OBSTETRICS

Multiple Repeat Caesarean Sections:Complications and OutcomesAdekunle Sobande, FRCOG, Mamdoh Eskandar, FRCSDepartment of Obstetrics and Gynaecology and Reproductive Medicine, College of Medicine, King Khalid University, Abha, Saudi Arabia

Abstract

Objective: To compare the complications and outcomes ofCaesarean section (CS) in women who have had three or moreprevious lower segment Caesarean sections with those in womenwith one previous CS.

Methods: We performed a retrospective study of 371 patientsundergoing repeat CS. Of these, 115 (31%) had previously hadthree or more Caesarean sections (group 1) and 256 (69%) hadpreviously had one CS (group 2). All 371 patients had the repeatCS performed at Abha Maternity Hospital, Saudi Arabia betweenJune 2002 and May 2004. Demographic data, complications, andoutcomes were compared using the Student t and chi-square tests.

Results: There were statistically significant differences between thetwo groups with respect to mean maternal age, parity, gestation atdelivery, and experience of the surgeon (P < 0.05). CS wasperformed as an emergency in 38 (32.9%) and 186 (72.6%) ofpatients in groups 1 and 2 respectively (P < 0.05).The consultant was involved in the decision to perform CS in215 (84.6%) of patients with one previous CS. There weresignificant differences between the two groups in the type of skinincision, the presence of dense adhesions during surgery, andbladder injury (P < 0.05). There were no statistically significantdifferences in birth weight, stillbirth rate, low Apgar score, bloodloss during surgery, duration of surgery, or the duration ofpostoperative hospital stay.

Conclusion: The prevalence of dense intra-abdominal adhesions andof bladder injury during CS was higher in women with a history ofthree or more previous CS than in women with one previous CS.Placenta previa and Caesarean hysterectomy occurred with equalfrequency in each group, and wound dehiscence and uterinerupture were rare.

Résumé

Objectif : Comparer les complications et les issues de la césarienne(CS) entre les femmes ayant déjà subi trois césariennes dusegment inférieur ou plus et les femmes n’ayant déjà subi qu’uneCS.

Méthodes : Nous avons mené une étude rétrospective portant sur371 patientes qui subissaient une CS de nouveau. De ce groupe,115 (31 %) avaient déjà subi trois CS ou plus (groupe 1) et256 (69 %) n’avaient déjà subi qu’une CS (groupe 2). Toutes les

371 patientes ont subi leur nouvelle CS au Abha Maternity

Hospital, en Arabie saoudite, entre juin 2002 et mai 2004. Lesdonnées démographiques, les complications et les issues ont étécomparées au moyen du test de Student et du test de chi carré.

Résultats : Nous avons constaté des différences significatives sur leplan statistique entre les deux groupes de patientes en ce qui atrait à l’âge maternel moyen, à la parité, à la gestation au momentde l’accouchement et à l’expérience du chirurgien (P < 0,05). LaCS a été effectuée d’urgence chez 38 (32,9 %) et 186 (72,6 %)des patientes des groupes 1 et 2, respectivement (P < 0,05). Leconsultant a participé à la décision de recourir à la CS chez215 (84,6 %) des patientes ayant déjà subi une CS. Nous avonsconstaté des différences notables entre les deux groupes en ce quiconcerne le type d’incision cutanée, la présence d’adhérencesdenses décelée au cours de la chirurgie et les lésions vésicales(P < 0,05). Nous n’avons constaté aucune différence significativesur le plan statistique en ce qui a trait au poids de naissance, autaux de mortinaissance, à la faiblesse de l’indice d’Apgar, à laperte sanguine au cours de la chirurgie, à la durée de la chirurgieou à la durée de l’hospitalisation postopératoire.

