Introduction
Hysterectomy is divided into 3 types: supracervical, total, and radical. A supracervical hysterectomy involves the removal of the uterine body itself. A total hysterectomy is the removal of the uterus along with the cervix. A radical hysterectomy is the removal of the uterus with the cervix and adjacent tissues. Each of these surgeries can be performed using different surgical methods: abdominal, vaginal, laparoscopic, or robotic.
The full names of the types of hysterectomy, along with the abbreviations used in the paper, are shown in Table 1.
Table 1
Types of hysterectomy and their abbreviations
According to the International Continence Society’s definitions, urinary incontinence (UI) is a complaint of involuntary loss of urine, while stress urinary incontinence (SUI) is a complaint of involuntary loss of urine upon effort or physical exertion including sporting activities, or upon sneezing or coughing. The World Health Organisation considers UI to be one of the most important health problems of the 21st century. It probably affects more than 200 million people worldwide, the vast majority of whom are women. The incidence of UI increases with age, so as the population ages, this problem will have an increasing impact on society’s quality of life [1].
Many studies show that hysterectomy is a risk factor for UI [2–4]. Abdominal hysterectomy (AH), along with older age and high body mass index (BMI), was among the top 3 etiological factors for SUI in women in Poland [5]. Undergoing a hysterectomy is associated with an increased risk of subsequent surgical treatment of UI [6–8]. Hysterectomy has an impact on the severity of this condition. Severe SUI was 6.3 times more common in women who had undergone a hysterectomy than in those who had not [9]. Stress urinary incontinence is more common in women who perform physical work and those who have had vaginal births. Long labor and multiple births further increase the risk of SUI [5].
The aim of this study was to review scientific articles on the correlation between different types of hysterectomy and the incidence of SUI.
Material and methods
Original open-access articles in English language describing the effects of different types of hysterectomy on the incidence of SUI were sought. The PubMed, Google Scholar, and ScienceDirect databases were used. Publications from the period 2004–2024 were considered.
Results
The type of hysterectomy affects the incidence of stress urinary incontinence
The influence of the surgical method of hysterectomy on the incidence of stress urinary incontinence
A cohort study (Forsgren et al.) including 118,601 women after hysterectomy and 578,200 women without hysterectomy was conducted in Sweden to assess the risk of SUI after hysterectomy. Vaginal hysterectomy (VH) was associated with a higher risk of SUI surgery than AH. The risk of SUI surgery was 5–6 times higher after a hysterectomy compared to women without hysterectomy [10].
An analysis of 5000 patients 10 years after hysterectomy by Tulokas et al. in Finland indicates that the risk of SUI surgery after VH is more than double that of AH. The association between VH and the occurrence of SUI remained significant even after other risk factors were considered and after an analysis including only women without preceding pelvic organ prolapse surgery. The study also showed that removal of the adnexa during hysterectomy does not affect post-operative UI [11].
Lakeman et al. conducted a prospective study of 430 patients in the Netherlands, following either VH (112) or AH (318) for benign indications, excluding pelvic organ prolapse. The women completed questionnaires before the surgery, and 6 months and 3 years after the surgery. Six months after hysterectomy, UI developed more frequently after VH than AH, and this difference increased with time [12].
In a Swedish cohort study (Bohlin et al.) including 16,182 patients, hysterectomy resulted in reduced UI symptoms after surgery, AH to a greater extent than VH. Improvement of UI symptoms after surgery was shown more often in women with a large uterus [13].
In a study conducted by Chmaj-Wierzchowska et al., 260 patients aged 27–82 years suffering from SUI without pelvic organ static abnormalities and no history of childbirth were observed. Patient history, and urogynaecological and urodynamic examination were conducted. It was unequivocally shown that age, BMI, and AH have the greatest impact on the incidence of SUI in pre- and post-menopausal women [5].
The study by Jędrzejczyk et al. involved 200 women aged 33–85 years who presented for urodynamic examination due to UI. In addition to the urogynaecological history, a urogynaecological examination and a full-profile urodynamic examination were performed. Among patients with a history of gynaecological-obstetric surgery (71 patients), AH was the most frequently performed procedure (29), followed by vaginal wall repair (18), caesarean section (11), surgery on adnexa (10), and the least frequently performed – VH (3). Stress urinary incontinence was found in 22 of 29 women after AH and in 2 of 3 women after VH. It has been recognised that among gynaecological-obstetric surgeries, AH predisposes most strongly to UI, especially SUI [14].
