SEARCH

SEARCH BY CITATION

Keywords:

  • Circumferential;
  • incontinent;
  • obstetric fistula;
  • surgery;
  • urine

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Obstetric vesicovaginal fistula is a tragic injury that is widespread in any country where access to emergency obstetric care is limited. More and more people are now interested in treating and caring for these women, but little is known about surgical management, and most operators have developed their own methods through years of experience. One of the more challenging cases to get a functional repair is in those women who have a circumferential defect in the bladder/urethra. This article analyses a series of 77 consecutive circumferential obstetric fistula cases operated in the Barhirdar Hamlin Fistula Centre by the author.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

The circumferential vesicovaginal or urethra–vaginal fistula is a common condition that requires considerable experience to obtain a functional and satisfactory surgical result. There is very little if anything written about the characteristics of these women and again very little about the methods of repair.

The circumferential fistula (as the name implies) has the whole circumference of the bladder and/or urethra destroyed so that the urethra is completely detached from the bladder (Figures 1 and 2).

image

Figure 1. Circumeferential fistula, the bladder is completely detached from the urethra. Adapted from: First Steps in Vesico-vaginal Fistula Repair. Brain Hancock, 2005, p5. With permission of The Royal Society of Medicine Press.

Download figure to PowerPoint

image

Figure 2. Circumferentail fistula, a 2cm gap between the proximal urethra and entry into the bladder.

Download figure to PowerPoint

The circumferential injury most often involves the urethra, and during the ischaemic process that developed the injury from a long unrelieved obstructed labour, the continence mechanisms are invariably destroyed. An anatomical closure is quite easy to obtain, but a more functional continent repair is more difficult. Currently, there are a variety of ways to repair them, and each method is practised by different surgeons across the developing world. The first option is merely to mobilise the posterior wall of the bladder off the vagina until enough mobilisation is obtained to close the bladder tissue directly to the posterior symphysis pubis and to the urethra. If the fistula remains closed after the operation, then the urethra, which is usually short, does not have any physiological function. Superior to the anterior portion of the urethrs, a thin membrane of epithelia lies over the bone and cartilage of the symphysis pubis. This now constitutes the anterior wall of the lower bladder and upper urethra. In these cases, up to 100% of women will have severe urethral incontinence via an incompetent urethra (Figure 3).1

image

Figure 3. Circumferential fistula, the posterior bladder/urethra repaired. Adapted from: First Steps in Vesico-vaginal Fistula Repair. Brain Hancock, 2005, p5. With permission of The Royal Society of Medicine Press.

Download figure to PowerPoint

The second method is to dissect the bladder free from the vagina and also from lateral and anterior attachments and thus advance the bladder in a circumferential fashion to the urethra and anastomose it directly onto it. This often has the dilemma of suturing a large defect in the bladder onto a narrow urethra, and it still remains that a short urethra will render the woman incontinent after repair (Figure 4).

image

Figure 4. Circumferentail fistula, the large bladder defect attached directly to the shortened urethra. Adapted from: First Steps in Vesico-vaginal Fistula Repair. Brain Hancock, 2005, p5. With permission of The Royal Society of Medicine Press.

Download figure to PowerPoint

The third option is similar to the second and is practised more widely by experienced fistula surgeons. The bladder is mobilised circumferentially, and the anterior bladder is wrapped around the urethra to anastomose it, with the remaining defect repaired in a longitudinal fashion (Figure 5).

image

Figure 5. Circumferentail fistula, the large bladder defect wrapped around the urethra and remaining defect repaired in a longitudinal manner. Adapted from: First Steps in Vesico-vaginal Fistula Repair. Brain Hancock, 2005, p5. With permission of The Royal Society of Medicine Press.

Download figure to PowerPoint

The fourth option is used for women who have more extensive injuries in which less than 1 cm of urethra remains, with at least the entire proximal three-quarters of the urethra destroyed in a circumferential manner. In these cases, a flap of anterior bladder is developed and tubularised to form a new urethra and either anastomised to the urethral remnant or to the new meatus if none available.2,3

This series used the third and fourth options where appropriate using the principles of attempting to maintain the urethral length and supporting the urethra with a sling of pubococcygeus muscle.1

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

All women with obstetric fistula admitted to the Barhirdar Hamlin Fistula Centre from February 2006 until September 2006 were involved in the review. All women had their demographic details recorded, such as age, parity, age at delivery, length of labour and delivery method. Results were statistically analysed with either the chi-square test or two-sample t test.

The fistula was classified according to the Goh classification system4 (Table 1), and attributes of the fistula were recorded, urethral involvement, size of fistula and scarring, if there was a concurrent rectovaginal fistula and if the bladder was completely destroyed.