Conclusion : La prévalence des adhérences intra-abdominalesdenses et des lésions vésicales au cours de la CS était plusélevée chez les femmes ayant déjà subi trois CS ou plus que chezles femmes n’ayant déjà subi qu’une CS. Le placenta praevia etl’hystérectomie par césarienne ont présenté une fréquence égaledans les deux groupes; de plus, la déhiscence de la plaie et larupture utérine se sont avérées rares.

J Obstet Gynaecol Can 2006;28(3):193–197

INTRODUCTION

The incidence of lower segment Caesarean section (CS)

has increased worldwide in the last three decades.1-3

Although maternal death as a result of CS is now rare,

reports of the short- and long-term consequences of the ris-

ing CS rate on the childbearing population are conflicting.4

At the same time, it is as yet not clear if the increase in CS

rate has resulted in more favourable fetal outcomes. In

developed countries where small family size is the norm,

pregnant women requiring CS will consider having a tubal

ligation after the second or third delivery.

In a country like Saudi Arabia, where having a large family isencouraged by social and cultural influences, it is notunusual to see women planning for their sixth or seventhCS. One major complication of repeat CS is uterine scar

MARCH JOGC MARS 2006 � 193

OBSTETRICS

Key Words: Multiple, repeat Caesarean sections, complications,fetal outcome

Competing Interests: None declared.

Received on September 23, 2005

Accepted on December 20, 2005

rupture with subsequent adverse fetal and maternal conse-quences.5,6 Because of the paucity of clinical trials, there isno consensus on the maximum number of Caesarean sec-tions that a woman can undergo with safety.

Some authors have found no increase in maternal morbidityin women with a history of multiple CS and have thereforeencouraged them to pursue further pregnancies.7,8 At thesame time, anecdotal case reports of women having morethan ten Caesarean sections have been documented.9 Withthe improved safety of anaesthesia, the availability of safeblood transfusion, and the use of prophylactic antibiotics,many Caesarean sections are performed uneventfully.

This study was conducted to compare the short-term com-plications and outcomes of CS in women who had a historyof three or more previous Caesarean sections with thosewho had a history of one previous lower segment CS.

MATERIALS AND METHODS

We reviewed the hospital records of women who had a CSperformed after three or more previous Caesarean sectionsand those who had a CS for the second time at Abha Mater-nity Hospital from June 2002 to May 2004. The womenwho met these criteria were identified from the operatingroom logbook. In total, 371 patients were identified, ofwhom 115 (31%) had undergone three or more previousCaesarean sections (group 1) and 256 (69%) had undergoneone previous CS (group 2). Women with a history of classi-cal CS or of inverted T uterine incision were excluded fromthe study.

Women with three or more previous Caesarean sectionswere admitted for planned elective CS. Women with thesame history who had no plan made for elective CS andwho were admitted in labour had emergency CS performed.Women with a history of one previous CS performed for a

non-recurring indication had a trial of vaginal delivery afterproper maternal and fetal assessment. Hospital departmentpolicy dictated that induction of labour was contraindicatedin patients with previous CS. Emergency CS was performedfor fetal distress, or failure to progress in labour, in patientswith previous CS who were admitted in labour.

All patients received prophylactic antibiotic therapy (intra-venous cefoxitin 2 g 30 minutes before surgery, followed by1 g every 4 hours for 24 hours) and a Foley catheter wasinserted in the bladder prior to surgery. The choice of trans-verse or vertical subumbilical skin incision was made at thesurgeon’s discretion.

Data extracted from the patients’ files included maternalage, parity, number of abortions, booking status (bookedpatients had at least two antenatal clinic visits, andunbooked patients were seen for the first time in the emer-gency room in labour or at term), antenatal complications,placental location on ultrasound, gestation at delivery,duration of surgery, experience of surgeon, type of skinincision, presence of severe adhesions, intraoperative andpostoperative complications, preoperative and postopera-tive hemoglobin concentration, fetal outcome, and numberof postoperative days in the hospital. The data were coded,tabulated, and entered into an IBM compatible computer.Statistical analyses were carried out using the SPSS Version9 for Windows (SPSS, Chicago, IL).

