As shown in the flowchart (Fig. 1), we enrolled 2900 participants at baseline, data from 2462 (85.1%) participants were ultimately entered into our data analyses. The age distribution ranged from 20 to 46 years, with a mean age of 32.4 years (±3.8 years). Overall, 1274 participants (51.7%) had vaginal deliveries (VDs), 692 (28.1%) women had cesarean section (CS) deliveries (not in labor), 236 women (9.6%) had CS deliveries (in labor), and 260 (10.6%) women had forceps deliveries (FDs). A total of 525 (21.3%) women had delivery histories. Table 1 shows the sociodemographic characteristics of the participants.
3.1. Prevalence of lower urinary tract symptoms
The frequencies of individual LUTS are shown in Table 2. The prevalence of any LUTS was 70.6% and varied with different delivery modes (P< 0.001). Storage symptoms were more prevalent than voiding symptoms (65.4% vs. 23.0%). Storage LUTS were more prevalent in women with VDs and FDs than in those with CS deliveries (70.3% with VDs, 73.5% with FDs, 53.4% with CS deliveries who were in labor and 57.4% with CS deliveries who were not in labor). Nocturia was the most common symptom (35.4%) and varied with different delivery modes (P< 0.001, 34.8% with VDs, 46.5% with FDs, 38.1% with CS deliveries who were in labor and 31.5% with CS deliveries who were not in labor), followed by frequency (25.6%). However, there was no significant difference in frequency between those with VDs and CS deliveries. Urgency (25.3%) and SUI (20.8%) were also prevalent and varied with different delivery modes (P< 0.001). Straining was the least common symptom (6.1%) and varied with different delivery modes (P=0.01) (Table 2).
Women who had VD or FD histories were more susceptible to different storage LUTS, such as frequency (P=0.0019) and UI (P < 0.001), than those with a CS history (Table 2; Fig. 2). However, the difference in voiding symptoms was not notable. Young patients were at high risk for voiding symptoms such as pain/burning (P=0.003), while SUI was most common in women between the ages of 30 and 39 years (22.4%).
3.2. Symptom-specific bother due to lower urinary tract symptoms
The extent to which the study participants were bothered by each urinary symptom is summarized in Table 3 and Fig. 3. Only 6.0–20.2% of participants with bothersome LUTS reported a moderate to severe impact on their quality of life; 1.3–15.6% reported severe bother. Some of the most prevalent symptoms were also reported as the most bothersome, such as UUI (87.3%) and SUI (85.5%). Some of the symptoms that were most likely to be rated as bothersome were not necessarily the most likely to cause moderate or severe bother, such as pain/burning (81.8% vs. 10.7%). UUI was the most likely symptom to cause severe (15.6%) or moderate to severe bother (18.1%), followed by SUI (15.3% and 15.5%, respectively). Nocturia was less likely to be bothersome but the most likely to cause moderate to severe bother (38.0% and 20.2%). Frequency, which caused more moderate or severe bother (21.5% and 17.6%), was the storage symptom that was least commonly reported to be bothersome.
Voiding symptoms were less likely to cause moderate or severe bother than storage symptoms. Straining was the most frequent symptom that caused bother (83.3%). Straining and intermittency were more likely to cause moderate or severe bother (16.0% and 13.8%, respectively) (Table 3, Fig.3).
3.3. Potential risk factors for lower urinary tract symptoms
The risk factors for LUTS according to logistic regression analysis are shown in Tables 4–6 (a-d). VDs and FDs increased the odds of every kind of LUTS. Women who had VDs and FDs were more susceptible to any LUTS and bothersome LUTS than those who had CS deliveries, especially any bothersome LUTS (2.11-fold (95% CI: 1.69-2.63), 2.55-fold (95% CI: 1.86-3.50)), moderate to severely bothersome LUTS (3.17-fold (95% CI: 2.03-4.92), 4.14-fold (95% CI: 2.42-7.07)), any severely bothersome LUTS (4.80-fold (95% CI: 2.22-10.39), 6.65-fold (95% CI: 2.76-16.03)), storage symptoms (1.84- fold (95% CI: 1.48-2.29), 2.18-fold (95% CI: 1.55-3.06)), any UI (2.79-fold (95% CI: 2.16-3.60), 2.89-fold (95% CI: 2.04-4.09)), urgency (1.70-fold (95% CI: 1.32-2.19), 1.76-fold (95% CI: 1.24-2.50)), UUI (3.80-fold (95% CI: 2.54-5.71), 5.10-fold (95% CI: 3.11-8.36)), SUI (2.84-fold (95% CI: 2.13-3.80), 2.68-fold (95% CI: 1.81-3.97)) and MUI (4.00-fold (95% CI: 2.43-6.62), 4.78-fold (95% CI: 2.59-8.83)). A history of VD (VD or FD history) was a strong predictor for both moderate to severely bothersome LUTS and any severely bothersome LUTS and UI (P ≤ 0.05). Participants with a perineal laceration had increased odds of any bothersome LUTS (1.90-fold (95% CI: 1.21-2.98)) and any SUI (2.27-fold (95% CI: 1.33-3.86)) (P ≤ 0.05). Voiding symptoms were less influenced by delivery mode and delivery histories.
3.4 Pelvic floor muscle function
A significant relationship between different delivery modes and PFM values was found (P < 0.001). The CS (not in labor) group had significantly better sEMG parameters than the VD group in terms of the total score (P < 0.001), pretest average mean amplitude (P<0.001), flick contraction average peak amplitude (P < 0.001), flick contraction time after peak (P<0.001), tonic contraction average mean amplitude (CS vs. VD, P < 0.001; CS vs. FD, P=0.002), tonic contraction mean amplitude variability (P < 0.001) and posttesting average mean amplitude (P < 0.001). Women who underwent CS deliveries in labor also showed superior pelvic floor muscle function compared to women who experienced VDs based on the total score (P < 0.001), pretest average mean amplitude (P < 0.001), flash contraction average peak amplitude (P < 0.001), tonic contractions average mean amplitude (P < 0.001), tonic contractions mean amplitude variability (CS vs. VD, P=0.006; CS vs. FD, P=0.002), and posttesting average mean amplitude (P < 0.001). FDs had a more negative impact on PFMs than VDs. The amplitudes of flick and tonic contractions on sEMG were 29.3 (14.2) and 19.1 (10.1), respectively, in the FD group, in which the results showed the significantly lowest amplitude compared with the other groups (P < 0.001). There were also significant differences in the total score (P=0.001) between the VD and FD groups. Whether a woman was in labor before undergoing CS had no obvious effect on the total score, flash contraction average peak amplitude or posttesting average mean amplitude.
In total, 503 women had a delivery history. The total sEMG scores were significantly different among the different groups (P=0.038). Women with a VD history had significantly lower flick contraction average peak amplitude (P=0.015) and pretest and posttest average mean amplitudes (P=0.001 and P=0.009) than women with a CS delivery history (Table 7). We did not find any significant differences between women with or without lateral episiotomies or perineal lacerations.