Turkish Neurosurgery 9: 113 - 122, 1999 Kerman: Evaluatioii of Clinical Results

Evaluation Of Clinical Results In 40 Patients WithBasal Skull Fracture

40 Kaide Kirigi Olgusunda Klinik Sonuçlarin Degerlendirilmesi

MEMDUH KERMAN, AHMET DAGTEKIN, VAROLAYDIN, N. NEFI KARA, HACi KAYMAZ

Süleyman Demirel University Faculty of Medicine, Department of Neurosurgery, Isparta, Turkey (MK, AD, VA, NNK, HK)

Gelis Tarihi: 22.10.1998 Ç:> Kabul Tarihi: 4.2.1999

Abstract: Fractures of the basis cranii are usually the resultof extension of a vault fracture. The most importantcomplications associated with these fractures arecerebrospinal fluid (CSF) fistula, related infection, andpneumocephalus with fistula. CSF fistula involves theleakage of fluid from the subarachnoid space to theextraarachnoidal space through a defect in thearachnoidea, the dura, or the epithelial tissue. Althoughthis leakage can occur along the cerebrospinal axis, it mostoften appears clinically as otorrhea and rhinorrhea. Eightypercent of CSF fistulae is caused by trauma, and thiscomplication is seen in 2-5% of closed head trauma cases.With regard to outcome, 85% of rhinorrhea cases and 95%of otorrhea cases improve spontaneously within a week.Meningitis is the most important complication of CSFfistulae. We retrospectively studied 40 basis cranii fracturecases that were treated at our clink between November

1994 and December 1997. Thirty-four of the patients weremale (85%) and six were female (15%), and their agesranged from 2 to 70 years. The most common cause of thefractures was traffic accident. The patients' Glasgow ComaScale (GCS) scores ranged from 3 to 15. üf the 40individuals studied, 19had otorrhea and 5 had rhinorrhea.Tension pneumocephalus occurred in two cases duringthe time they were receiving treatment. Posttraumaticmeningitis developed in three of our patients, and themortality rate associated with meningitis was 5% (2/40patients). üur study also highlighted the importance ofother complications that may accompany basis fracture,namely, posttraumatic intracranial infection and tensionpneumocephalus.

Key Worrls: Basal skull fracture, CSF fistula, meningitis,tension pneumocephalus

118

Özet: Kaide kiriklari, siklikla kafa çatisinda görülenkiriklarin tabana dogru uzanmasi sonucu olusur.Kaide kiriklarinin en önemli komplikasyonu BüSfistülü ve buna bagli olusan enfeksiyonlar ilepnömosefaluslardir. BüS fistülü, BüS'un araknoid,dura veya epitel dokudaki defekte bagli subaraknoidmesafeden extraaraknoid mesafeye kaçisidir.Serebrospinal aks in herhangi bir yerinde görülse deen sik otore ve ri no re seklinde ortaya çikar. BüSfistüllerinin %80'nin nedeni travmalardir ve kapalikafa travmalarinin %2-5'inde BüS fistülü görülür.Rinorelerin %85'i, otorelerin ise %90-95' ilk bir haftaiçinde kendiliginden kesilir. BüS fistülünün en önemlikomplikasyonu menenjittir. Klinigimizde kasim 1994­Aralik 1997 yillari arasinda tedavi edilen 40 kaidekirigi olgusu retrospektif olarak degerlendirildi.Hastalarin 34'ü (%85) erkek, 6'si (%15) kadindi. Enküçügü 2 ve en büyügü ise 70 yasindaydi. Trafik kazasien sik travmatik nedendi. Glasgow Koma Skalasi(GKS) puanlari 3-15 arasinda degismekteydi. 19hastada otoraji ve 5 hastada rinore vardi. Izlemsirasinda 2 hastada tansiyon pnömosefalus gelisti. 3hastada posttravmatik menenjit görüldü. Menenjitebagli mortalite orani %5'di. Sonuç olarak bu çalismadakaide kirigi komplikasyonlarindan olan posttravmatikintrakranial enfeksiyon ve tansiyon pnömosefalusunönemi vurgulandi.

