dissected for the levatorscapulae (p < .05) but not for other muscles (Table 2). Myofascial pain wasnot related to radiation therapy (Table 2). Pain was present at thesternoclavicular joint in 24 patients (16%), and in the acromioclavicular jointin 37 patients (24%). Pain was significantly more frequently present on thesurgical side (p < .01). On the nonsurgical side the sternoclavicular joint waspainful in five patients (4%), and the acromioclavicular joint in six patients(5%). Joint pain was not significantly related to the number of levelsdissected or radiation therapy (Table 2). Shoulder pain was closely related tomyogenic pain and joint pain (Table 2).

Chapter 6

80

Table 1 Neck pain, hyperaesthesia and allodynia in relation to radiation therapy, total of levels dissected during neck dissection, and shoulder pain (n=153).

Frequency

Radiation therapyYes No(n= 107) (n= 46)

Levels dissectedUpper* Upper + lower**(n= 72) (n= 81)

Shoulder painYes No(n = 57) (n = 96)

Neck pain Yes 51 (33 %)No 102 (67 %)

41 (38%) 10 (22%)66 (62%) 36 (78%) p < .046

21 (29%) 30 (37%)51 (71%) 51 (63%) n.s

34 (60%) 17 (18%)23 (40%) 79 (82%) p < .01

Hyperpathia Yes 49 (32 %)No 104 (68 %)

36 (33%) 13 (28%)71 (66%) 33 (72%) p < .064

23 (32%) 26 % (32%)49 (68%) 55 % (68%) n.s.

23 (40%) 26 (27%)34 (60%) 70 (73%) n.s.

Allodynia Yes 20 (13 %)No 133 (87 %)

15 (33% ) 5 (11%)92 (66%) 41 (89%) p < .056

9 (13 %) 11 (14%)63 (88 %) 70 (86%) n.s.

13 (23%) 7 (7%)44 (77%) 89 (93%) p < .01

Percentages refer to column percentagesAbbreviations: n.s: not significant.* Upper McFee are all supraomohyoid neck dissections ** Upper and lower McFee are radical, modified radical, and posterolateral neck dissections

Chapter 6

81

Table 2 Myofascial pain and pain of the sternoclavicular joint and acromioclavicular joint related to radiation therapy, number of dissected areas, and shoulder pain.

Location of pain Pain ontheoperatedside(n=153)

Radiation therapy

Yes No (n=107) (n=46)

Levels dissectedUpper* Upper + lower**(n= 72) (n= 81)

Shoulder pain

Yes No(n = 57) (n = 96)

Myofascial pain -Trapezius pars desc.-Levator scapulae-Rhomboideus-Pectoralis major

54 (35%)70 (46%)47 (31%)11 (7%)

39 (37%) 15 (33%) n.s.54 (50%) 16 (35%) n.s.34 (32%) 13 (28%) n.s.8 (8%) 3 (7%) n.s.

24 (33%) 30 (38%) n.s. 26 (36%) 44 (54%) p < .05 19 (26%) 28 (35%) n.s.3 (4%) 8 (10%) n.s.

33 (58%) 21 (22%) p < .0148 (84%) 22 (23%) p < .0138 (68%) 9 (9%) p < .018 (14%) 3 (3%) p < .05

Joint pain -Sternoclavicular joint 24 (16%) 18 (17%) 6 (13%) n.s. 10 (14%) 14 (17%) n.s. 16 (28%) 8 (8%) p < .01-Acromioclavicular joint 37 (24%) 27 (25%) 10 (22%) n.s. 13 (18%) 24 (30%) n.s. 26 (46%) 11 (12%) p < .01For statistical analyses chi-square test was used. Percentages are column percentages.* Upper McFee are all supraomohyoid neck dissections** Upper and lower McFee are radical, modified radical, and posterolateral ne ck dissections

Chapter 6

82

Ninety-nine patients (65%) experienced loss of sensation. This loss was mostfrequently observed in area E (upper anterolateral neck), area A (the lowerhalf external ear), and area F (lower posterior neck) (Table 3). Loss ofsensation was significantly related to the number of levels dissected andradiation therapy (Table 4). Age and follow-up were not significantly relatedto loss of sensation.

Table 3 Frequencies and percentages of regions with loss of sensation after four types of neck dissection

Reg

ion SOHND

(n=72)PLND(n=22)

MRND(n = 54)

RND(n = 5)

Total group(n=153)

A:B:C:D:EF:

10 (14%)6 (8%)6 (8%)6 (8%)

20 (28%)3 (4%)

12 (55%)2 (9%)1 (5%)

9 (41%)18 (82%)13 (59%)

32 (59%)5 (9%)

9 (17%)28 (52%)43 (80%)31 (57%)

5 (100%)2 (40%)2 (40%)4 (80%)

5 (100%)5 (100%)

59 (38%)15 (10%)18 (12%)47 (31%)86 (56%)52 (34%)

Abbreviations: SOHND: Supraomohyoid neck dissection, PLND Postero-lateralneck dissection, MRND Modified radical neck dissection, RND Radical neckdissectionClinical note: supraomohyoid neck dissection has a lower degree of loss of sensationin almost all regions.

The descriptive statistics of range of motion of the cervical spine fordifferent age groups are presented in Table 5. To analyze the relationshipbetween active range of motion and the number of levels dissected andradiation therapy, we performed a linear regression (stepwise forward).Lateral flexion away from the dissected side was significantly related to age(ß: –0.3), number of levels dissected (ß: -2.7), and radiation therapy (ß –7.9).Active lateroflexion towards the surgical side, active rotation towards thesurgical side, and active flexion were only significantly related to age (Table6).

Chapter 6

83

The sternocleidomastoid muscle was sacrificed in 22 patients. No significantrelationship was found between sacrificing the sternocleidomastoid muscleand a reduced range of motion of the neck, although rotation away from thesurgical side was averagely 10° less in patients for whom thesternocleidomastoid muscle was sacrificed.

Table 4 Results of linear regression of analysis to predict the number of regions with loss of sensation, as a result of head and neck cancer therapy

Variable ß 95 % CI interval for ß R square

Levels dissected 0.9 (0.6 to 1.1)Radiation therapy 1.2 (0.7 to 1.7)Constant -2.5 (-3.3 to -1.5 ) .39In the regression analyse the following variables were entered stepwise, number oflevels dissected (level 1 to 6) and radiation therapy (yes/no), age, ge nder andfollow-up.

Chapter 6

84

Table 5 Active range of motion of the cervical spine in degree’s and standard deviation, classified by age, in patients (n = 153) after radical, modified radical or selective neck dissection.

Age Number ofpatients

Rotation awayfrom operatedside

Rotation tooperatedside

Lateral flexionaway fromoperated side

Lateralflexion tooperatedside

Flexion Extension

< -40 n=6 70° (15) 74° (16 ) 35° (12) 35° (11) 66° (11) 53° (10)40-49 n=15 61° (21) 70° (14) 31° (13) 35° (12) 69° (15) 50° (17)50-59 n=51 61° (15) 60° (14) 26° (9) 29° (8) 52° (13) 49° (15)60-69 n=48 59° (13) 59° (13) 24° (10) 29° (10) 52° (11) 47° (15)70-79 n=23 58° (7) 55° (10) 23° (8) 27° (8) 53° (12 ) 48° (11)80- > n=10 52° (14) 52° (10) 19° (7) 20° (10) 45° (9) 37° (21)

Chapter 6

85

Discussion Neck pain (33%) and loss of sensation (65%) were important aspects of neckmorbidity after head and neck cancer therapy. Contrary to Talmi et al. 8 wefound neck pain in 33 % of the patients. Similar to the results of Sist et al. 7

we found that neuropathic pain, hyperpathia (96%), and allodynia (39%)were present in most patients with neck pain. In daily life, patientsexperienced neuropathic pain during shaving or when they were exposed towind or low temperatures. Neck pain was significantly related to radiationtherapy. Although not significant hyperpathia (p < .064) and allodynia of theneck (p < .056) seem to be present more frequently after radiation therapy,suggesting a trend. Myofascial pain was most frequently present in the levator scapulae (46%).This percentage is higher than the percentages of patients that reported neckpain (33%). Several patients did not experience myofascial painspontaneously but only during palpation, and myofascial pain was often feltas shoulder pain. Myofascial pain in the levator scapulae was related to thenumber of dissected levels. For the other types of myofascial pain norelationship was found with the number of levels dissected. Myofascial painwas strongly related to shoulder pain. Postoperative myofascial shoulder pain may be associated withpostoperative shoulder drop as a consequence of spinal accessory nervedysfunction.20 But in this group in approximately 50 % of the patients withshoulder pain, the spinal accessory nerve was functioning normally.21 Inthese cases, myofascial shoulder pain could not be attributed to spinalaccessory nerve dysfunction. So shoulder pain with an intact spinalaccessory nerve should be attributed to other causes, such as psychologicalfactors.22 The relationship between shoulder pain and neck pain issignificant, probably because most of the tested structures span both neckand shoulder region. It seems, however, that neck pain was more associated with neuropatic painand shoulder pain more with myofascial pain.

Chapter 6

86

Table 6 Linear regression to predict range of motion of the cervical spine after head and neck cancer therapy

VariableDependent

ßIndependent 95 % CI of ß

Rsquare

Active lateroflexion to the operated side-Age-Constant

-0.344.0

(-0.4 to – 0.1) (34.4 to 35.6) .12

Active lateral flexion away from operated side-Age-Levels dissected-Radiation therapy -Constant

-0.3-2.7-7.959.4

(-0.4 to – 0.1)(-4.8 to – 0.5)

(-15.7 to – 0.3) ( 44.0 to 74.8) .22

Active rotation to the operated side-Age-Constant

-0.483.5

(-0.6 to – 0.2)(69.2 to 97.8) .13

Active flexion -Age-Constant

-0.479.4

(-0.6 to - 0.2)(65.8 to 93.0) .14

In the regression analyses, the following variables were entered stepwise; age,number of levels dissected during neck dissection, and radiation therapy (yes/no). In the other movements of the cervical spine no significant influence of age, neckdissection or radiation therapy was found.

Loss of sensation was seen most in area E, A, and F. These findings weresimilar to the findings described by Saffold et al.12 Many patients in ourstudy group m entioned recovery of loss of sensation in the firstpostoperative year; after this first year, loss of sensation was not related tofollow-up. Loss of sensation was not likely to recover more than 1 year aftersurgery. The assessment of loss of sensation could be influenced byreconstruction in the surgical area. In our population, however,reconstruction did not significantly influence loss of sensation. Range ofmotion of the cervical spine was affected by radiation therapy and thenumber of dissected levels in lateral flexion away from the surgical side.Scars and radiation therapy could cause fibroses of tissues in the neck.

Chapter 6

87

Movements for which elasticity of soft tissues is needed seem to beespecially affected. Other movements were not significantly related toradiation therapy or the number of levels dissected. Therefore, we concludethat the effects of neck dissection and radiation therapy on range of motionof the neck are limited. In the literature, normative data of range of motionof the cervical spine of similar patients assessed with the same type ofinclinometer were not available. In patients with a sacrificed sternocleidomastoid muscle, a nonsignificantmean reduction of 10° in active rotation away from the surgical side wasfound. The difficulty in analyzing the influence of sacrificing thesternocleidomastoid muscle on neck range of motion was that we only had asmall group, and almost all (81%) underwent a radical or modified radicalneck dissections.In this study, 220 patients were invitated to participate. Participation in thestudy was high (70%). Reasons for not participating were lack of time, lackof interest, or the belief that the investigation was too strenuous.Participation might have been higher in patients perceiving pain and or lossof sensation. Therefore incidences described in this study may be anoverestimation of the incidence of neck morbidity in the total populationafter head and neck cancer therapy. The implications for the clinical practice for patients after head and neckcancer therapy are that in the management of neck pain, a distinction shouldbe made between different types of pain. Neuropathic pain should be treateddifferently than myofascial pain. Neuropathic pain may be prevented bystandard adequate perioperative pain management.23 Furthermore patientscan be informed, before head and neck cancer therapy, about the chances ofloss of sensation and reduced range of motion. It is important to informpatients both, preoperative and postoperative, about possible morbidity aftertherapy. Especially in patients who fear for tumor recurrence, educationabout pain and the source of pain can be of great value, because thesepatients have an increased body perception.24 Furthermore, standardexercises to preserve range of motion of the neck may be important,although so far no studies have been performed to analyze the effects ofexercises on range of motion of the neck in these patients. On the basis of our results, we conclude that neck pain after head and neckcancer therapy was present in 33 % of the patients and is explained primarily

Chapter 6

88

by neuropathic pain. Neck pain is closely related to shoulder pain. Shoulderpain is present in 37% of the patients and is related to myofascial pain. Lossof sensation was present in 65% of the patients, and was related to thenumber of dissected levels and radiation therapy. Range of motion, thelateral flexion away from the surgical side, was decreased significantly bysurgery, radiation therapy or both.

Chapter 6

89

References1. Netherlands Cancer Registry. Head and Neck tumours in the Netherlands

1989-1995. Utrecht: Association of comprehensive cancer centres; 1998.2. Chaplin JM, Morton RP. A prospective, longitudinal study of pain in head

and neck cancer patients. Head Neck 1999;21:531-537.3. Dijkstra PU, van Wilgen CP, Buijs RP. et al. Incidence of shoulder pain after

neck dissection: A clinical explorative study for risk factors. Head Neck2001;23:947-953.

4. Kuntz AL, Weymuller-EA J. Impact of neck dissection on quality of life.Laryngoscope 1999;109:1334-1338.

5. Morton RP. Evolution of quality of life assessment in head and neck cancer.J.Laryngol.Otol. 1995;109:1029-1035.


7. Sist T, Miner M, Lema M. Characteristics of postradical neck pain syndrome:a report of 25 cases. J.Pain Symptom Manage. 1999;18:95-102.

8. Talmi YP, Horowitz Z, Pfeffer MR. et al. Pain in the neck after neckdissection. Otolaryngol.Head Neck Surg. 2000;123:302-306.

9. Krause HR. Shoulder-arm-syndrome after radical neck dissection: its relationwith the innervation of the trapezius muscle. Int.J.Oral Maxillofac.Surg.1992;21:276-279.

