Impact of the implementation of an obstetric sepsis protocol on maternal and neonatal outcomes

Abstract Objective Maternal sepsis is a leading cause of preventable maternal morbidity and mortality in the United States. We assessed the implementation of a standardized early‐goal‐directed sepsis care bundle on maternal and neonatal outcomes. This bundle was activated by a set of vital sign criteria in combination with providers’ concern for infection and included immediate rapid response team evaluation and standardized orders for fluid resuscitation, antibiotics, and labs. Methods A retrospective cohort study was performed at a single tertiary care center between January 1, 2012 and December 31, 2018. Individuals were included if they were pregnant or within 6 weeks of delivery with a singleton gestation and had a recorded temperature measurement of >100.4°F (38°C) or <96.8°F (36°C) in addition to a heart rate >100 beats/min or a respiratory rate >20 breaths/min within 30 min of the documented temperature during an inpatient stay or triage visit. Groups were categorized as pre‐implementation (January 2012–September 2016) and post‐implementation (October 2016–December 2018) of the “code sepsis” protocol. The primary outcome was a composite of maternal sepsis‐related outcomes: (1) intensive care unit admission, (2) acute renal failure (defined as a creatinine > 1.1 mg/dL), and (3) maternal death. Multiple secondary outcomes were analyzed. Outcomes were compared across groups using generalized estimating equation models and adjusted for significant characteristics. Results Overall, 5368 pregnancies met criteria for inclusion: 56% pre‐implementation and 44% post‐implementation. Post‐implementation, 126 code sepsis protocols were activated after providers had clinical concern for infection. An identified infection diagnosis was found in 12.2% of the post‐implementation and 12.1% of the pre‐implementation group with urinary tract infection being the most common. The primary outcome occurred in 7.0% of the pre‐implementation group and 4.6% of the post‐implementation group (odds ratio [OR], 0.66; 95% confidence interval [CI], 0.52–0.83; adjusted OR [aOR], 0.64; 95% CI, 0.48–0.83). Stratified analysis by temperature >100.4°F (38°C) or temperature <96.8°F (36°C) revealed an association when criteria were met with hypothermia (OR, 0.55; 95% CI, 0.39–0.78; aOR, 0.48; 95% CI, 0.33–0.70), but not hyperthermia. When activated, the protocol had >92% adherence. Conclusion Implementation of an early‐goal‐directed sepsis care pathway led to improved maternal outcomes in an obstetric population with a 36% reduction in the primary outcome.