Numbers and percentages were calculated for qualitativevariables, and mean and standard deviation were calculatedfor quantitative data. Comparisons between mean values ofquantitative variables were calculated using the Student t

test, and chi-square was used for qualitative data. The test ofsignificance was set at the 0.05 level.

OBSTETRICS

194 �MARCH JOGC MARS 2006

Table 1. Maternal demographic characteristics and birth weight

Characteristic Group 1

n = 115

Group 2

n = 256

Significance

P

Maternal age in years (mean ± SD) 31.8 ± 4.7 30.0 ± 5.6 0.006

Parity (mean ± SD) 4.3 ± 1.7 3.4 ± 3.0 0.005

Blood loss in mL (mean ± SD) 436 ± 191 415 ± 229 NS

Gestational age at delivery in weeks (mean ± SD) 37.3 ± 2.7 38.7 ± 2.4 0.013

Postoperative hemoglobin in g/L (mean ± SD) 104.4 ± 15.7 103.9 ± 14.9 NS

Hospital stay in days (mean ± SD) 6.8 ± 2.0 6.3 ± 3.2 NS

Birth weight in grams (mean ± SD) 2972 ± 818 2941 ± 747 NS

Number of abortions (mean ± SD) 0.9 ± 1.2 0.8 ± 1.2 NS

Duration of surgery in minutes (mean ± SD) 55.2 ± 15.7 51.7 ± 12.8 NS

SD: standard deviation; NS: not significant.

RESULTS

During the study period, there were 11 228 deliveries withan overall CS rate of 21.1%. The women who constitutedthe study population accounted for 15.6% of the total num-ber of Caesarean sections performed during the studyperiod.

Of group 1 patients, 79 (68.6%) had three previous Caesar-ean sections, 24 (20.8%) had four, 10 (8.7%) had five, andtwo (1.7%) had six previous Caesarean sections. The mater-nal demographic characteristics and birth weights areshown in Table 1. There were statistically significant differ-ences between the two groups regarding mean maternalage, parity, and gestational age at delivery (P < 0.05), but nodifference in the number of abortions, blood loss duringsurgery, duration of surgery, and birth weight. Theintraoperative and postoperative complications are shownin Table 2. Caesarean section was performed as anemergency in 38 (32.5%) and 186 (72.6%) of group1 and 2 patients, respectively (P < 0.05). There were no sig-nificant differences in the rates of bowel injury and Caesar-ean hysterectomy, but the rates of bladder injury and

presence of severe adhesions were higher in group 1 thanin group 2 (P < 0.05). No differences were found in othervariables. Other characteristics and fetal outcomes in thetwo groups of patients are shown in Table 3. The rates ofplacenta previa and placental abruption were similar in thetwo groups. Patients in group 1 were significantly morelikely to have a consultant (senior specialist) surgeon thanpatients in group 2.

DISCUSSION

It is not uncommon in Saudi Arabia for pregnant women tohave had more than three Caesarean sections. The highprevalence of grand multiparity in the community may berelated to the cultural views of the community, whichcelebrates the male gender and therefore regards a womanwithout male offspring as a failure. Paradoxically, thewoman without female offspring feels out of place in thefamily.

Abha Maternity Hospital is a secondary referral obstetriccentre that provides care for a population of about two mil-lion, and therefore infrequently admits complicated

Multiple Repeat Caesarean Sections: Complications and Outcomes

MARCH JOGC MARS 2006 � 195

Table 2. Intraoperative and early postoperative complications

Maternal complication Group 1

n = 115 (%)

Group 2

n = 256 (%)