Anahtar Sözcükler: Kaide kirigi, BüS fistülü, menenjit,tansiyon pnömosefalus

Tiirkisl, Neiirosiirgery 9: 113 - 122, 1999

INTRODUCTION

Basal skull fraetures usually result fromextension of a vault fraeture. Typieally, thesefraetures involve the perinasal sinuses and mastoidair eells (14). Basal skull fraetures are divided intotwo subgroups. Fractures that traverse the petrouspyramid at right angles are ealled "transversefraetures," and hematotympanum is a eommonfinding in these cases. Longitudinal fraetures runpara llei to the long axis of the petrous bone andrepresent 70% to 90% of temporal bone fraetures.These fraetures often spare the nerves but disruptthe ossieular ehain (14). Basal skull fraetures canresult in injury to eranial nerves and arteries (23).eN VII and/or VIII injuries are assoeiated withtemporal bone fraeture. Olfaetory nerve injury oftenoecurs with anterior fossa basis fraetures, and resultsin anosmia. Injury to CN VI can oeeur with fraeturesof the eliyus (14). Basal skull fraetures are associatedwith traumatie earotid-eavernous fistulae, traumatieaneurysms of the petrous and eavernous portions ofthe earotid artery, and earotid artery oeclusion0,18,21,27).

Cerebrospinal fluid (CSF) fistula is the mostimportant eomplieation associated with basa i skullfraetures. These fistulae are classified in two majorgroups, as traumatie and nontraumatie (26). CSFfistulae may be deteeted either in the first week afterthe trauma (aeute), or after months, and even years,posttrauma (delayed). This life-threateningcomplieation was first described by Bidloo and Elderin the 17th century (26). Miller identifiednontraumatie rhinorrhea in 1826, and proved thatthis eondition oeeurs as a result of increased CSFpressure (26). In 1884, Chiari was able todemonstrate a postmortem fistula between theethmoid sinuses and a pneumatoeele of the frontallobe in a patient who had had meningitis withrhinorrhea (12).

Another important eomplieation, espeeially oftraverse of paranasal mastoid fraeture, ispneumoeephalus. In addition to head trauma, othereauses of pneumoeephalus include infeetion, tumor,eongenital eranial defect, shunt plaeement, and theuse of nitrous oxide during anesthesia(3,4,7,20,24,28).

Especially in the past 20 years, our ability todiagnose and treat basal skull fraetures and CSFfistulae has improved with the advent of newteehniques. The leakage site can now be easily

Kerinail: Evaliiatioii of Cliiiical Resiilts

identified using radionudide eisternography,eomputed tomography (CT),and magnetie resonanceimaging (MR!).CT eisternography with metrizamideis the best diagnostie method (13,19).

CLINICAL MATERIAL AND METHOD S

From November 1994 to Deeember 1997,280head trauma patients were treated in our department.Forty (15%)of these patients had basal skull fraetures,and we retrospeetively studied these eases withregard to eomplieations eneountered, diagnosis, andtreatment.

RESULTS

Of 280 head-injured patients,40 were diagnosedwith basal skull fracture. The ages of these patientsranged from 2 to 70 years, and the mean age was30.5 years (Tabie 11).Thirty-four patients were mal e(85%) and six were female (15%). Twelve patients(30%) were children and 28 patients (70%) wereadults. The eauses of trauma were traffie aeeident(28 patients), falIing (11 patients), and blow to thehead (1 patient) (Table 11).In eaeh ease, neurologiealeondition was evaluated according to the GlasgowComa Seale (GCS), and patients were eategorized inone of three head trauma groups: severe (GCS 3-8),moderate (GCS 9-12), and mild (GCS 13-15). Thirty­five patients (88%) had mild, two patients (5%) hadmoderate, and three patients (7%) had severe headtrauma. Nineteen of the patients had otorrhea andfive had rhinorrhea. Craniography and eranial CTwere routinely performed on all of the patients. Thefracture line was identified in 35% of the eases byeraniography, and in 50% of the eases by eranial CT.In addition, eranial CT led to diagnoses ofpneumoeephalus in 10 patients, brain eontusions in7, brain edema in 2, and aeute subdural hematomain one patient.

