This is the first study to investigate the risk factors of UI in female adult patients with chronic cough. In this study, we found that the incidence of UI in adult female chronic cough patients was 50.1%. In addition, older age, more severe cough, combining with chronic sinusitis, cough easily triggered by exercise, and abdominal muscle pain due to cough were risk factors for UI in adult female patients.
Previous studies have proven that older age was an important risk factor for UI [13, 17, 18, 25,26,27,28,29], which was consistent with our observation. It might be resulted in the decline of urethral sphincter tension as the increase of age . Univariate analysis showed daytime CSS, night-time CSS, and cough VAS scores in patients with UI were significantly higher than that in patients without UI. However, when the above three values were enrolled in multivariate logistic regression analysis, it was found that only the cough VAS score could be an independent risk factor. Hence, the cough VAS score might reflect the cough severity generally, and might be possible to correct for the daytime and nighttime CSS.
There are two possible reasons for “cough easily triggered by exercise” (defined as intentional exercise beyond daily activities in this study) being a risk factor for UI [31,32,33,34,35,36]. Firstly, muscle tightness could increase abdominal pressure during daily exercise, especially during moderate-to-high intensity exercise. Meanwhile, cough triggered by exercise could continuously increase abdominal pressure then increase the risk of UI. Secondly, mild-to-moderate physical exercise can train the function of pelvic floor muscles. A normal pelvic floor muscle tension is the key to offset the excessive increase in abdominal pressure after exercise, which is helpful to prevent UI. If the cough is easily induced by exercise, patients will be resistant to do exercises, which makes it difficult to train the pelvic floor function adequately. Additionally, with the increased abdominal pressure due to prolonged coughing, the risk of UI will be certainly increased.
This study also showed that “chronic sinusitis” was an independent risk factor for UI. That chronic sinusitis is often characterized by nasal congestion, runny nose, dysosmia and facial pain, is common in patients who have been diagnosed with allergic rhinitis, asthma, chronic obstructive pulmonary disease, and other respiratory diseases . But up to now, there is no direct evidence that chronic sinusitis was a risk factor of UI. Bekele, et al. reported that the incidence of UI in pregnant women with respiratory diseases such as asthma, allergic rhinitis and rhinosinusitis is higher than those without respiratory diseases . A few studies have shown that severe nasal symptoms such as sneezing and runny nose in women could induce UI [19, 39, 40]. However, in our study, we found that there was no significant difference in the incidence of sneezing and nasal congestion between the patients with UI and patients without UI, we speculated that this is because of the high prevalence of nasal symptoms in Chinese population and the serious air pollution in China, yet there is still lacking of epidemiological researches in this area . We also compared the prevalence of UACS between the patients with UI and patients without UI and found that there was no significant difference in it (4.8% vs. 5.6%, p = 0.651). A large-scale epidemiological survey of chronic sinusitis covering 7 cities in China shows that age is a risk factor for chronic sinusitis . Therefore, a large proportion of elderly patients in UI group might increase the incidence of chronic sinusitis, which may also be the reason why chronic sinusitis was a risk factor for UI in this study. However, "chronic sinusitis" was not corrected by "age" in the multivariate analysis. Xie, et al. also found that "rhinitis" was a risk factor for UI in females, but there was still no sufficient evidence to explain the result . Thus, whether there was any overlapping pathogenesis between chronic sinusitis and UI needed further study.
The study also found that abdominal muscle pain due to cough was a risk factor for UI in female patients with chronic cough. The median VAS score of patients with abdominal muscle pain due to cough was significantly higher than that in those without this complication (60 vs. 50, p = 0.002), which indicated that patients who had abdominal muscle pain often presented more severe cough symptom, However, "abdominal muscle pain due to cough" was not corrected by "cough severity (VAS)" and remained as an independent risk factor in multivariate analysis. We wondered if it was attributed to the weakness and the lack of tension of the abdominal wall muscles and the pelvic floor muscles in these patients so that persistent cough was more likely to cause more severe damage to the muscles, ultimately lead to UI. Nevertheless, the specific mechanisms and interconnections are needed to be confirmed by further studies.
We analyzed the relationship between the distribution of cause of chronic cough and the occurrence of UI, and found that neither single nor compound causes could significantly increase the risk of UI. Chronic coughers with any cause showed similar susceptible to developing UI, and if the risk factors mentioned above were considered together, the occurrence of UI would increase. With univariate analysis, it has been found that the patients with UI were more sensitive to environmental allergens such as dust, oil fume, cigarette smoke, cold air, but all of these factors were wiped off the list after using multivariate analysis. Thus, no significant correlation between allergen-induced cough and UI was found in our study.
In this study, the reasons why BMI failed to enter the multivariate model might be related to the subjects, the numerical distribution, and, most importantly, the large number of missing data. Yet, previous studies have shown that high BMI was a strong predictor of the presence of UI in female [13, 17, 18, 25,26,27,28,29]. Noblett et al. considered that overweight and obesity could cause the pelvic floor in a chronic state of increased pressure, thereby increase the risk of UI, by illustrating the relationship between BMI and intra-abdominal and intravesical pressure . Even if BMI was not included in the multivariate analysis at the end, we can still believe that overweight and obesity were very important for UI developing by combining the results of the univariate analysis.
There were also some limitations in this study. First of all, some potential factors which could affect female pelvic floor dysfunction including surgical history (e.g. delivery and gynecological surgery), and the number of deliveries were not recorded, it might make an influence on the comprehensiveness of the result of the study. Additionally, the severity and duration of UI were not evaluated, resulting in nonsignificant effect of the risk factors. It has been reported that stress UI was more commonly presented [43, 44], thus it would be worth analyzing whether there were differences in the risk factors of different types of UI, and future study could focus on this question. Thirdly, the details of the treatment history were not recorded, which made it difficult to analyze the impact of treatment on UI. Meanwhile, this study was a single-center retrospective study, and some missing variables could not be verified, the relationship of the demographic, clinical factors and UI might not be comprehensive and accurate. However, considering the large sample size in this study, our findings could provide a helpful and referenced guiding for the management of chronic coughers. Lastly, as our sample consists of patients visiting a specialist cough clinic, it might not be representative to chronic cough patients in the community. But in our study, most of the patients were presented with the common causes of chronic cough, and the proportion of unexplained cough or chronic refractory cough was similar as previous reports . Furthermore, we found that the CSS and cough VAS were similar among different causes in this study. The etiological distribution did not play an predominent role in UI. This study still provided an important implication in clinical, but further study aimed at patients from the community should be conducted to confirm it in the future.