Admission Rate, Indications, Maternal Outcomes, and Factors Associated with Mortality Among Obstetric Intensive Care Unit Admissions at a Tertiary Referral Hospital in Tanzania.

Background About 0.07% to 8% of parturients will require intensive care unit (ICU) admission during pregnancy, childbirth or puerperium, with great regional variations. When this occurs, it increases morbidity, mortality, long-term physical and psychological sequelae as well as cost. This study sought to examine a five-year admission rate, indications and outcomes of parturients with severe morbidities admitted in the ICU at a tertiary referral facility and teaching hospital in Tanzania. Methodology A retrospective cross-sectional study was conducted at Muhimbili National Hospital in Dar es Salaam city that included all women who were admitted to the obstetric ICU between 2019 and 2023. Data were retrospectively collected using a pre-tested checklist from all 1526 obstetric ICU admissions, out of 29391 obstetric unit admissions, during the study period. Sources of data were both case files and electronic registries in the ICU and labour ward. Multivariable logistic regression was used to identify factors associated with maternal mortality, with results presented as adjusted odds ratios (aOR) while indications of ICU admission and maternal outcomes were presented descriptively by using frequencies and tables. Results Out of 29391 obstetric unit admissions, 1526 required ICU, resulting into obstetric ICU admission rate of 5.2%. Hypertensive disorders were the leading cause of admission (73.7%) followed by obstetric haemorrhages (15%). Frequent ICU interventions included; mechanical ventilation (50%), anticonvulsants (26%), blood transfusion (15%), inotropes (4.2%), and dialysis (3.5%). Maternal fatality rate was 14%. In multivariable regression analysis, vaginal delivery was associated with 47% higher odds of mortality compared to caesarean section (aOR 1.47, 95% CI 1.03-2.11, p=0.036). A moderate ICU stay of 3-4 days showed a 91% protective effect relative to stays of ≤2 days (aOR 0.09, P < 0.001). 5-7day stays were also protective (aOR 0.32, p=0.006). Anticonvulsant use was strongly protective against mortality (aOR 0.02, p < 0.001). Conclusion This five years review of obstetric ICU admissions (2019-2023) showed that, approximately one in 20 obstetric patients required critical care, with one in seven of ICU admissions resulting in maternal death. Hypertensive disorders of pregnancy accounted for majority of ICU admissions, and deaths occurred predominantly during the early phase of critical care. These findings underscore the need to strengthen antenatal risk identification, optimise referral and emergency obstetric care, ensure timely magnesium sulphate administration, and establish obstetric high dependency units to facilitate earlier interventions. Such measures could reduce the burden on ICUs while improving maternal outcomes.