10. Cantlon GE, Gluckman JL. Sternoclavicular joint hypertrophy followingradical neck dissection. Head Neck Surg. 1983;5:218-221.

11. Remmler D, Byers R, Scheetz J. et al. A prospective study of shoulderdisability resulting from radical and modified neck dissections. Head NeckSurg. 1986;8:280-286.

12. Saffold SH, Wax MK, Nguyen A. et al. Sensory changes associated withselective neck dissection. Arch.Otolaryngol.Head Neck Surg. 2000;126:425-428.

13. Schuller DE, Reiches NA, Hamaker RC. et al. Analysis of disability resultingfrom treatment including radical neck dissection or modified neck dissection.Head Neck Surg. 1983;6:551-558.

14. Haribhakti VV, Kavarana NM, Tibrewala AN. Oral cavity reconstruction: anobjective assessment of function. Head Neck 1993;15:119-124.

15. Robbins KT, Medina JE, Wolfe GT, Levine PA, Sessions RB, Pruet CW.Standardizing neck dissection terminology. Official report of the Academy'sCommittee for Head and Neck Surgery and Oncology.Arch.Otolaryngol.Head Neck Surg. 1991;117:601-605.

Chapter 6

90

16. Merskey H, Bogduk N. Classification of chronic pain, second edition. Seattle:IASP press; 1994.

17. Travell JG, Simons DG. Myofascial Pain and Dysfunction. Baltimore:Williams and Wilkins; 1982.

18. Frisch H. Systematic musculoskeletal examination. Berlin:Springer-Verlag;1995. 280 p.

19. Mellin G, Olenius P, Setala H. Comparison between three differentinclinometers. Physiotherapy. 1994;80:612-614.

20. Nori S, Soo KC, Green RF, Strong EW, Miodownik S. Utilization ofintraoperative electroneurography to understand the innervation of thetrapezius muscle. Muscle Nerve 1997;20:279-285.

21. Van Wilgen CP, Dijkstra PU, van der Laan BFAM, Plukker JT, RoodenburgJLN. Shoulder complaints after neck dissection is the spinal accessory nerveinvolved? Br. J. Oral Maxillofac. surg. 2003;41:7-11.

22. Zaza C. Baine N. Cancer pain and psychological factors: a critical review ofthe literature. J. Pain Symptom Manage. 2002;5:526-542.

23. McQuay HJ, Moore RA. Pre-emptive analgesia: a systematic review ofclinical studies:1950-94. In: McQuay HJ, editor. An evidence based resourcefor pain relief. Oxford:Oxford University Press: 1998. p 164-171.

24. Jones C, Humphris G, Dixon R, Hatcher MB. Fear of cancer recurrence – Aliterature review and proposed cognitive formulation to explain exacerbationof recurrence fears. Psycho-Oncology 1997;6:95-105.

Chapter 6

91

CHAPTER 7SHOULDER AND NECK MORBIDITY IN QUALITY OF

LIFE AFTER SURGERY FOR HEAD AND NECK CANCER

C.P. van Wilgen, PT 1,2 , P.U. Dijkstra, PT,MT,PhD 1,2 B.F.A.M. van der Laan, MD, PhD 3 . J.Th. Plukker, MD,PhD, 4J.L.N. Roodenburg, DDS, PhD 11. Department of Oral and Maxillofacial Surgery, 2. Department ofRehabilitation 3. Department of Otorhinolaryngology, Head & NeckSurgery 4.Department of Surgical Oncology.

Accepted Head and Neck (februari 2004)

92

AbstractQuality of life has become a major issue in determining the outcome oftreatment in head and neck surgery with curative intent. Aim of our studywas to determine which factors in the post-operative care, especiallyshoulder and neck morbidity, are related to quality of life, and how theseoutcomes where related to a control group. Depression scores contributed significantly to all domains of quality of life.Reduced shoulder abduction, shoulder pain, and neck pain are related toseveral domains of quality of life. The patient group scored significantlyworse for social functioning and limitations due to physical problems, butscored significantly better for bodily pain and health changes.

Chapter 7

93

IntroductionThe last decade quality of life has become a major issue in determining theoutcome of treatment in head and neck surgery with curative intent.1 Qualityof life is a multidimensional construct with contributions of severaldomains.2,3 Morton stated that ten domains can be assessed investigatingquality of life in head and neck cancer patients.3 These domains includephysical aspects, psychological aspects, social well being and eveneconomic, occupational and domestic/family aspects.2 Because quality of lifehas no clear theoretical model, the outcome of research on thismultidimensional construct depends on the assumptions made by theresearchers, and the assessment tools used. In head and neck cancer researchthese assumptions often include outcome in the medical domains. Medicalaspects that have been described to be of significant influence on quality oflife are: site of tumour, TNM-stage,4 type of neck dissection,5 resection orpreservation of the spinal accessory nerve,6 reconstruction throughmyocutaneous flaps, and post-excisional defects.7

In the long term after surgery the medical domains are less important. In thisphase patients have to learn to cope with the consequences of the cancertreatment.8 Some authors have described consequences that have asignificant influence on aspects of quality of life on the long term: physicalfunction, fatigue,9 shoulder discomfort, neck tightness,10 speech, eating,11

and also depression has a significant influence on quality of life.12 However,in none of these studies the findings of a physical examination together withthe findings of the assessment of emotional factors were used to analysequality of life on the long term after head and neck surgery.The purpose of this study was to explore the impact of shoulder and neckmorbidity, following head and neck cancer treatment, beside other domainslike depression, education, age etc. on quality of life. (figure 1) We analysedphysical symptoms, psychological symptoms and social and functional wellbeing as a part of quality of life at least a year after surgery. Furthermore, weanalysed the differences in quality of life of patients after head and necksurgery and that of a control group.

Chapter 7

94

Figure 1 Model used to assess the influence of shoulder and neck morbidity and of depression on quality of life in head and neck cancer patients

Materials and methodsPatients with a neck dissection in their medical history with a standardappointment on the department of Oral and Maxillofacial Surgery,Otorhinolaryngology Head & Neck Surgery, or the department of SurgicalOncology were informed about the study by means of a letter. This letterwas send a week before they visited the outpatient department. During theappointment patients were asked by the physician to participate in the studyand to fill out an informed consent.

Chapter 7

Head and neck cancer

Survival

DeathTreatment: surgery (and radiotherapy)

Depression

QUALITY OF LIFERAND-36

Shoulder andneckmorbidity

Socio-demographicvariables

95

All patients were operated in the period 1994 to 2000, by ourmultidisciplinary Head and Neck Oncology Group. Patients with recurrenceof the tumor, or patients unable to understand Dutch were excluded from thestudy. All patients were at least one year after neck dissection. From the medical records of the participating patients and during astandardised interview the following data were retrieved; socio-demographicdata (gender, age, education, marital status, employment or social welfare,disablement insurance), and follow-up. Patients were physically examined,measuring range of motion of shoulder on operated and non-operated side(abduction, forward flexion) and neck (lateral-flexion to, and away from theoperated side). Range of motion of the shoulder and neck were measuredusing an inclinometer according to a standardised protocol. Sensibility wasmeasured according to the anatomic levels at the lateral side of the head andneck described by Saffold.13 Pain was assessed with a numbered visualanalogue scale (VAS). Patients were asked to indicate the mean painintensity over the last week for the head, neck (both sides), shoulders andboth arms. For patients operated bilaterally the (most) painful side wasreported as operated side or, if no pain was present the dominant side wasdetermined to be operated side.Depression was assessed with the Center for Epidemiological StudiesDepression scale (CES-D). This instrument measures depression in a non-psychiatric population.14,15 The questionnaire consists of 20 items describingsomatic and psychological symptoms of depression, and is translated anvalidated for a Dutch population.16 The CES-D is suitable for healthypopulations as well as for patients with cancer.17 A cut off score of 16 orhigher (range 0 – 60) is used an indicator for possible clinical depression.14

Quality of life was assessed using the RAND-36 questionnaire 1,18 this is theDutch version of the SF-36 extended with the domain health changes. TheRAND-36 contains 36-items from which 9 domains can be calculated:physical functioning, social functioning, role limitations due to physicalproblem, role limitations due to emotional problem, general mental health,vitality, bodily pain, general health perception, and health changes19

Responses are calculated to percentages from 0 % (poor health) to 100 %(excellent health). Domains of the RAND-36 were calculated as described inthe questionnaires manual. Quality of life of our study population was

Chapter 7

96

compared with the data of a control group (age 55-64) described in themanual.18 The questionnaires were administered by the investigator.Correlation’s (Pearson) were analysed between depression, shoulderabduction, shoulder pain (operated and non-operated side) lateroflexion ofthe neck, away from the operated side, neck pain, age, follow-up andsensibility, and the domains of the RAND-36.The following items were entered (method stepwise forward) as independentvariables into the linear regression: age, follow-up, gender (male/female),education (no education, elementary school/higher education), social support(living alone/married, living with somebody else), employment (working,house wife, volunteer / no employment), depression (CES-D), results of theVisual Analog Scale (0-10) for pain in the head, neck both sides, shouldersand arms, range of motion of the neck and shoulders, and sensibility(number of area’s, 0-6). The dependent variable were the RAND-36domains. Statistical analyses were performed with Social Package SocialScience 10.0 (SPSS, Inc. Chicago). Pearsons correlations (two-tailed) wereused to analyze correlation’s between several variables and the RAND-36domains.For the analyses of the two groups an independent sample t-test was used,and a 95 % confidence interval of the differences was calculated.

ResultsA letter informing patients about the study was send to 220 patients. Onehundred and fifty five patients (70%) were included in the study; mean age(SD) 61,3 (11,9). One patients did not understand several questions of thequestionnaire and therefor only 154 sets of questionnaires were included. Inthe analyses 104 male and 51 female patients were assessed, the meanfollow-up was 3.0 (SD: 1.7) years. The following neck dissections wereperformed; 7 radical neck dissections, 54 modified radical, 22 posterolateraland 72 supraomohyoid neck dissections. Neck dissections were performed in61 patients on the left side, 62 on the right side and 32 bilateral. One hundredand seven patients received radiation therapy of whom 90 postoperatively.

Chapter 7

97

Table 1: Pearsons correlation’s between the domains of the RAND 36 and the significant outcome of the regression analyses.

RAND 36+CES-D C

ES-D

tota

l

Shou

lder

abdu

ctio

n

Shou

lder

pai

nop

erat

ed si

de

Shou

lder

pai

nno

n-op

erat

edsi

de

Late

rofle

xion

from

ope

rate

dsi

de

Nec

k pa

inop

erat

ed si

de

Age

Follo

w-u

p

Sens

ibili

ty

-Physical functioning -.38** .55** -.39** -.17* .29** -.39** -.20* -.05 -.22**-Social functioning -.63** .09 -.18* -.05 .07 -.26** -.05 -.05 -.06-Limitations due to physical problems

-.45** .44** -.42** -.08 .28** -.39** -.02 -.01 -.23**

-Role limitations due to emotional problems

-.65** .02 -.27** -.10 -.07 -.31** .11 .06 -.01

-General mental health -.74** -.02 -.22** -.07 -.02 -.27** .01 -.01 -.09-Vitality -.75** .23** -.26** -.06 .06 -.45** .12 .02 -.06-Bodily pain -.46** .45** -.68** -.36** .29** -.52** .03 -.04 -.12-General health perception -.51** .26** -.30** .04 .10 -.29** .05 .03 -.01-Health changes -.19* .03 .01 -.03 -.06 -.13 -.20* -.19 .17*

CES-D total -.15 .31** .09 -.04 .38** -.10 .01 .02= p < 0.05, ** = p < 0.01 (Two-tailed) Chapter 7

98

The correlation’s between the RAND 36 domains and the post-operativeoutcomes are presented in Table 1. According to the calculations thedepressions scores, but also shoulder and neck pain are significantly relatedto almost all domains for quality of life. (Table 1)Sixteen percent (n=25) of the patients had a score of 16 or higher on theCES-D, which might indicate a depression. The results of the multivariatelinear regression of the domains of the RAND-36 are presented in Table 2.Depression scores contributed significantly to all domains in the physicaldomains as well as in the emotional domains. Furthermore shoulderabduction, neck pain, shoulder pain, and age are significantly related toseveral domains of quality of life. (Table 2) Patients after head and neck surgery scored similar to the control group in 5domains. Social functioning and limitations due to physical problem scoredsignificantly lower in the patient group, but the patient group scoredsignificantly better for bodily pain and health changes. (Table 3)

DiscussionScores on the CES-D are related to all domains of quality of life on the longterm after head and neck surgery. Head and neck morbidity (decreasedshoulder abduction, pain in neck and shoulder) and age are also related toseveral domains of quality of life. Patients, at least 1 year after head andneck surgery, differ little from a control group with respect of their quality oflife. They score significantly worse for social functioning and limitations dueto physical problems, but significantly better for bodily pain and healthchanges. In literature of patients following head and neck surgery severalpost-operative factors have been described that interfere with post-operativequality of life: depression12, physical function, and fatigue9 shoulderdiscomfort, neck tightness10 speech and eating.11 Most findings are similar tothe findings in our study, although we did not include fatigue, speech andeating in our regression analyses.

