Urinary and faecal incontinence following delayed primary repair of obstetric genital fistula

Transcription

1 BJOG: an International Journal of Obstetrics and Gynaecology July 2002, Vol. 109, pp Urinary and faecal incontinence following delayed primary repair of obstetric genital fistula Christine Murray, Judith T. Goh, Michelle Fynes, Marcus P. Carey* Objective To evaluate: (1) the factors associated with the development of obstetric genitourinary fistula, (2) the incidence of urinary and faecal incontinence following closure of the fistula and (3) the urodynamic findings in women with persistent urinary incontinence. Design An observational clinical study. Setting A specialised fistula unit in a developing country. Population Women following successful anatomical closure of obstetric genitourinary fistula. Methods Fifty-five women were enrolled from the Fistula Hospital in Ethiopia, following obstetric fistula repair. Their case records were reviewed and details regarding (1) antecedent obstetric factors, (2) the site, size and type of fistula and (3) pre-operative bladder neck mobility and vaginal scarring were recorded. All women were questioned regarding symptoms of faecal and urinary incontinence. Women reporting urinary incontinence following fistula repair underwent urodynamic investigations. Main outcome measures Clinical and urodynamic assessment. Results The mean age of the women was 23 years (range years). The fistula in 38 women (69%) followed the first delivery and in 17 women (31%) following a subsequent delivery. The mean duration of labour was four days (range 1 9 days). Forty-four women (80%) had an isolated vesico-vaginal fistula and 11 (20%) had a combined vesico-vaginal and recto-vaginal fistula. The mean diameter of the fistula was 2.9 cm (0.5 6 cm). Successful repair occurred in all women. Thirty women (55%) reported persistent urinary incontinence and 21 (38%) altered faecal continence at follow up. In the former group, urodynamic investigations identified genuine stress incontinence in 17 women (31%), detrusor instability in two (4%) and mixed incontinence in 11 (20%). Conclusion This study demonstrates the high rate of successful closure of the fistula in a specialised fistula unit, but highlights the problem of persistent urinary incontinence following closure. INTRODUCTION In East Africa, the maternal mortality rate is estimated at per 100,000 live births and fewer than 15% of these women receive antenatal care 1. Genitourinary fistula is a common complication of childbirth, occurring in 3 4 per 1000 deliveries 2 (Fig. 1). The most common risk factors leading to obstetric fistula are first delivery and prolonged labour 1 4. The Fistula Hospital in Addis Ababa was established by Drs Reginald and Catherine Hamlin in 1975 (Fig. 2). This unit consists of 60 inpatient and 50 hostel beds. The latter are used to accommodate women who require nutritional supplementation pre-operatively or prolonged post-operative care. Over 15,000 obstetric vesico-vaginal or recto-vaginal fistulae have been repaired since this unit opened 1,2. The majority of these have been repaired vaginally under regional anaesthetic, and the success rate is more than 90% 1. Despite this high success rate, persistent urinary and faecal incontinence is commonly reported following surgery 2. The aims of this study were to evaluate (1) the factors associated with the development of obstetric genitourinary fistula, (2) the incidence of persistent urinary incontinence following closure of the fistula and the findings on urodynamic investigation in those affected and (3) the incidence of altered faecal continence in women following repair of the fistula. METHODS Department of Urogynaecology, The Mercy Hospital for Women, Melbourne, Australia The Fistula Hospital and Royal Women s Hospital, Addis Ababa, Ethiopia * Correspondence: Dr M. Carey, Frances Perry House, Suite D, Level 10, 286 Cardigan Street, Carlton 3053, Victoria, Australia. D RCOG 2002 BJOG: an International Journal of Obstetrics and Gynaecology PII: S (02) This was an uncontrolled series of cases studied during a five-week period at the Fistula Hospital in Addis Ababa, Ethiopia in All women who were more than four weeks and less than three months following delayed primary closure of the fistula during this time were enrolled. The women were recruited from the outpatient clinic and the

