Ghosts in the machine: a multi-institutional comparison of laparoscopic and open pyloromyotomy.

INTRODUCTION: The purpose of this study was to compare laparoscopic and open pyloromyotomy using data from multiple centers in the United States. METHODS: Children's hospitals that have a predominant surgical approach to pyloromyotomy were identified in the Pediatric Health Information System database. Using 2005 data, institutions were stratified into open (OPEN) and laparoscopic (LAP) groups. Patients with significant comorbid conditions were excluded. Group differences were compared using t tests and Mann-Whitney nonparametric tests for continuous variables and exact tests for categorical variables. RESULTS: A total of 3 hospitals were in the LAP group (n = 207), and 4 hospitals were in the OPEN group (n = 357). The LAP group had a longer mean length of stay (LAP = 2.5 +/- 1.7, OPEN = 2.1 +/- 1.4 days; P = .02). Mean total hospital charges were similar in both groups (LAP = $11307 +/- 9499, OPEN = $11245 +/- 4841; P = .93), but there was significant skewness of the distribution for the LAP group. Nonparametric analysis demonstrated a statistically significant difference in charges (LAP median = $9727, min = $5075, max = $94323, OPEN median = $10001, min = $1614, max = $46461; P = .004). Four patients in the LAP group had charges ranging from approximately $56000 to $94000, which may have resulted from surgical complications. CONCLUSION: Prolonged length of stay and skewed hospital charge data in patients undergoing laparoscopic pyloromyotomy may be the result of rare but serious complications associated with the laparoscopic approach.