Chapter 7

99

Table 2 Outcome of the regression analyses of the domains of the RAND-36RAND 36 Variable ß 95 %CI interval

for ßR

squarePhysical functioning Shoulder abduction

DepressionNeck pain

Age Constant

0.3-0.7-9.3-0.370.8

(0.2 to 0.3)( -1.0 to -0.3)(-16.6 to –2.1)

(-0.6 to -0.1)(48.1 to 93.4) 0.44

Social functioning DepressionConstant

-1.995.0

(-2.3 to -1.5)(90.5 to 99.6) 0.39

Limitations due tophysical problems

DepressionShoulder abduction

Shoulder painConstant

-1.70.3

-14.938.8

( -2.4 to –1.1 )(0.2 to 0.5)

(-28.0 to –2.0)(14.7 to 62.8) 0.37

Role limitations due toemotional problems

DepressionConstant

-2.3104.1

(-2.8 to –1.9)(98.7 to 109.5) 0.41

General mental health DepressionNeck flexion to

operated sideAge

Constant

-1.5-0.4

-0.2112.9

(-1.7 to -1.3)(-0.6 to -0.2)(-0.4 to -0.0)

(99.0 to 126.9) 0.58

Vitality DepressionNeck pain

Constant

-1.7-9.383.3

(-1.9 to -1.4)(-14.5 to –4.0)(80.1 to 86.5) 0.59

Bodily pain Shoulder painDepression

Shoulder pain *Neck pain

Shoulder abductionConstant

-19.2-0.6

-12.4-8.70.1

84.8

(-25.2 to –13.3)(-0.9 to -0.3)

(-18.9 to –5.9)(-14.5 to –2.9)

(0.0 to 0.2)(74.2 to 95.4) 0.62

General healthperception

DepressionShoulder abduction

Social supportConstant

-1.40.1

-7.865.7

(-1.8 to –1.0)( 0.0 to 0.2)

(-15.4 to –0.3)(52.0 t0 81.4) 0.31

Health changes AgeDepressionSensibilityFollow-up

Constant

-0.4-0.62.6

-2.789.7

(-7.2 to -0.1)(-1.0 to -0.2)

(0.6 to 4.6)(-4.8 to -0.5)

(69.2 to 110.2) 0.15All physical outcome were on the operated side except for shoulder pain marked *Abduction is degrees (0-180), neck flexion (degrees 0-120) depression (CES-D 0-60), pain (VAS 0-10), age (years), sensibility (area’s with lost sensibility 0-6),follow-up (years), social support (martial status 0-1).

100

Of the patients 16 % showed symptoms of depression, this outcome isalmost similar to the percentages described by de Leeuw et al. (2000) whoalso used the CES-D and described a prevalence of 21 % a year aftertreatment.20 It can be discussed in what way the scores of the CES-D areaffecting head and neck cancer patients. Disturbances in eating, speaking,and fatigue are symptoms which may indicate a depression however thesedisturbances are also direct consequences of cancer treatment. For instancepatients with a tracheostoma do have more problems with speaking, andafter radiotherapy patients may have more eating problems due to reducedsaliva production. Thus patients with scores of 16 or higher on the CES-D,indicating a depression, may have a high score because of a depressed moodbut may as well have a high score because of physical sequela related to thecancer treatment. In this view it can be discussed if cut-off scores fordepression in head and neck cancer patients should be higher or thatdepressed mood items and the somatic items should be presentedseparately.21 Additionally this view may have consequences for the post-operative care. Maybe patients with mainly somatic items on the CES-Dbenefit more from a physical rehabilitation program, and patients with a highscore based on a depressed mood a psycho-social intervention seems moresuitable. Furthermore the psychometric construction of the RAND-36 and the CES-Dhave strong correlation’s, especially for mental health and vitality.14 Despitethis discussion it is clear that depression is strongly related to quality of lifein many patients and therefor physicians should pay attention to signs of adepressed mood in the post-operative care.Reduction in shoulder abduction and shoulder pain were significantly relatedto the outcome of 4 domains (physical functioning, limitations due tophysical problems, bodily pain, general health perception) of the RAND 36.Shoulder morbidity is often described, especially after neck dissection whenthe spinal accessory nerve is involved.22,23 In the standard post-operative carerange of motion of the shoulder and pain must be evaluated, and whenindicated a specific rehabilitation program may be prescribed.24

Chapter 7

101

Table 3 Comparison between patients after neck dissection (mean age 61,3) and the at random chosen control group (age 55-64). The mean differences and 95% Confidence interval.

RAND 36 Head and Neckoncology patients(n = 154)

Control group(n=140)Age 55-64

Differences between groups

Mean SD Mean SD Meandifference #

95-% ConfidenceInterval

Physical functioning 78.1 (24.4) 72.7 (24.4) -5.4 -11.0 to 0.2Social functioning 79.2 (26.8) 86.6 (21.4) 7.4* 1.8 to 13.0Limitations due to physical problems 63.0 (41.8) 76.5 (38.1) 13.5* 4.3 to 22.7Role limitations due to emotionalproblems

84.6 (32.4) 90.1 (24.5) 5.5 -1.1 to 12.1

General mental health 78.9 (18.5) 77.1 (18.7) -1.8 -6.1 to 2.5Vitality 66.5 (21.9) 67.0 (21.3) 0.5 -4.5 to 5.5Bodily pain 80.8 (23.0) 74.7 (25.0) -6.1* -11.6 to -0.6General health perception 65.1 (24.4) 64.4 (22.2) -0.7 -6.1 to 4.7Health changes 56.5 (23.9) 48.7 (15.4) -7.8* -12.5 to -3.1# mean difference, a negative result indicates a better outcome for the neck dissection group. *significant

Chapter 7

102

The significantly worse outcome for limitations due to physical problemscompared to the control group shows that physical rehabilitation may beimportant, but the reduced social functioning and importance of depressionshows that besides training of physical skills, psychosocial rehabilitationshould also be addressed.25 Notable was the significantly better outcome in the pain domain, comparedto the control group. Although about 35 % of the patients had pain related tothe surgery, pain seems to have less impact in the patient group. Maybe headand neck cancer patients accept pain, following treatment, more easily as aside effect of a life saving therapy. The better result on health changes in thehead and neck cancer group can be expected after surgery and possibleradiotherapy. Follow-up has a negative effect on health changes, becausehealth changes decrease when the follow-up gets longer. Recovery ofsensibility has a significant influence on health changes. This is oftennoticed by patients, for instance recovery of numbness in the earlobe.We performed a cross-sectional study more than one year after surgery. Aprospective study would have been preferable to identify, possible, pre-operative morbidity. The results of our study however give a good indicationin the relationships between quality of life and physical, emotional andsocial aspects after head and neck surgery. Furthermore the results showlong term consequences which may need attention in the post-operative care.We choose to analyse quality of life with the consequences of the cancertreatment. It is known that variables related to surgery (TNM status, place ofprimary tumour, type of neck dissection etc.) are of importance in theoutcome of quality of life. These variables however can not be influenced bya rehabilitation program. We choose in our view the most importantvariables that beside shoulder and neck morbidity could interfere withquality of life. But in a multidimensional construct like quality of life it canbe discussed which variables should be added to a regression analyses andwhich not.In conclusion, beside the medical domains a physician should pay attentionto non-medical factors in the long term after surgery. Shoulder and neckmorbidity are, beside a depressed mood, important outcomes in quality oflife. Outcomes on physical problems and social functioning are, compared toa control group, significantly lower for patients with head and neck tumoursafter treatment.

Chapter 7

103

References1. Aaronson NK, Acquadro C, Alonso J, Apolone G, Bucquet D, Bullinger M,

Bungay K, Fukuhara S, Gandek B, Keller S, et a. International Quality ofLife Assessment (IQOLA) Project. Qual.Life Res. 1992;1:349-351.

2. Smith KW, Avis NE, Assmann SF. Distinguishing between quality of lifeand health status in quality of life research: a meta-analysis. Qual.Life Res.1999;8:447-459.

3. Morton RP. Evolution of quality of life assessment in head and neck cancer.J.Laryngol.Otol. 1995;109:1029-1035.

4. Weymuller EA, Yueh B, Deleyiannis FW, Kuntz AL, Alsarraf R, ColtreraMD. Quality of life in patients with head and neck cancer: lessons learnedfrom 549 prospectively evaluated patients. Arch.Otolaryngol.Head NeckSurg. 2000;126:329-335.

5. Kuntz AL, Weymuller-EA J. Impact of neck dissection on quality of life.Laryngoscope 1999;109:1334-1338.

6. Terrell JE, Welsh DE, Bradford CR, Chepeha DB, Esclamado RM,Hogikyan ND, Wolf GT. Pain, quality of life, and spinal accessory nervestatus after neck dissection. Laryngoscope 2000;110:620-626.

7. Konstantinovic VS. Quality of life after surgical excision followed byradiotherapy for cancer of the tongue and floor of the mouth: evaluation of78 patients. J.Craniomaxillofac.Surg. 1999;27:192-197.

8. Pourel N, Peiffert D, Lartigau E, Desandes E, Luporsi E, Conroy T. Qualityof life in long term survivors of oropharynx carcinoma. Int.J.RadiationOncology Biol.Phys. 2002;54:742-751.

9. de Graeff A, de Leeuw J, Ros WJ, Hordijk GJ, Blijham GH, Winnubst JA.A prospective study on quality of life of patients with cancer of the oralcavity or oropharynx treated with surgery with or without radiotherapy.Oral Oncol. 1999;35:27-32.


11. Karnell LH, Funk GF, Hoffman HT. Assessing head and neck cancerpatient outcome domains. Head Neck 2000;22:6-11.

12. D'Antonio LL, Long SA, Zimmerman GJ, Peterman AH, Petti GH,Chonkich GD. Relationship between quality of life and depression inpatients with head and neck cancer. Laryngoscope 1998;108:806-811.

13. Saffold SH, Wax MK, Nguyen A, Caro JE, Andersen PE, Everts EC, CohenJI. Sensory changes associated with selective neck dissection.Arch.Otolaryngol.Head Neck Surg. 2000;126:425-428.

Chapter 7

104

14. Bouma J, Ranchor AV, Sanderman R, van Sonderen E. CES-D. Noordelijkcentrum voor gezondheidsvraagstukken, Groningen, 1995.

15. Orme JG, Reis J, Herz EJ. Factorial and discriminant validity of the Centerfor Epidemiological Studies Depression (CES-D) scale. J.Clin.Psychol.1986;42:28-33.

16. Hanewald GJ. CES-D the Dutch version: a study of the reliability andvalidity. 1987.

17. Hann D, Winter K, Jacobsen P. Measurement of depressive symptoms incancer patients: evaluation of the Center for Epidemiological StudiesDepression Scale (CES-D). J.Psychosom.Res. 1999;46:437-443.

18. van der Zee KI, Sanderman R. RAND-36. Noordelijk centrum voorgezondheidsvraagstukken, Groningen, 1993.

19. Ware-JE J, Sherbourne CD. The MOS 36-item short-form health survey(SF-36). I. Conceptual framework and item selection. Med.Care1992;30:473-483.

20. de Leeuw JR, de Graeff A, Ros WJ, Blijham GH, Hordijk GJ, WinnubstJA. Prediction of depressive symptomatology after treatment of head andneck cancer: the influence of pre-treatment physical and depressivesymptoms, coping, and social support. Head Neck 2000;22:799-807.

21. Visser MR, Smets EM. Fatigue, depression and quality of life in cancerpatients: how are they related? Support.Care Cancer 1998;6:101-108.

22. van Wilgen CP, Dijkstra PU, van der Laan BF, Plukker JT, Roodenburg JL.Shoulder complaints after neck dissection; is the spinal accessory nerveinvolved? Br.J.Oral Maxillofac.Surg. 2003;41:7-11.

23. Dijkstra PU, van Wilgen PC, Buijs RP, Brendeke W, de Goede CJ, Kerst A,Koolstra M, Marinus J, Schoppink EM, Stuiver MM, van de Velde CF,Roodenburg JL. Incidence of shoulder pain after neck dissection: a clinicalexplorative study for risk factors. Head Neck 2001;23:947-953.

24. Salerno G, Cavaliere M, Foglia A, Pellicoro DP, Mottola G, Nardone M,Galli V. The 11th nerve syndrome in functional neck dissection.Laryngoscope 2002;112:1299-1307.

25. Ronson A, Body JJ. Psychosocial rehabilitation of cancer patients aftercurative therapy. Support.Care Cancer 2002;10:281-291.

Chapter 7

105

CHAPTER 8MEASURING SOMATIC SYMPTOMS WITH THE CES-D

TO ASSESS DEPRESSION IN CANCER PATIENTS AFTER

TREATMENT, VALID OR NOT?COMPARISON BETWEEN PATIENTS WITH HEAD AND NECK,

GYNAECOLOGICAL, COLO-RECTAL, AND BREAST CANCER

C.P. van Wilgen 1,2,4, P.U. Dijkstra 1,2, R.E. Stewart 3 , A.V. Ranchor 3 , J.L.N. Roodenburg 1

1. Department of Oral and Maxillofacial Surgery 2. Department ofRehabilitation 3. Northern Centre for Health Care Research (NCG) 4. PainExpertise Centre University Hospital Groningen.

Submitted

106

SummaryThe prevalence of depression after cancer treatment is high. In literatureseveral authors have raised questions about assessing somatic symptomsafter cancer treatment to explore depression. They stated that the somaticsequela are a consequence of cancer treatment and are therefor causing thehigh depression rates in cancer patients. In this study we analysed thesomatic domain on a depression questionnaire (CES-D) in cancer patientsafter treatment, in comparison with a control group, and we comparedbetween cancer groups. Data of 566 cancer patients (head and neck,gynaecological, colo-rectal and breast cancer), and 255 randomly chosencontrol patients were analysed. The total score of the CES-D on the domainsomatic retarded activity is significantly (p < 0.01) different for cancergroups and control group. But the cancer groups score lower (colo-rectalcancer) as well as higher (head and neck, breast) than the control group onthe somatic domain. We conclude that cancer patients are not a homogenousgroup of patients concerning somatic sequela and therefor we can not findevidence te remove somatic items from depression questionnaires forpatients after cancer treatment.