2 URINARY AND FAECAL INCONTINENCE AFTER REPAIR OF FISTULA 829 Table 1. et al. 5 ). Faecal continence scoring system (modified from Pescatori Faecal incontinence Never Monthly Weekly Daily Always Faecal staining Poor flatal control Frank faecal incontinence Interference with daily life Fig. 1. Urethral sound demonstrating a large vesico-vaginal fistula. hospital hostel. These women were still in the hospital for a number of reasons including awaiting relatives or transport and continuing therapy for other medical problems, such as footdrop. The Fistula Hospital in Addis Ababa has kept a detailed filing system since it opened in Each woman is required to complete a form with their demographic, obstetric and medical details, using an interpreter if necessary. Information about the operation was added by the medical staff, in particular (1) antecedent obstetric factors, (2) the site, size and type of fistula and (3) the degree of bladder neck mobility and scarring of the vaginal wall. Successful closure or not was also recorded. All of the women were interviewed by a urogynaecology nurse practitioner (CM) with an interpreter as none of the women spoke English. We asked questions on postoperative urinary and faecal incontinence and judged the severity of faecal incontinence using the scoring system derived from Pescatori et al. 5 (Table 1). This scoring system was modified to exclude questions on wearing a pad, as pads were not available. Women with post-operative urinary incontinence underwent standard urodynamic assessment. Subtracted voiding cystometry was performed using a Browne MiniPro H 2 O Urodynamics Monitor 8100 (Timm Medical Technologies, Maine, USA), with a 3.3 mm double lumen urethral catheter and a 4.7 mm rectal catheter (Timm Medical Technologies). Urodynamic diagnosis was established according to the criteria of the International Continence Society 6. Fig. 2. The post-operative ward at the Fistula Hospital in Addis Ababa, Ethiopia.

3 830 C. MURRAY ET AL. Statistical analysis was performed using the SPSS for Windows statistical software package (IBM, Redmond, Washington, USA). RESULTS The mean age of the women was 23 years (range years). Thirty-eight fistulae (69%) followed the first delivery and 17 (31%) followed the second or subsequent delivery. Seven women (13%) had six or more previous deliveries. Forty-nine fistulae (89%) followed spontaneous vaginal delivery, three (5%) emergency caesarean section and three (5%) instrumental delivery for obstructed labour. The mean duration of labour was four days (range 1 9). Forty-four (80%) women had an isolated vesico-vaginal fistula and 11 (20%) had a combined vesico-vaginal and recto-vaginal fistula. None of the women had an isolated recto-vaginal fistula. The mean diameter of the vesicovaginal fistula was 2.9 cm (range cm). Thirty-nine (71%) had marked scarring of the vagina. The women underwent repair of the fistula under regional anaesthesia. In 52 women (95%), a Martius graft was used; while in three women (5%), appositional repair was performed. All of the latter group had a vesico-vaginal fistula more than 0.5 cm in diameter. All the women remained in the hospital on bedrest for 14 days with a urethral catheter on free drainage. On the 14th day, successful repair was established by clamping the catheter and ensuring that there was no leakage at the site of the fistula. If there was any doubt, a three swab test using methylene blue was performed at the bedside. Successful repair occurred in all the women. The median follow up time was eight weeks. Of the 55 women, 30 (55%) complained of persistent urinary incontinence and 21 (38%) had altered faecal continence. Of these, 13 women (24%) had symptoms of both urinary and faecal incontinence. Post-operative urinary incontinence was not more common following repair of a urethral compared with a vesical fistula. There was also no significant difference in the mean diameter of the fistula in women with post-operative urinary incontinence (mean 2 cm, range cm) compared with women with no urinary incontinence (mean Table 2. Faecal incontinence symptoms. Symptom Poor flatal control 18 Faecal staining 9 Frank incontinence 9 Faecal urgency <5 minutes 8 Defaecatory straining 9 Defaecatory pain 7 Bleeding on defaecation 2 n Table 3. Post-operative urodynamic findings. GSI ¼ genuine stress incontinence; DI ¼ detrusor instability. Voiding cystometry Incontinent (n = 30) 1.5 cm, range cm) ( P ¼ 0.3, Wilcoxon signed rank sum test). Twenty-one (38%) women with symptoms of altered faecal continence reported loss of flatus or faecal staining rather than frank faecal incontinence (Table 2). The mean faecal incontinence score in this group of women was 7 (range 2 16). Of the 21 women with post-operative