This was a single-center, ambispective (symptoms occurring during pregnancy were collected retrospectively at postnatal consultations; postpartum symptoms were assessed prospectively at Day 2, Day 15, and Day 45), descriptive and analytical study conducted at a level III hospital in the suburban area of Dakar, from February 1 to April 30, 2025.

All patients who delivered at CHNP — vaginally or by cesarean section — a pregnancy of at least 22 weeks of gestation, and who agreed to participate in the survey during the study period, were included. Patients who did not provide consent were excluded, as were those who could not be reached by phone during postpartum follow-up.

Data were collected during three scheduled postnatal consultations at Day 2, Day 15, and Day 45 after delivery. A standardized case report form was completed for each patient, including the following variables: age, gravidity, parity, history of UI or AI, mode of delivery (vaginal or cesarean), type of delivery (spontaneous or instrumental), presence of episiotomy or perineal tear, newborn birth weight, occurrence of urinary leakage or involuntary loss of gas or stool, their context of occurrence (on exertion, at rest, or mixed), and their management.

Symptoms occurring during pregnancy were collected retrospectively at postnatal consultations, while those occurring after delivery were assessed prospectively up to 45 days postpartum. Screening was based on a standardized interview conducted by a gynecologist. Participants were systematically questioned about involuntary loss of urine, gas, or stool during pregnancy and the postpartum period.

Persistence of UI or AI was defined as symptoms lasting beyond 45 days after delivery. Spontaneous remission was defined as the complete disappearance of symptoms within 45 days postpartum without pelvic floor rehabilitation.

Data were entered on the KoBoToolbox platform by a single investigator (gynecologist) to limit interviewer bias, information bias, and inter-observer variability.

Macrosomia was defined as birth weight ≥ 4000 g. Multiparity referred to 4 to 5 prior deliveries, and grand multiparity to more than 5. Mixed incontinence refers to the coexistence of UI and AI in the same patient.

Statistical analyses were performed using R software version 4.4.1. Quantitative variables were described using mean, median, standard deviation, and extreme values. Qualitative variables were described using frequencies and percentages. For bivariate analysis, Pearson’s chi-square test, Fisher’s exact test, and bivariate logistic regression were used according to applicability conditions. A multivariate binary logistic regression analysis was conducted to study significant associations. Variables with p < 0.2 in bivariate analysis or recognized in the literature were included in the models. Firth’s regression was applied in cases of small sample sizes. Model quality was assessed using the Hosmer-Lemeshow test and the ROC curve (AUC = 0.837 for the UI model; AUC = 0.840 for the AI model). The statistical significance threshold was set at p < 0.05. Results are expressed as adjusted odds ratios (aOR) with 95% confidence intervals (95% CI).

During the study period, 864 women delivered at the Mother-Child Center of CHNP. After applying exclusion criteria (2 refusals, 17 unreachable, 49 unavailable phone numbers), 796 patients were included in the analysis.

The mean age was 28.4 ± 6.1 years (median: 28 years; range: 14 - 47 years). The 20 - 29 age group was the most represented (51.4%). The mean gravidity was 2.73 ± 1.7. Paucigravidae constituted the majority category (41.8%). Nulliparous women accounted for 34.3% of the study population.

Of all included patients, 57.5% (n = 458) delivered vaginally and 42.4% (n = 338) by cesarean section. Among vaginal deliveries, 94.1% were spontaneous. Episiotomy was performed in 27.9% of patients and perineal tear occurred in 21.6% of cases. Birth weight was normal in 67.7% of cases; 9.1% of newborns presented with fetal macrosomia.

The overall frequency of incontinence during pregnancy was 42.7% (n = 340), distributed as follows: isolated UI (36%; n = 287), mixed incontinence (5.9%; n = 47), and isolated AI (0.75%; n = 6). Stress incontinence was the predominant type, accounting for 91.9% of antepartum incontinence cases (Table 1).

In the postpartum period, 14.3% (n = 114) of patients presented with pure UI, occurring mainly between Day 3 and Day 15 (93.5%). Isolated AI affected 0.75% (n = 6) of patients, appearing predominantly between Day 3 and Day 15 (85.1%). Mixed postpartum incontinence was found in 11.2% (n = 88). In total, 26.1% of patients (n = 208) presented with postpartum incontinence.

Table 1. Frequencies of antepartum and postpartum incontinence (N = 796).

For postpartum UI, spontaneous remission was observed in 54.5% (n = 110) of patients, pelvic floor rehabilitation was required in 25.5% of cases, and symptom persistence beyond 45 days was found in 17.3% of patients. One case of surgical treatment was reported.

For postpartum AI, spontaneous remission occurred in 51% of cases, pelvic floor rehabilitation was required in 30.5%, and symptom persistence was observed in 18.5% of patients.