Chapter 8

107

IntroductionSurvival rates of several cancer types have become higher because of earlierdetection and better treatment properties (Brenner, 2002). As a consequencemore patients have to cope with the physical and emotional consequences ofthe diagnosis cancer and the side effects of the treatment. In the last decademore attention has been given to this group of patients. Important topics inpost-treatment cancer research are quality of life (Aaronson et al., 1992),coping (Hassanein et al., 2001; Petticrew et al., 2002) depression (Hjerl etal., 2003; Krishnan et al., 2002) somatic morbidity (Saffold et al., 2000; vanWilgen et al., 2003), pain (Caraceni and Portenoy, 1999; Portenoy, 1992;Zaza and Baine, 2002) and fatigue (de Jong et al., 2002; Morrow et al.,2002). This research lead to a better understanding of the physical andemotional problems after cancer treatment, and as a consequence treatmentprograms have been developed, such as: rehabilitation programs (DeLisa,2001), psycho-social interventions (Owen et al., 2001; Ronson and Body,2002), pain management, (Portenoy and Lesage, 1999) and multidisciplinaryprograms (Van Weert, 2004). An important outcome after cancer treatment is depression. The prevalenceof depression after cancer treatment is about 24 % (range 1.5 % to 50%) (McDaniel et al., 1995). Depression affects quality of life, survival, length ofhospital stay, and therapy compliance (Bottomley, 1998, Hjerl et al., 2003;Mc Daniel et al., 1995). It is therefor of clinical importance to recognisedepression in the post-treatment phase. According to the DSM IV a depressive episode is assumed if five (or more)out of nine symptoms (Table 1) are present, additionally the symptomsshould be present during the same 2-week period and represent a changefrom previous functioning (American Psychiatric Association, 1994).Symptoms of a clinical depression in patients after cancer treatment can beassessed through a clinical interview, a (semi)standardised interview or aquestionnaire. Questionnaires are used most frequently in cancer research. The prevalence of depression in cancer patients is related to type of cancer,follow-up, medical illness, gender, and method of assessment (Beeber et al.,1998; Bottom ley, 1998; Mc Daniel et al., 1995; Stommel et al., 1993).Several authors have suggested that the high prevalence of depression aftercancer treatment is due to the fact that the somatic symptoms of depressionare identical with the somatic symptoms caused by the cancer treatment

Chapter 8

108

(Beeb er et al., 1998; Dugan et al., 1998; Krishnan et al., 2002; Visser andSmets, 1998). They state that somatic symptoms should not be measuredwhen assessing depression in cancer patients. Some authors even removedsomatic items from depression questionnaires when measuring cancerpatients (Dugan et al., 1998; Visser and Smets, 1998). In the HospitalAnxiety and Depression Scale somatic symptoms are not added to thequestionnaire (Zigmund and Snaith, 1983). Hard evidence for the hypothesisthat somatic symptoms should be removed in a depression questionnairewhen assessing cancer patients is virtually missing, although some studieshave been performed to explore this hypothesis. In one of the studies, toexplore the role of somatic items in cancer patients, the Zung self ratingscale was divided into a questionnaire with and without somatic items. Theoutcome on the questionnaire with somatic items was about 5 % higher. Theauthors stated that the Zung self rating scale has 5% more false-positivedepressed cancer patients when somatic items are assessed (Dugan et al.,1998). After a factor analyses of the Zung Self Depression scale, Passik et al.stated that fatigue is the only somatic item that is typically considered to be asymptom of depression in cancer patients (Passik et al., 2000). While norelationship between depression and fatigue was found by Visser et al. Theystated that fatigue is no valid criterion for depression if patients had receivedradiation therapy (Visser and Smets, 1998). After analysing a structuredinterview with cancer patients Akechi et al. described that eating disordersand concentration problems are strongly related to depression while sleepdisorders and fatigue are not related to the depression but to the somaticsequela of cancer treatment (Akechi et al., 2003). Several eating related side effects are described before (pain) and duringradiotherapy (xerostomia, dysphagia, pain) in head and neck cancer patients,which may affect outcome on a depression questionnaire (Epstein et al.,2001; Sehlen et al., 2003). From these performed studies no clear statement about the somatic items, inthe assessment of depression in cancer patient, can be made. The statementsabout fatigue are even conflicting.Aim of this study was to analyse the influence of somatic sequela on adepression questionnaire (CES-D) in cancer patients after treatment incomparison with a control group, and to analyse the different cancer types.

Chapter 8

109

Figure 1 DSM IV criteria for depression:

(1) depressed mood most of the day, nearly every day, as indicated byeither subjective report (e.g., feels sad or empty) or observation madeby others.

(2) markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day (3) significant weight loss when not dieting or weight gain (e.g., a change

of more than 5% of body weight in a month), or decrease or increase inappetite nearly every day.

(4) insomnia or hypersomnia nearly every day(5) psychomotor agitation or retardation nearly every day (observable by

others, not merely subjective feelings of restlessness or being sloweddown)

(6) fatigue or loss of energy nearly every day(7) feelings of worthlessness or excessive or inappropriate guilt (8) diminished ability to think or concentrate, or indecisiveness, nearly

every day (9) recurrent thoughts of death (not just fear of dying), recurrent suicidal

ideation without a specific plan, or a suicide attempt or a specific planfor committing suicide

Furthermore:B. The symptoms do not meet criteria for a mixed episode.C. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.D. The symptoms are not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication) or a general medical condition (e.g., hypothyroidism).E. The symptoms are not better accounted for by bereavement, i.e., after the loss of a loved one, the symptoms persist for longer than 2 months or are characterised by marked functional impairment, morbid preoccupation with worthlessness, suicidal ideation, psychotic symptoms, or psychomotor retardation.

Chapter 8

110

Methods and material The CES-D is a short self-report scale designed to measure depressivesymptomatology (Radloff, 1977). The questionnaire is translated to Dutchand contains 20 items divided in four domains: somatic retarded activity (7items), depressed affect (5 items), positive affect (4 items), and interpersonalaffect (2 items) and two single items that complete the total score (Bouma etal., 1995; Hanewald, 1987). The total score ranges between 0 – 60, a score of16 or higher indicates a depressed symptomatology. The CES-D is oftenassessed in patient with cancer. The psychometric properties of the CES-Din cancer patients are described by several authors. Hann et al. (1999) foundan internal consistence of alpha .89 and the test-retest reliability was .51 (p <0.001). The outcome in the validity analyses were all satisfactory. Concludedwas that the CES-D is appropriate for the use on clinical psycho-socialresearch (Hann et al., 1999). Although the relation with somatic symptomsin cancer patients was, on our knowledge, never investigated with the CES-D. In our study the CES-D was administered in patients at least a year afterthe first cancer treatment and in a control group. Patients with recurrence ofthe tumour were excluded. The control patients and patients after breast,colo-rectal and gynaecological cancer were obtained from the database of theNorthern Centre for Health Care Research (NCG). The control group wasmatched for gender and age with the cancer group and lived in the same areaas the patients with cancer. Patients with head and neck cancer were assessedin the University Hospital Groningen on the Department of oral andmaxillofacial surgery, the Department of otorhinolaryngology head & necksurgery or on the Department of surgical oncology. The CES-D was assessedduring a regular appointment on the out patients clinic. Patients received aletter, a week before their appointment in which the study was explained.The medical doctor asked the patients to participate. If patients were willingto participate an informed consent was signed, and the CES-D was filled out.For the statistical analyses SPSS 10.0 was used. Of the CES-D the totalscore, the somatic retarded activity score and the depressed affect score wereanalysed. Because the other domains, positive affect and interpersonal affect,were not relevant for this study these were not analysed.

Chapter 8

111

An ANOVA multi-comparison between groups was performed for thecontrol and cancer group on the domain somatic retarded activity. AnA NOVA multi-comparison between groups with an Bonferroni post hocanalyses was performed for the control group and four cancer groups. AKruskal-Wallis tests and median tests was performed. These ordinal testswere performed as control for the outcome on the ANOVA multicomparison tests because the outcomes on the CES-D are not normallydivided data. Also regression analyses were performed of the CES-D totalscore, the somatic retarded activity score and the depressed affect score.

Table 1 Descriptive data of the control group and the cancer groups, and the

number of patients at risk for depression according to the CES-D.

Control Breastcancer

Colo-rectalcancer

Gynae-cological

cancer

Head andneck cancer

Number ofsubjects

255 206 136 69 155

Female 68% (173) 99% (205) 44 % (60) 100 %(69) 33 %(51)Age mean (SD) 58 (15) 55 (13) 66 (12) 53 (16) 61(12) Surgery - 99% (203) 100% (136) 96 % (66) 100% (155) Radiation therapy - 55% (114) 10% (13) 45 % (31) 69% (107)Chemotherapy - 46% (94) 10% (13) 20 % (14) 0% (0)*Patients atrisk fordepression

11% (28) 21% (43) 5% (7) 20 % (14) 16% (25)* Patients with a score of 16 or higher on the CES-D are at risk for depression.

Chapter 8

112

ResultsData of 566 patients and of 255 randomly chosen control patients wereanalysed. The descriptive data are presented in Table 1. The breast andgynaecological patients were mainly females. The patients at risk for apossible depression according to the CES-D (score ≥16) are presented inTable 1.In table 2 the results of the ANOVA multi-comparison between groups arepresented. The cancer group differs significantly from the control group onthe domain somatic retarded activity, and depressed affect. The four cancergroups and control group were also significantly different on total score,somatic retarded activity, and depressed affect score (Table 2). The ANOVAwith a post-hoc bonferroni analyses (Table 3) shows the differences betweenthe control group and four cancer groups on the domain somatic retardedactivity. The colo-rectal patients score significantly lower compared to thebreast and head and neck patients. Because the outcomes on the CES-D arenot normally divided, we performed a Kruskal-Wallis test to analyse thedifferences between the cancer groups and the control group. In the rankingof the Kruskal Wallis the colo-rectal patients scored the lowest on alloutcome. Head and neck patients and breast patients scored highest forsomatic retarded activity. The total score, the somatic retarded activity scoreand depressed affect score are significantly (p < 0.01) different for cancergroups and control group. This significant difference was also found after amedian test. The outcome on the Kruskal Wallis test and the median testwere similar to the outcome on the ANOVA multi comparison .

Chapter 8

113

Table2 Mean scores on the CES-D total, the domains somatic retarded activity (7 items), and depressed affect (5 items). Presented are the ANOVA multi comparison between groups (control – cancer) and an ANOVA multi comparison between groups for the control and four cancer types.

CES-D Total

F p SomaticRetarded

F p Depressedaffect

F p

Control group 8.3 (6.4) 2.3 (2.6) 1.0 (1.7)Cancer group total 9.0 (8.5) 1.3 .26 3.0 (3.4) 7.5 .006 1.6 (2.5) 12.1 .001

Control group 8.3 (6.4) 2.3 (2.6) 1.0 (1.7) Breast 10.5 (8.3) 3.3 (3.2) 1.9 (2.6) Colo-rectal 7.0 (6.7) 1.7 (2.4) 0.9 (1.8) Gynaecological 9.4 (9.6) 2.6 (3.5) 1.9 (3.0) Head and Neck 8.3 (8.9) 4.2 .002 3.8 (4.0) 8.7 .000 1.5 (2.4) 7.1 .000

Table 3 An ANOVA multi-comparison with post-hoc analyses with a Bonferroni correction was performed for the four cancer types and the control group for the domain somatic retarded activity. In the cells the p values for the differences are shown.

Control Breast Colo-rectal GynaecologicalBreast .02 < 0.01 1.0Colo-rectal 1.0 < 0.01 .88Gynaecological 1.0 1.0 .88Head and Neck < 0.01 1.0 < 0.01 .20

114

In the regression analyses we estimated the mean CES-D total score, somaticretarded activity and depressed affect as dependent of the variables cancertypes, age and gender (Table 4). Somatic retarded activity is related to breast cancer, and head and neckcancer as well as gender. Gender was related to all outcome. Colo-rectalcancer was excluded in all analyses. Table 4 Regression analyses for CES-D total, somatic

retarded activity and depressed affect. Significant data (p < 0.05) were entered.

Breast GenderConstant

CES-D tot1.6-2.28.3

BreastHead and neckGender Constant

Somatic retarded0.7 1.8-0.93.1

BreastGynaecologicalHead and neckGenderConstant

Depressed affect0.70.70.9-0.91.6

DiscussionCancer patients are not a homogeneous group of patients with respect to theoutcome of the CES-D and the domains somatic retarded activity anddepressed affect. In comparison with a control group cancer patients scorelower (colo-rectal) as well as higher (breast, and head & neck) on thedomain somatic retarded activity. Therefor removing somatic items in adepression questionnaire to assess cancer patients seems not valid. The

Chapter 8

115

influence of gender is also in cancer patients important and strongly relatedto cancer type (Weissman et al., 1993). We can assume that regarding to somatic sequela “the cancer patient” doesnot exist. Cancer patients have similarities, like everyone has to cope with alife threatening disease, interruption of normal daily life, and being exposedto treatments, but there are also differences like type of treatment, place ofsurgery and radiation therapy, the severity of the cancer, the extensiveness ofthe treatment and the extent of sacrificed structures. As a consequence thesomatic morbidity after cancer treatments is different. Especially the place ofthe tumour seems important. For instance the surgical removal of a tumourof the tong including removing of salivary glands, a neck dissection andradiation therapy of the neck, has more somatic consequences than removingthe uterus or a part of colon without radiation therapy. These differences intreatment have consequences on somatic morbidity and probably also on theresults on the somatic items assessed by a depression questionnaire.

In the analyses between the complete cancer group and the control group thehypothesis that somatic morbidity is affecting the prevalence of depressionafter cancer treatment seems true. In the multi comparison although onlyhead and neck cancer and breast cancer score high on the somatic domain.Somatic morbidity might increase the prevalence of depression in thesegroups. This could implicate that in these groups the somatic domain orspecific somatic items should be removed. This opportunity seems not valid,because we don’t know which specific items should be removed andremoving all somatic items seems conflicting with the construction ofdepression. Breast and gynaecological patients are mainly women, the female gender hasan important impact on the CES-D total scores. The percentages of patientsat risk for depression are the highest in these two patient groups. Also thescores on depressed affect are the highest in these two groups. Head andneck patients score significantly higher on the somatic retarded activitydomain. Patients with head and neck cancer often have to undergo extensivesurgery. This might contain sacrificing part of the glottis, removing salivaryglands, removing the upper trachea (getting a tracheostoma), or removing apart of the mandibula possible restored with part of the fibula bone. Theseextensive operation seem to clarify the high somatic morbidity rates.

Chapter 8

116

The use of questionnaires to assess depression in the medical setting seemsvaluable because of the high prevalence of depression. Physicians are notwell trained in recognising depressions (Passik et al., 2000), therefor aquestionnaire can be a helpful tool to explore a possible depression duringthe medical follow up. If patients score high on a CES-D a referral to apsychologists, social worker or a peer group must be considered to thepatient. If a score is around 16 the physician should further explore thepatient by asking specific questions that focus on a depressed mood.

