Abstract:

Background: Worldwide the overall incidence of obstetric hysterectomy is 0.05%. In Tanzania two studies reported an incidence of 0.1% to 0.3% in Dodoma and 2/1000 birth to 3.5/1000 birth in Dar es Salaam city. Studies have reported varying outcomes among patients who underwent obstetric hysterectomy. There are several factors reported to predict the outcomes of this procedure, however there is limited information on predictors of poor outcomes among patients who had obstetric hysterectomy at Bugando Medical Centre.

Objectives: To determine the outcomes and predictors of poor outcomes of patients who underwent obstetrics hysterectomy at Bugando Medical Centre, Mwanza Tanzania.

Method: This was the retrospective study between January 2013 and December 2017 .Case files were retrieved from Bugando medical records and information such as patient identification, place of delivery, mode of delivery, cadre of surgeon who perform a surgery, time elapsed between onset of complication and hysterectomy, timing of hysterectomy, blood products availability during and after hysterectomy, day of which surgery was done, vitals sign before surgery,haemoglobin level before surgery, indication of hysterectomy and outcomes after hysterectomy were extracted and filled in the pre-tested data collection tool with respect of anonymity and confidentiality. Data was analyzed with STATA version 13.

Results: During the study period a total 124 case files of patients who underwent obstetric hysterectomy were enrolled. The median age was 27 [IQR 23 – 34] years and median parity was 3 [IQR 2 – 4] numbers of deliveries with median number of living children 1 [IQR 1–3] children. Fifty two percent of these women delivered at BMC. The main documented indication was necrotic uterus 65 (52.4%) followed by ruptured uterus 35 (28.2%). Complication occurred in 68 (54.8%) patients, leading complication was wound sepsis 38 (55.9%), followed by vesicovaginal fistula and burst abdomen 7 (10.3%) respectively. There were 13 maternal deaths, giving case fatality rate of 10.5%. Diagnosis of necrotic uterus was the strong predictor of adverse outcome (OR 13.54; 95% CI [3.59 – 51.02]; p-value=0.000).

Conclusion: Half of patients recovered with complication while one third recovered without complication. Less than one third died after obstetric hysterectomy. Half of patients sustained wound sepsis post surgery. Intraoperative diagnosis of necrotic uterus is only strong predictor of poor outcome. The main causes of death post obstetric hysterectomy at BMC are septicaemia and haemorrhagic shock.