A retrospective study by Davidson et al. compared laparoscopic hysterectomy (LH), robot-assisted hysterectomy, and VH. Differences between minimally invasive sacrocolpopexy with simultaneous laparoscopic-assisted supracervical hysterectomy (LASH) or robotic-assisted supracervical hysterectomy and total laparoscopic hysterectomy (TLH) or total vaginal hysterectomy were analysed. The study included 161 women, 116 after supracervical hysterectomy (SCH) and 45 after VH, and data were collected through questionnaires distributed to patients. Vaginal hysterectomy was faster and was associated with a lower risk of pelvic organ prolapse recurrence but a higher risk of SUI after surgery [15].
A multicentre, retrospective cohort study by Cao et al. was conducted using data from 203 patients after hysterectomy for cervical cancer in stages IA2-IIB. The patients underwent urodynamic examinations. The analysis showed that LH reduces the risk of developing SUI compared to AH [16].
In opposition is a study by Wang et al. from China. The study included 169 patients after radical hysterectomy for cervical cancer. Urinary incontinence was identified using the International Consultation on Incontinence questionnaire. Urinary incontinence occurred more frequently after LH than after AH [17].
The influence of the extent of hysterectomy on the incidence of stress urinary incontinence
Aleixo et al. conducted a meta-analysis of studies on hysterectomy for benign indications: total hysterectomy (TH) and SCH. They included 4 studies with a total of 566 women. The follow-up period after surgery was 5–14 years, and data were collected using questionnaires. Analysis of 3 studies involving 384 women showed a higher incidence of UI after SCH. The same comparison result was also achieved in each of these studies separately. Stress urinary incontinence also occurred more frequently after SCH than TH, as analysed from 4 studies involving a total of 450 patients [18].
Andersen et al. conducted a randomised clinical trial in Denmark on patients who underwent hysterectomy for benign indications (excluding pelvic organ prolapse). The results were analysed at 1, 5, and 14 years after the procedure [19–21].
Follow-up one year after hysterectomy indicated a higher risk of UI after subtotal abdominal hysterectomy (SAH) than after total abdominal hysterectomy (TAH). Total abdominal hysterectomy was associated with a higher rate of symptom resolution and a lower rate of new symptom onset [19].
Follow-up of 234 patients 5 years after hysterectomy also showed a higher risk of UI after SAH than after TAH. However, more women after TAH than SAH reported that they had never experienced UI before [20].
Follow-up of 197 women 14 years after hysterectomy was included in the meta-analysis presented above. A higher risk of SUI was noted among women after SAH than after TAH. There were no differences in quality of life after SAH and TAH, but women with UI had lower quality of life [21].
A retrospective study by Shin et al. included 289 patients: 160 after TH for uterine fibroids, 57 after radical hysterectomy (RH) for cervical cancer, and 72 after RH along with radiation therapy for cervical cancer. A medical history and urodynamic examination were conducted, and 3-day micturition diaries were collected from the patients. It was shown that SUI was more common after TH than after RH, and it was least common after RH with radiation therapy. Patients after TH were more often operated on for UI than were patients after RH, and least often after RH with radiation therapy [22].
Gilbaz et al. collected micturition diaries, pad tests, urodynamic examinations, bladder wall thickness measurements, and King’s Health Questionnaires before hysterectomy, and 6 and 18 months after surgery. It was shown that TH does not lead to lower urinary tract symptoms, and even that the symptoms occurring before surgery can decrease after TH. Radical hysterectomy alone and combined with radiotherapy led to the development of UI [23].
Yuk et al. analysed the correlation between the type of hysterectomy performed due to uterine fibroids and SUI. The method of operation (AH, LH) and its scope (TH, RH) were considered. The study was conducted using health insurance data in South Korea, including 81,373 women after hysterectomy observed for an average of 7.9 years and 81,373 women in the control group observed for an average of 7.8 years. Stress urinary incontinence was defined as the occurrence of SUI surgery. Abdominal hysterectomy without adnexa removal was associated with a higher risk of subsequent UI surgery. Abdominal hysterectomy and LH with adnexa surgery were not associated with increased rates of UI surgery. Laparoscopic hysterectomy did not increase the likelihood of UI surgery, regardless of whether it was combined with the removal of adnexa or not [6].