References1. Aaronson NK, Acquadro C, Alonso J, Apolone G, Bucquet D, Bullinger M,

Bungay K, Fukuhara S, Gandek B, Keller S. 1992. International Quality ofLife Assessment (IQOLA) Project. Qual. Life Res. 1: 349-51.

2. Akechi T, Nakano T, Akizuki N, Okamura M, Sakuma K, Nakanishi T,Yoshikawa E, Uchitomi Y. 2003. Somatic symptoms for diagnosing majordepression in cancer patients. Psychosomatics 44: 244-8.

3. American Psychiatric Association. 1994. Diagnostic and statistical manualof mental disorders. American Psychiatric Association: Washington DC,320-27.

4. Beeber LS, Shea J, McCorkle R. 1998. The Center for EpidemiologicStudies Depression Scale as a measure of depressive symptoms in newlydiagnosed patients. J Psychosoc. Oncol. 16: 1-20.

5. Bottomley A. 1998. Depression in cancer patients: a literature review. Eur.J. Cancer care. 7: 181-91.

6. Bouma J, Ranchor AV, Sanderman R, van Sonderen E. CES-D. 1995.Groningen, Noordelijk centrum voor gezondheidsvraagstukken,Rijksuniversiteit Groningen.

7. Brenner H. 2002. Long-term survival rates of cancer patients achieved bythe end of the 20th century: a period analysis. Lancet 360: 1131-5.

8. Caraceni A, Portenoy RK. 1999. An international survey of cancer paincharacteristics and syndromes. IASP Task Force on Cancer Pain.International Association for the Study of Pain. Pain 82: 263-74.

9. de Jong N, Courtens AM, Abu-Saad HH, Schouten HC. 2002. Fatigue inpatients with breast cancer receiving adjuvant chemotherapy: a review ofthe literature. Cancer Nurs. 25: 283-97.

10. DeLisa JA. 2001. A history of cancer rehabilitation. Cancer 92: 970-4.

Chapter 8

117

11. Dugan W, McDonald MV, Passik SD, Rosenfeld BD, Theobald D,Edgerton S. 1998. Use of the Zung Self-Rating Depression Scale in cancerpatients: feasibility as a screening tool. Psycho-oncol. 7: 483-93.

12. Weert van E, Hoeksta-Weebers JEHM, Grol BMF, Otter R, Arendzen JH,Postema K, Schans van der CP. Physical functioning and quality of lifeafter cancer rehabilitation 2004. Int J Rehabil Res.27: 27-35.

13. Epstein JB, Robertson M, Emerton S, Phillips N, Stevenson-Moore P.2001. Quality of life and oral function in patients treated with radiationtherapy for head and neck cancer. Head Neck 23: 389-98.

14. Hanewald GJ. 1987. CES-D the dutch version: a study of the reliability andvalidity. Amsterdam: University press.

15. Hann D, Winter K, Jacobsen P. 1999. Measurement of depressivesymptoms in cancer patients: evaluation of the Center for EpidemiologicalStudies Depression Scale (CES-D). J. Psychosom. Res. 46: 437-43.

16. Hassanein KA, Musgrove BT, Bradbury E. 2001. Functional status ofpatients with oral cancer and its relation to style of coping, social supportand psychological status. Br. J. Oral Maxillofac. Surg. 39: 340-5.

17. Hjerl K, Andersen EW, Keiding N, Mouridsen HT, Mortensen PB,Jorgensen T. 2003. Depression as a prognostic factor for breast cancermortality. Psychosomatics 44: 24-30.

18. Krishnan KR, Delong M, Kraemer H, Carney R, Spiegel D, Gordon C,McDonald WI, Dew MA, Alexopoulos G, Buckwalter K, Cohen PD, EvansD, Kaufmann PG, Olin J, Otey E, Wainscott C. 2002. Comorbidity ofdepression with other medical diseases in the elderly. Biol Psychiatry 52:559-88.

19. McDaniel JS, Musselman DL, Porter MR, Reed DA, Nemeroff CB. 1995.Depression in patients with cancer. Diagnosis, biology, and treatment.Arch. Gen. Psychiatry 52: 89-99.

20. Morrow GR, Andrews PL, Hickok JT, Roscoe JA, Matteson S. 2002.Fatigue associated with cancer and its treatment. Support Care Cancer 10:389-98.

21. Owen JE, Klapow JC, Hicken B, Tucker DC. 2001. Psychosocialinterventions for cancer: review and analysis using a three-tiered outcomesmodel. Psycho-oncol 10: 218-30.

22. Passik SD, Donaghy KB, Theobald DE, Lundberg JC, Holtsclaw E, DuganWM. 2000. Oncology staff recognition of depressive symptoms onvideotaped interviews of depressed cancer patients: implications for designing a training program. J. Pain Symptom Manage. 19: 329-38.

Chapter 8

118

23. Passik SD, Lundberg JC, Rosenfeld B, Kirsh KL, Donaghy K, Theobald D,Lundberg E, Dugan W. 2000. Factor analysis of the Zung Self-RatingDepression Scale in a large ambulatory oncology sample. Psychosomatics41: 121-7.

24. Petticrew M, Bell R, Hunter D. 2002. Influence of psychological coping onsurvival and recurrence in people with cancer; systematic review. BMJ 325:1066-76

25. Portenoy RK. 1992. Cancer pain: pathophysiology and syndromes. Lancet339: 1026-31

26. Portenoy RK, Lesage P. 1999. Management of cancer pain. Lancet 353:1695-1700.

27. Radloff LS. 1977. The CES-D Scale: A self-report depression scale forresearch in the general population. Appl Psychol. Measurem 1: 385-401.

28. Ronson A, Body JJ. 2002. Psychosocial rehabilitation of cancer patientsafter curative therapy. Support Care Cancer 10: 281-91.

29. Saffold SH, Wax MK, Nguyen A, Caro JE, Andersen PE, Everts EC,Cohen JI. 2000. Sensory changes associated with selective neck dissection.Arch. Otolaryngol Head Neck Surg 126: 425-8.

30. Sehlen S, Lenk M, Herschbach P, Aydemir U, Dellian M, Schymura B,Hollenhorst H, Duhmke E. 2003. Depressive symptoms during and afterradiotherapy for head and neck cancer. Head Neck 25: 1004-18.

31. Stommel M, Given BA, Given CW, Kalaian HA, Schulz R, McCorkle R.1993. Gender bias in the measurement properties of the Center forEpidemiologic Studies Depression Scale (CES-D). Psychiatry Res 49: 239-50.

32. van Wilgen CP, Dijkstra PU, van der Laan BF, Plukker JT, Roodenburg JL.2003. Shoulder complaints after neck dissection; is the spinal accessorynerve involved? Br J Oral Maxillofac Surg 41: 7-11.

33. Visser MR, Smets EM. 1998. Fatigue, depression and quality of life incancer patients: how are they related? Support Care Cancer 6: 101-8.

34. Weissman MM, Bland R, Joyce PR, Newman S, Wells JE, Wittchen HU.1993. Sex differences in rates of depression: cross-national perspectives. JAffect Disord 29: 77-84.

35. Zaza C, Baine N. 2002. Cancer pain and psychosocial factors. A criticalreview of the literature. J Pain Symptom Manage 24: 526-42.

36. Zigmund AS, Snaith RP. 1983. The hospital anxiety and depression scale.Acta Psychiatr Scand 67: 361-70.

Chapter 8

119

CHAPTER 9GENERAL DISCUSSION

120

General DiscussionHead and neck cancer treatment leads to considerable morbidity; physically,emotionally, and socially. After interviewing and physically examining over150 patients we became more aware of the fact that all patients are havingtheir own specific problems and their own specific way of coping with theirproblems. We examined patients with severe changes in the externalappearance, spinal accessory nerve dysfunction, and severe shoulderdisfigurement without complaints, while other patients with no visibleexternal changes and without physical loss, were not able to get on with theirlives because of pain or a severe depression.

In this thesis the incidences of shoulder complaints after radical, modifiedand selective neck dissections that occur after head and neck cancertreatment are described. In the pilot study the conclusion was that aftersupraomohyoid neck dissection still 28% of 52 patients experienced shouldercomplaints, but activities of daily life were hardly restricted. In this study wegathered data through a self developed questionnaire, although the questionswere used in other questionnaires, our questionnaire was not tested for itsreliability. From this data we could not analyse what kind of pain patientsexperienced neither could we conclude if patients had a “shouldersyndrome” as a consequence of spinal accessory nerve dysfunction or thatother pain mechanisms were underlying. The results of this pilot studyprompted us into further research.In the clinical multicentre study the complaints of 177 patients were assessedduring hospital stay. Non-selective neck dissections were a risk factor forpost-operative shoulder pain and a reduced range of motion of the shoulder.Patients were on average 13 days after surgery. In this period it is difficult todistinguish scar pain after surgery from shoulder pain as a consequence ofspinal accessory nerve dysfunction. Effects of radiation therapy were notincluded in that study. Therefor the effects found in that study can not begeneralised to the total patient population. It would have been interesting toexecute a follow up study of this specific group to measure the morbidity ayear after operation, and to analyse the risk factors again. Currently we areperforming a prospective multicentre study to analyse head and neck cancerpatients before and after surgery and after 3 months follow up.

Chapter 9

121

In a cross-sectional study design 155 patients were included and their post-operative complaints were analysed. A prospective study would have given abetter estimation of the prevalence of complaints after head and neck cancertreatment, because patients at risk could have been identified more easily.However a better understanding in several aspects of post-operativemorbidity was obtained. The role of the spinal accessory nerve in shouldercomplaints after treatment was investigated. The spinal accessory nervedysfunction was assumed to be present if two of three physical signs werepresent: trapezius pars descendens muscle atrophy, shoulder drop and achanged scapula position. These signs were chosen because they can easilybe assessed during a physical examination and are strongly related totrapezius dysfunction. In about 50% of the patients after neck dissection thespinal accessory nerve was involved in shoulder complaints. This implicatesthat shoulder complaints after neck dissection can no longer solely beattributed to spinal accessory nerve dysfunction. Another important outcomeof that study was that of the patients after supraomohyoid neck dissectionthat perceived shoulder complaints, a spinal accessory nerve dysfunction ispresent in only 6%.Also the neck itself was examined in our study. The neck is directly affectedby neck dissection and often exposed to radiation therapy. It was concludedthat several patients suffer from neck pain as well as loss of sensation. Neckpain was often accompanied by shoulder pain. Notable was that neck painwas strongly associated with neuropathic pain (hyperpathia and allodyia)while shoulder pain was more often associated with myofascial pain. Thesefindings have consequences for the choice of treatment. The treatment forneuropathic pain is medication, while for myofascial pain physical therapyor a multidisciplinary program seems to be the treatment of choice. Radiation therapy has influence on sensation and range of motion, and seemsto have influence on pain. In this study all patients underwent surgery, thusradiation therapy could not be evaluated solely. Further radiation therapywas often performed on patients with extensive operations (radical ormodified radical neck dissection) which makes it impossible to analyse theinfluence of the treatment modalities separately. The decrease in range ofmotion is related to fibrosis. In a study in which only radiation therapy isprovided as cancer treatment the influence of radiation therapy on range ofmotion can be analysed further.

Chapter 9

122

In the quality of life study a remarkable outcome was the relatively highscores (meaning a good quality of life) of patients on the RAND 36,compared to the control group. Patients scored even better on perceived pain.It would be interesting to further investigate this outcome. Maybe patientswho survive cancer may also have positive feeling after their treatmentperiod after surviving the disease. As a consequence they might live moreintense, which improves their quality of life. The most important outcome in the quality of life study was depression.Depression has a multidimensional construct with physical and emotionalitems and it is of importance to assess depression during and after cancertreatment. But also physical complaints have an important influence inseveral domains in quality of life, such as limitations due to physicalproblems, and bodily pain. In this quality of life study many other importantconsequences were not assessed like fatigue, physical condition, fear ofrecurrence, digestive problems, sleep problems, reintegration in work, sexualproblems etc. All these post-treatment problems may have impact on patientsquality of life. In a new study much more aspects of morbidity after cancertreatment must be investigated.In our last study we focussed on depression partly because of the results ofour quality of life study. The somatic morbidity after cancer treatmentdepends on type of cancer and type of treatment. Patients after head andneck cancer treatment score high in the somatic domain, when assessingdepression with the CES-D the influence of somatic morbidity must be takeninto account.

Clinical implicationsThrough this study a better understanding in the diversity of complaints inpatients after head and neck treatment was achieved. Knowledge aboutproblems on which a surgeon, physical therapist, nurse or psychologist canfocus during a medical check up or treatment. This diversity in patients andproblems makes a good communication between physician and patientmandatory to identify these specific problems. How can patients, after head and neck cancer therapy, benefit from theresults of this thesis. Together with the Dutch Physical Therapy Neck Dissection Study Group(Nederlandse Fysiotherapie Halsklierdissectie Studie Groep, NFHSG),

Chapter 9

123

physical therapy guidelines for patients after neck dissection were developed(appendix 1). The guidelines include diagnostic criteria, information forcolleagues, an exercise program for shoulder problems after neck dissection,treatment advises and a research protocol to assess patients prospectively.Currently we are implementing this protocol, together with the study group,in a new prospective multi-centre trail to further analyse patients complaintsbefore and after cancer treatment. We sincerely hope that patients, andphysical therapists, will benefit from the guidelines. Some research into the effects of physical therapy has been done (appendix2). Several authors report good results of physical therapy programs but sofar only one controlled trial has been executed and has shown somebeneficial effects of physical therapy. Physical therapy programs should, aspart of a multidisciplinary approach, be further tested on its effectiveness.In the follow up after surgical treatment, attention must be given to apossible accessory nerve dysfunction. Currently patients are often asked toshrug their shoulders to assess trapezius muscle dysfunction. From our studyit became clear that for a quick assessment of trapezius muscle function thepatient should be asked to abduct both arms on the same time, and look fordifferences in range of motion. When differences are found or when patientsreport shoulder pain an inspection of the shoulder region is of importance toassess the trapezius muscle and to investigate the type of pain.Furthermore psychological problems, especially depression, are ofimportance to assess in the post-clinical phase because of its influence onquality of life. Integrating a questionnaire or standard questions to assessdepression into the follow-up in the medical care after cancer treatment,must be considered. Pain is one of the most important forms of morbidity. We conclude thatseveral types of pain are present in patients after head and neck cancertherapy. These types of pain may change during the different stages aftersurgery or radiation therapy. A distinction should be made in nociceptive,neuropathic, and myofascial pain. However in all types of pain the influenceof psychological or social problems must be considered. Neuropathic painseems to be related to neck pain while myofascial pain seems to be morepresent in shoulder muscles. Unknown is the pain related to the resection ofthe primary tumour.

