Discussion:
Main Findings: This retrospective study examines risk factors for patients who underwent surgical repair of urogenital fistulas at a Fistula Hospital in Masaka, Uganda. The data demonstrate that the incidence rate of blood transfusions in our patient population is 6.2%. Consistent with our hypothesis, there are specific patient characteristics that predict a need for perioperative blood transfusions. Specifically, these risk factors include abdominal approach and timing of fistula repair was borderline.
Strengths and Limitations: Strengths of this study include well-defined outcomes and their importance given the lack of availability of blood in many areas of the world. This is the first study to our knowledge looking at this important outcome. Furthermore, we include a population that is generalizable to obstetric fistula populations in the developing world. The surgeries were completed by a group of well-trained fistula surgeons from both Uganda as well as other countries who worked collaboratively to care for these women in a dedicated Fistula Hospital. Limitations include a retrospective nature of this study and therefore missing data in some cases that prevent us from looking at other risk factors that may not have been collected at the time or even known by the patients themselves. Some of these include other patient comorbidities or history of previous non-fistula surgeries for example.
Interpretation: Fistulas approached abdominally were approximately four time more likely to need a blood transfusion compared to a vaginal approach. This is consistent with existing research advocating for vaginal approach as the preferred management option for surgical fistula repairs.7 A vaginal approach is typically preferred when feasible due to its minimally invasive nature and association with shorter operative times, decreased blood loss and shorter length of stays in hospitals.8,9 Previous studies have demonstrated higher rate of complications associated with an abdominal approach, including urinary tract infections, sepsis, blood transfusions, and hospital readmissions.6,7 Results from our study are consistent with published data advocating for vaginal approach in regards to risk of perioperative blood transfusion as well. However, there are circumstances in which an abdominal approach may be indicated, such as for ureteric involvement, limited vaginal access, significant associated scarring or failed prior repairs.7,8,10 Ultimately, approach is largely determined by patient characteristics, size and location of fistula and expertise of surgeon. Although an abdominal approach may be unavoidable at times, our data may help surgeons identify these higher risk patients in preoperative surgical planning and ensuring availability of resources such as sufficient blood.
In our study, timing of repair was also identified as a borderline risk factor with patients who underwent fistula repair before three months being approximately half as likely to receive a blood transfusion. Although this was a borderline relationship in our study, the authors feel it is notable and may require future studies to fully investigate. Timing of repair is often left to the surgeon’s discretion, based on evaluation of the tissue quality and maturity of the fistula. Traditional teaching typically dictates waiting at least three months after childbirth before surgically repairing a fistula to allow time for healing and inflammation to subside as well as recovery from central necrosis, edema, or infection.4,7,11 Additionally, conservative management for women who present within the first three months for repair may help the fistula spontaneously close.4 However, some experts would argue, that it is reasonable to repair before the 3-month mark so long as the fistula tract is mature and the tissue appears healthy. Furthermore, studies have demonstrated earlier closures within 1-2 weeks of injury may have similar success rates to delayed repairs.8,12,13 Our results suggest that there may be additional benefits to delayed repair in regards to decreasing associated risk for perioperative blood transfusions particularly in low resource settings where blood may not be readily available.
In our study, timing of repair was also identified as a borderline risk factor with patients who underwent fistula repair before three months being approximately half as likely to receive a blood transfusion. Although this was a borderline relationship in our study, the authors feel it is notable and may require future studies to fully investigate. Timing of repair is often left to the surgeon’s discretion, based on evaluation of the tissue quality and maturity of the fistula. Traditional teaching typically dictates waiting at least three months after childbirth before surgically repairing a fistula to allow time for healing and inflammation to subside as well as recovery from central necrosis, edema, or infection.4,7,11 Additionally, conservative management for women who present within the first three months for repair may help the fistula spontaneously close.4 However, some experts would argue, that it is reasonable to repair before the 3-month mark so long as the fistula tract is mature and the tissue appears healthy. Furthermore, studies have demonstrated earlier closures within 1-2 weeks of injury may have similar success rates to delayed repairs.8,12,13 Our results suggest that there may be additional benefits to delayed repair in regards to decreasing associated risk for perioperative blood transfusions particularly in low resource settings where blood may not be readily available.
Adequately powered prospective case-control studies would provide the best estimates of risk factors and potentially identify other characteristics that serve as risk factors for needing a perioperative blood transfusion, but such trials would require a large sample size. In the interim, we believe our study represents important evidence for consideration in guiding surgical practice in low-resource settings.