Single- Versus Multi-Stage Pull-Through for Neonatal Hirschsprung's Disease: Practice Trends and Outcomes

Sunday, October 27, 2013: 9:28 AM
Windermere Ballroom W (Hyatt Regency Orlando, formerly the Peabody)
Jason P. Sulkowski, MD, Jennifer N. Cooper, PhD, Anthony Congeni, Benedict C. Nwomeh, MD, MPH, Peter C. Minneci, MD, MHSc and Katherine J. Deans, MD, MHSc, Nationwide Children's Hospital, Columbus, OH


Despite limited evidence directly comparing outcomes from single- and multi-stage procedures many surgeons have adopted the single-stage pull-through procedure for Hirschsprung’s disease (HD).  The purpose of this study was to create and validate a multi-institutional cohort of neonatal HD patients and categorize their surgical treatment in the Pediatric Health Information System (PHIS) database to determine practice trends and outcomes.


Patients treated at children’s hospitals contributing to the PHIS database between 1999 and 2009 were included if they were inpatients prior to 60 days of life with the ICD-9-CM diagnosis code for HD (751.3) and a procedure code for a pull-through procedure by age 1 year.  Patients who had an ostomy procedure prior to a pull-through were classified as “multi-stage” and those who had a pull-through procedure without or prior to an ostomy procedure were classified as “single-stage”.   The diagnosis and classification algorithm was validated by chart review at our institution.  Comparisons are based on data present during the admission in which the surgical treatment was initiated (i.e. the ostomy for multi-stage and pull-through for single-stage). Readmissions and reoperations were calculated subsequent to this admission excluding the pull-through procedure (and the associated admission) in the multi-stage group. Multivariable logistic regression using generalized estimating equations was used to compare outcomes between treatment groups.


Of the 1555 patients identified with HD, 1201 (77.2%) were classified as single-stage and 354 (22.8%) as multi-stage. There was no misclassification of disease or surgical treatment based on chart review at our institution. The percentage of patients treated with a single-stage procedure increased over time (p=0.03). Compared to single-stage patients, multi-stage patients had higher rates of  prematurity (8.8% vs. 5.3%, p=0.01), enterocolitis (15.8% vs. 7.2%, p<0.001), preoperative mechanical ventilation (15.3% vs. 7.3%, p<0.001), lower birth weight (3190g ±635g vs. 3324g ±536g, p=0.001) and had shorter pre-operative length of stay (LOS) (4, IQR 1-7 vs. 5, IQR 2-8, p=0.02) but longer total LOS (12, IQR 8-20 vs. 11, IQR 7-17, p=0.001) during that admission.  Multi-stage patients also had higher rates of readmission (58.5 vs. 37.9%, p<0.001) and reoperation (38.7% vs. 26.0%, p<0.001). Results were similar after adjustment for risk factors that differed between treatment groups.


Our algorithm accurately identified and classified patients with neonatal HD and their surgical treatments in the PHIS database. This large multi-institutional cohort can be used to better understand treatments and outcomes for this rare disease. The current analysis demonstrates increasing treatment of neonatal HD with single-stage pull-through over time. Patients undergoing multi-stage procedures were more severely ill and had higher rates of readmission and reoperation. Future analyses of this cohort, including a propensity score matched analysis, may reveal further differences in practice patterns and outcomes.