PREVALENCE OF ANAL INCONTINENCE AMONG ‘MOTHERS OF MULTIPLES’ & ANALYSIS OF RISK FACTORS
 
R.P. Goldberg, MD, C. Kwon, MD, S. Gandhi, MD, L. Atkuru, M. Sorensen, P. Sand, MD

   Objective: To determine the risk of anal incontinence among women with previous multiple pregnancy and childbirth, and assess the impact of demographic and obstetrical factors.

   Materials and Methods: The Evanston-Northwestern Mothers of Multiples Survey was administered to a cohort of 769 women with a history of multiple pregnancy and childbirth, attending the annual meeting of the National Organization of Mothers of Twins Club. The 77-item questionnaire was designed to elicit prevalence rates of pelvic floor symptoms, along with extensive obstetrical, medical and demographic data. Univariate techniques for the analysis of risk factors included chi-square and student t-tests. Forward stepwise multiple logistic regression analyses were used to control for potentially confounding covariates.
  
   Results: 733 of 769 registered attendees (95.3%) completed the survey. The sample was 94% White, 2.3% African American, 1% Hispanic and 0.6% Asian American. Median age was 37 (22-75), mean weight 160.4 lbs (92-375), parity 3.0 (2-12), and time since delivery 7.6 yrs (SD 9.5, 0.3-67.3). 93.4% delivered twins, 5.3% triplets and 0.4% quadruplets. 401 women had at least one vaginal birth, and 332 had delivered by cesarean only; 76.3% of women had 2-3 children, and 23.6% had 4 or more. For 54.4% of women, the ‘multiples’ represented their first childbirth event; 45.4% reported one or more singleton deliveries preceding their multiple pregnancy and delivery.
Fecal incontinence was reported by 10% (73), defined as ‘liquid stool only’ for 5.9%, and ‘solid and liquid’ for 1.6%. Among women reporting fecal incontinence, their symptoms first arose during pregnancy for 6.3%, after delivery for 40.6%, and were ‘unrelated to childbirth’ for 53.1%. Fecal incontinence was reported by 5.8% of women delivering by cesarean only, and by 11% with previous vaginal delivery. In the univariate analysis, ‘cesarean only’ appeared to be significantly protective against fecal incontinence (chisq=5.28, p=.02 OR = 0.44, 95% CL=0.23, 0.88). However, in the multiple regression model incorporating age and parity, delivery mode did not remain statistically significant. Age remained predictive of symptoms(p=0.0001), with the odds of fecal incontinence increasing by a factor of 2.0 per ten years of age. The mean age (years) corresponding with the reported onset of fecal incontinence was 42.7 yrs – representing a mean interval of 12.9 yrs since childbirt! h. No differences in fecal incontinence outcomes were observed between women whose multiples represented their first childbirth, versus those with a preceding singleton delivery.
Flatal incontinence was reported by 25.2% (183) – starting during pregnancy for 21.2%, after delivery for 30.3%, and unrelated to childbirth for 48.5%. The univariate analysis demonstrated flatal incontinence to be associated with older age (p=0.0001), total parity (p=0.01), and previous hysterectomy (OR 1.85, 95%CL 1.08, 3.20). The risk of flatal incontinence was increased for women with more than 3 children (O.R. 1.69, 95%CL 1.12, 2.5). Flatal incontinence was not associated with delivery mode (p=0.30) or BMI (p=0.19). According to the stepwise logistic regression analysis, the risk of flatal incontinence was associated only with increasing age (p=0.0001), with the odds of flatal incontinence increasing by a factor of 1.78 per ten years of age. The mean time interval between multiple childbirth and the onset of flatal incontinence was 7.8 yrs. The mean age (years) corresponding with the reported onset of flatal incontinence was 38.8 for flatal incontinence and 42.7 for ! fecal incontinence.
Fecal soiling was reported by 10% (73) of the cohort. As with flatal and fecal incontinence, age was the only significant correlate of reported symptoms (p=0.0001). The odds of soiling increased by a factor of 1.69 per ten years of age.

   Conclusions: The impact of multiple pregnancy and childbirth on subsequent anal incontinence has not been previously investigated. Within this cohort of 733 ‘mothers of multiples’, substantial rates of fecal incontinence (10%) and flatal incontinence (25.2%) were observed. The prevalence of all anal incontinence types increased with age. Delivering 'only by cesarean' is not fully protective against subsequent anal incontinence symptoms within this population. These findings highlight the substantial prevalence of post-reproductive anal incontinence symptoms within this steadily expanding female population. Understanding the most effective and rational strategies for primary prevention of these disorders will require further research.