The type of hysterectomy does not affect the incidence of stress urinary incontinence
The type of surgical method of hysterectomy does not influence the incidence of stress urinary incontinence
The study by Atakul et al. included 196 patients after hysterectomy for benign indications: 149 after AH and 47 after VH. Twelve months post-surgery, patients were asked about UI via phone, followed by confirmation using a cough test with a filled bladder (250 ml). De novo UI was diagnosed in 41 patients (20.9%), with similar rates after AH and VH. The frequencies of different types of UI were comparable in both groups [24].
Several studies compared UI frequency after AH and LH
Kluivers et al. conducted a randomised controlled trial with 38 patients per group (AH and LH). Before and 3 months after surgery, patients were asked about UI, and one year after surgery, they completed the Urogenital Distress Inventory, Incontinence Impact Questionnaire, and Defecatory Distress Inventory questionnaires. The results showed that 12 weeks after surgery, the incidence of UI decreased in both groups, with no significant differences between them [25].
Sukgen et al. found similar results in 52 post-AH and 34 post-LH patients, with UI symptoms decreasing post-surgery, but without statistical difference. Before and after the operation, patients filled out the Urogenital Distress Inventory form, were examined with the Q-tip test, and were evaluated according to the POP-Q scale [26].
A Chinese multicentre retrospective comparative study of 4895 patients (3539 LH, 1356 AH) also showed no significant difference in SUI incidence between these 2 methods. The procedures were performed for benign gynaecological diseases, CIN3, and endometrial hyperplasia. Patients completed the ICIQ-FLUTS questionnaire before and 12 months after the surgery [27].
A Swedish prospective clinical cohort study evaluated robotic-assisted total laparoscopic hysterectomy (RTLH), TAH, and TLH in 242 women undergoing hysterectomy for benign conditions. Pelvic floor function improved similarly across all techniques at 6 months and one year after surgery, based on Pelvic Floor Impact Questionnaire and Pelvic Floor Distress Inventory results [28].
In Germany, a study compared long-term satisfaction between 97 LH patients and 42 RTLH patients (Da Vinci-X system) for benign disease or early endometrial cancer. Using the Female Sexual Function Index and the Health-related Quality of Life Questionnaire by the EuroQol Group, no statistically significant differences in SUI incidence or severity were found between the groups [29].
Several studies compared multiple surgical methods.
In Sri Lanka, a multicentre randomised controlled clinical trial of 139 patients compared TAH, non-descent vaginal hysterectomy, and TLH, finding no significant differences in UI symptoms after 6 months. Urinary symptoms were assessed using the ICIQ-VS and ICIQ-FLUTS questionnaires [30].
At Wuhan University, a study of 260 patients across 5 surgical methods (46 TAH, 59 TLH, 42 VH, 78 abdominal intrafascial hysterectomy, and 35 laparoscopic intrafascial hysterectomy) showed no statistically significant differences in SUI 6 and 12 months after hysterectomy. Patients underwent pelvic examination, pelvic floor strength testing, were assessed on the POP-Q scale, and were asked to complete the Pelvic Floor Distress Inventory-short Form 20 questionnaire [31].
A retrospective cohort study assessed bladder function in women undergoing minimally invasive surgery for early-stage cervical cancer. The groups included laparoscopic-assisted vaginal radical hysterectomy (45 women), laparoscopic/robotic-assisted vaginal radical hysterectomy (61 women), and laparoscopic total mesometrial resection (25 women), with control groups of gynaecological patients (24) and urogynaecological patients (63). All surgical groups had higher bladder dysfunction rates than the control groups, and the incidence of SUI did not differ by method of surgery [32].
The extent of hysterectomy does not influence the incidence of stress urinary incontinence
A meta-analysis conducted by Robert et al. covered randomised trials with a minimum of one year of follow-up concerning the effects of SAH and TAH on UI. The MEDLINE, EMBASE, CINAHL, Biological Abstract, and Cochrane Library databases 1966–2007 were searched, and abstracts from major urogynaecology conferences from 2003–2005 and a bibliography of the resulting articles were analysed. After exclusion of ineligible studies, 3 studies remained, whose analysis showed that there was no statistically significant difference in the risk of developing SUI after SAH or TAH [33].