faecal incontinence, three followed combined repair of a vesicovaginal and recto-vaginal fistula and 18 had a vesicovaginal fistula repair alone. Ten women who had a fistula repair at the vaginal vault reported altered faecal incontinence, compared with eight following bladder neck and three following urethral fistula closure. The mean post-operative stool frequency was one bowel movement (range 1 3) per day. Of the 21 women with altered faecal continence, nine reported normal stool frequency, nine had reduced stool and three reported alternating stool consistency. Nine women reported postoperative defaecatory difficulty, with six needing to evacuate digitally (Table 2). All 30 women with persistent post-operative urinary incontinence underwent urodynamic investigation (Table 3). Seventeen women (57%) had genuine stress incontinence, two (7%) had detrusor instability and 11 (37%) had mixed incontinence. Twenty-two of the 30 women (73%) had bladder hypersensitivity, four (13%) had voiding difficulty and 25 (83%) had a fixed bladder neck. DISCUSSION GSI (n ¼ 17) DI F GSI (n ¼ 13) First desire (ml) 110 (40 300) 94 (20 200) Strong desire (ml) 145 (0 300) 124 (50 270) Capacity (ml) 222 ( ) 198 ( ) Pressure rise on filling (cm H 2 O) 6.6 (2 12) 34 (15 80) Obstetric fistula is a common problem in countries with limited medical resources and access to antenatal care 1,2.In Ethiopia, this problem is compounded by many social and cultural factors which lead to delay in seeking obstetric intervention when labour is obstructed or prolonged 4. Specific cultural problems include marriage at an early age, often before skeletal maturity, illiteracy and lack of contraception, all of which result in high parity and poor nutritional status 1 4. This problem is exacerbated by the rural nature of Ethiopia, with lack of public transportation. Urinary and faecal incontinence is a devastating social handicap common in countries with limited medical resources 4. The aim of this study was to evaluate the

4 URINARY AND FAECAL INCONTINENCE AFTER REPAIR OF FISTULA 831 factors associated with obstetric genital fistulae and the incidence of urinary and faecal incontinence following delayed primary repair. The most common risk factors were prolonged labour and first delivery, a smaller number of fistulae being associated with emergency caesarean section and instrumental delivery for obstructed labour. Although the number of women in this study is small, our findings agree with those of other authors and suggest that the pathogenesis of genital fistula is pressure necrosis and ischaemic injury rather than direct mechanical or surgical trauma 1 4,7 11. Isolated vesico-vaginal fistula was more common than combined vesico-vaginal and recto-vaginal fistula, which also agrees with other reports 7 10,12. Forty-five of the women in our study had a vesico-vaginal fistula involving the bladder neck or vault. The high rate of bladder neck and vault fistula is most likely due to cephalopelvic disproportion 2,8 10. This problem may arise due to a combination of factors, including early age at first delivery, poor nutrition, skeletal dysmaturity and an android pelvis 2,9. Observational studies from Africa report successful closure of obstetric fistula in 75 95% of delayed women with primary repair 13,14. Despite the high rate of success, persistent urinary incontinence following fistula surgery has long been recognised, occurring in 10 12% of women 15. There have been no previous reports of the risk of post-operative faecal incontinence. A large study of women following fistula surgery with adequate follow up would allow a precise estimate of the prevalence of post-operative incontinence to be made. However, lack of infrastructure, limited hospital resources, poverty and illiteracy will prevent such a study in the developing world. In this study, all the women attending the follow up clinic for their first post-operative visit, or who were still in the hostel, were investigated. Although the number of women in the study was small, we think that the women are representative of all women attending the Fistula Hospital in Addis Ababa. This study gives reliable information on the frequency of urinary and faecal incontinence, at least in the short term, following successful repair of vesico-vaginal and recto-vaginal fistula. Twenty-eight of the 30 women with post-operative urinary incontinence had genuine stress incontinence 17 with genuine stress incontinence alone and 11 with mixed stress and urge incontinence. With the exception of three women who had a