Developing a Feasible Scoring System for Critically Ill Obstetric Patients in Intensive Care Units in Resource Limited Settings

Background: Maternal mortality is a challenge to the health care system especially in developing countries. One of the strategies of reducing maternal mortality is the care of the critically ill obstetric patients in the Intensive Care Unit (ICU). The clinical state of these patients at admission remains a major determinant of their outcome. Evidence based assessment tools such as the Acute Physiologic Assessment and Chronic Health Evaluation 11 (APACHE 11) and the Sequential Organ Failure Assessment (SOFA) scores have been used to stratify and prognosticate obstetric patients in the ICU but are limited by some shortfalls. The Modified Obstetric Early Warning Signs scoring system (MOEWS) has been suggested to serve this purpose devoid of the shortfalls.

Objective: This study retrospectively compared the MOEWS to the APACHE II and SOFA scores in predicting the outcome of critically ill obstetric patients in the ICU with the aim of developing a scoring tool adaptable to a resource limited setting.

Method: An evaluation of the scoring systems as prognostic tools was done. The primary outcome was the predicted maternal mortality by the scoring tools as against the actual maternal mortality observed while the secondary outcomes were mechanical ventilation, use of vasopressors and length of ICU stay.

Results: The area under receiver operator characteristic (AUROC) curve of MOEWS, SOFA, and APACHE II for prediction of maternal mortality was 0.794 (95% CI, 0.649–0.818), 0.684 (95% CI, 0.594–0.772), and 0.724 (95% CI, 0.641–0.828), respectively. The standardized mortality rate (SMR) indicated that all the three scoring systems were comparable in predicting maternal mortality. The MOEWS correlated more significantly with the length of ICU stay (p = 0.029, R = 0.138) and number of hours on ventilation (p = 0.025, R = 0.228). Statistical Analysis of ICU parameters showed that hospital stay (p = 0.011), and ventilation days (p = 0.014) were significant predictors of maternal outcome. Age (p = 0.789) and ICU stay (p = 0.701) were not significant.

Conclusion: The MOEWS scoring system which is comparable to other commonly used scoring systems will be of value in resource limited settings in developing management protocols for obstetric patients in the ICU.