Ala-Nissilä et al. conducted 2 prospective studies to compare the long-term outcomes of SAH with TAH. The first study (19 years after surgery) included 193 patients. The patients’ hospital records were reviewed. Questionnaires were sent out, including questions about possible further surgeries or urinary dysfunctions. The second study (33 years after surgery) included 95 women (48 SAH and 47 TAH) from the first study. Urinary symptoms were assessed using the Urinary Incontinence Severity Score, Detrusor Instability Score, Incontinence Impact Questionnaire (short form), and Urogenital Distress Inventory (short form). Urinary incontinence severity was measured using a visual-analogue scale. At the follow-up visit, a gynaecological examination was performed according to the Pelvic Organ Prolapse Quantification system. They found that the types of hysterectomy studied were associated with similar post-operative outcomes in terms of subjective symptoms and subsequent gynaecological surgery at 33-year follow-up [34].
No differences between TH and SCH were shown in the above-mentioned studies by Lakeman et al. and Bohlin et al. [12, 13].
Ceccaroni et al. conducted a study to compare nerve-sparing hysterectomy with conventional radical hysterectomy for cervical cancer (FIGO IA2-IIB). Fifty-six women were included in the study, 25 after nerve-sparing hysterectomy and 31 after RH. Data for the study were collected through questionnaires. Stress urinary incontinence occurred with comparable frequency in both study groups [35].
Neither the surgical method nor the extent of hysterectomy influences the incidence of stress urinary incontinence
Skorupska et al. in Poland conducted a study to evaluate the impact of different types of hysterectomy on UI symptoms. The study included 500 women, 121 of whom underwent VH, 171 LASH, 96 SAH, 68 TAH, and 44 TLH. The main indications for the procedure were uterine fibroids, abnormal uterine bleeding, and pelvic organ prolapse. Before surgery and 12 months after surgery, patients completed the Urogenital Distress Inventory-6, the Incontinence Impact Questionnaire-7, and the International Consultation on Incontinence Questionnaire. The results showed that the surgical method and extent of hysterectomy did not affect the symptoms of UI [36].
The same author conducted another study involving 392 patients after hysterectomy for benign indications, including LASH, TAH, SAH, or VH. Women who claimed to have de novo UI 12 months after surgery were subjected to a cough test and a 24-hour pad test, and completed the King’s Health questionnaire. The results showed that 38% of patients developed UI and that there was no statistically significant difference between the incidence of UI according to the type of hysterectomy [37].
A single-centre retrospective study was conducted in Germany involving 981 women who had, at least 6 months prior, hysterectomy surgery for benign tumours – AH, VH, LASH, laparoscopic-assisted vaginal hysterectomy (LAVH), or TLH. Data were collected through questionnaires. The study found that long-term UI rates did not differ between the above types of hysterectomy [38].
Ellström Engh et al. analysed the Swedish national register for gynaecological surgery (1997–2002) to assess the impact of hysterectomy on UI. They compared TH (TAH/TLH, n = 198/116 and VH/LAVH, n = 265/7) and SCH (SAH/LASH, n = 163/86) with a control group of patients who underwent endometrial destruction (endometrial ablation, endometrial balloon treatment). Questionnaires were completed before and after surgery, and the follow-up period was 2 years. De novo UI (urgency UI, SUI, mixed UI) occurred in all groups, but there were no statistically significant differences based on the surgical method or extent of the procedure [39].
Discussion
The results of the analysed studies are divergent. Some of them find correlations between the type of hysterectomy and SUI, while others show no statistically significant differences.
The cited articles show that VH is associated with a higher risk of SUI than AH. These are good quality studies conducted on a large cohort of women undergoing hysterectomy – 118,601 in the Forsgren et al. study, 5000 in the Tulokas et al. study, and 430 in the Lakeman et al. study. However, they all have some limitations. The limitation of the studies by Forsgren et al. and Tulokas et al. is the recognition of the number of operations performed for SUI as a measure of the prevalence of SUI. Using such a method not all women with SUI are included. A limitation of the study by Lakeman et al. is the diagnosis of SUI based on questionnaires, a subjective method of assessment [10–12].
