Quality of life after tension-free vaginal tape surgery for female stress incontinence

Objective. To undertake a long-term follow-up evaluation of the quality of life (QOL) of women who had undergone a tension-free vaginal tape (TVT) procedure. Material and methods. During the period 1995–2001, 970 women with urinary stress incontinence underwent TVT surgery at the Department of Obstetrics and Gynecology, Falun Hospital. A questionnaire was mailed on average 5.7 years after the TVT procedure. Two incontinence-specific QOL instruments—the Incontinence Impact Questionnaire-7 (IIQ-7) and the Urogenital Distress Inventory-6 (UDI-6)—were administered. An additional questionnaire included general questions and questions about chronic diseases that may be associated with urinary incontinence. Results. The mean age at surgery was 58.7 years (range 29–89 years). Of 913 eligible women, 768 (78.9%) responded. Mean IIQ-7 and IDU-6 scores as estimated by the women improved dramatically at follow-up as compared to preoperative values: from 43.7 to 11.5 for the IIQ-7 and from 54.2 to 24.0 for the UDI-6 on a scale from 0 to 100 (p = 0.0001 for both). There were few differences in mean QOL scores even 8 years after TVT surgery, compared to those determined a shorter time after the operation. Women with diabetes, chronic constipation, chronic bronchitis and preoperative recurrent urinary infections had a relative improvement in QOL of the same magnitude as that of the remaining study population. Advanced age was negatively associated with an improvement in QOL scores. Conclusions. Improvements in measures of QOL after TVT surgery are dramatic and persist for years. Women with concomitant diseases that may be associated with urinary incontinence can be assured that there is a good chance of success with TVT surgery.

Keywords: Incontinence Impact Questionnaire-7; quality of life; stress incontinence; tension-free vaginal tape surgery; Urogenital Distress Inventory-6

Quality of life (QOL) is linked to health as defined by the WHO [1], i.e. "not merely the absence of disease, but complete physical, mental and social well-being". In this respect, health is dependent on the patient's own perspective and is influenced by personal and cultural values. Medical testing that defines an "objective cure rate or improvement" is at best similar to the patient's own opinion. This does not mean that medical tests are of limited value. On the contrary, they are useful and practical in many situations, such as follow-ups of results and comparisons between different treatment methods. Without asking the patient, however, it is difficult to evaluate the efficacy of a medical procedure or find out if it is useful at all.

The diagnosis of urinary incontinence defined by Abrams et al. [2] as "a condition in which involuntary loss of urine is objectively demonstrable and is a social or hygienic problem" can hardly be made without measuring QOL. Impairment of QOL is "the inevitable consequence of urinary incontinence" [3]. In general, it is the patient who is seeking help, and his/her perspective should consequently be the main "instrument" used to evaluate clinical success.

QOL is generally measured using validated questionnaires, the scores of which can be estimated with simple formulas [4]. Visual analog scales have proved less useful. The questionnaire should be simple, relevant and capable of rapid completion. Usually, questions are divided into domains, such as daily activities, social interaction and self-perception. Questions concern everyday activities, feelings, etc. and are easy to answer. In clinical practice it is important to limit the number of questions. Condition-specific questionnaires usually include no more than 30 questions. Scores are generally given for each domain, together with a total score [3].

There are generic and condition-specific instruments. The former are designed to measure QOL in general and can be used for different diseases as well as for healthy individuals. Their advantages are that they allow for comparisons between different patient groups and between patients and healthy individuals. However, they must be applicable for many conditions and thus their sensitivity for specific conditions is reduced. They can also be extensive and impractical in routine use.

A variety of QOL instruments designed for urinary incontinence have been used. These include validated and reliable instruments, but also homemade questionnaires designed only for a specific study [4]. Shumaker et al. [5] designed and introduced the Incontinence Impact Questionnaire (IIQ) and the Urogenital Distress Inventory (UDI). The UDI consists of a list of symptoms associated with lower urinary tract dysfunction. The IIQ has a total of 30 items relating to the degree to which urinary incontinence affects a range of activities, mainly of a social nature. The questionnaires were evaluated and showed good validity and reliability. Later, the same group made short forms of the IIQ and UDI, namely the IIQ-7 and UDI-6, respectively, the names indicating the number of questions [6]. The UDI-6 can be divided into three subscales, measuring irritative, stress and obstructive/discomfort symptoms. The results of both short instruments showed good correlations with their longer versions.

Tension-free vaginal tape (TVT) is currently the most widely used surgical intervention for female urinary incontinence [7]. The aim of this study was to evaluate QOL by means of long-term follow-up of the TVT procedure in a large cohort of women. The size of the study would possibly enable evaluation of women with concurrent diseases. The instruments used were the IIQ-7 and UDI-6 (see Appendix).