Table 1.  Goh classification of genitourinary fistula
Site
Type 1Distal edge of fistula >3.5 cm from the external urinary meatus
Type 2Distal edge of fistula 2.5–3.5 cm from the external urinary meatus
Type 3Distal edge of fistula 1.5 to <2.5 cm from the external urinary meatus
Type 4Distal edge of fistula <1.5 cm from the external urinary meatus
Size
ASize <1.5 cm in the largest diameter
BSize 1.5–3 cm in the largest diameter
CSize >3 cm in the largest diameter
Scarring
INone or only mild fibrosis (around fistula and/or vagina) and/or vaginal length >6 cm, normal capacity
IIModerate or severe fibrosis (around fistula and/or vagina) and/or reduced vaginal length and/or normal capacity
IIISpecial consideration, e.g. postradiation, ureteric involvement, circumferential fistula, previous repair

All women were operated under spinal anaesthesia, with a single dose of antibiotics given at the time of anaesthesia. All operations were performed vaginally, and if the length of the urethra was less than 3 cm, a sling was placed underneath the urethra (Goh classification types 2, 3 or 4). This sling was formed from pubococcygeal muscle or scar tissue from the medial aspect of the inferior pubic ramus if all the muscle had been destroyed.1 None of the repairs were performed using a Martius graft.5 The vagina was packed for 24 hours postoperatively, and women were mobilised after the vaginal pack was removed. The bladder was left on free drainage for 14 days for all anastomoses and 16 days if a new urethra was created from an anterior bladder flap. Upon removal of the catheter, the woman was examined vaginally and a history taken with regards to symptoms. The outcomes were recorded as 1) cured, no urethral incontinence; 2) fistula closed but leaking on stressful activity, such as coughing, bending, lifting; 3) fistula closed but leaking on walking; 4) fistula closed but leaking all the time but able to void; 5) fistula closed but leaking all the time not voiding; 6) urinary retention and 7) fistula broken.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

A total of 321 women with an obstetric vesicovaginal fistula were operated from February 2006 until September 2006 at the Barhirdar Hamlin Fistula Centre. Seventy-seven (24%) of these defects were circumferential and 244 were noncircumferential defects (76%).

The women with circumferential injuries were significantly younger, sustained the injury with their first delivery, were more likely to deliver at home, were less likely to have a caesarean delivery, spent longer in labour, had larger defects, had more severe vaginal scarring, were more likely to have complete bladder loss from the labour and more often had a concurrent rectovaginal fistula than those women with noncircumferential defects (Table 2).

Table 2.  Attributes of noncircumferential and circumferential obstetric vesicovaginal fistulae
 Noncircumferential, n= 244Circumferential, n= 77P value
Age at causative delivery (years)24.521.70.007
Primiparous, n (%)92 (37.7)50 (64.9)<0.001
Caesarean delivery, n (%)54 (22.1)4 (5.2)0.001
Home delivery, n (%)107 (43.9)50 (64.9)0.002
Urethral involvement, n (%)133 (54.5)74 (96.1)<0.001
Significant scarring, n (%)33 (13.5)46 (59.7)<0.001
Concurrent rectovaginal fistula, n (%)9 (3.7)17 (22.1)<0.001
Size of defect (cm)2.213.08<0.001
Days in labour3.13.790.005
Bladder completely destroyed, n (%)0 (0)5 (6.5)0.001

Of the 77 women with circumferential fistulae, five were not operated, as there was no bladder tissue left to repair, and these women were given the option of a urinary diversion operation. Of the 72 women operated, two had a breakdown of the vesicovaginal fistula (2.7%) and 34 had some residual incontinence (44.1% of all women, 47% of women actually operated upon). Of these 34 women, 13 were then completely continent using the urethral plug.6 Nineteen of the remaining 21 women with incontinence had mild symptoms such as incontinence when standing up, bending or even slight incontinence on walking. All these 19 women were happy with the improvement to their condition from the operation and were discharged with advice on pelvic floor exercises and were asked to return for a possible stress incontinence procedure in 6 months of time. The remaining two women had severe symptoms that were not improved even by the use of the urethral plug. These two women had a new urethral reconstruction (see below). Six women (7.8%) had urinary retention after the procedure and were discharged self-catheterising and were also asked to return for follow up in 6 months of time (Table 3).

Table 3.  Outcomes from surgical repair for noncircumferential and circumferential obstetric vesicovaginal fistulae
Postoperative outcomesNoncircumferential, n= 244Circumferential, n= 72P value
Residual urethral incontinence, n (%)24 (9.8)34 (47.2)<0.001
Urinary retention, n (%)14 (5.7)6 (7.8)0.59
Failed repair, n (%)6 (2.4)2 (2.6)1.0

Of the noncircumferential fistulae, all were able to be repaired, that is none of the women had the bladder completely destroyed making a repair impossible. Six repairs failed (2.4%) and 24 successfully closed repairs had some residual urethral incontinence (9.8%). Of the latter, six were continent using the urethral plug and the remaining 18 women described their incontience as mild and not needing further management with the plug. Fourteen women had urinary retention after the repair (5.7%) and were continent with clean intermittent self-catheterisation.

