Maternal Intensive Care Unit Admission as an Indicator of Severe Acute Maternal Morbidity: A Population-based Study. Issue 3 (24th September 2022)

• Record Type: Journal Article
• Authors: Godeberge, C.; Deneux-Tharaux, C.; Seco, A.; Rossignol, M.; Chantry, A.; Bonnet, M.
• Is Part Of: Obstetric anesthesia digest. Volume 42:Issue 3(2022)
• Page Start: 117
• Page End: 117
• Publication Date: 2022-09-24
• Subjects: Critical Care -- Maternal Morbidity and Mortality; Anesthesia in obstetrics -- Periodicals; Anesthesia in obstetrics -- Abstracts -- Periodicals; 617.968205
• DOI: 10.1097/01.aoa.0000853492.49420.0f
• Languages: English
• ISSNs: 0275-665X

Abstract

Abstract : ( Anesth Analg . 2022;134:581–591) Severe acute maternal morbidity (SAMM) is an important indicator of maternal care due to the lowering rates of maternal deaths. One criterion for the diagnosis of SAMM is admission to the intensive care unit (ICU). Other aspects that would meet the diagnosis of SAMM is obstetric bleeding, eclampsia, severe preeclampsia, pulmonary embolism, cerebrovascular events, ICU admission, laparotomy after delivery, maternal death, or disorders of the hepatic, hematologic, respiratory, cardiovascular, renal, or neurologic systems. This study investigated if ICU admission serves as a valid indicator of SAMM. Using the Data from the EPIMOMS study which examined 182, 309 women from 6 regions of France, 2540 met the criteria for SAMM or were admitted to the ICU. From this cohort, maternal characteristics, preexisting medical conditions, pregnancy characteristics, length of ICU stay, and organ dysfunction were recorded. The cohort was divided into 3 different groups: those with SAMM not admitted to an ICU, SAMM patients admitted to the ICU, and patients admitted to the ICU not meeting the criteria for SAMM. Statistical analysis was performed and univariable and multivariable multinomial logistic regression models to account for variables in all groups, histograms for continuous variables, standard deviation for normally distributed variables, Kruskal-Wallis test for ICU length of stay, and Fisher exact tests for categorical variables. The study used multiple imputations with chained equations to ascribe missing multivariable analysis model data values followed by a complete case multivariable logistic regression analysis. The results were at a α=0.05 significance level, a statistical power of 80%, an odds ratio of ≥1.4 or ≥2.0.