Chapter 9

124

In our post-clinical study only 17 (11%) of the patients used pain medicationof which 15 used paracetamol or NSAID’s. Specific medication forneuropathic pain was hardly used. Forty-one (26%) patients were treated bya physical therapist after the clinical period, but the indication for physicaltherapy was unknown. According to the amount of patients with pain in thisstudy, pain management needs more attention after cancer treatment.

The radical neck dissection and it’s modifications have lead to improvementof survival of head and neck cancer patients. The morbidity of thesetreatments is complex and fragmentarily studied. Therefor there is noevidence based treatment strategy. Now we have got more insight in thiscomplex morbidity, treatment protocols must be directed to the individualproblems of patients and evaluated in prospective studies.

Chapter 9

125

CHAPTER 10SUMMARY

SAMENVATTING

DANKWOORD

APPENDIX

LIST OF PUBLICATIONS

126

SummaryIn the introduction (chapter 1) a brief history of the neck dissection, in headand neck oncology treatment, is described based on the manuscript of Crilein 1906. About 50 years later, in 1952, Ewing described the postoperativemorbidity after radical neck dissection: shoulder complaints, pain,disfigurement, loss of sensation, functional loss, restricted daily activities,and loss of strength. The reported prevalence of shoulder complaints afterradical neck dissections are high (47% to 100%). Because of these post-operative morbidity and the gained insight in the biological behaviour ofhead and neck tumours modified and selective neck dissections weredeveloped. The most important modification was preservation of the spinalaccessory nerve. Prevalence rates of shoulder complaints after modifiedradical neck dissections (18 % to 61 %) and after selective neck dissections(29 % to 39 %) seem to decrease. The occurrence of shoulder complaints is strongly related to spinal accessorydysfunction. The spinal accessory nerve normally innervates the trapeziusmuscle, if this connection is disintegrated atrophy of the trapezius musclewill be the consequence. As a result of this the scapula, which is normallystabilised on the thorax by the trapezius muscle, will glide laterally and theglenoid fossa will point more caudally. A restricted range of motion of theshoulder and loss of function will be the consequence. Pain may occur as aconsequence of this, although it is still scarcely described what causes pain. The occurrence of shoulder complaints after modified or selectiveprocedures with preservation of the spinal accessory nerve seems less wellunderstood.Aim of this thesis was to obtain a better understanding about physicalcomplaints (shoulder complaints, pain, range of motion, loss of sensation)function (activities of daily life, shoulder function), psychological factors(depression), and quality of life after head and neck cancer treatment.

In chapter 2 the pilot study is described. The purpose of that pilot study wasto asses shoulder morbidity; pain and disability in daily activities one yearafter unilateral or bilateral supraomohyoid neck dissection. In total 52patients were included. These patients filled out a questionnaire assessingpain and daily activities. Of these patients 14 (28%) complained ofipsilateral shoulder pain following supraomohyoid neck dissection. The

Chapter 10

127

disability, because of shoulder complaints, perceived during daily life wasminor. In only two patients the pain and disability led to dependency ofothers for heavy household activities. It was concluded that despite the factthat the supraomohyoid neck dissection was also developed to reduceshoulder morbidity, 28% of the patients experience some type of shoulderpain following supraomohyoid neck dissection. The degree of disability dueto shoulder complaints was minor.

In chapter 3 the incidence of shoulder pain and reduced range of motion ofthe shoulder after neck dissection in the clinical period was studied.Additionally risk factors for the development of shoulder pain and arestricted shoulder range of motion were identified. Clinical patients whounderwent a neck dissection completed a questionnaire assessing shoulderpain the day before discharge from the hospital. Range of motion of theshoulder was measured, abduction and forward flexion. Information aboutsurgery and type of neck dissection was retrieved from the medical records.Of the 177 patients included 70% experienced shoulder pain, which mainlyoccurred during moving the shoulder (31%), and lying on the affectedshoulder (30%). Forward flexion and abduction of the operated side wasseverely reduced compared to the non-operated side. Non-selective neckdissection was a risk factor for the development of shoulder pain and arestricted shoulder abduction. Reconstruction after tumour resection was arisk factor for a restricted forward flexion of the shoulder.

In chapter 4 the prevalence of shoulder complaints after nerve sparing neckdissection with or without radiation therapy at least one year after surgery isdescribed. Patients were interviewed for shoulder complaints, and filled outthe shoulder disability questionnaire to evaluate shoulder disability in dailyactivities. In total 137 patients; 51 after modified radical neck dissection(MRND), 21 after postero-lateral neck dissection (PLND), and 65 aftersupraomohyoid neck dissection (SOHND) were analysed. After MRND33.3% of the patients experienced shoulder complaints, after PLND 66.7 %,and after S OHND 20 % of the patients experienced shoulder complaints.Type of neck dissection was significantly (p < 0.001) related to shouldercomplaints. Radiation therapy was not significantly related to shouldercomplaints and disability. Outcome on the shoulder disability questionnaire

Chapter 10

128

also showed a significa nt (p< 0.01) difference in outcome for type of neckdissection. Age and follow up were not of significant influence on shouldercomplaints.It was concluded that a SOHND prevents for shoulder complaints and thatdisability in daily activities are the lowest after SOHND.

In chapter 5 the relationship between shoulder morbidity (pain and range ofmotion), and the spinal accessory nerve function after neck dissection wasanalysed. Spinal accessory nerve dysfunction was made operational by threeclinical signs: atrophy of the trapezius muscle, shoulder drop and anobjectified changed scapula posture. If two out of three signs were present aspinal accessory nerve dysfunction was assumed. Shoulder pain wasmeasured with a numbered VAS over the last week. Shoulder abduction wasassessed, and a difference of ≥ 40° compared to the non-operated side was assumed to be clinicalrelevant. In total 112 patients after neck dissection (73 male/ 39 female) wereincluded. Of this group 39 patients complained of shoulder pain. Of thepatients with shoulder complaints 20 (51%) had dysfunction of the spinalaccessory nerve. Of the total 29 group of patients (26%) had dysfunction ofthe spinal accessory nerve of whom 20 (69%) had shoulder pain. Shoulderpain and a difference in active shoulder abduction of ≥ 40° are significantlyrelated to dysfunction of the spinal accessory nerve (p <0.001). Shoulderpain after neck dissection can only be attributed in 50% of the patients todysfunction of the spinal accessory nerve.

In chapter 6 the consequences of head and neck cancer therapy on the neckitself are described. Patients who underwent surgery, including neckdissection, with or without radiation therapy, at least one year before thestudy, were asked to participate. Neck pain, loss of sensation, range ofmotion of the cervical spine, and shoulder pain was assessed. Of the 220 patients who were invited 153 (70 %) participated in the study.Neck pain was present in 33% (n=51) of the patients, and shoulder pain in37% (n=57). Neck pain and shoulder pain were significantly related (p<0.001). Of the patients who experienced neck pain 20 (39 %) had allodyniain the neck, and 49 (96%) experienced hyperpathia, the mean VAS intensity

Chapter 10

129

of neck pain was 3.5 (sd 2.3). Of the study group 32% (n=49) experiencedneuropathic pain, 46% (n=70) experienced myofascial pain, and 24% (n=37)experienced joint pain. Loss of sensation of the neck was present in 65%(n=99) and was related to type of neck dissection and radiation therapy.Range of motion of the neck was significantly decreased, in lateral flexionaway from the operated side, due to the neck dissection and or radiationtherapy. Morbidity of the neck after cancer therapy occurred frequently.

In chapter 7 the impact of neck and shoulder complaints on quality of lifeare described. Quality of life has become a major outcome in determiningtreatment effects in head and neck surgery with curative intent, as well asmany other medical domains. Aim of this study was to determine whichfactors were related to quality of life, and how these outcomes where relatedto a at randomly chosen control group. As outcome measurement theRAND-36 (SF-36) was assessed. The following items were entered asindependent variables into the linear regression: age, follow-up, gender(male/female), education (no education, elementary school/highereducation), social support (living alone/married, living with somebody else),employment (working, house wife, volunteer / no employment), depression(CES-D), Visual Analog Scale (0-10) for pain in the head, neck both sides,shoulders and arms, range of motion of the neck and shoulders, andsensibility (number of area’s, 0-6).Depression scores contributed significantly to all domains of quality of life.Reduced shoulder abduction, shoulder pain, and neck pain were related toseveral domains of quality of life. The patient group scored significantlyworse for social functioning and limitations due to physical problems, butscored significantly better for bodily pain and health changes than thecontrol group.It was concluded that head and neck cancer patients score relatively good onquality of life. Depression has the strongest relationship with severaldomains but also physical complaints are significantly related to severaldomains.

In chapter 8 a study is described in which the validity of measuring somaticitems in a depression questionnaire in patients after cancer treatment isdescribed. Several authors describe that measuring somatic items in patients

Chapter 10

130

after cancer treatment is not valid because the somatic problems assessed ina questionnaire are strongly related to the somatic morbidity caused by the cancer treatment. These authors propose to remove the somatic items in adepression questionnaire when assessing cancer patients. In total 206patients with breast cancer, 136 patients with colo-rectal cancer, 69 withgynaecological cancer, and 155 with head and neck cancer were assessed. Asa control group 255 patients randomly chosen Dutch people were assessedmatched for age and gender out of the same region as the cancer patients. Alpatients were about 15 months after treatment. Cancer patients do scorehigher on the somatic domain compared to the control group. But cancerpatients were not a homogenous group with respect to somatic morbidity.Colo-rectal patients score much lower then the control group while head andneck and breast cancer patients score significantly higher then the controlgroup. Gender (female) has an important influence in all domains. It seemsnot valid to remove somatic items from the CES-D for measuring cancerpatients.

General discussion head and neck cancer treatment leads to considerablemorbidity; physically, emotionally, and socially. In this thesis the incidencesof shoulder complaints after radical, modified and selective neck dissectionsthat occur after head and neck cancer treatment are described.The clinical implications are that through this study a better understanding inthe diversity of complaints in patients after head and neck treatment wasachieved. This knowledge can be implemented in care of surgeons, physicaltherapists, nurses, and psychologists.

How can patients, after head and neck cancer therapy, benefit from theresults of this thesis. Together with the Dutch Physical Therapy Neck Dissection Study Group(Nederlandse Fysiotherapie Halsklierdissectie Studie Groep, NFHSG),physical therapy guidelines for patients after neck dissection were developed(appendix 1). The guidelines include diagnostic criteria, information forcolleagues, an exercise program for shoulder problems after neck dissection,treatment advises and a research protocol to assess patients prospectively. Inthe clinical assessment of patients after neck dissection currently patients areasked to shrug their shoulders to assess trapezius muscle dysfunction. From

Chapter 10

131

our study it became clear that for a quick assessment of trapezius musclefunction the patient should be asked for to abduct both arms on the sametime, and look for differences in range of motion. When differences arefound or when patients report shoulder pain an inspection of the shoulderregion is of importance to assess the trapezius muscle and to investigate thetype of pain.Furthermore psychological problems, especially depression, are ofimportance to assess in the post-clinical phase because of its influence onquality of life. Pain is one of the most important forms of morbidity. Adistinction should be made in nociceptive, neuropathic, and myofascial pain.However in all types of pain the influence of psychological or socialproblems must be considered.

Chapter 10

132

SamenvattingIn de inleiding (hoofdstuk 1) wordt de voorgeschiedenis van de hoofd-halsoncologie beschreven. De eerste publicatie over de radicalehalsklierdissectie werd beschreven door Crile in 1906. Crile voegde deradicale halsklierdissectie, naast het verwijderen van de primaire tumor, toeaan de operatie procedure. Deze toevoeging had een enorme positieveimpact op de overlevingskansen van patiënten. De morbiditeit die optrad nadeze operatie procedure werd echter niet door Crile beschreven. Pas in 1952beschreef Ewing de vele postoperatieve complicaties na een radicalehalsklierdissectie zoals: pijn, houdingsafwijkingen, gevoelsverlies, functievermindering, beperkingen in dagelijkse activiteiten en spierkrachtverlies.Schouderklachten na de radicale halsklierdissectie werden in verschillendeonderzoeken beschreven, het aantal keren dat schouderklachten voorkwamenna een radicale halsklierdissectie lag tussen de 47% en 100%. Het optreden van schouderklachten wordt toegeschreven aan het uitvallenvan de nervus accessorius. De nervus accessorius is de 11e hoofdzenuw dieloopt vanaf het achterhoofd door het gebied van de halsklieren naar demusculus trapezius (monnikskapspier). De nervus accessorius wordtopgeofferd tijdens een radicale halsklierdissectie waardoor de musculustrapezius zijn functie verliest. Een belangrijke functie van de musculustrapezius is het stabiliseren van het schouderblad op de borstkas. Bij uitvalvan de musculus trapezius glijdt het schouderblad zijwaarts naar buiten ennaar beneden. De schouderkom komt hierdoor naar beneden gericht te staan.Door deze veranderde stand wordt de schouder beperkt in zijn bewegingenen kunnen pijnklachten ontstaan. Schouderklachten na een halsklierdissectiewerden dus toegeschreven aan uitval van de nervus accessorius. Doordat schouderklachten na de radicale halsklierdissectie veel voorkwamenen door het toegenomen inzicht in het uitzaaiingpatroon van hoofdhalstumoren werd er door de jaren heen geprobeerd steeds meer selectievehalsklierdissecties uit te voeren. Hiervoor werden de gemodificeerde radicalehalsklierdissecties en de selectieve halsklierdissecties ontwikkeld. Bij degemodificeerde halsklierdissectie worden alle halsklieren verwijderd maarwordt de nervus accessorius gespaard, terwijl bij een selectievehalsklierdissectie slechts een deel van de halsklieren wordt verwijderd enook de nervus accessorius wordt gespaard. Schouderklachten nagemodificeerd radicale halsklierdissecties komen voor tussen de 18 % en 61

Chapter 10

133

% en bij selectieve halsklierdissecties tussen 29% en 39 %. De studies dieverricht zijn naar de radicale en gemodificeerd radicale halsklierdissectieszijn talrijk, terwijl naar de selectieve halsklierdissecties slechts drie studieszijn verricht met geringe patiënten aantallen. Een belangrijke vraag blijft hoe er ondanks het sparen van de nervusaccessorius, toch schouderklachten kunnen ontstaan. Niet alleenschouderklachten zijn van belang na een behandeling voor kanker in hethoofd hals gebied, maar ook andere factoren als functieverlies,psychologische problemen, nekklachten, angst voor het terug komen vankanker en kwaliteit van leven zijn belangrijke uitkomstmaten. Doel van ditonderzoek is om meer inzicht te krijgen in het optreden vanschouderklachten maar ook andere klachten te onderzoeken zoalsnekklachten, mobiliteitsverlies, sensibiliteitsstoornissen, functieverlies,depressiviteit en kwaliteit van leven na een behandeling voor kanker(operatie en radiotherapie) in het hoofdhalsgebied. De behandeling bestaatuit een operatie met een halsklierdissectie met of zonder aanvullenderadiotherapie.