The study includes 970 consecutive TVT procedures performed at the Department of Obstetrics and Gynecology, Falun Hospital between October 1995 and December 2001. Initially, one surgeon, trained at the Department of Obstetrics and Gynecology, University Hospital, Uppsala, where the TVT procedure was initiated and developed, performed the operations. Under his guidance, experienced gynecologists soon began to practice the TVT procedure. Today, most of our gynecological surgeons perform the operation. In total, ≈10 surgeons account for the TVT operations included in this study, but most of them were performed by just three surgeons.

Preoperatively, all women with a history of stress incontinence only underwent a routine assessment including a gynecological history, physical examination and stress test. Cystoscopy and cystometry, including determination of detrusor pressure, were performed as appropriate, generally when there was a history of mixed incontinence. A complete urodynamic evaluation was made in 156 patients (16.4%), in general because of urgency or suspicion of neurological disorder.

TVT surgery was carried out according to a previously described procedure [8]. In brief, patients were operated on under local anesthesia. When necessary, small doses of sedatives were given intravenously. When concomitant vaginal repair surgery was done, spinal anesthesia was used. The vaginal anterior wall was incised at the level of the mid-urethra. After minimal para-urethral dissection, a Prolene tape (Ethicon®) covered with a plastic sheath was introduced on both sides of the urethra using a previously described needle instrument [8]. The Prolene tape was brought retropubically through the abdominal wall above the pubic bone. Cystoscopy was performed to confirm an intact bladder. The plastic sheath was removed and, with the bladder filled with 300 ml of saline, the tape was adjusted to check continence while the patient coughed. The abdominal ends of the tape were cut and no fixation was performed. The TVT procedure was ambulatory, i.e. the patients were released within 24 h postoperatively when there were no complications and the patient could void twice with <100 ml of residual urine.

The UDI-6 and IIQ-7 were mailed to all women in early 2004 and a prepaid, addressed envelope was included. In addition, questions about chronic diseases, such as diabetes, neurological disorders, bronchitis and constipation, smoking, recurrent urinary tract infections, sex life and dyspareunia were asked. Non-respondents received a reminder 1 month after the initial request. The women were asked to recall the situation before their surgery and describe their present situation on separate questionnaires.

IIQ-7 and UDI-6 scores have four different response levels, describing how concerned patients are with different symptoms or situations, as follows: 0, not at all; 1, slightly; 2, moderately; and 3, greatly. Scores were calculated as the average score for all questions multiplied by a factor of 33.33, to give a best score of 0 and a worst score of 100. Continuous variables, such as IIQ-7 and UDI-6 scores, were compared using a t-test. Nominal variables, such as comparison of answers for individual questions, were analyzed by means of a χ2 test. When women with concomitant diseases or specific complaints were compared with the remaining study population (Tables III and IV), odds ratios and 95% CIs were adjusted for age and estimated by logistic regression. Percentage differences in QOL scores were compared as the relative change in score after compared to before TVT surgery, e.g. an improvement in QOL score from 50 to 25 represented a 50% improvement.

The study was approved by the Research Ethical Committee, Uppsala University.

The mean age at the time of the TVT procedure was 58.7 years (SD 12.3 years; range 29–89 years) and when the questionnaires were distributed it was 64.4 years, giving an average follow-up period of 5.7 years after surgery. Forty-nine patients had died and eight were too ill to answer the questionnaire, in general because of dementia. The questionnaires were completed by 768 of the 913 eligible patients, giving a response rate of 78.9%. The mean ages of the non-responders and responders were 62.9 and 64.5 years, respectively (p=0.14).

The mean IIQ-7 score before surgery was 43.7, compared to 11.5 at follow-up (p=0.0001). When the women were subgrouped in terms of the length of time since surgery (2–8 years), few differences were found between the subgroups (Figure 1).

Graph: Figure 1. QOL after TVT surgery, measured using the IIQ-7 and UDI-6, as a function of the length of time since the operation.

The questionnaire responses "not at all" or "slightly" were summarized into one variable, while the responses "moderately" and "greatly" were analyzed separately in the analyses of individual variables shown in Tables I and II. Detailed answers for all the questions in the IIQ-7 are given in Table I. The majority of the women were moderately or greatly affected by their urinary leakage during physical recreation preoperatively but it is notable that, in > 40% of them, entertainment activities and participation in social activities were also affected. Almost half of the women had a feeling of frustration before TVT surgery. At follow-up, only 10% of women had complaints. All comparisons between the situation before TVT surgery and at follow-up were highly significant.

Table I.  QOL (IIQ-7) before and after TVT surgery for female urinary stress incontinence (n=768).

