Recurrence of Severe Maternal Morbidity and Transfusion During Delivery Hospitalisations: A Retrospective Cohort Study

Objective: To determine risks for non-transfusion severe maternal morbidity and transfusion during a second delivery hospitalisation based on clinical risk factors and obstetric complications from an index, first delivery hospitalisation. Design: Retrospective cohort. Population: Delivery hospitalisations in the 2010–2017 New York State Inpatient Database. Methods: Patients with a first index delivery hospitalisation followed by a second delivery hospitalisation during the study period were included. Clinical risk factors and obstetric complications were obtained from the first index delivery hospitalisation. Adjusted logistic regression models for non-transfusion severe maternal morbidity during the second delivery were performed with adjusted (aORs) odds ratios as measures of effect. These analyses were then repeated for the outcome of transfusion. Results: Of 624 500 paired delivery hospitalisations to 312 250 women, severe maternal morbidity occurred among 0.85% of second deliveries (n = 2672). When adjusted analysis was performed, several clinical factors were associated with severe maternal morbidity in a subsequent pregnancy, including severe maternal morbidity during the index pregnancy (aOR 8.4, 95% CI 7.0, 9.9), transfusion (aOR 2.0, 95% CI 1.6, 2.4) and pregestational diabetes (aOR 2.2, 95% 1.6, 2.9). When analyses were repeated for transfusion, several factors were associated with increased risk, including severe maternal morbidity (aOR 1.5, 95% CI 1.2, 1.8), index transfusion (aOR 6.3, 95% CI 5.6, 7.0), chronic heart disease (aOR 1.6, 95% 1.4, 1.9) and pregestational diabetes (aOR 1.7, 95% 1.3, 2.2). Conclusion: Many obstetric complications and chronic conditions identified during an index delivery hospitalisation are associated with severe morbidity during a second, subsequent delivery. Index severe maternal morbidity is associated with the highest odds. These findings may be of use in patient counselling and risk stratification.