In hoofdstuk 2 worden de resultaten van de pilot-studie beschreven. Hetdoel van deze studie was om de incidentie van schouderklachten na eenselectieve (supraomohyoidale) halsklierdissectie te onderzoeken. Onderzochtwerd pijn en functieverlies een jaar na een éénzijdige of dubbelzijdigesupraomohyoidale halsklierdissectie. In totaal 52 patiënten werdengeïncludeerd, alle patiënten vulden een vragenlijst in gericht op hetvastleggen van pijn en functieverlies in activiteiten van de schouder. Van de52 patiënt en hadden 14 (28%) schouderklachten aan de geopereerde zijde.Slechts twee patiënten hadden vanwege hun schouderklachten zoveel moeitemet zware huishoudelijke activiteiten dat ze hulp nodig hadden van anderen.Uit de pilot-studie kan geconcludeerd worden dat ondanks de selectievehalsklierdissectie 28 % van de patiënten pijnklachten heeft, maar dat erweinig functieverlies optreedt.

In hoofdstuk 3 wordt het onderzoek beschreven dat is uitgevoerd in deklinische setting. Doel van deze studie was om te onderzoeken in welke matepijnklachten en een verminderde mobiliteit voorkomen in de klinische fase.Daarnaast is gekeken welke risicofactoren er zijn voor het ontstaan van

Chapter 10

134

klachten. De studie is uitgevoerd in een multicentre-trial waarbij fysiotherapeuten uit zeven andere ziekenhuizen in Nederland hebbendeelgenomen. Patiënten vulden een vragenlijst in en de schoudermobiliteitwerd op een gestandaardiseerde wijze gemeten. Informatie betreffende deoperatie procedures zijn uit de medische status verkregen. In totaal werden177 patiënten geïncludeerd van wie 70% schouderklachten had. Pijnklachtenkwamen vooral voor bij bewegen van de schouder (31%), en het liggen opde aangedane schouder (30%). Anteflexie (voorwaarts bewegen) en abductie(zijwaarts bewegen) zijn significant afgenomen ten opzichte van de nietgeopereerde zijde. Risicofactor voor het ontstaan van schouderpijn enmobiliteitsverlies is het hebben ondergaan van een niet selectievehalsklierdissectie. Een reconstructie (reconstrueren van wondgebied metspierlappen soms aangevuld met vrij bot) tijdens chirurgie is een risicofactorvoor een verminderde anteflexie (24.5° verschil) van de schouder tijdens deklinische fase.

In hoofdstuk 4 wordt beschreven hoe vaak schouderklachten na eenhalsklierdissectie met sparen van de nervus accessorius, tenminste één jaarna de operatie, voorkomen. Daarbij werd tevens gekeken naar de invloed vanradiotherapie op schouderklachten. Patiënten werd een interview afgenomen om schouderklachten teinventariseren, daarnaast werd de Schouder Beperkingen Vragenlijst (SDQ)afgenomen om beperkingen in activiteiten gedurende het dagelijks leven teinventariseren. Honderdzevenendertig patiënten zijn geïncludeerd; 51 na eengemodificeerd radicale halsklierdissectie (GRHKD), 21 na eenposterolaterale halsklierdissectie (PLHKD) en 65 na een supraomohyoidalehalsklierdissectie (SOHKD). Na een GRHKD heeft 33.3% van de patiëntenschouderklachten, na PLHKD 66.7 %, en na een SOHKD 20 % van depatiënten. Type halsklierdissectie was significant (p < 0.001) van invloed ophet krijgen van schouderklachten. De uitkomsten op de SchouderBeperkingen Vragenlijst (SDQ) gaven dezelfde significante verschillentussen de verschillende halsklierdissectie typen (p< 0.01). Leeftijd en duurvan de follow-up waren niet significant van invloed op het voorkomen vanschouderklachten. Schouderklachten komen ook na het sparen van de nervus accessoriusveelvuldig voor. De prevalentie van schouderklachten na een SOHKD is

Chapter 10

135

laag en de klachten in het algemeen dagelijks leven zijn beperkt aanwezig.Radiotherapie heeft geen significante invloed op het krijgen vanschouderklachten of beperkingen.

In hoofdstuk 5 wordt de relatie tussen schouderklachten en de functie vande nervus accessorius beschreven. Doel van deze studie was te onderzoekenof er bij patiënten die na een halsklierdissectie schouderklachten hebben,sprake is van uitval van de nervus accessorius. Uitval van de nervusaccessorius werd geobjectiveerd middels drie kenmerken: atrofie van demusculus trapezius, schouderlaagstand en een standsverandering van descapula. Als twee van de drie kenmerken aanwezig waren dan werdaangenomen dat er sprake was van een nervus accessorius uitval. Verderwerd gekeken naar schouderabductie ten opzichte van de niet geopereerdezijde, en pijnklachten (geobjectiveerd met een genummerde VAS pijn (0-10)over de afgelopen week).Honderdentwaalf patiënten (73 mannen/39 vrouwen) participeerden in destudie. Een lichamelijk onderzoek werd bij 111 patiënten verricht.Negenendertig hadden schouderklachten van wie 20 (51%) uitval had van denervus accessorius. Van de 29 patiënten met uitval van de nervus accessoriushadden 20 (69%) schouder pijn. Schouderpijn en een verminderde abductievan ≥ 40° waren significant (p < 0.001) gerelateerd aan disfunctie van denervus accessorius. Geconcludeerd kan worden dat schouderklachten dieoptreden na een halsklierdissectie voor slechts 50% kunnen wordentoegeschreven aan uitval van de nervus accessorius. Een abductievermindering van ≥ 40° ten opzichte van de niet geopereerde zijde was eengoede voorspeller voor nervus accessorius disfunctie. In hoofdstuk 6 worden de gevolgen van hoofd-hals kanker behandelingenop de nek zelf beschreven. In deze studie werden nekpijn,sensibiliteitsverlies, mobiliteit van de nek en schouderpijn onderzocht.Honderddrieënvijftig patiënten hebben geparticipeerd in het onderzoek.Drieëndertig procent had nekpijn en 37% schouderpijn. Nekpijn enschouderpijn waren sterk gerelateerd (p < 0.001). Van de patiënten die nekpijn hadden waren er 20 (39%) met allodynie en 49 (96%) met hyperpathie,de gemiddelde VAS score voor pijn was 3.5 (sd 2.3). Van de patiënten had32% neuropathische pijn, 46% myofasciale pijn en 24% gewrichtspijn.

Chapter 10

136

Sensibiliteitsverlies was aanwezig bij 65% en was significant gerelateerd aantype halsklierdissectie en radiotherapie. De mobiliteit van de nek was nachirurgie en of radiotherapie significant vermindert bij lateroflexie van degeopereerde zijde af. Voor de andere bewegingen werden geen significanteveranderingen gevonden.

In hoofdstuk 7 wordt ingegaan op de invloed van schouder en nekklachten,ontstaan na de hoofdhals kanker behandeling, op kwaliteit van leven. Deuitkomstmaten van de studie werden vergeleken met een a-select gekozencontrole groep. Als uitkomstmaat is de RAND-36 gebruikt, deze vragenlijstheeft 9 domeinen (fysiek functioneren, sociaal functioneren, rol beperkingenals gevolg van fysieke problemen, rol beperkingen als gevolg vanemotionele problemen, algemene mentale gezondheid, vitaliteit, pijn,gezondheidsbeleving en gezondheidsveranderingen). De volgende itemswerden ingevoerd in de regressieanalyse als onafhankelijke variabelen:leeftijd, scholing, sociale ondersteuning, werk, depressiviteit, VAS pijn(hoofd, nek beiderzijds, beide schouders en armen), range of motion van denek en beide schouders en de sensibiliteit van het gelaat, de nek enschouders. Depressie had met alle 9 domeinen een significante relatie. Een verminderdeabductie, schouderpijn en nekpijn waren ook gerelateerd aan verschillendedomeinen van kwaliteit van leven. De patiënten groep scoorde alleen op dedomeinen sociaal functioneren en rol beperkingen als gevolg van fysiekeproblemen significant slechter dan de controle groep. De patiënten groepscoorde beter op de domeinen pijn en gezondheidsveranderingen dan decontrole groep. Geconcludeerd wordt dat patiënten na een hoofd hals oncologischebehandeling relatief goed scoren op kwaliteit van leven vergeleken met eencontrole groep. Depressie heeft de meeste impact op kwaliteit van levenmaar ook fysieke klachten na de behandeling hebben op veel domeinen eensign ificante invloed.

In hoofdstuk 8 is onderzocht in welke mate het meten van somatische items,bij het meten depressiviteit middels de CES-D, invloed heeft op de totaalscore bij patiënten na een kankerbehandeling. Verschillende auteurs stellendat de relatie tussen de somatische gevolgen van de kanker behandeling en

Chapter 10

137

de somatische items van een depressie vragenlijst zo strek gerelateerd zijndat het niet valide is somatische items te meten voor het meten van depressie. In dit onderzoek werden 206 patiënten na borstkanker, 136patiënten na colo-rectaal kanker, 69 met gynaecologische kanker, en 155 nahoofdhals kanker geïncludeerd. Deze werden vergeleken met 255 at randomgekozen patiënten die diende als controle groep. De samenstelling van decontrole groep was qua geslacht en leeftijd overeenkomend met de patiëntengroep en wonen in dezelfde regio. Kankerpatiënten scoren hoger op hetsomatische domein vergeleken met de controle groep, echter kankerpatiënten blijken hierin geen homogene groep te zijn. Sommige groepenscoren veel lager op het somatische domein (colo-rectaal) terwijl anderenkanker groepen veel hoger scoren (hoofd-hals, borst). Geslacht (vrouwelijk)heeft een sterke relatie met een verhoogde prevalentie. Geconcludeerd werddat somatische items op basis van een kanker behandeling niet bij allekanker patiënten uit de CES-D verwijderd mogen worden. Dit doet geenrecht aan het construct depressie en daarnaast is de prevalentie vansomatische morbiditeit bij verschillende kanker typen erg wisselend.

Algehele discussie de behandeling van hoofdhals tumoren kan leiden totaanzienlijke morbiditeit in zowel fysieke, emotionele als sociale zin. In ditproefschrift wordt het optreden van schouderklachten na radicale-,gemodificeerd radicale- en selectieve halsklierdissecties na de kankerbehandeling beschreven. Daarnaast worden gevolgen beschreven zoalsnekklachten, kwaliteit van leven en depressiviteit. De klinischeconsequenties zijn dat we door deze studie een beter inzicht hebbenverworven in de diversiteit van klachten en de diversiteit van patiënten. Dezekennis kan worden toegepast door de chirurg, verpleegkundige,fysiotherapeut en psycholoog in de diagnostiek en behandeling bij patiëntenmet hoofdhals kanker.Hoe kunnen patiënten profiteren van de resultaten van onze studies. Naast dit proefschrift zijn er, mede op basis van deze verworven kennis,richtlijnen ontwikkeld door de Nederlandse Fysiotherapie HalsklierdissectieStudie Groep (NFHSG) (zie Appendix 1) voor de behandeling van patiëntenna hoofd-hals oncologie. In de richtlijnen staan beschreven: diagnostischecriteria, informatie voor fysiotherapeuten, fysiotherapeutische overdracht,behandeladviezen, een oefenprogramma voor schouderklachten na een

Chapter 10

138

halsklierdissectie en een research protocol om patiënten prospectief teonderzoeken. Momenteel wordt dit onderzoeksprotocol door de leden vanhet NFHSG uitgevoerd. Verder worden in de discussie adviezen gegeven voor de diagnostiek bijpatiënten na een halsklierdissectie om te bepalen of de nervus accessoriusintact is. Daarnaast wordt het belang van een goed pijnonderzoekbeschreven, om een onderscheid te maken in verschillende pijntypes vooreen gerichte pijn behandeling.

Chapter 10

139

DankwoordAls beginnend fysiotherapeut werd ik in de kliniek “geconfronteerd” metpatiënten met hoofd-hals kanker. Schouderklachten traden na een operatie(halsklierdissectie) regelmatig op en waren voor veel patiënten, net als voormij, een lastig probleem. Maar het probleem daagde me uit me erin teverdiepen. Nu een aantal jaren later heeft dit geresulteerd in dit proefschrift.Bij dit leerzame proces heb ik veel te danken gehad aan verschillendepersonen die mij hebben gestuurd, gecoached, gefaciliteerd, gemotiveerd,gecorrigeerd soms afgeremd maar bovenal gestimuleerd!

Allereerst wil ik alle patiënten bedanken die vrijwillig en veelal enthousiasthebben geholpen dit onderzoek mogelijk te maken. Het spreken en metenvan een zo grote, zeer diverse groep patiënten heeft geleid tot inzicht inklachten maar ook tot “intieme” gesprekken over hoe te moeten leven metkanker. Alleen deze gesprekken waren al de moeite waard.

