Colectomy for constipation: time trends and impact based on the US Nationwide Inpatient Sample, 1998-2011

BackgroundCurrent guidelines include subtotal colectomy as treatment for refractory slow transit constipation. AimTo use the US Nationwide Inpatient Sample (NIS) (1998-2011) and longitudinal data from the State Inpatient Database (2005-2011), comparable to NIS, to examine colectomy rates, in-hospital morbidity and emergency department (ED) visits or readmissions among patients treated for constipation. MethodsColectomies for any reason were identified based on the primary procedural code (ICD-9-CM 45.8x). Index hospitalisations were defined by the primary diagnosis of constipation (ICD-9-CM 564.x) associated with the primary procedural code for colectomy (ICD-9-CM45.8x) after exclusion of other diseases associated with colectomy. Demographic variables, comorbidities, complications and adverse events during the hospitalisation were captured, and ED visits and admissions were recorded for periods before and after colectomy. ResultsNationally, colectomies for constipation rose from 104 procedures in 1998 (1.2% of annual colectomies) to 311 in 2011 (2.4% of annual colectomies). While there were no perioperative deaths, perioperative complications occurred in 42.7% of patients during the index hospitalisation. Longitudinal data were analysed for 181 patients, with similar perioperative complications and a readmission rate of 28.9% within the first 30 days after the index hospitalisation. Resource utilisation was tracked for a median time of 630 (0-2386) before and 463 (0-2204) days after colectomy with unchanged ED visits (median: 2 vs. 2, P = 0.21), but increased hospitalisations (median: 1 vs. 2, P = 0.003). ConclusionsColectomy rates for constipation are rising, are associated with significant morbidity and do not decrease resource utilisation, raising questions about the true benefit of surgery for slow transit constipation.