Six women with circumferential defects had such extensive loss of the urethra that a new urethra was constructed in its entirety from a flap of anterior bladder. Of these, all were successfully closed and four women (66%) were continent, although three of these four women had urinary retention and were continent with self-catheterising. The remaining two women were incontinent per urethra and were not voiding at all. The urethral plug did not help them.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

The circumferential fistula occurs more often in the urethra, and it would seem likely that the presenting part is impacted against the symphysis for the duration of the obstructed labour, impinging and destroying the urethra. There are only a small proportion of circumferential fistulae that occur just above the urethra, and each time, the fistula was attached to the superior margin of the symphysis pubis, presumably the presenting part was obstructed at the pelvic inlet, destroying the bladder just superior to the bladder neck. It may have been technically easier to repair these from an abdominal route but still comfortably possible from the vaginal route.

The circumferential fistula occurs significantly after a longer and, perhaps, more severe obstructed labour. Although not proven, it is the opinion of several fistula surgeons that if the obstetric fistula occurs after the first labour, it is more likely to be a more severe injury (Hancock B, pers. comm.). This would certainly fit with the description here, as the circumferential injury involves more tissue loss and destruction and occurs more frequently in primiparous women. Perhaps it is because that an obstructed labour in multiparous women more easily results in a uterine rupture than in a primiparous women and perhaps a maternal death instead of the woman surviving with a fistula. This has been the observation of several obstetricians working in the developing world (Breen M and Hamlin EC, pers. comm.). Also the fact that more women with circumferential fistulae deliver at home rather than making their way to an institution and having their obstructed labour relieved by a caesarean section also fits the picture of a longer, more difficult delivery.

The only women in this series who had their bladder completely destroyed were those who sustained a circumferential fistula. Anecdotally, it seems that total loss of bladder occurs more often after a caesarean section for a long obstructed labour. The caesarean section performed after some days in labour, and the operation is made difficult as the bladder is oedematous, even necrotic, may have ruptured and therefore easy to inadvertently amputate during the caesarean section. Interestingly, none of the women who had no bladder remaining delivered by caesarean section; all delivered vaginally either at home or in a health centre after labours lasting from 2 to 7 days.

The other markers of the presence of significant vaginal scarring, concurrent rectovaginal fistula and all the bladder having been destroyed also point to a longer, more severe obstructed labour and extensive tissue loss.

Although not ideal, the method described affords a superior method of closure and continence to a mere apposition of the fistula margins or a simple end-to-end anastomosis. The author also has noted that a number of breakdowns have occurred in circumferential fistulae when the repair was not performed with a circumferential anastomosis. In these instances, the breakdowns occurred to the sides of the repair where the posterior bladder was sutured to the posterior symphysis pubis. This created a challenge to repair it a second time, and most commonly, the bladder was incised and then the whole repair performed again in a formal circumferential manner. The results from the few repairs that have been performed in this manner were successful.

The creation of a new urethra also did not yield ideal results. It was interesting to note that those still completely incontinent did not receive any benefit from the size 3 urethral plug. It could be that the author is still on a learning curve with this procedure, and he created the new urethra with a lumen too wide for the plug and certainly too wide for the sides of the urethra to oppose and hold urine inside the bladder. Refining the technique could hold some benefit. It is planned that these women return for a secondary procedure to narrow and support the urethra.

The usage of self-catheterisation for urinary retention and the urethral plug for severe continuing incontinence are useful but not ideal ways of management in the setting described. Most women live in far-off places from where they have to travel for days to reach the Barhirdar Hamlin Fistula Hospital, which is the only centre that supplies these devices. Using them in the rural villages of Ethiopia is also problematic as clean water is not available, and the woman is asked to boil these devices each day, but realistically, they probably are not doing so. Thus infections, trauma to the urethra and heamaturia would be common in this woman group. The women are given 6 months’ supply of either device and asked to return. Few have returned so far. Only two discharged women using the urethral plug have returned, one was completely continent and no longer needed the plug, and the other woman was still using the plug, continent with it and happy to continue using it. Her urinalysis showed microscopic heamaturia only. So far, no women discharged with self-catheterisation have returned for follow up.

Ideally, an earlier intervention during labour via caesarean section should decrease and hopefully eliminate the number of these types of injuries around the world. Unfortunately, this is a distant dream. The region in which this article was written is an area of 10 million people, and only two gynaecologists are performing emergency obstetrics, which is a common scenario repeated across the developing world. Steps are being taken to educate the population about the importance of antenatal care, delivery in an institution or at least early seeking of help in labour, but still most women live in places of more than a day’s travel from the nearest health institution, which is often ill equipped to help labouring women. Until there are more accessible obstetric services across the world, there will be many women suffering from this condition, and therefore, more research and development is needed to care for women with an obstetric fistula in an appropriate manner.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

The circumferential obstetric vesicovaginal and urethra–vaginal fistula still presents a challenge for the fistula surgeon. It appears that it does occur after a more severe injury, and certainly, the prognosis is less encouraging. Certainly, further research is needed to refine techniques which will render these women completely continent without having to rely on devices such as the urethral plug and self-catheterisation which are not without their morbidities and also are not readily available across the developing world.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

The author thank Julie Morris, Head statistician, University Hospital of South Manchester, for her help with the statistical analysis and also Dr Catherine Hamlin and the staff of the Addis Ababa Fistula Hospital, Ethiopia, for their support.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References