Background: To gauge the feasibility of a pediatric trial of a drug to prevent post‐operative ileus, we sought to estimate the number of children who underwent bowel resection with a primary anastomosis. We initially queried the Kids' Inpatient Database (KID), the largest all‐payer pediatric inpatient care database in the US. While KID can provide population‐based estimates, it can only identify surgical procedures using ICD‐9‐CM codes (hereafter ICD‐9), and a substantial proportion of bowel resections could not be fully characterized. To better characterize these procedures, we sought to extrapolate from a claims database that included the more granular CPT codes, in addition to ICD‐9 procedure codes.
Objectives: Among patients for whom ICD‐9‐ procedure codes indicated only nonspecific small or large bowel resection, to determine the proportion of surgeries that CPT codes could reclassify as involving a primary anastomosis. Methods: We used the inpatient admissions data from the MarketScan® Commercial Claims and Encounters database, which includes both ICD‐9 and CPT procedure codes. Pediatric patients <19 years of age) admitted and discharged in 2012 with an ICD‐9‐CM procedure code for bowel resection but without any additional code to distinguish primary anastomosis from diverting ostomy were included. For each eligible bowel resection procedure, we further assessed the type of surgery based on CPT codes.
Results: Of 1,174 pediatric patients who had bowel resection surgery based on ICD‐9 procedure codes, we could not determine whether 802 (68.3%) procedures involved primary anastomoses vs. diverting ostomies. Using information from CPT coding, 274 (34.2%) bowel resection surgeries could be further classified as primary anastomoses and 194 (24.2%) as ostomies.
Conclusions: Bowel resection procedures that involve primary anastomoses are incompletely identified in children in administrative data using only ICD‐9 procedure codes. Use of CPT codes improves classification. Our results suggest that when ICD‐9 procedure codes failed to specify the type of bowel resection, about one‐third of patients actually had a primary anastomosis.
