Caesarean section is protective against stress urinary incontinence: an analysis of women with multiple deliveries


*Dr C. Chaliha, Department of Urogynaecology, St Mary's Hospital, Imperial College School of Medicine, London, UK.


All women who had three elective caesarean sections were selected from a database of 40,000 women delivering between 1977 and 1998, and age-matched with women having three vaginal births. They all completed a (validated) urinary and bowel symptom questionnaire. Women who had vaginal births had a significantly higher prevalence of stress incontinence but not other urinary or faecal symptoms compared with those delivered by caesarean section. The prevalence of faecal incontinence was lower than the prevalence of urinary incontinence. Although the prevalence of faecal incontinence was lower after caesarean delivery, this was not statistically different. These data have shown that caesarean section was associated with a lower risk of urinary incontinence, although a protective effect on development of faecal symptoms was not seen.


The development of pelvic floor disorders such as urinary and faecal incontinence and pelvic organ prolapse have been associated with pregnancy and vaginal delivery. The prevalence of urinary incontinence has been reported to be up to 34% after a vaginal delivery and is associated with denervation injury to the pelvic floor or mechanical trauma to the urethral sphincter mechanism.1–3 Anal incontinence is reported in 10% of women at 12 weeks postpartum and has been related to vaginal delivery and anal sphincter injury.4,5 It is unclear how protective an elective caesarean section is for the development of these disorders as after multiple caesarean births the prevalence of urinary incontinence is similar to women having vaginal births.1 There are very few data on women who have had multiple elective caesarean sections alone, and it may be that the risk of postpartum incontinence is different in those who have had a history of prelabour (elective) caesarean section alone compared with those who have undergone a caesarean section in labour.

The aim of our study was to investigate whether multiple deliveries by elective caesarean section reduced the risk of urinary and faecal incontinence compared with a group of women who had spontaneous vaginal deliveries.


Forty women who had three elective caesarean sections were identified from a database of 40,000 women delivering between 1977 and 1998, and age-matched with 80 women having three spontaneous vaginal deliveries. The database identified 256 women who had three caesarean sections but on hand searching the notes only 40 had three elective caesarean sections. A validated urinary symptom questionnaire (King's Health Questionnaire6) was posted to all women, to assess the presence and severity of urinary symptoms and their impact on quality of life. Additional questions regarding incontinence of flatus and solid and liquid stool were also asked (derived from a validated questionnaire, the Manchester Health Questionnaire).7

All terms and definitions are in accordance with the International Continence Society. Data are reported as number (%) or mean [SD]. Analysis was performed using SPSS software (version 10.0, SPSS). Fisher's exact test was used to compare urinary and faecal symptoms and Mann–Whitney U test to compare the total domain scores in the King's Health Questionnaire.


Thirty-three (82.5%) of women in the caesarean section group, and 50 (62.5%) in the vaginal delivery group responded. Women in the caesarean section group were older at first delivery than those who had a vaginal delivery (27 ± 4.5 years vs 24 ± 5.4). They were also older at the time of completion of the questionnaire (37 ± 4.5 years vs 34 ± 5.2 years).

There was a significantly higher prevalence of stress incontinence in the vaginal delivery group versus caesarean section group (Table 1). Stress incontinence was reported by 18 (36%) women in the vaginal delivery group compared with one woman (3%) in the caesarean section group. There was no significant difference in reporting of other urinary symptoms. Faecal urgency was reported by 5 (10%) of women in the vaginal delivery group versus 7 (21%) in the caesarean section group. Faecal incontinence was reported by 3 (6%) in the vaginal delivery group versus 1 (3%) in the caesarean section group. This was not statistically different.

Table 1.  Urinary and faecal symptoms reported by responders. Data are presented as n (%).
 Spontaneous vaginal delivery (n= 50)Caesarean section (n= 33)P
Stress incontinence18 (36)1 (3)0.0001
Urge incontinence6 (12)4 (12)0.99
Frequency11 (22)5 (15)0.44
Urgency8 (16)3 (9)0.36
Nocturia12 (24)12 (36)0.22
Nocturnal enuresis000.33
Intercourse incontinence01 (3)0.08
Faecal incontinence3 (6)1 (3)0.54
Faecal urgency5 (10)7 (21)0.15

The mean score of the incontinence impact domain from the King's Health Questionnaire was higher in the vaginal delivery group compared with the caesarean group (24.6 vs 11.1; P= 0.046). No other domain scores were significantly different.


This study has demonstrated that symptoms of both urinary and faecal incontinence are reported after vaginal and elective caesarean delivery. However, the prevalence of stress urinary incontinence was far greater after multiple vaginal deliveries, suggesting that elective caesarean section reduces this risk.

The lack of significant differences in other urinary symptoms and faecal symptoms between vaginal delivery and elective caesarean section delivery may be due to our small sample size. It may also be that these symptoms are not wholly related to damage to the pelvic floor during vaginal delivery but to changes that occur in pregnancy.

This study suggests that elective caesarean section may protect against the development of urinary incontinence, but the risk of faecal incontinence and other urinary symptoms including urge incontinence may not be reduced. The clinical relevance of the high prevalence of occult anal sphincter injuries after vaginal delivery of up to 35%4 is also questioned as clearly this does not result in a similar high prevalence of faecal symptoms even after multiple vaginal births. Decisions regarding promoting elective caesarean delivery must take into account that this may not protect against all pelvic floor morbidity and may be associated with an overall increase in maternal morbidity and mortality.

Accepted 3 March 2004