///2019 Abstract Details
2019 Abstract Details2019-07-13T07:45:15-05:00

The Obstetric Comorbidity Index (OB-CMI) as a predictor for admission to high dependency or intensive care unit

Abstract Number: T1A-77
Abstract Type: Original Research

Ilai Ronel BMBS BMedSci1 ; Tahel Ronel Phd2; Brian T Bateman MD3; Eyal Egoz B.Sc4; Idit Matot MD5; Carolyn F Weiniger MB ChB6

Introduction: Almost 4% of laboring women will experience a near-miss event associated with morbidity and risk of death. We aimed to assess the obstetric comorbidity index (OB-CMI) [1] as a predictor for high dependency unit (HDU) or intensive care unit (ICU) admission in a 12,000 annual deliveries tertiary center.

Methods: Retrospective case-control study of women delivering Jan 2016-Jun 2018. The study group comprised all obstetric admissions to the HDU or ICU in the immediate post-partum period or within the peri-partum period - 6 weeks antepartum to 6 weeks postpartum, for at least 4 hours. The controls were 2000 women who underwent uneventful deliveries in the same time period. Demographic, medical and obstetric characteristics were retrieved. Specific co-morbidities were noted and the OB-CMI score calculated. The difference between OB-CMI scores for study vs. control group was evaluated using Mann-Whitney U test. The area under the receiver operating curve (AUC), reflecting OB-CMI as a predictor for HDU or ICU admission, was calculated using R© programming language.

Results: 1062 study group women identified, 963 (90.7%) HDU and 99 (9.3%) ICU admissions. Demographics and obstetric characteristics are summarized in Table 1. The most common co-morbidities in both groups were previous cesarean delivery or myomectomy, maternal age (35-44 years), multiparity and pre-eclampsia (Table 2). The median (range) OB-CMI scores for study and control groups were 1 (0-17) and 0 (0-6), respectively. The OB-CMI score distributions differed significantly (Mann-Whitney U=1440300, p-value<0.0001, effect size r=0.314). The AUC was 0.678 (95% confidence interval 0.658-0.698). The OB-CMI scores of multiple samples of 1062 controls were randomly selected and compared with scores from the study group. This resulted in highest mean accuracy of 0.652, achieved at OB-CMI scores of ≥2. The associated sensitivity was 0.482 and specificity was 0.822.

Discussion: In our population, we report a significantly higher OB-CMI score for women admitted to HDU or ICU vs. women with uneventful deliveries. Using OB-CMI score cut-off of 2 or above, nearly half of women with adverse outcomes were identified, while still maintaining high specificity. However, due to the low sensitivity, many women with low scores would not be identified using this score.

[1] Bateman BT et al. Obstetrics and Gynecology. 2013;122(5):957–65

SOAP 2019