small fistula (<0.5 cm) and who underwent direct appositional repair, all of the remaining women had a Martius graft performed. The use of a vascular pedicle has increased the rate of successful closure from 70% to more than 90% 4. Hilton 15 suggests that interposition of this labial fat graft reduces the risk of post-operative stress incontinence by adding bulk to the bladder neck. This has not been our experience with the 26 women who had a Martius graft and who developed post-operative genuine stress incontinence. Our study shows the need for the urodynamic assessment in women with persistent urinary incontinence following fistula surgery. This would allow appropriate selection of women who require additional surgery for genuine stress incontinence. Only two women had detrusor instability alone. This is in contrast to the findings of other authors who have found detrusor instability in 20% of women following fistula surgery 14. This may reflect differing aetiology as previous studies have included women with fistulae secondary to gynaecological surgery and radiotherapy rather than purely obstetric causes 14. Faecal incontinence has been reported in association with repair of a vesico-vaginal fistula but the frequency of this problem is unclear 3,8,9,13. We have identified altered faecal continence in 21 women (38%) following repair of obstetric fistula, including women with frank faecal incontinence (Table 2). The mechanism of injury leading to faecal incontinence is unclear and may be different from that of postpartum faecal incontinence in the developed world. Several authors have noted footdrop in one-quarter of African women who have obstetric fistula 2,4. This finding, linked to the increased frequency of high obstruction in labour, may suggest a neuropathic rather than mechanical aetiology 1,2,8,9. This hypothesis remains to be tested. Obstetric genital fistula is still a major problem in countries with limited health resources, with 50, ,000 new cases reported each year 1,2. This study confirms the high rate of successful repair that can be achieved, but highlights the poor functional outcome, with a high frequency of faecal and urinary incontinence. While urodynamic assessment has shown that genuine stress incontinence is usually the cause of persistent urinary incontinence, the mechanism of faecal incontinence is poorly understood. Further research is required to investigate the mechanisms of faecal incontinence in these women. Acknowledgements We would like to thank Dr Catherine Hamlin for the opportunity to contribute to the work of the Fistula Hospital in Addis Ababa. We would also like to acknowledge her enormous contribution and lifelong dedication to the prevention and treatment of obstetric genitourinary fistula in Ethiopia. References 1. Goh JT. Genital tract fistula repair on 116 women. Aust N Z J Obstet Gynaecol 1998;38(2): Margolis T, Mercer LJ. Vesicovaginal fistula. Obstet Gynecol Surv 1994;49: Danso KA, Markey JO, Wall LL, Elkins TE. The epidemiology of genitourinary fistulae in Kumasi, Ghana, Int Urogynaecol J Pelvic Floor Dysfunct 1993;7(3): Mathelaer JJ, Williams G. Extraurethral urinary incontinence after incompetent vaginal obstetrics. Br J Urol 1999;84(1):10 13.

5 832 C. MURRAY ET AL. 5. Pescatori M, Anastasio G, Bottini C, Menbsti A. New grading and scoring for anal incontinence. Evaluation of 335 patients. Dis Colon Rectum 1992;35: International Continence Society. The standardisation of terminology of lower urinary tract function. Obstet Gynecol 1990;56: Steiner AK. The problem of post-partum fistulas in developing countries. Acta Trop 1996;62: Goh JT, Krause HG Brown Craig Travelling Fellowship destination: Ethiopia. Aust N Z J Obstet Gynaecol 1996;36(3): Gharoro EP, Abedi HO. Vesicovaginal fistula in Benin City, Nigeria. Int J Obstet Gynaecol 1999;64(3): Amr MF. Vesico-vaginal fistula in Jordan. Eur J Obstet, Gynaecol Reprod Biol 1998;80(2): Elkins TE. Fistula surgery: past, present and future directions. Int Urogynaecol J Pelvic Floor Dysfunct 1997;8(1): Hilton P, Ward A. Epidemiological and surgical aspects of urogenital fistula: a review of 25 years experience in south east Nigeria. Int Urogynaecol J Pelvic Floor Dysfunct 1998;9(4): Woo HH, Rosario DJ, Chappler CR. The treatment of vesico-vaginal fistulae. Eur Urol 1996;29(1): Hilton P. Urodynamic findings in patients with urogenital fistulae. Br J Urol 1998;81(4): Hilton P. Debate: postoperative urinary fistulae should be managed by gynaecologists in specialist centers. Br J Urol 1997;80(Suppl 1): Accepted 12 February 2002