The epidemiology of sepsis related to childbirth has been studied in the United Kingdom, but not in the United States. Dr. Melissa E. Bauer, Department of Anesthesiology, University of Michigan Health System, Ann Arbor, MI and colleagues used the Nationwide Inpatient Sample database to examine this problem in the United States. Their work is published in the current issue of Anesthesia & Analgesia in the article “Maternal Sepsis Mortality and Morbidity During Hospitalization for Delivery: Temporal Trends and Independent Associations for Severe Sepsis.” They examined hospitalizations for delivery between 1998 and 2008.
Of over 40 million hospitalizations for delivery in the United States, the authors identified over 4,000 patients with severe sepsis. The frequency of sepsis varied little during the study period; the odds of becoming septic and the odds of dying each increased by 10% per year. Independent risk factors for developing sepsis included age >35 years, African American race, Medicaid insurance, retained products of conception, premature preterm rupture of membranes, congestive heart failure, chronic liver failure, chronic renal failure, human immunodeficiency virus infection, SLE, multiple gestation, and cerclage. Using logistic regression, the authors determined that stillbirth, preterm delivery, postpartum hemorrhage, and caesarean delivery during labor were significantly associated with sepsis. Elective caesarean delivery was protective against sepsis. Hemodialysis rates did not change during the study period, though the mechanical ventilation rate decreased. During the study period, severe sepsis increased 112% and sepsis-related death increased 129%. Organisms associated with sepsis included E. Coli, staphylococcus, streptococcus, and gram-negative organisms. The increases in sepsis and sepsis-related death are consistent with what has been observed in the UK.
The Nationwide Inpatient Sample database is limited in terms of data contained for each patient. The findings depend on the accuracy of coding. In addition, the data cannot be verified. It is also not possible to develop a timeline for what happened and at what time. End organ injury, for example, could have preceded sepsis.
The study does not give the clinician an idea of what can be done to decrease the incidence of sepsis. Given the increased prevalence, we need to be vigilant in managing parturients and treat patients with symptoms of sepsis as early as possible.