1 Emergency peripartum hysterectomies at Muhimbili National ...

Tanzania Journal of Health Research Volume 14, Number 1, January 2012

DOI: http://dx.doi.org/10.4314/thrb.v14i1.7

Emergency peripartum hysterectomies at Muhimbili National Hospital, Tanzania: a review of cases from 2003 to 2007 ANDREA B. PEMBE*, PETER J.T. WANGWE and SIRIEL N. MASSAWE Department of Obstetrics and Gynaecology, School of Medicine, Muhimbili University of Health and Allied Sciences, P.O. Box 65117, Dar es Salaam, Tanzania

______________________________________________________________________________

Abstract: A retrospective review of all cases of emergency peripartum hysterectomy performed between January 1, 2003 and December 31, 2007 at Muhimbili National Hospital was done to determine the incidence, indications and complications, background characteristics, antenatal care attendance, referral status, and maternal and foetal outcomes. There were 55,152 deliveries during the study period and 165 cases of emergency peripartum hysterectomy, giving the incidence of emergency peripartum hysterectomy of 3 per 1000 deliveries. The main indication was uterine rupture (79%) followed by severe post-partum haemorrhage due to uterine atony (12.7%). The case fatality rate was 10.3% where as perinatal mortality rate was 7.7 per 1000 deliveries. The common complication identified intraoperatively was severe haemorrhage which accounted for 39.4% where as intensive care unit admissions (14.4%) and febrile morbidity (12.4%) were common after the operation. Blood was ordered in all cases but in 31 cases it was indicated that it was not available. Seventy nine patients received blood transfusion with the maximum number of units given to one patient being eight. Twenty two patients were given fresh frozen plasma (FFP), the median number of units given was two (range = 1– 6). In conclusion, emergency peripartum hysterectomy is a life saving procedure and very common at MNH. The most common indication was ruptured uterus followed by severe postpartum haemorrhage. More than half of the patients underwent emergency peripartum hysterectomy were referred from other health facilities with ruptured or suspected ruptured uterus. The procedure was associated with unacceptably high maternal and perinatal morbidity and mortality.

______________________________________________________________________________

Key words: Emergency, peripartum hysterectomy, uterine rupture, maternal, morbidity, mortality, Tanzania

Introduction Emergency peripartum hysterectomy involves removal of the uterus after vaginal delivery or during the caesarean section. The operation is commonly made without delay when a decision is reached to save the life of the mother when conservative methods have failed and the mother’s condition is deteriorating. The incidence of emergency peripartum hysterectomy varies from 2 to 10 per 1000 deliveries (Engelsen et al., 2001; Okogbenin et al., 2003; Zamzami, 2003; El-Jallad et al., 2004; Kwame-Aryee et al., 2007). Some studies have reported excessive haemorrhage in the third stage of labour due to uterine atony or placenta praevia as the common indication for emergency peripartum hysterectomy (Zamzami, 2003; Ezechi et al., 2004) while other studies have

*

Correspondence: Andrea B. Pembe; E-mail: [email protected]

