Factors Associated With Early Deaths Following Neonatal Circumcision

Purpose: Neonatal male circumcision is the most commonly performed surgical procedure in the United States. There is currently uncertainty about the balance between the risks and benefits of circumcision, in part because of a lack of sufficient data concerning the scope and magnitude of each. Although the most common risks associated with neonatal circumcision are generally understood to be mild (e.g., temporary bleeding), estimates of the likelihood of more serious complications, up to and including death, are both variable and conflicting. In this study, we attempted to quantify early deaths in subjects who underwent neonatal circumcision, and to identify the main factors associated with such mortality. Methods: We performed a retrospective analysis of all patients who underwent circumcision while hospitalized during the first 30 days of life from the years 2001 to 2010 using the National Impatient Sample (NIS). The NIS is the largest publicly available all-payer inpatient health care database in the United States, yielding national estimates of hospital inpatient stays. Weighted data from the NIS estimates more than 36 million hospitalizations nationally with data drawn from 44 States representing more than 95% of the U.S. population. Multiple patient and hospital level factors were analyzed with descriptive statistics relative to early mortality as the outcome of interest, defined here as deaths that occurred during the primary admission. Data are presented as percentages or Odds Ratio [OR] and 95% confidence intervals. This study was considered exempted by our Institutional Review Board. Results: Over the course of 10 years, there were 200 recorded early deaths among 9,899,110 subjects who underwent circumcision in a hospital inpatient setting (1 death in every 49,166 circumcisions). Subjects who died soon after circumcision were predominantly white (63.9% vs 68.2% among survivors), treated at teaching hospitals (80.1% vs 49.9% among survivors), and had medical care funded by private insurance (52.1% vs 61.4 among survivors) or Medicaid (39.7% vs 31.7% among survivors). Compared to survivors, subjects who died after circumcision were also more likely to have associated co-morbid conditions, such as cardiac disease (OR: 697.8 [378.5-1286.6] p < 0.001), coagulopathy (OR: 159.6 [95.6-266.2] p < 0.001), fluid and electrolyte disorders (OR: 68.2 [49.1-94.6] p < 0.001), or pulmonary circulatory disorders (OR: 169.5 [69.7-412.5] p < 0.001). Conclusion: Data from this large national cohort of infants who underwent circumcision in the first 30 days of life can help to inform estimates of the risk of early death following the procedure, especially in patients with co-morbid conditions. Such conditions include cardiac disease, coagulopathy, fluid and electrolyte disorders, and pulmonary circulatory disorders. Advance recognition of these factors could inform parental decisions regarding the procedure, thereby potentially reducing associated risks.

• Copyright © 2018 by the American Academy of Pediatrics