1. Prevalence of pre- and post-surgical UI – a comparison by incontinence classification
Many previous studies have used the Urogenital Distress Inventory (UDI) as a tool for the evaluation of UI classifications. Although the UDI can be used to classify UI, it cannot be used to evaluate the frequency or volume of UI. On the other hand, the ICIQ-SF is a new questionnaire developed by the International Consultation on Incontinence (ICI). Since it can also be used to evaluate the frequency and volume of UI, it can be used for all UI patients regardless of sex or age. The final version, after verification of its reliability and validity, was released in 2001, with a Japanese version developed by Goto et al. . The prevalence of UI among women in Japan has been reported to range between 26% and 53.7% [21-23].
In the present study, evaluation using the ICIQ-SF showed that the prevalence of UI pre-surgery was 33.8%, which is comparable to the previously reported figures. On the other hand, the prevalence of UI post-surgery was 52.4%, with a significant increase noted after surgical intervention. Hazewinkel et al. reported the prevalence of UI post-surgery to be 24%  based on a questionnaire survey distributed by mail to 146 cervical cancer patients after radical hysterectomy, a value much lower than that observed in the present study. The reasons for this discrepancy are thought to be that the subjects in their study were younger than those in the present study, the median time from surgery to questionnaire was 6 (range, 1 to 11) years, which was longer than that in the present study, and the method of UI evaluation also differed.
As mentioned above, the prevalence of UI post-surgery in the present study was 52.4%, of which 34.4% of patients newly experienced UI post-surgery. The study by Hazewinkel et al.  did not compare subjects with UI pre- and post-surgery, and the present study is the first to identify patients newly experiencing UI post-surgery.
With regard to incontinence classification, the ICIQ-SF was used for classification in the present study. Results showed that 87.7% of those with UI had stress incontinence, 10.2% had mixed incontinence, 2.0% had urge incontinence, and 0.0% had overflow incontinence pre-surgery, with the majority of patients having stress incontinence. Araki et al. conducted a questionnaire survey of working women  and reported that the prevalence of women with UI was 16.7%, categorized as 72.7% with stress incontinence, 12.1% with urge incontinence, and 9.9% with mixed incontinence. Their results were quite similar to the present results. On the other hand, the rates by classification were 57.1%, 7.8%, 31.3%, and 2.6% for stress, urge, mixed, and overflow incontinence, respectively. The rate for stress incontinence was markedly lower, whereas that for mixed incontinence was markedly higher than the pre-surgery values.
Stress incontinence accounts for the majority of cases of UI among women in general, and it has been reported to be caused by aging (over 40 years of age) and the tendency for pelvic floor muscles to become weaker due to obesity . However, there was no significant difference in age between those with and without UI either pre-or post-surgery in the present study, and this is likely because the UI in the present study population was due to surgical invasion rather than age. On the other hand, in gynecological cancer patients, tissues supporting the cervix, such as the vesico-uterine ligament and the cervico-uterine ligament, are separated from the uterine cervix with removal of the uterus , leading to collapse of the pelvic mechanism balance and the bladder or urethra, relaxation of the pubococcygeus muscle, and insufficient closure of the urethra, resulting in the onset or exacerbation of stress incontinence . In addition, it is thought that, in cases where the hypogastric nerve (a sympathetic nerve) is damaged during hysterectomy , the pelvic nerve (a parasympathetic nerve) becomes dominant, resulting in the occurrence of urinary urgency ; i.e., OAB symptoms, subsequently leading to mixed (including stress) incontinence.
2. Prevalence of pre- and post-surgical OAB
The OAB-q  and OABSS  are two evaluation tools for OAB for which the reliability and validity have been verified. The OAB-q consists of 8 items regarding symptoms and 25 items for QOL, but it has a major drawback in that the evaluation is time-consuming. The OABSS is a symptom-focused questionnaire developed by Homma and colleagues in Japan. Since it consists of just 4 questions, evaluation can be performed in a much shorter time than for the OAB-q. The OABSS was used in the present study to avoid placing too much of a burden on the subjects.
The prevalence of OAB pre-surgery in the present study was 4.1%, which was almost the same as that in women in the general population aged over 40 years reported in a previous study (8.1%) . The present results showed, however, that the OAB rate increased significantly to 13.1% post-surgery, which is more than double that in the general population. Francesco et al. conducted urodynamic tests of 15 patients after total hysterectomy, and they reported the postoperative prevalence of OAB to be 27% . It is difficult to directly compare the results of their study with those from the present study due to differences in the evaluation method; however, the results are consistent in terms of the rate of OAB increasing post-surgery. On the other hand, the present study showed that the proportion of those in whom OAB was recognized for the first time post-surgery was 10.8%. However, there are no previous reports on the proportion of newly developed UI post-surgery, so the results in this study represent a new finding.