Thirty-six (90%) of the patients were treatedmedieally and four (LO%) underwent surgery.Medical therapy included the following: 1) elevationof the head, 2) prevention of increases in intraeranialpressure (use of eough preventatives, laxatives,sedatives), 3) eontinued lumbar CSF drainage, 4)drugs to reduee CSF produetion (dexamethasone,aeetazolamide, furosemide), 5) prophylaetieantibioties (eephtriaxon and ornidazole), and 6) useof antiepilepties. In all of the patients who hadotorrhea and who reeeived medical treatment, CSFleakage stopped within 1 week. The same was truefor all but one patient with rhinorrhea, whose CSF

119

Turkish Neurosurgery 9: 113 - 122, 1999 Kerman: Evalim/ioii of Cliiiical Results

NO.ofCases48810226

Table i. Number of cases of head trauma by agegroup.

Age Group0-910-1920-2930-3940-4950-5960 and over

contained 2,400 polymorphonuelear leukocytes permicroliter. StreptocOCCltS pneumoniae was grown ona CSF culture, which confirmed the diagnosis ofposttraumatic meningitis postmortem. Anotherpatient, who had a traumatic corneallaceration, wastreated at the ophthalmology elinic. When fever,signs of meningeal irritation, and diminishedconsciousness developed, this patient was diagnosedwith posttraumatic meningitis. Third patient whoreceived prophylactic antibiotics developedmeningitis.

Table II. Causes of head trauma.

Cause of traumaTraffic accidentMotor Vehiele

BicyeleFallingBlow to head

NO.ofCases28

244

111

The hospitalization period for our patientsranged from 3 to 29 days (mean, 8 days). One patientdied due to disseminated pneumocephalus and braincontusions. Another developed subarachnoidalhemorrhage and exophthalmos. In this patient,digital subtraction angiography revealed a carotid­cavernous fistula.

leakage continued for more than 2 weeks.

Table III. Glasgow Coma Scale scores for the 40patients.

Three of the 40 basal skull fracture of patientsdeveloped posttraumatic meningitis. One of theseindividuals had been hospitalized for 3 days atanother facility due to trauma, and his level ofconsciousness deteriorated 2 days after discharge.The patient died 1 hour af ter admittance to ouremergency elinic. There were no signs of pathologyon the patient's cranial CT. His CSF was eloudy, and

Four patients underwent surgical treatment.Surgery was indicated when there was rapiddeterioration in a patient's level of consciousness anda worsening prognosis. In two patients, cranial CTdemonstrated tension pneumocephalus, whichrequired emergency surgical treatment. In one ofthese patients (GCS 7), the air was released througha drilled burr hole. The other patient underwentcraniotomy and duraplasty. A third individual hadrhinorrhea that continued for over 2 weeks, and thispatient was treated with duraplasty. The fourthpatient's cranial CT revealed an acute subduralhematoma, and this required drainage viacraniotomy.

Basal skull fractures are elinically associatedwith CSF fistulae (otorrhea and rhinorrhea),hematotympanum, postauricular ecchymosis(Battle's sign), periorbital ecchymosis (raccoon'seyes), cranial nerve injury, and vessel injury. Thereported incidence of basal skull fractures due tocranial trauma ranges from 10-25% (6,10,16). Cranialtrauma is considered to cause 80% of all CSF fistulae

(20), and this problem is seen in 2-5% of cranialtrauma cases (10,12,26). Of 280 head trauma patientsadmitted to our elinic, 40 had basal skull fracturesand 24 of those with basa i skull fractures had CSF

fistulae. These proportions correspond fairly wellwith those reported in the literature.