Significance

Bladder injury 2 (1.7) 0 (0.00) �2= 5.16

P = 0.023

Bowel injury 1 (0.86) 0 (0.00) �2= 2.23

P = NS

Caesarean hysterectomy 1 (0.86) 1 (0.39) �2= 2.23

P = NS

Wound infection/dehiscence 4 (3.4) 5 (1.9) �2= 0.78

P = NS

Blood transfusion 4 (3.4) 10 (3.9) �2= 0.04

P = NS

Deep vein thrombosis 1 (0.86) 2 (0.78) �2= 0.01

P = NS

Other* 2 (1.7) 2 (2.7) �2= 0.033

P = NS

Scar dehiscence or rupture 0 (0.0) 1 (0.39) �2= 0.39

P = NS

Skin incision

Lower transverse

Vertical midline

53 (46.0)

62 (54)

214 (83.5)

42 (16.5)

�2= 40.75

P = 0.000

Emergency Caesarean section 38 (33.0) 186 (72.6) �2= 45.7

P � 0.001

*Urinary tract infection, chest infection, atelectasis

NS: not significant.

obstetric cases. The major complications of repeat CSinclude rupture of the scarred uterus, placenta accrete, andintraoperative complications such as bladder or bowelinjury. Unfortunately, there are no guidelines regarding themaximum number of Caesarean sections a woman mayundergo before she risks having serious complications.

In our study, scar dehiscence (defined as the presence of awindow in part of the uterine scar with intact membranes)was found incidentally in only one patient (0.26%), who wasundergoing a second elective CS. There were no instancesof symptomatic uterine rupture, nor was there any maternalmortality in this study. In the review by Kirkinen,10 27% ofpatients with three or more previous Caesarean sectionshad fenestration of the uterine scar, but recent studies havedescribed rates ranging from 1% to 10% in women under-going anywhere from a fifth to a ninth CS.11,12

It is unclear why the incidence of scar dehiscence in ourstudy was low despite some patients with more than threeprevious Caesarean sections being admitted in early labourfor emergency CS. Although the sample size was relativelysmall, at least one study has reported a higher incidence ofscar dehiscence in a study population smaller than ours.10

It is notable that risk factors for scar dehiscence such asmultiple pregnancy and polyhydramnios were not presentin our patients, and induction of labour was not carried outin patients with one previous CS. The risk of uterine rupturein patients with one previous CS has been shown toincrease with induction of labour, depending on the agentused for the induction.13

The incidence of bowel and bladder injury in our series isconsistent with other reports.12 It is difficult to determinewhether the type of skin incision makes bowel or bladderinjury more likely. In the present study, the majority ofpatients with a history of three or more Caesarean sectionshad a vertical midline skin incision. We have observed that abladder that is drawn towards the uterine fundus by scarringis more likely to be damaged after a suprapubic transverseincision, but small bowel that is adherent to the parietalperitoneum is likely to be damaged after a vertical incision.In general, careful and meticulous entry into the peritonealcavity is the key to reducing injury to these organs.

In our study, there were significantly more adhesions in thepatients with three or more Caesarean sections than inthose with one previous CS, as found in other reviews.12

This is not unexpected because repeated surgery may beassociated with postoperative infection and subsequentadhesion formation. It must be stressed that factors such asmeticulous surgical technique, gentle tissue handling, andthe general health of the patient affect tissue healing andadhesion formation. We also found no difference in theincidence of placenta previa and placenta accreta betweentwo surgical groups. Although Rashid et al.11 found no dif-ference in the incidence of placenta previa and placentaaccreta between higher order (5–9) repeat CS and lowerorder (3–4) repeat CS, others have demonstrated that theincidence of placenta previa and placenta accreta increaseswith the number of Caesarean sections.12,14 In a recentreview by Makoha et al.,15 the third CS did not define athreshold for increased maternal risk, but abnormal

OBSTETRICS

196 �MARCH JOGC MARS 2006

Table 3. Antenatal complications, maternal characteristics, and fetal outcome

Characteristic Group 1

n = 115 (%)

Group 2

n = 256 (%)