The effects of aging are thought likely to be the major reason for OAB pre-surgery. The aging mechanisms thought to give rise to OAB include a decrease in the bladder relaxation response and a weakening of the pelvic floor due to disturbance of blood flow to the bladder . However, some consideration should also be given to the possible compression of the bladder and autonomic nerves by the tumor associated with the primary disease, and the tumors were localized in all cases.
With regard to the development of OAB postoperatively, we considered that the parasympathetic pelvic nerve was dominant because the sympathetic hypogastric nerve was damaged due to surgical stress from the invasive intervention, and it is therefore possible that α receptors led to the relaxation of the neck of the bladder, and β receptors induced contraction of the body of the bladder .
3. Effects of UI/OAB on QOL
UI and OAB are both pathological conditions known to greatly impair QOL, particularly in women, who are known to experience adverse physical, emotional, and social effects. In a previous study that used the IIQ-7 to evaluate UI in 28 cervical cancer patients, the total score pre-surgery was 4.7±0.8, and that at 6 months postoperatively was 10.9±1.0, indicating that patient QOL was significantly worsened after surgery . In the present study, IIQ-7 scores were compared by UI classification for patients with UI post-surgery. This is the first report of such a comparison, and the present results showed the scores for UI classifications to be in the following order: mixed < urge < stress < overflow incontinence, with the IIQ-7 score for mixed incontinence being significantly lower than that for stress incontinence. The reason for this is that the frequency of UI in cases of stress incontinence can be reduced to a certain extent through one’s own behavior, whereas urgency cannot be controlled to a similar degree in cases of urge incontinence. A previous study that compared QOL by UI classification in the general female population reported that the QOL for mixed incontinence was lower than that for stress incontinence, which supports our hypothesis.
Furthermore, all subscale scores were high for stress, urge, and mixed incontinence; however, for overflow incontinence, the subscale scores were high for travel/outing/social life, but zero (0) for physical activity and emotional impact. Overflow incontinence involves an increase in the volume of the bladder content, and this stored urine leaks out, resulting in “overflow”. The increased content can be compensated for by increasing the frequency of urination, so the patients consider the problem to be less severe, and the impact on their QOL is reduced.
The relationships of many factors, such as the frequency and extent of physical activity, working conditions, and the ability to cope with UI, are yet to be established, and further study is needed.
4. Risk factors for the onset of UI post-surgery
In the present study, the number of Cesarean sections and the days of urinary bladder catheterization post-surgery were identified as risk factors for UI post-surgery. Studies of the general female population showed that women with vaginal delivery have a higher frequency of UI compared to nullipara or those delivering by Cesarean section [33, 34]. It is thought that the increased rate of onset observed in the vaginal delivery group is due to neural damage to the pubococcygeus muscle during delivery  and injury to the pudendal nerve . On the other hand, a study of UI in 505 pregnant women followed for 3 months after delivery showed that the incidence of UI was significantly lower in the Cesarean section group than in the vaginal delivery group. However, it was reported that there was no significant difference in the rate of UI between those in the vaginal delivery group and women having 3 or more Cesarean sections . The fact that a higher number of Cesarean sections leads to a higher incidence of UI can be explained by the invasion of the abdominal wall during surgery. Repeated surgical invasion of the abdominal wall reduces the activity of the abdominal muscles, which then become unable to support the abdominal wall, resulting in lumbar lordosis. The condition in which the abdomen is extended due to the lumber lordosis acts to lower the pressure in the urethra, leading to UI .
With regard to the days of urinary bladder catheterization post-surgery, a longer period of catheterization can lead to urethral mucosal irritation or bladder irritation due to urinary tract infections. Such bladder irritation causes the bladder to suppress uncontrolled contractions, thereby resulting in urine leakage.
5. Limitations and future issues
This study was conducted by postal questionnaire survey, with subjects responding to items covering frequency and volume of UI and the frequency of daytime and nighttime urination based on their situation pre-surgery, so the accuracy and reproducibility could be low. The time between surgery and questionnaire response was quite long (mean 839.6 ± 48.6 days), and this could cause some bias with regard to the patients’ ability to recall pre-surgery symptoms or overvalue post-surgery symptoms. In addition, the necessary population size of the UI classification post-surgery of the subjects who had incontinence pre-surgery for this study calculated using G*Power version 184.108.40.206 for Windows was n = 48. However, the statistical analysis of the UI patients was based on a population of only 34, so that the potential for type 2 error cannot be excluded. Thus, it was not possible to conduct a meaningful subgroup analysis by cancer type. Further, many of the patients who did not return their responses may have failed to do so due to feelings of shame about their current situation regarding UI. In order to accurately understand the situation regarding the onset and causes of UI pre-surgery, it is necessary to carry out future prospective studies to observe patients pre-surgery. Nevertheless, the present study identified patients who newly experienced UI post-surgery, and this is the first study to show such findings.