From my experience in Zaria, northern Nigeria, I agree with Dr Browning1 that the provision of a hostel adjacent to maternity units is essential as part of the antenatal care of women with previous vesicovaginal fistula (VVF) repair. In Zaria, only a small minority of the pregnant women who had VVF in a previous pregnancy used the available hostel when they needed it. Unlike the situation reported by Browning in Ethiopia, in Zaria, previous VVF repair was by itself not regarded as an absolute indication for delivery by elective Caesarean section in subsequent pregnancies. There, the size of the maternal height and bony pelvis changed markedly between the first and subsequent pregnancies in teenage girls who had not finished growing during their first pregnancies. Growth spurts even occurred during the first pregnancy in adolescents who took malarial chemoprophylaxis and iron and folic acid supplements.2,3 Furthermore, in some, the VVF resulted from the gishiri cut performed by traditional birth attendants, for reasons unrelated to prolonged obstructed labour. Therefore, in a subsequent pregnancy, if clinical and radiological pelvimetry at 36 weeks showed the pelvis to be suitable for vaginal delivery, and if there were no other contraindications, including vaginal fibrosis and stenosis, then vaginal delivery assisted by episiotomy was allowed. Otherwise, delivery was by elective Caesarean section at term.
Of 22 774 consecutive births that took place at the Ahmadu Bello University Teaching Hospital, Zaria, between January 1976 and July 1979, 144 had previous repair of VVF with or without previous Caesarean section, and 79 had acquired VVF not in their previous pregnancies but in the course of labour at the end of their index pregnancies.4
We compared pregnancy outcomes following previous VVF repair (144), previous surgery for ectopic pregnancy (41), previous Caesarean section (866), and the remainder without previous operations. Those with previous VVF repair stood out as the group with the worst possible reproductive results: the highest incidence of short maternal stature, the lowest rates for literacy, the lowest rate of child survival from previous births, the highest rate of operative delivery, the highest rate of low birth weight babies, and the worst fetal results in their current pregnancies. The principal indication for operative delivery was cephalopelvic disproportion and its complications, including uterine rupture.
Of the 79 women found to have a fresh VVF when they were admitted in labour or reported to hospital shortly after delivery at home, 12 died, all from complications of advanced obstructed labour. Only five of the 66 singleton babies delivered in hospital survived; the rest were mostly intrauterine deaths before arrival in hospital, necessitating abdominal deliveries and craniotomies in the presence of gross genital tract infection and severe anaemia. Furthermore, the surviving babies were on the average lighter (mean 2.77 kg) than the perinatal deaths (mean 3.14 kg). Hence, in terms of birth weight distribution, the babies who failed to survive were potentially superior to the babies who did survive. Although the numbers are small, they suggest that in places where disproportion is common and most deliveries are unsupervised, the surviving babies may not necessarily be the best babies. This pattern of reproduction carries serious implications for the future well-being of the population.
In as much as health reforms including curative centres for fistula repairs are important, prevention is the priority. Firefighting in hospital by itself will not break the cycle of death and destruction: political intervention and socio-economic reforms, beginning with compulsory universal formal education, will.