Prognostic value of the obstetric specific scoring systems and four years' experience of a tertiary center.

BACKGROUND: Obstetric sepsis is a significant and preventable cause of maternal mortality. The obstetric modified SOFA (omSOFA) score was created to diagnose sepsis in this population, and the sepsis in obstetrics score (SOS) was developed to predict the need for ICU care. This study investigated the prognostic value of these scores in patients treated in the ICU for obstetric sepsis, analyzed the patients' characteristics, and examined the perinatal outcomes associated with sepsis. METHODS: The study was retrospectively designed, and patients hospitalized in the ICU due to maternal sepsis were evaluated. These patients were divided into two groups based on the length of ICU stay. An ICU stay exceeding 72 h is considered a prolonged stay. The data, including demographics, laboratory results and obstetric outcomes, were compared between the groups. Receiver operating characteristic (ROC) curve analyses were performed to investigate the performance of omSOFA and SOS scores in predicting prolonged ICU stay and septic shock. RESULTS: The study included 25 women with maternal sepsis. Sepsis developed during pregnancy in 40% of patients, after cesarean section in 28%, after abortion in 20%, and after vaginal delivery in 12%. Septic shock developed in 12% (n = 3) of patients during ICU hospitalization. SOS showed better performance in predicting the need for ICU longer than 72 h (cut-off: 3, AUC:0.865, CI:0.721-1.000, sensitivity:75%, specificity:100% PPV:100% NPV:92.7% p = 0.003). At the same time, omSOFA performed better in predicting septic shock (cut-off: 5, AUC:0.985, CI: 0.941-1.000, sensitivity:100% specificity:95%, PPV:73.2%, NPV:100% p = 0.007). DISCUSSION: Considering that prolonged stay in the ICU and development of septic shock are important determinants of morbidity and mortality, SOS and omSOFA scores could serve as valuable prognostic tools in obstetric sepsis cases.