A different conclusion was presented by Jędrzejczyk et al., who claimed it is AH that is associated with a higher risk of SUI than VH. However, it is a work of lower quality. The study group consisted of 200 patients who presented with a referral for urodynamic examination due to UI. Selection into the group was determined only by the order in which the women applied. The group varied in age (33–85 years). Their obstetric history and UI risk factors were not included in the study. Of the 71 patients who had a history of gynaecological-obstetric surgery, the most common procedure was AH (29), and the least common was VH (3). Stress urinary incontinence was found in 22 out of 29 women after AH and in 2 out of 3 women after VH. Based on these data, the authors concluded that AH most strongly predisposes to UI, especially SUI. The study has the advantage of making a diagnosis of SUI based on urodynamic examination, an objective method of assessment. However, due to the above-described limitations of the study, the incomparable size of the study groups, the extremely small group of women after VH (3 women), and the fact that AH is by far the more commonly performed method of hysterectomy in Poland, the conclusion put forward seems unreliable [14].
There are also papers that found no statistically significant differences between the frequency of SUI after AH and VH [24, 30, 31, 36–38].
AH reduced symptoms of SUI to a greater extent than VH in women who already had them before the procedure. This effect may not have been related to the method itself, but to the presence of a large uterus that caused pressure on the bladder, and a large uterus was usually an indication for AH [13]. Also, Lakeman et al. observed that patients undergoing VH had, on average, a smaller uterine size than those undergoing AH [12].
There are scientific reports showing that AH is associated with a higher risk of SUI than LH. In a study by Yuk et al., it was found that LH, unlike AH, does not increase the risk of UI surgery in the future. This study has the advantage of an exceptionally large study group: 81,373 women after hysterectomy. A limitation of the study is defining SUI as the occurrence of SUI surgery [6]. In a study by Cao et al., LH also proved superior. The advantage of this study is the group of 203 women undergoing hysterectomy and the use of urodynamic examination [16]. The more favourable effects of LH are explained by the possibility of potentially sparing autonomic nerves during such surgeries due to better visualisation of the operated area [6, 16]. In addition, AH is more often chosen for surgery of large uterine fibroids, which can result in greater tissue and nerve damage [6].
In the article by Wang et al., a completely opposite conclusion was presented. It was shown that it is LH that is associated with a higher risk of UI after surgery than AH. This was a questionnaire-based study involving 169 post-hysterectomy patients, which may provide weaker evidence [17]. There are also studies that found no statistically significant differences between LH and AH [25–28, 30, 31, 36–39]. Some of them are good-quality trials with a large study group. In one study, in which an improvement in UI symptoms was observed after hysterectomy, with no statistically significant difference between the methods by which the procedure was performed, patients were examined 3 months and one year after surgery. During the second examination, although statistical significance was not obtained, the improvement seemed more pronounced in the group of women after LH [25]. This suggests that perhaps with longer follow-up, a statistically significant result would have been obtained, as in a study in which patients were examined between 3 and 24 months after surgery [16].
When examining cervical cancer patients, Cao et al. observed that AH was connected with a higher risk of SUI than LH, while Wang et al. conversely noted a lower risk of SUI [16, 17]. Nevertheless, currently, according to the European Society of Gynaecological Oncology, the European Society for Radiotherapy and Oncology, and the European Society of Pathology, laparotomy is the standard approach for the surgical treatment of cervical cancer patients because it is associated with better patient survival [40].
Performing hysterectomy in patients with stage IIB cervical cancer in some studies is controversial [16, 35]. Current guidelines recommend surgical treatment for stages IB1-IIA1, while stage IIB is not considered an indication for surgery [40].
The results of studies comparing TH and SCH are inconclusive. A meta-analysis of randomised trials conducted by Aleixo et al., and other studies, show fewer UI complications after TH than after SCH [18–21]. The authors of one study suggest that this may be due to vaginal suspension, which is performed during most TH surgeries and is not a standard procedure during SCH. It is suggested that this additional procedure provides more support for the bladder neck and reduces its mobility, hence the lower risk of UI after surgery [21].
In contrast, a meta-analysis of randomised trials by Robert et al. and other studies found no statistically significant differences between TH and SCH [12, 13, 33, 34, 36–39]. The papers claiming the superiority of TH over SCH, as well as most of those finding no difference between these types of hysterectomy, use questionnaires as a method of assessment, providing similar strength of evidence and conducted on comparably large groups of women [12, 18–21, 33, 34, 36, 38, 39].
