Infection rate correlated with time to repair of open neural tube defects (myelomeningoceles): an institutional and national study.

BACKGROUND: The optimal time to closure of a newborn with an open neural tube defect (NTD-myelomeningocele) has been the subject of a number of investigations. One aspect of timing that has received attention is its relationship to repair site and central nervous system (CNS) infection that can lead to irreversible deficits and prolonged hospital stays. No studies have evaluated infection as a function of surgical timing at a national level. We hypothesized an increase in wound infection in those patients with delays in myelomeningocele repair when evaluated in both a single-center and national database. METHODS: Treatment outcomes following documented times to transfer and closure were evaluated at Children's Hospital of Los Angeles (CHLA) for the years 2004 to 2014. Data of newborns with a myelomeningocele with varying time to repair were also obtained from non-overlapping abstracts of the 2000-2010 Kids' Inpatient Database (KID) and Nationwide Inpatient Sample (NIS). Poisson multivariable regression analyses were used to assess the effect of time to repair on infection and time to discharge. RESULTS: At CHLA, 95 neonates who underwent myelomeningocele repair were identified, with a median time from birth to treatment of 1 day. Six (6 %) patients were noted to have postrepair complications. CHLA data was not sufficiently powered to detect a difference in infection following delay in closure. In the NIS, we identified 3775 neonates with repaired myelomeningocele of whom infection was reported in 681 (18 %) patients. There was no significant difference in rates of infection between same-day and 1-day wait times (p = 0.22). Wait times of two (RR = 1.65 [1.23, 2.22], p < 0.01) or more days (RR = 1.88 [1.39, 2.54], p < 0.01), respectively, experienced a 65 % and 88 increase in rates of infection compared to same-day procedures. Prolonged wait time was 32 % less likely at facilities with increased myelomeningocele repair volume (RR = 0.68 [0.56 0.83], p < 0.01). The presence of infection was associated with a 54 % (RR = 1.54 [1.36, 1.74], p < 0.01) increase in the length of stay when compared to neonates without infection. CONCLUSION: Myelomeningocele closure, when delayed more than 1 day after birth, is associated with an increased rate of infection and length of stay in the national cohort. High-volume centers are associated with fewer delays to repair. Though constrained by limitations of a national coded database, these results suggest that early myelomeningocele repair decreases the rate of infection.