Prof. Dr. J.L.N. Roodenburg, Hooggeachte Promotor, beste Jan.Nog goed kan ik me herinneren dat ik één van de eerste keren, op jouwkamer zat (tussen de stapeltjes, die voor het zitten gaan wordenweggeschoven of opgestapeld tot grotere stapels) we hadden net een pilot-studie gedaan en het idee was doelmatigheidsgeld aan te vragen voor vervolgonderzoek. Je zei toen met genoeglijke blik: “nou dan moet je ook maar gaanpromoveren”. Zelf had ik er eigenlijk nog niet over nagedacht maar jou zoaanhorende leek er al geen weg meer terug. Daar hoefde ik dan niet meerover na te denken. Bedankt voor dit vertrouwen. Verder bedankt voor jepersoonlijke benadering, wijsheid, je “hoe gaat ie’s” en humor tijdens hetvoor mij zeer prettige promotie traject. Dr. P.U. Dijkstra, zeer geleerde co-promotor, beste PieterTot het moment dat ik als fysiotherapeut stage ging lopen op het AZG namik “het allemaal niet zo serieus”. In de stage werd me door jou en een aantalander collega’s duidelijk gemaakt dat het belangrijk is “het vak” serieus opte pakken. Toen ik daarna een baan kreeg in het AZG heb je me bij de handgenomen en me veel laten ontdekken van fysiotherapie, wetenschap,schrijven en vele andere dingen. En nu ik los kan lopen vraag jij mij zelfs afen toe naar de weg. Voor dit proces kan ik je niet genoeg bedanken!! Pieter

Chapter 10

140

ik hoop dat we nog lang mogen samenwerken, dat je nog veel zakmessengaat kopen op onze gezamenlijke successen en dat je daarmee nog veelappeltjes voor me gaat schillen in de trein.

Professoren J.H.B. Geertzen, G.J. Hordijk, J.M.H.K Wierda bedankt datjullie hebben willen optreden als leden van mijn beoordelingscommissie,bedankt voor de geïnvesteerde tijd en de prettige samenwerking.

Paranimfen Gerbrig Versteegen en Doeke Keizer, het is erg leuk om dezegebeurtenis met jullie te mogen delen. Een uitvloeisel van de prettige,productieve en creatieve manier waarop wij in het Pijncentrum samenwerken. Hopelijk gaan er nog flink wat jaren en Nobelprijzen volgen.

De afdeling fysiotherapie/revalidatie, Paul Nijkrake/Prof. Klaas Postema, encollega fysiotherapeuten die het mogelijk gemaakt hebben me in ditonderwerp te verdiepen en hebben meegewerkt aan de detachering naar deafdeling Mondziekten, Kaakchirurgie en bijzondere Tandheelkunde

Prof. dr. L.G.M. de Bont en Dhr. R.M. Rolvink, bedankt dat jullie hetmogelijk hebben gemaakt dat ik mijn promotietraject, op deze prettige wijze,op de afdeling heb mogen uitvoeren.

Alle oncologie verpleegkundigen en administratie personeel van depolikliniek Mondziekten, Kaakchirurgie en bijzondere Tandheelkunde(Jenny van den Akker, Corriet Hagenus, Esther Wartena, Wadia Ganesh,Linda Tuzla-Schuurman, Piet Haanstra, Miranda Been en Carol van Ark) de(voormalig) oncologie verpleegkundigen van de afdeling KNO (HildaBakker, Antje Havenga, Ellen Spijkstra, Karin Hempenius, Marianne Duits,Herma Spoelman) en de medische administratie van de chirurgischeoncologie en de afdeling Revalidatie. De succesvolle includering vanpatiënten in het onderzoek is voor een groot deel te danken aan deoplettendheid van jullie als “het ondersteunend personeel”.

De Raad van bestuur van het Academisch Ziekenhuis Groningen voor hetgeven van huisvesting in een zeer prettig ziekenhuis en het verschaffen vanfinanciële ondersteuning in de vorm van doelmatigheidsgelden.

141

Alle mede-onderzoekers: Gerreke, Diana, Monique, Jurjen, Justin, Arnoud,Bauke, Geerten, Christiaan en Pepijn en mede-gangbewoners Karin, Nienkeen Harrie op de derde. Bedankt voor de afleiding, humor (hoewel degewelddadige ontvoering van mijn lievelingssmurf mijns inziens wel wat verging, hij heeft nog steeds last van PTSS: posttraumatische smurf stoornis)etentjes en gezamenlijke lunchpauzes.

Alle collega’s van de Nederlandse Fysiotherapie Halsklierdissectie StudieGroep. Bedankt voor jullie bijdrage in het onderzoek, het ontwikkelen vande richtlijnen en de vele (informele) vergaderuurtjes in Utrecht. Hopelijkwordt ons huidige onderzoek ook weer een succes.

Marten van Wijhe en alle collega’s van het Pijncentrum waar ik veel opdirecte en indirecte aan te danken heb.

Collega fysiotherapeut Jettie Nomden, bedankt voor het belangeloos vertalenvan dat voor mij belangrijke Franse artikel.

Brother Gerald, vanaf de wieg gingen we elk onze eigen weg maar nu wasdaar toch ineens een gezamenlijk project. Bedankt voor je bereidheid voorhet corrigeren van mijn artikelen op schrijfstijl en Engels.

Mijn ouders, lieve Pa en Ma. We hebben net het einde van jullie “carrière”mogen vieren. Nu maar genieten van het pensioen, echter van jullie carrièreals ouder ben je nooit verlost! Waarschijnlijk kunnen jullie er langzaamaangerust op zijn dat het allemaal wel goed gaat komen. Op naar de volgendegeneratie. Het blijft een heerlijk gevoel dat jullie er altijd zijn voor ons en dekids.

Het Thuisfront, lieve Margriet, lieve Myrthe en lieve Roos; mijn bloementuin!! Bedankt voor jullie aanwezigheid.

Chapter 10

142

Appendix 1 Physical therapy guidelines developed by the Nederlandse Fysiotherapie Halsklierdissectie Studie Groep (NFHSG).

The physical therapy guidelines are in Dutch. They are developed primarily toinform colleagues in the primary care. It contains information about a neckdissection and the possible morbidity in neck and shoulder region. It further containsinformation for diagnostics and treatment for physical therapists who treat patientsafter neck dissection, and a handout with exercises for patients. Furthermore itcontains a protocol for scientific research that is currently used by the study groupfor further research. The guidelines are free to download fromhttp://home.wanadoo.nl/m.stuiver/index.htm

De richtlijn "Fysiotherapie bij halsklierdissectie" is primair bedoeld voor diefysiotherapeuten die geconfronteerd worden met patiënten die een HKD hebbenondergaan en als gevolg daarvan klachten in de nek- en schouderregio ervaren. Derichtlijn geeft informatie over wat is een halsklierdissectie en de mogelijke vormenvan morbiditeit die kunnen optreden. De richtlijn beoogt een praktische handreikingte zijn voor de fysiotherapeutische diagnostiek en behandeling voor patiënten dieeen halsklierdissectie hebben ondergaan. In de richtlijnen staan adviezen voor hetfysiotherapeutisch onderzoeken, behandeladviezen en een folder met oefeningen.Naast het klinische deel bevat deze richtlijn een wetenschappelijk deel waarin eenprotocol voor onderzoek staat besschreven die momenteel door de werkgroep wordtuitgevoerd. De richtlijnen zijn vrij te downloaden vanhttp://home.wanadoo.nl/m.stuiver/index.htm.

∗ De leden van de Nederlandse Fysiotherapie Halsklierdissectie Studiegroep zijn(alfabetisch): W. Brendeke, Rijnstate Ziekenhuis, Arnhem, R.P. Buijs, Erasmus MedischCentrum, Rotterdam,. dr PU Dijkstra, Academisch Ziekenhuis, Groningen, drs CJT de Goede,Vrije Universiteit Medisch Centrum, Amsterdam, A. van Opzeeland, MCL zuid, Leeuwarden,A. Kerst, Universitair Medisch Centrum, Utrecht, M. Koolstra, Vrije Universiteit MedischCentrum, Amsterdam, MM Stuiver, Nederlands Kanker Instituut /Antoni van Leeuwenhoek,Amsterdam, CP van Wilgen, Academisch Ziekenhuis, Groningen, drs.M.Wilschut , ErasmusMedisch Centrum, Rotterdam

Chapter 10

"FYSIOTHERAPIE BIJ HALSKLIERDISSECTIE"Een richtlijn voor onderzoek en behandeling van

schouderklachten na halsklierdissectie

C.P. VAN WILGEN(Namens de Nederlandse Fysiotherapie Halsklier Dissectie Studiegroep)

143

Appendix 2

Physical therapy for shoulder complains after neck dissection.

Authors,Publication year

Zilkens et al (1976) Johnson et al(1978)

Fialka et al(1988)

Grade evidence C/D C/D C/D

Study type Cohort Cohort Cohort

Population(incl. number ofpatients)

Radical neckdissection n = 32

??? n = 16

Radical neckdissection.n = 18

Intervention(including duration,frequency)

-active and passiverange of motion-muscle strengthtraining-relaxation-ball games

-infra red light-muscle strengthtraining-active range ofmotion-stretching

-electrotherapy-massage neckshoulder-exercises

Control -

Result (effect size,follow-up)

-pain-abduction-forward-flexion-exorotationfollow-up: ??

-pain-arm function-range of motion-well beingfollow-up: 2 years

-abduction-scapula posture-scapula spinedistance-pain follow-up: 4months

Result -pain reduction-increased abduction-increased forwardflexion

-pain reduction-arm functionincreased-range of motionincreased-wellbeingincreased

-pain reduction-reduction inscapula spinedistance mean 5mm

Comments Only describingresults

Only describingresults

Chapter 10

144

Physical therapy for shoulder complains after neck dissection.

Authors,Publication year

Radtke et al (1992) Salerno et al(2002)

Grade evidence C/D B

Study type Cohort Cohort

Population(incl. number of patients)

??? n = 45

Neck dissection withsparring of the spinalaccessory nerven=60

Intervention(including duration,frequency)

Physical therapy (notfurther described)

Exercises (n=30)3 x a weekTotal treatmentunknownAverage 97 days

Control No exercises (n=30)

Result (effect size,follow-up)

-forward flexion-abduction-scapula spine distancefollow-up: 3-14months

Modified ConstantScore (range 0-85) Higher score means abetter functioning

Result Increased abduction±15 °-reduction in scapulaspine distance with amean of 30%

Mean scoreintervention group 77.4and control group 56.2

Comments No differences betweengroups beforeintervention

Chapter 10

145

List of publications

-Dijkstra PU, Roodenburg JLN, van Oort RP, van Wilgen CP.Fysiotherapeutische en gnathologische consequenties van behandeling vantumoren in het hoofd-hals gebied. NVGPT,1998:4-7.

-van Wilgen CP, Pijn een veelzijdig probleem.’t Web,2000,4:16-19.

-van Wilgen CP, Dijkstra PU, Meyler WJ. Herkenning van chronische pijn.Fysiopraxis,2000,5:12-15.

-van Wilgen CP, Aanen HA, Anderegg Q, Koke A, Oosterhof J. Chronischepijn als kwaliteitsproject in een IOF. Fysiopraxis, 2000,9:27-28.

-Dijkstra PU, van Wilgen CP, Buijs RP, Brendeke W, de Goede CJT, KerstA, Koolstra M, Marinus J, Schoppink EM, Stuiver MM, van de Velde CF,Roodenburg JLN. Incidence of shoulder pain after neck dissection: a clinicalexplorative study for risk factors. Head & Neck, 2001,NOV:947-953

-van Wilgen CP, Dijkstra PU, Versteegen GJ. De fysiotherapeutischecognitief-gedragsmatige behandeling bij chronische pijn.Fysiopraxis,2001,10:8-11.

-van Wilgen CP, Geertzen JHB, van Wijhe M, Dijkstra PU. Complexregionaal pijnsyndroom type I, behandeld als een chronisch pijnsyndroom.Nederlands tijdschrift voor fysiotherapie, 2002,112:69-76.

-van Wilgen CP. Opzetten van een multidisciplinair programma voorfibromyalgie patienten in de eerste lijn. Nederlands tijdschrift voor pijn enpijnbestrijding, 2002, 22 (12):9-12.

-van Wilgen CP, Versteegen GJ. Het bio-psycho-sociaal model in eenpijncentrum. Pijnperiodiek, 2003,oktober:2-4.

Chapter 10

146

-van Wilgen CP, Dijkstra PU, Nauta JM, Vermey A, Roodenburg JLN.Shoulder pain and disability in daily life, following supraomohyoid neckdissection: a pilot study.Journal of CraniomaxillofacialSurgery.2003,31(3):183-6.

-van Wilgen CP, Dijkstra PU, van der Laan BFAM, Plukker JT, RoodenburgJLN. Shoulder complaints after neck dissection; is the spinal accessory nerveinvolved? British journal of oral and maxillofacial surgery, 2003,41:7-11.

-van Wilgen CP, Akkerman L, Wieringa J, Dijkstra PU. Muscle strength inpatients with chronic pain. Clinical Rehabilitation. 2003,17(8):885-9.

-van Wilgen CP, Dijkstra PU, van der Laan BFAM, Plukker JT, RoodenburgJLN. Shoulder complaints after nerve sparring neck dissection. Internationaljournal of oral and maxillofacial surgery. 2003:33(3):253-257.

-van Wilgen CP, Dijkstra PU, van der Laan BFAM, Plukker JT, RoodenburgJLN. Shoulder and neck morbidity in quality of life after surgery for headand neck cancer. Head and Neck (accepted december 2003)

-van Wilgen CP, Dijkstra PU, van der Laan BFAM, Plukker JT, RoodenburgJLN. Shoulder and neck morbidity in quality of life after surgery for headand neck cancer. Head and Neck (accepted februari 2004)

Chapter 10

Morbidity after neck dissection in head and neck1 Morbidity after neck dissection in head and neck...

Documents

Transcript of Morbidity after neck dissection in head and neck1 Morbidity after neck dissection in head and neck...