The meta-analysis conducted by Aleixo et al. covered 4 randomised trials comparing SUI after TH and SCH [18]. It included 3 studies that found no statistically significant differences between these surgical techniques – the study by Greer et al., Thakar et al., and Persson et al., as well as one study by Andersen et al. described 3 articles, of which only one achieved a statistically significant difference [21, 41–45]. Interestingly, although most of the studies included in the meta-analysis did not yield statistically significant correlations, the meta-analysis showed a higher incidence of SUI after SCH than after TH, and it was a statistically significant result.
The results of one study were analysed twice using different statistical methods and published as 2 papers with opposite conclusions. One argued that TH is less likely to cause UI than SCH, and another argued that there are no statistically significant differences between the techniques [21, 41]. The rates of SUI in this study were slightly higher in the SCH group. Initially, this difference was statistically significant, but after multiple imputation analyses, the correlation was not confirmed, and the final conclusion was that there were no statistically significant differences between TH and SCH [41].
Some of the patients participating in this study were enrolled in the next stage. There were no differences in the results of the PFDI-20 questionnaire, the micturition diary, the incontinence test, or other tests for SUI between the groups. It was noted, however, that significantly more women with UI symptoms were enrolled in the follow-up study than among women who completed the first initial questionnaire, which may also have an influence on the final outcome [42].
Thus, reports suggest the superiority of TH over SCH, but given the studies that have not found such correlations, such a conclusion should be approached with caution.
Opinions concerning the impact of adnexa removal on UI are divided. Yuk et al. showed that AH without the removal of the adnexa is associated with a higher risk of subsequent UI surgery, as opposed to AH and LH combined with the removal of the adnexa. The authors suggest that adnexa preservation may be associated with higher serum oestrogen levels, which may promote the development of SUI. This survey has the advantage of a very large group of patients (81,373). The limitation is defining SUI as the occurrence of SUI surgery [6]. In contrast, in a study by Tulokas et al., the removal of the adnexa did not affect the occurrence of post-operative UI. This is a good-quality study with a large cohort (5000), in which, as in the previous one, the number of UI operations was assessed. According to the authors, the error that appears in univariate studies concerning the removal of adnexa is because they are removed during transabdominal access via laparotomy or laparoscopy, and, according to the paper cited above, the transabdominal method alone is associated with a lower risk of UI [11].
No statistically significant differences were found between LH and RTLH, although one might have expected that robotic procedures, given their even greater precision, would cause less tissue damage and thus fewer complications in the form of UI [28, 29]. The authors suggest that this may have been the result of less surgeons’ experience in robotic surgery. In addition, the study groups were small and unequal in number, with 97 women subjected to LH and 42 to RTLH. It was noted that rates of SUI were higher in the laparoscopic group, although not statistically significant. Perhaps in a study with a larger group undergoing robotic surgery and with more experienced operators, this correlation would be statistically significant in favour of robotic surgery [29].
More good-quality studies on the subject are necessary: randomised studies on a large cohort of patients, using multiple diagnostic methods simultaneously, preferably objective ones. Patients should be examined before surgery, as well as several times after surgery during long-term follow-up. In addition to original research, large meta-analyses of methodologically similar studies are necessary.
Conclusions
Most of the reviewed studies suggest the following:
Vaginal hysterectomy is associated with a higher risk of SUI than abdominal hysterectomy.
Abdominal hysterectomy is associated with a higher risk of SUI than laparoscopic hysterectomy.
It is likely that supracervical hysterectomy is associated with a higher risk of SUI than total hysterectomy. However, some scientific reports do not show significant differences in SUI between the above types of hysterectomy, so the first 3 conclusions should be approached with caution.
It is uncertain whether the removal of the adnexa during hysterectomy increases the risk of SUI.
There were no differences between traditional laparoscopic hysterectomy and robotic-assisted laparoscopic hysterectomy.
Currently, there are no definitive recommendations for the optimal hysterectomy approach. The preferred method is to remove the uterus through the vagina. However, the evolution of minimally invasive laparoscopic techniques is changing the incidence of particular types of surgical procedures. A robust comparative analysis of postoperative complications associated with different techniques is crucial for formulating future surgical guidelines.
More good-quality studies on the effect of type of hysterectomy on the incidence of SUI are necessary.
