Initial clinical characteristics of gravid severe acute respiratory syndrome coronavirus 2-positive patients and the risk of progression to severe coronavirus disease 2019

BACKGROUND: Available data suggest that the obstetric population is particularly vulnerable to severe respiratory syndrome coronavirus 2 infection, with a variable clinical course leading to severe respiratory failure. However, established early warning scores designed to identify patients at risk of clinical deterioration were never validated in the obstetric population. OBJECTIVE: This retrospective cohort study sought to evaluate the initial clinical characteristics of pregnant patients diagnosed with severe acute respiratory syndrome coronavirus 2 infection and to develop a pregnancy-specific early warning score to identify patients at risk for clinical deterioration and requiring advanced respiratory support. STUDY DESIGN: This was a single center, retrospective cohort study of pregnant patients diagnosed with severe acute respiratory syndrome coronavirus 2 infection between April 2020 and December 2020. A total of 50 patients with severe acute respiratory syndrome coronavirus 2 infection between April 2020 and November 2020 were used to create the prediction model. Initial clinical characteristics identified at the time of diagnosis were compared between patients who required advanced respiratory support and those who were asymptomatic or had mild symptoms for those diagnosed during the period of April 2020 to November 2020. Risk factors associated with a requirement for advanced respiratory support were used to create the Obstetric Warning Score system. The Obstetric Warning Score system was then validated using 30 patients diagnosed with severe acute respiratory syndrome coronavirus 2 infection in December 2020. A receiver operating characteristic curve was generated to evaluate the test characteristics of the Obstetric Warning Score system compared with other scoring systems including the Early Warning Score, the National Early Warning Score 2, and the Maternal Early Warning Criteria. RESULTS: Women who required advanced respiratory support were more likely to present with dyspnea (100% vs 33.3%; P<.001), have a higher heart rate (113.4 beats per minute vs 93 beats per minute; P<.001), respiratory rate (23.5 breaths per minute vs 17.7 breaths per minute; P<.001), temperature (99.1 degrees F vs 98.3 degrees F; P=.004), and C-reactive protein level (7.4 mg/dL vs 2.4 mg/dL; P<.001). Furthermore, 88.2% of patients requiring advanced respiratory support showed chest x-ray findings consistent with pneumonia, compared with 20.0% of the patients not requiring advanced respiratory support (P<.001). All patients requiring advanced respiratory support presented with at least 1 coronavirus disease 2019 symptom, whereas only 51.5% of patients not requiring advanced respiratory support were symptomatic (P<.001). The Obstetrical Warning Score model allocated 1 point each for a hazard ratio of >100 beats per minute, temperature of >99.0 degrees F, C-reactive protein level of >2.0 mg/dL, respiratory rate between 20 and 24 breaths per minute, complaints of dyspnea, and a positive chest x-ray. A respiratory rate of >24 breaths per minute was assigned 2 points. The area under the curve for the Obstetric Warning Score system was 0.97 compared with 0.72 for the Early Warning Score system, 0.92 for the National Early Warning Score 2 system, and 0.85 for the Maternal Early Warning Criteria system. An Obstetric Warning Score of >= 3 was predictive of a requirement for advanced respiratory support with a sensitivity of 100%, specificity 64%, and a positive predictive value of 36%. CONCLUSION: The Obstetric Warning Score system presents a validated method for providers to identify pregnant patients who are at risk for respiratory failure and a requirement for advanced respiratory support.

Automated summary

This study aimed to evaluate the clinical characteristics of pregnant patients with COVID-19, develop and validate a pregnancy-specific early warning score. By comparing the presentations of different patient groups at diagnosis, a scoring system was successfully established to predict the need for respiratory support.