DISCUSSION

The detection of otorrhea and rhinorrhea is veryimportant with regard to diagnosing CSF fistula and

Basal skull fractures and traumatic CSF fistulaeare less common in children than in adults (17). Thisis due to the child's immature growth of the frontalsinuses, the presence of a cartilaginous-type ethmoidbone, and the more elastic basis cranii than adults',all of which result in better absorption of headtrauma. We found an expected smaller proportionof pediatric head trauma cases in our study group(30%). FaIling and traffic accidents have beenidentified as the major causes of basa iskull fracturesand CSF fistulae, and account for 80-90% of thesecases (20).Ninety-seven percent of our patients withskull fractures and fistulae had been injured in oneof these two ways.

NO.ofCases35

23

GCS ScoreMild (13-15)Moderate (9-12)Severe (3-8)

120

Tiirkis/i Neiirosiirgenj 9: 113 - 122, 1999

basal skull fracture. Following cranial trauma, fluidleakage from the nose and ear is assumed to be CSFwhen blood is visualized centrally, surrounded byclear fluid. However, this method of identification isnot 100% accurate. The best and most recent method

for identifying the origin of the leaked material isthe detection of b2 transferrin throughimmunoelectrophoresis (25). Concerning thediagnosis of CSF fistulae, various techniques havebeen used to date, including direct craniography,intrathecal injection of different dyes, andpneumoencephalography. The introduction of CThas made it very easy to detect basal skull fracturesand associated with CSF fistulae. In particular, thin­slice axial and coronal scanning allows accuratediagnosis and pinpointing of the anatomicallocationof the fistula and the fracture (2,11).

In 1977, Drayer (13) and Manelfa (19) reportedthat metrizamide cranial CT-cisternography was thebest technique for locating a CSF fistula. Recently,new techniques, including MR!, magnetic resonance­cisternography, digital subtraction cisternography,and positron emission tomography, have been usedto this end (15,22,29).Currently, metrizamide cranialCT-cisternography remains the best way to diagnosethese fistulae. Craniography and cranial CT weredone routinely on all of our patients. Basal skullfractures were identified by craniography in 35% ofthe cases, and by cranial CT in 50% of the cases. Inaddition, cranial CT revealed that 10 patients hadpneumocephalus, 7had brain contusions, 2 had brainedema, and 1 had an acute subdural hematoma.

One important complication of basal skullfracture is tension pneumocephalus, a problem thatmay require immediate surgery (7,20,28). Researchhas shown that patients with rhinorrhea and otorrheaare at greater risk of developing tensionpneumocephalus compared to other head traumapatients (24). In these patients, an increase innasopharyngeal pressure causes air to enter thecranial cavity through the dural defect and thenbecome trapped. Elevated intracranial pressure mayincrease the size of the defect and the patient'scondition may deteriorate due to the pressure exertedon the brain and the air accumulating inside thecranium. Pneumocephalus was seen in 20% of basalskull fractures. On the other hand 75-80% ofpneumocephalus caused by trauma (23).

Cranial CT led to the diagnosis ofpneumocephalus in 10 (25%) of our 40 patients withbasal skull fractures. Tension pneumocephalus

Kennaii: Eva/iiatioii of Cliiiical Resiilts

developed in two patients. The treatment for one(GCS 7) involved releasing the air through a burrho le, and for the other involved craniotomy andduraplasty.

Eighty-five percent of patients with rhinorrheaand 95% of those with otorrhea improvespontaneously within 1 week of diagnosis(2,14,17,25). CSF leakage stopped after 1 week in allof our patients who had otorrhea and was treatedmedically. The same was true for all but one patientwith rhinorrhea, whose CSF leakage continued formore than 2 weeks and required surgical treatment.