Significance

Placenta previa 4 (3.4) 8 (3.1) �2= 0.03

P = NS

Abruptio placenta 0 (0.0) 7 (2.7) �2= 3.21

P = NS

Stillbirth 4 (3.4) 2 (0.78) �2= 3.63

P = NS

Low Apgar score

(7 at 5 minutes)

6 (5.2) 8 (3.1) �2= 0.96

P = NS

Unbooked patients 19 (16.5) 130 (50.7) �2= 38.07

P � 0.001

Grade of surgeon

Senior specialist

Specialist

Resident

54 (46.9)

50 (43.4)

11 (9.7)

54 (21.0)

139 (54.2)

62 (24.8)

�2= 48.5

P � 0.001

placentation on a scarred uterus increased with successiveCaesarean sections.

Our study found no difference in the rates of Caesareanhysterectomy between the two groups. Both hysterectomiesperformed were carried out because of uncontrolled bleed-ing from a placenta accreta. The timely decision to performhysterectomy may have contributed to the favourablematernal outcome data.

Despite the limitations of our study, we conclude that scardehiscence and uterine rupture are not common in womenwho have a CS following either one or three or more Cae-sarean sections. Nonetheless, dense intra-abdominal adhe-sions and bladder injury occur more commonly in womenwho have had three or more previous Caesarean sections.Larger, prospective trials are required to confirm thesefindings.

REFERENCES

1. Chamberlain G. What is the correct caesarean section rate? Br J ObstetGynaecol 1993;100:403–4.

2. Macfarlane A, Chamberlain G. What is happening to caesarean sectionrates? Lancet 1993; 342:1005–6.

3. Mcllwaine GM, Colse SJ, Macnaughton MC. The rising caesarean sectionrate—a matter of concern? Health Bulletin 1985;43:301–5.

4. Tower CL, Strachan BK, Baker PN. Long-term implications of caesareansection. J Obstet Gynaecol 2000;20(4):365–7.

5. Khashoggi TY. Higher order multiple repeat cesarean sections: maternal

and fetal outcome. Ann Saudi Med 2003;23:278–82.

6. Leung AS, Leung EK, Paul RH. Uterine rupture after previous cesareansection delivery: maternal and fetal consequences. Am J Obstet Gynecol1993;169: 945–50.

7. Seidman DS, Pazz I, Nadu A, Dolberg S, Stephenson K,Gale R. Aremultiple caesarean sections safe? Eur J Obstet Gynecol ReprodBiol1994;57:7–12.

8. Soltan MH, Al Nuaim L, Khashoggi T, Chowdhury N, Kangave D, AdelusiB. Sequelae of repeat cesarean sections. Int J Gynaecol Obstet1996;52:127–32.

9. Piver M, Johnston P. The safety of multiple cesarean sections. ObstetGynecol 1969;34:690–3.

10. Kirkinen P. Multiple caesarean sections: outcomes and complications. Br J

Obstet Gynaecol 1988; 95:778–82.

11. Rashid M, Rashid RS. Higher order repeat caesarean section: how safe are

five or more? BJOG 2004;111:1090–4.

12. Juntunen K, Makarainen L, Kirkenen P. Outcome after a high number

(4–10) of repeated caesarean sections. BJOG 2004;111:561–3.

13. Lydon-Rochelle M, Holt VL, Easterling TR, Martin DP. Risk of uterine

rupture during labor among women with a prior cesarean delivery. N Engl J

Med 2001;345(1):3–8.

14. Clark SL, Koonings PP, Phelan JP. Placenta previa/accreta and prior

cesarean section. Obstet Gynecol 1985;66:89–92.

15. Makoha FW, Felimban HM, Fathuddien MA, Roomi F, Ghabra T. Multiple

cesarean section morbidity. Int J Gynaecol Obstet 2004:87:227–32.

Multiple Repeat Caesarean Sections: Complications and Outcomes

MARCH JOGC MARS 2006 � 197