AFJOG

ORIGINAL RESEARCH African Journal of Obstetrics and Gynaecology | Volume 3 | Issue 1 | 2025 | Maternal deaths from Obstetric Haemorrhage 2020-2022 PPH after vaginal delivery Late detection of PPH (see above) was a common problem. Improvement was noticed compared to previous reports in the first medical response to PPH. However, there was an inadequate approach to refractory or ongoing PPH with poor recognition of causes other than uterine atony. This occurred particularly at DHs where inadequate blood, inadequate support, lack of surgical and anaesthetic skills further limited the response treatment measures for refractory PPH. Although inverted uterus is a very rare cause of PPH and only accounted for three maternal deaths, it was poorly managed in all three women, with non-recognition of the diagnosis and no approach to treatment. Bleeding at/after CD The majority (119 of 173 deaths) were only recognised after the CD, although it is probable that the problem must have occurred during the CD and was not detected. Reluctance to re-operate at DHs with evidence of postoperative bleeding (due to inadequate blood, inadequate support, lack of skills in surgery and anaesthesia) was observed with patients being transferred out to a higher level of care and dying during the referral process. There was also poor prediction and recognition of cases of morbidly adherent placentae. Abruptio Placentae Many of these deaths had been inadequately resuscitated. Inappropriate CD was observed in cases where the fetus was thought to be still alive, but the cardiotocograph was actually detecting a maternal tachycardia rather than fetal heart. Also, CD was sometimes performed too soon when amniotomy and resuscitation would have allowed more time to achieve vaginal delivery. Placenta praevia Most of the deaths due to placenta praevia had the condition diagnosed antenatally by ultrasound. However, inadequate precautions were made prior to CD in terms of ensuring availability of blood and a skilled surgeon and anesthetist. A few cases of morbidly adherent placenta were not predicted (and thus not prepared for) despite the patients having anterior placenta praevia and several repeat CDs. Ruptured uterus These deaths occurred equally in women with scarred and unscarred uteri. With scarred uteri, the diagnosis was made appropriately but if it was at a DH, medical staff were unskilled to manage this even though uterine tourniquet, non-pneumatic anti shock garment (NASG) and referral were treatment options within the scope of medical officers. There was frequently delay in diagnosis of uterine rupture in women with unscarred uteri which meant that laparotomy was delayed, or the rupture was a postmortem diagnosis. Also, in several of these deaths, excessive doses of misoprostol were used for induction of labour (IOL). Sub-analyses of the MaMMAS data showed that induction of labour was performed in 2 (5.4%) of the 43 MDs with scarred uteri, and 16 (34%) of the 47 MDs with unscarred uteri. This confirms the risks associated with IOL. The above themes are illustrated in the following case summaries/vignettes. MATERNAL DEATH FROM PPH AFTER VAGINAL DELIVERY (VBAC) 34yrs, para 2, previous CD x1. Booked, HIV positive with suppressed viral load, No anaemia. Vaginal Birth After Caesarean Delivery (VBAC) at district hospital (DH). Normal labour progress with vaginal delivery of live baby and placenta. Blood loss 1000mls detected at 30 mins. Diagnosis: Uterine atony vs Cervical tears? Treated with 20 units oxytocin infusion, Syntometrine im, Tranexamic acid iv, iv fluids, Voluven, 3 units blood, 2 units fresh dried plasma (FDP). Considered taking patient to Operating Theatre (OT) for exploration and possible laparotomy, but second doctor not confident to perform General Anaesthetic. Decided to refer to Regional hospital (4 hours drive away), patient accepted. Ambulance collected her after 90 minutes. Died in ambulance on route and returned to DH. Postmortem excluded uterine rupture, trauma or retained products, but showed uterine atony with uterus full of blood. Assessment Primary Obstetric Cause: PPH after vaginal delivery/ uterine atony. Final cause: hypovolaemic shock. Contributory: Coagulopathy (DIC) Avoidable factors: Patient: Nil. Administrative: Ambulance delay; Lack of appropriate skill on site at DH. No NASGs. Medical care: Poor problem recognition/late detection PPH; Substandard care, Patient should have been taken to OT at DH, Patient needed NASG for transfer. Suggestions for prevention. E MOTIVE approach for early detection PPH and early escalation of care for refractory PPH. Doctors at DHs must have skills for postpartum laparotomy and general anaesthesia, and be supported by regional hospital. NASG to be available at DHs. Should VBACs be performed at DHs? Bleeding at/after emergency CD following IOL at district hospital 28yrs, P2G4. Booked. HIV pos with viral load suppressed. Induction of labour for prolonged pregnancy. Misoprostol 200mcgms po 2 hourly. Emergency CD for fetal distress. Difficult delivery of baby, fresh still birth. Excessive bleeding from tears in right broad ligament and lower segment. Applied B-Lynch compression suture, sutured tears and ligated right uterine artery. Shortage of blood. Planned to transfer to RH, but RH declined, advised observation. Accepted on second phone call. Patient died 4 hours post operatively awaiting ambulance. Assessment Primary obstetric cause: Bleeding associated with caesarean delivery/trauma. Final cause: hypovolaemic shock. Contributory: Coagulopathy (DIC) Avoidable factors: Patient: Nil. Administrative: Shortage blood products, No NASG Medical Care: Excessive doses misoprostol for IOL leading to precipitate labour, trauma and still birth: RH should have advised relaparotomy with tourniquet prior to referral and been more supportive. Suggestions for prevention. Correct dosage regimens for IOL. Blood product availability. Skills development for DH doctors with support from RHs Placenta praevia/ morbidly adherent placenta 34yrs, P2 G4. Booked at regional hospital. HIV pos with viral load suppressed. Previous CDx2 Ultrasound (early) showed major placenta praevia plus possible percreta. African Journal of Obstetrics and Gynaecology | Volume 3 | Issue 1 | 2025 | 10