1



shown ruptured uterus to be the common indication (Chew & Biswas, 1998; Okogbenin, et al., 2003). Tanzania is among the countries with a high maternal mortality ratio, currently estimated at 454 per 100,000 live births (TDHS, 2010). A study in Dar es Salaam city by Urassa et al. (1995) found that the major causes of the maternal deaths were haemorrhage, postpartum infection, hypertensive disorders, obstructed labour and complications of abortion. The majority of the deaths occurred in women who had contact to the modern health care system and spent reasonable time receiving treatment. This could suggest inadequate quality of services provided by the health care system. The incidence, indications and complications of emergency peripartum hysterectomy may reflect the quality of antenatal, intrapartum and referral care provided in the country. We conducted a five year review of cases of emergency peripartum hysterectomy at the tertiary and university teaching hospital in Dar ea Salaam, Tanzania, in order to describe the incidence, indications, complications, background characteristics, antenatal care attendance, referral status, and maternal and foetal outcomes. Material and Methods Study setting and design This was a retrospective review of all cases of emergency peripartum hysterectomy performed between January 1, 2003 and December 31, 2007 at Muhimbili National Hospital (MNH). MNH is the largest referral and University teaching Hospital in Tanzania. The hospital is located in Dar-es-Salaam city which has an estimated population of 2.5 million (2002 census). The city has three municipal (district) hospitals, 16 private hospitals, 10 health centres and more than 60 dispensaries which are government and non-government owned. MNH provides antenatal care in the clinics conducted from Monday to Friday. In average 40 pregnant mothers attend the clinic in a day. The hospital has a maternity building which has seven wards. Four wards each with 40 beds capacity are for antenatal and postnatal mothers. The other three wards are the labour ward, eclampsia ward and the last ward save for neonatal care. The hospital has one intensive care unit (ICU) with 12 beds. Patients needing intensive care are transferred to the unit after interdepartmental consultation of the doctors in the obstetrics and gynaecology and those in the ICU under the department of Anaesthesia. MNH attends all pregnant women, received and not received antenatal care at the hospital irrespective of their clinical state. It receives women referred from the municipal hospitals, district hospitals surrounding Dar-es-Salaam city, private hospitals, health centres and dispensaries. Some pregnant mothers are coming directly from home. The annual total number of deliveries is about 11,000. Most obstetrical surgical procedures including hysterectomies are performed by the residents in obstetrics and gynaecology under supervision of the consultants. During the study period MNH was undergoing major rehabilitation and there was one obstetric theatre which was used for both emergency and elective cases. In circumstances when there were too many obstetric

2



cases the gynaecological theatre had to be used thus gynaecological procedures were halted. Data collection In this review emergency peripartum hysterectomy was defined as hysterectomy performed after delivery of the baby because of obstetric complication. The register, report books and patients’ clinical records in the obstetric theatre and postnatal wards were searched to identify cases of emergency peripartum hysterectomy. The first author and one research assistant extracted information from the patient’s clinical records. The information retrieved includes maternal age, parity, gestational age at delivery, presence of uterine scar, mode of delivery, type of hysterectomy whether sub-total or total, as well as complications of the procedure such as intra-operative hypotension, mean blood loss, blood transfusion and intraoperative and postoperative complications. The indication of emergency peripartum hysterectomy was obtained in the operative notes. Data analysis Data entry was done using Epi-Info 6 computer programme and subsequently analysed by Statistical Package for Social Sciences (SPSS) version 14.0. Frequency distribution and measure of location were used to summarize data. Ethical considerations Ethical clearance was given by the Senate, Research and Publication Committee of Muhimbili University of Health and Allied Sciences (MUHAS) and permission to conduct the study was given by the Executive Director of MNH. Retrieval of information was conducted confidentially by researchers and no name of the patients was used in the data collection protocol. Results During the five year period there were 55,152 deliveries and 165 cases of emergency peripartum hysterectomy of which 151 (92%) were total – and 14 (8%) were sub-total abdominal hysterectomy. Among the 165 cases, 28 were performed in women who had vaginal deliveries, 43 after caesarean section and 94 during laparatomy due to ruptured uterus. The overall incidence of emergency peripartum hysterectomy was 3 per 1000 deliveries. The annual incidence of emergency peripartum hysterectomy increased from 2 per 1000 deliveries in the year 2003 to 3.5 per 1000 in 2004, thereafter there was no much change in the incidence between 2004 to 2007 (Figure 1).

3



14

Number

12 10 8 6 4 2 0 2003

2004

2005

2006

2007

Year Deliveries (x1000) Incidence of emergency peripartum hysterectomy per 1000 deliveries

Figure 1: Total deliveries and annual incidence of emergency peripartum hysterectomy

The median age of the cases was 30 years (range= 15-40). The median parity was 3 (range = 1-12). The mean gestation age at delivery was 38 ± 1.96 weeks. Eleven per cent of the cases had previous caesarean section scar (Table 1). Table 1: Demographic, obstetrics and referral characteristics of emergency peripartum hysterectomy cases (N=165) Characteristic Age group (years)