101 ap<0.0001 for all pre- and postoperative comparisons.

Table II.  QOL (UDI-6) before and after TVT surgery for female urinary stress incontinence (n=768).

102 ap<0.0001 for all pre- and postoperative comparisons.

The mean UDI-6 score was 54.2 preoperatively and 24.0 at follow-up (p=0.0001). When the women were subgrouped in terms of the length of time since surgery (2–8 years), few differences were found between the subgroups (Figure 1). There was a small tendency towards a decline in mean UDI-6 score at 6 years since TVT surgery, but this was not comparable to the preoperative scores.

As expected, highly significant differences in response to all questions in the UDI-6 at follow-up compared to preoperatively were also found. This was also true for preoperative mixed incontinence. However, improvements in the two symptoms that were not directly related to incontinence, i.e. difficulties in emptying the bladder and lower abdominal or genital discomfort, were not numerically dramatic. Twenty-five women (3.3%) developed postoperative discomfort. It was not possible to establish the postoperative period at which discomfort appeared. These women had a similar mean age at TVT surgery (61.9 years; SD 12.3 years; range 36–85 years) compared to the remaining study population.

Tables III and IV analyze IIQ-7 and UDI-6 scores after the TVT procedure in defined subgroups of women. The Tables also compare relative changes, i.e. the percentage change in the scores at follow-up compared to before TVT surgery, in these subgroups of women compared to the remaining study population. A direct comparison of the scores enabled analysis of the actual differences in QOL between the subgroups of women and the remaining women. The analyses were adjusted for age.

Table III.  Change in QOL (IIQ-7a) in patients with other complaints before and after TVT surgery for female urinary stress incontinence and comparison with the remaining study population.

103 aThe IIQ-7 score was dichotomized into <0 and ≥0 (median) in order to estimate odds ratios and 95% CIs. The mean difference was 72.0%. bOdds ratios, 95% CIs and p-values were adjusted for age. OR = odds ratio;% diff.=percentage difference.

Table IV.  Change in QOL (UDI-6a) in patients with other complaints before and after TVT surgery for female urinary stress incontinence and comparison with the remaining study population.

104 aThe UDI-6 score was dichotomized into < 15 and ≥ 15 (median) in order to estimate odds ratios and 95% CIs. The mean difference was 53.1%. bOdds ratios, 95% CIs and p-values were adjusted for age. OR = odds ratio;% diff.=percentage difference.

Advanced age was not significantly correlated with postoperative QOL scores. Chronic diseases that were possible to analyze were diabetes, constipation and bronchitis. There were no significant differences between diabetics and constipated women, in contrast to those with chronic bronchitis, and the remaining study population with regard to both IIQ-7 and UDI-6 scores after TVT surgery. Some of these subgroups of women were rather small and the actual mean score may, therefore, differ substantially from that for the remaining study population. Improvements after TVT surgery were, however, of the same magnitude in all subgroups of women. Smoking was not significantly associated with a lower QOL score or with improvement after TVT surgery.

Finally, pre- or postoperative recurrent urinary tract infections and dyspareunia were significantly negatively associated with IIQ-7 and UDI-6 scores, while an improvement in sex life was associated with a good QOL. The impact of favorable postoperative IIQ-7 and UDI-6 scores was also reflected in a significantly improved QOL with improved sex life and a relatively lower QOL with dyspareunia, as compared to that for the remaining study population.

The main strengths of this study were the large study population, the high response rate and the long follow-up period. In contrast to previous studies these enabled us both to study differences as a function of time since the TVT procedure and to study subgroups of patients with specific diseases or postoperative complaints that could be associated with incontinence, which formed the basis of our main findings. We could also analyze improvements and deteriorations that may have occurred years after the TVT procedure and report the answers to each specific question in the IIQ-7 and UDI-6.

A weakness of the pre- and postoperative comparisons was that the IIQ-7 and UDI-6 were not given to the patients preoperatively, but were answered at the same time as the follow-up questionnaires. This may have introduced a recall bias. On the one hand, it is not likely that the women would have forgotten the problems for which they sought medical care. On the other, one can speculate that there may have been a tendency for those women who were cured to exaggerate their preoperative problems, and for those with persistent complaints to ignore them. In contrast, opposite results were reported in a study of men with early-stage prostate cancer who were followed for up to 3 years postoperatively [9]. As none of those patients had symptoms of advanced cancer, women with urinary incontinence share some similarities in terms of urinary symptoms and sexual function with those men. As compared using two preoperative QOL questionnaires, one disease-specific and one general, the men significantly romanticized their preoperative problems for any measured score when the same questionnaires were used at postoperative follow-ups. The recall bias thus estimated the preoperative problems to be less when estimated postoperatively. We have not found a similar study that concerned women with urinary incontinence, but the male study indicates that the improvements found in our study were not exaggerated.