Meningitis is the most important problemassociated with CSF fistulae, and causes highmorbidity and mortality, even when antibiotictherapy is used. The incidence of meningitis inpatients with trauma-induced CSF fistulae rangesfrom 3-50% (6,10).S. pneumoniae is the most commoncausal agent in meningitis (14). Eight percent (3/40)of our patients with basal skull fracture developedmeningitis. There is controversy regarding the useof prophylactic antibiotics in patients with CSF fistulaand basal skull fracture (6,14,17). Some reports havestated that such treatment does not effectively reducethe risk of meningitis in patients with traumatic CSFfistulae (5,14,17,23). Choi et aL.(8), who studied 293cases of traumatic CSF leakage, found the incidenceof meningitis was significantly higher in patients whoreceived prophylactic antibiotic therapy than in thosewho did not receive preventive antibiotic therapy.In contrast, Brodie (7) investigated 324 cases ofposttraumatic CSF fistula, and found a lowerincidence of meningitis in those who receivedprophylactic antibiotic therapy than in those who didnot use preventive antibiotics. Of all our patients withbasal skull fractures and CSF fistulae who were givenantibiotic therapy, onlyone developed meningitis.On the other hand, two of the patients who were notgiven prophylactic antibiotics developed thisinfection. Based on these findings, we advise the useof preventive antibiotic treatment in patients withbasa i skull fracture and CSF fistula.

Correspondence: Memduh KermanSüleyman Demirel ÜniversitesiTip Fak. NörosirürjiAnabilim Dali32040/ISPARTA

Tel:O-246-2326657/179

121

Turkis/i Neurosurgery 9: 113 - 122, 1999

REFERENCES

1. Aarabi B and McQueen JD: Traumatic internal carotidocelusion at the base of the skull. Surg Neuroll0: 233­236, 1978

2. Aarabi B, Leibrock LG: Neurosurgical approaches tocerebrospinal fluid rhinorrhea. ENT 71: 300-305, 1991

3. Altinörs N, Arda N, Kars Z, Türker A, Senveli E,Çepoglu C, Çinar N: Tension pneumocephalus aftertranssphenoidal surgery: Case report. Neurosurgery23:416-518, 1988

4. Aoki N: Air in acute epidural hematomas. J Neurosurg65: 55-56, 1986

5. Baltas I, Tsoulfa S, Sakellariou P, Vogas V, FylaktakisM, Kondodimou A: Posttraumatic meningitis:Bacteriology, hydrocephalus and outcome.Neurosurgery 35: 422-427, 1994

6. Brawley B, Kelly W: Treatment of skull fractures withand without cerebrospinal fluid fistula. J Neurosurg26: 57-61, 1967

7. Bret PH, Kzaiz M, Guyotat J, Fischer G, Zannini c: Lapneumatocele intracranienne sous pression. Une causepossible d'aggravation post operatoire enneurochirurgie 10observation. Neurochirurgie 33: 209­215, 1987

8. Brodie HA: Prophylactic antibiotics for posttraumaticcerebrospinal fluid fistulae. A meta-analysis. ArchOtolaryngol Head Neck Surg 123:749-752, 1997

9. Choi D, Spann R: Traumatic cerebrospinal fluidleakage: risk factors and the use of prophylacticantibiotics. Br. J Neurosurg 10:571-575, 1996

10. Couldwell WT, Weiss MH: Cerebrospinal FluidFistulas. In Apuzzo MLJ (ed). Brain surgery. New York,Churchill Livingstone Inc. 1993, pp 2329-2342

11. Creamer MJ, Blendonohy P,Katz R, Russell E: Coronalcomputerized tomography and cerebrospinal fluidrhinorrhea. Arch Phys Med Rehabil 73: 599-602, 1992