On multivariate analysis, factors independently associated with postpartum UI were (Table 2): history of UI (aOR = 3.26; 95% CI: 1.76 - 6.10; p < 0.001), presence of UI or AI during the current pregnancy (aOR = 6.49; 95% CI: 4.16 - 10.4; p < 0.001), instrumental delivery (aOR = 73.1; 95% CI: 8.90 - 999; p < 0.001), and fetal macrosomia (aOR = 2.6; 95% CI: 1.40 - 4.91; p = 0.002).

Multiparity was associated with an increased risk of postpartum AI without reaching statistical significance (aOR = 2.22; 95% CI: 0.54 - 10.3; p = 0.XX). Factors independently associated with postpartum AI were: history of UI (aOR = 2.18; 95% CI: 1.04 - 4.53; p = 0.039), presence of incontinence during the current pregnancy (aOR = 4.66; 95% CI: 2.61 - 8.68; p < 0.001), instrumental delivery (aOR = 20.9; 95% CI: 2.93 - 260; p = 0.002), perineal tear (aOR = 2.57; 95% CI: 1.23 - 5.36; p = 0.012), and macrosomia (aOR = 2.31; 95% CI: 1.19 - 4.43; p = 0.013).

Table 2. Factors independently associated with postpartum urinary and anal incontinence (multivariate analysis).

* p < 0.05; ** p < 0.01; *** p < 0.001. UI: urinary incontinence; AI: anal incontinence; aOR: adjusted odds ratio; 95% CI: 95% confidence interval.

The frequency of UI during pregnancy in our study (41.9%) is consistent with international data. Dinç (2018) reports a prevalence of 40% among 750 pregnant women in Turkey [1], and Bø et al. (2012) found 41.7% in a multi-ethnic Norwegian cohort of 823 patients [2]. Regarding antepartum AI, our frequency of 6.6% (isolated AI + mixed) is slightly lower than the 10.3% reported by Solans et al. (2010) in a cohort study of 949 women [5], which may reflect underreporting related to the taboo nature of this symptom.

In the postpartum period, our UI rate of 25.4% is comparable to the 26% (95% CI: 21% - 30%) reported by the meta-analysis of Dai et al. (2023) encompassing 28,303 women [6], as well as the 33% from the systematic review by Thom and Rortveit (2010) [7]. The postpartum AI frequency of 11.8% falls within the 4% - 39% range reported at 6 weeks postpartum in the systematic review by Villot et al. [8].

Instrumental delivery emerged as the factor most strongly associated with postpartum UI (aOR = 73.1) and AI (aOR = 20.9). These results, although characterized by wide confidence intervals due to small sample sizes (n = 27), are consistent with the literature. The meta-analysis by Barca et al. (2021), including over one million women, demonstrates that vaginal delivery increases the risk of UI (OR ≈ 2.17) and AI (OR ≈ 1.53) compared to cesarean section [9]. A recent prospective Nigerian study confirms that vaginal delivery multiplies the risk of UI by 2.8 and that of AI by 2.1 compared to cesarean section [10]. The JAMA study by Blomquist et al. (2018) reports an increased risk of sphincter disorders after instrumental delivery (HR ≈ 1.75 for AI) [11].

Perineal tear was significantly associated with postpartum AI (aOR = 2.57; p = 0.012), confirming the deleterious role of obstetric injuries on sphincter function. Episiotomy, however, was not significantly associated with incontinence in our study, likely because the majority were median or mediolateral episiotomies without sphincter involvement.

Fetal macrosomia was significantly associated with antepartum UI (OR = 7.44; 95% CI: 3.61 - 15.33) and postpartum UI (OR = 4.96). These findings are consistent with those of Zhu et al. (2023), who showed in 1575 women that high birth weight was correlated with increased UI in the third trimester [12]. Yang and Liao (2022) reported that birth weight ≥ 4000 g multiplied the risk of stress UI by 2.59 and that of AI by 3.01 in the early postpartum period [13].

The pathophysiological mechanism involves increased mechanical pressure on the pelvic floor during gestation, greater stretching of the pudendal nerve during expulsion, and a longer second stage of labor, increasing the risk of muscular and nerve injuries.

Antepartum, nulliparous women constituted the most affected category by UI (34.43%), which is consistent with Dinç et al., who found a higher frequency in nulliparous women (41.7%) than in multiparous women (20.3%) [1]. This observation is explained by the initial exposure of the pelvic floor—which has never undergone distension—to the constraints of pregnancy. Postpartum, however, pauciparous and multiparous women became the most represented, reflecting the dose-dependent effect of successive pregnancies and deliveries on the weakening of perineal structures.

Our study has several strengths: a large sample size (n = 796), a rigorous analytical approach including multivariate analysis with Firth’s regression, and being the first Senegalese study to jointly assess UI and AI during the perinatal period.

Its limitations include: 1) underreporting bias related to the taboo nature of the subject; 2) data relying on interviews without validated objective tools (pad test, anorectal manometry, standardized questionnaires such as the ICIQ); 3) absence of adjustment for certain potential confounding factors (BMI, gestational diabetes, duration of labor, tear grade); 4) absence of longitudinal follow-up beyond 45 days postpartum.