The magnitudes of preoperative and follow-up IIQ-7 and UDI-6 scores in this study were also very similar to those of other researchers [7], [10]. Older women seemed to show a relatively smaller improvement in QOL scores compared to younger women. These results are difficult to interpret as our investigation was conducted almost 6 years after the TVT procedure. The results could be biased by a natural decrease in QOL with increasing age, which is not so evident in younger women.

In this study we were also able to analyze some conditions and behaviors that may be important for the outcome of TVT surgery or are associated with urinary incontinence, such as diabetes, chronic bronchitis and constipation, and smoking. We believe that this facilitated interpretation of the results, when relative improvements were analyzed. It was encouraging to find that none of these conditions significantly affected the relative improvements in IIQ-7 (with the exception of chronic constipation) and UDI-6 scores at follow-up compared to preoperatively. As far as we know this is a novel finding. Women with postoperative recurrent urinary infections and dyspareunia, in contrast to those with an improved sex life, were, however, less improved at follow-up. The reasons for these impairments are unclear and we cannot discount the fact that they may have been caused by the TVT procedure.

A QOL questionnaire must fulfill some criteria. Content validity (are the questions relevant?), construct validity (is there an appropriate difference between asymptomatic and symptomatic patients?), criterion validity (is there an agreement with objective symptoms?), reliability (is there internal consistency between questions) and test–retest reliability must have been evaluated. There are a number of generic and urinary incontinence-specific QOL questionnaires that have been validated and tested for reliability, with varying results [4].

In the discussion that follows, we refer only to studies which dealt with TVT surgery, or another sling procedure in one case, and which included at least 100 study subjects.

The Bristol Female Lower Urinary Tract Symptoms questionnaire [11] was found to have high validity and reliability. The questionnaire consists of 33 questions, but a score cannot be estimated. A total of 170 women who had undergone the TVT procedure were assessed after 6 months [12]. Good or excellent improvements were found for most variables.

The Kings Health Questionnaire [13] was used to evaluate the TVT procedure in women aged < and > 70 years. This questionnaire is divided into domains that measure general health, impact on life of urinary symptoms, the severity of urinary incontinence and different urinary symptoms. The improvement after the TVT procedure in social and physical limitations, as compared to preoperatively, was equal in the two age groups.

The IIQ-7 and UDI-6 were used to assess QOL in 162 women at least 9 months (mean 22 months) after TVT surgery [10]. The difference in pre- and postoperative scores was, as mentioned above, very similar to ours. The mean preoperative IIQ-7 and UDI-6 scores were 39.5 and 48.1, respectively, while the corresponding postoperative scores were 10.6 and 20.6. The size of the study would not have allowed for analyses of women with concomitant diseases, but the postoperative improvements were similar for different types of stress incontinence.

In another study of 102 women [14], TVT surgery was not performed, but a pubovaginal sling was used instead. The women were followed up using the IIQ-7 and UDI-6 annually for 4 years postoperatively. The IIQ-7 score declined from 55.1 preoperatively to 11.0 at 12 months, and the corresponding scores for the UDI-6 were 67.1 and 28.0. The authors claimed that the improvement persisted for up to 48 months. Of the 102 women, however, 76 were assessed at 12 months, while only 13 could be assessed after 48 months, which make the long-term results inconclusive.

The main finding of our study was that the good results of TVT surgery and its great impact on QOL seem to persist for many years. It was possible for us to make these analyses due to the large study population, and because the time period since TVT surgery ranged from 2 to 8 years. The high response rate was a prerequisite for the conclusions drawn. The size of the study also allowed for analyses of women with concomitant diseases that may worsen the cure rate. According to our results, women can be reassured that cure from incontinence will persist for many years, irrespective of concomitant conditions. In addition, we obtained similar pre- and postoperative IIQ-7 and UDI-6 scores as those in previous studies with sufficient power, despite a different distribution of questionnaires.

Has urinary leakage affected your: 1. Ability to do household chores (cooking, housecleaning, laundry)? 2. Physical recreation, such as walking, swimming or other exercise? 3. Entertainment activities (movies, concerts, etc.)? 4. Ability to travel by car or bus for >30 min from home? 5. Participation in social activities outside your home? 6. Emotional health (nervousness, depression, etc.)? 7. Feeling of frustration?

Do you experience and, if so, how much are you bothered by: 1. Frequent urination? 2. Urine leakage related to the feeling of urgency? 3. Urine leakage as a result of physical activity, coughing or sneezing? 4. Small amounts of urine leakage (drops)? 5. Difficulty emptying your bladder? 6. Pain or discomfort in the lower abdominal or genital area?