12. Dagi TF, George ED: Surgical management of cranialcerebrospinal fluid fistulas: In: Schidek HH, Sweet WH(eds). Operative Neurosurgical Techniques. Thirdedition. Philadelphia, WB Saund ers Company, 1995,pp117-131

13. Drayer BP, Wilkins RH, Boehnke M, Horton JA,Rosenbaum AE: Cerebrospinal fluid rhinorrheademonstrated by metrizamide CT cisternography. AmJ Roentgenol129: 149-151, 1977

14. Gade GF, Becker DP, Miller JD, Dwan PS: Pathologyand pathophysiology of head injury. In Youmans JR(ed): Neurological Surgery. Third Edition.Philadelphia, WB Saunders Company, 1990, pp: 1965­2016

15. Gammal TE, Brooks BS: MR Cisternography: Initialexperience in 41 cases. AJNR 15: 1647-1656, 1994

122

Kerman: Evaluaiion of Clinical Results

16. Hendrick EB, Harwood-Nash OC, Hudson AR: Headinjuries in children: A survey of 4,465 consecutive casesat the Hospital for Sick Children, Toronto. ClinNeurosurg 11: 46-65, 1964

17. Jones DT, MC Gill, Healy GB: Cerebrospinal fistulaein children .Laryngoscope 102: 443-446, 1992

18. KasselI NF, Boarini DJ, Adams HP: Intracranial andcervical vascular injuries. In Cooper PR (ed): HeadInjury. Second Edition. Baltimore, Williams andWilkins, 1987, pp: 313-354

19. Manelfe C, Guiraud B,Tremoulet M: Diagnosis of CSFrhinorrhea by computerized cisternography usingmetrizamide. Lancet 2: 1073, 1977

20. Monajati A, Cotanch WW: Subdural tensionpneumocephalus following surgery. J Comput AssistTomogr 6: 902-906, 1982

21. Morley TP: Appraisal of various forms of managementin 41 cases of carotid-cavernous fistula. In Morley TP(ed): Current Controversies in Neurosurgery.Philadelphia, WB Saunders Company, 1976, pp: 217­236

22. Papay FA, Maggiano H, Dominquez S, HassenbuschSJ, Levine HL, Lavertu P: Rigid endoscopic repair ofparanasal sinus cerebrospinal fluid fistulas.Laryngoscope 99: 1195-1201, 1989

23. Rouit RL and Murali R: Injuries of the Cranial Nerves.In Cooper PR. (ed): Head Injury. Second Edition.Baltimore, Williams and Wilkins, 1987, pp: 141-158

24. Ruge RJ, Cerullo LJ, Melone DG: Pneumocephalus inpatients with CSF shunt. J Neurosurg 63:532-536,1985

25. Ryall RG, Peacock K, Simpson DA: Usefulness of b2transferrin assay in the detection of cerebrospinal fluidleaks following head injury. J Neurosurg 77: 737-739,1992

26. Spetzler RF, Zambramski JM: Cerebrospinal fluidfistulae: their management and repair. In: YoumansJR (ed): Neurogical Surgery. Third edition.Philadelphia, WB Saunders Company, 1990, pp 2269­2289

27. Teal JS, Bergeron RT, Rumbaugh CL, Segall HO:Aneurysms of the petrous or cavernous portions of theinternal carotid artery associated with non-penetratinghead trauma. J Neurosurg 38: 568-574, 1973

28. Toung T, Donham RT, Lehner A, Alano J, Campbell J:Tension pneumocephalus after posterior fossacraniotomy: report of four additional cases and reviewof postoperative pneumocephalus. Neurosurgery 12:164-168,1983

29. Wakhloo AK, Van Velthoven V, Schumacher M, KraussJK: Evaluation of MR imaging, digital subtractioncisternography and CT cisternography in diagnosis ofCSF fistulae. Acta Neurochir (Wien) 111: 119-127,1991