///2016 Abstract Details
2016 Abstract Details2019-07-15T10:10:51-05:00

Evaluation of Maternal Early Warning Criteria at a Tertiary Obstetric Care Center

Abstract Number: SA-34
Abstract Type: Original Research

Kristin B Brennan MD1 ; Ashley M Tonidandel MD, MS2; Lynette C Harris BSN3

Intro: The National Partnership for Maternal Safety recommended Maternal Early Warning Criteria (MEWC) to promote earlier recognition of critical illness in the obstetric patient1. The simplicity and specificity of this single parameter risk assessment tool was endorsed for use in the US. The aggregate-weighted scoring of the Modified Early Obstetric Warning System (MEOWS) in the UK has a reported 89% sensitivity, 79% specificity, 39% positive predictive value (PPV), and 98% negative predictive value (NPV) for predicting morbidity2. The purpose of this study was to evaluate MEWC in our tertiary care obstetric center.

Methods: Data was collected retrospectively from electronic medical records of patients admitted to the labor and delivery unit over a month-long period. Morbidity was defined by MEOWS diagnostic criteria and retrieved from the admission note or discharge summary. Maximum and minimum values for systolic and diastolic blood pressure (BP), heart rate, respiratory rate, oxygen saturation (SaO2) and urine output (UO) were retrieved from vital sign flowsheets.

Results: To date, data retrieval is complete for 200 patients and continues for an additional 300. Thirty-four of 200 experienced some form of morbidity. The most common morbidity was pre-eclampsia (71%), followed by suspected infection (35%), hemorrhage (18%), pulmonary edema (5%), and acute asthma exacerbation (3%). No maternal deaths were reported and 1 patient was admitted to the ICU. By MEWC, 63.5% of patients would have triggered a bedside evaluation by a clinician. Like the MEOWS results, the most frequent trigger was high BP (42%), followed by tachycardia (27%), and low BP (15%). Respiratory rate, SaO2, and UO were the least frequent triggers (5%, 5% and less than 1%, respectively). Four patients with diagnosed morbidity did not meet MEWC. The overall sensitivity of MEWC in predicting morbidity was 88%, specificity 42%, PPV 24% and NPV 95%.

Discussion: Based on early analysis, the MEWC did not demonstrate specificity and PPV expected of a risk assessment tool in our obstetric population. Analyses of a more complete data set (to be presented at SOAP) will give a more definitive look at the feasibility of implementation. Our methodology is limited by the retrospective approach. Specifically, parameter documentation practices were not audited and abnormal triggers were not verified by repeat measurements. However, the current designated triggers in this single parameter scoring system may result in excessive calls for bedside evaluation by a clinician, exhausting limited resources and possibly leading to diminished attention to patient decompensation. Prior to implementation at a high risk center, the MEWC may need to be modified by institution or patient population to improve outcomes.

1)Mhyre et al. The maternal early warning criteria. Obstet Gynec2014;124:782-6

2)Singh et al. A validation study of the CEMACH recommended MEOWS. Anaesthesia2012;67:12-8

SOAP 2016