Is the Hartmann’s procedure for diverticulitis obsolete? National trends in colectomy for diverticulitis in the emergency setting from 1993 to 2018

Background Historically, Hartmann’s procedure (HP) has been the operation of choice for diverticulitis in the emergency setting. However, recent evidence has demonstrated the safety of primary anastomosis (PA) with or without diverting ileostomy. The purpose of this study was to evaluate the trends of, and factors associated with, HP compared to PA in emergency surgery for diverticulitis over 25 years. Methods Using the National Inpatient Sample database, we identified adult patients ≥ 18 years old who underwent emergency surgery for diverticulitis (HP or PA) between 1993 and 2018 using ICD-9 and ICD-10 codes. Patients with inflammatory bowel disease, gastrointestinal cancer or who underwent elective diverticulitis surgery were excluded. Trends in HP were analyzed using multivariable linear regression, and factors associated with HP were assessed with multiple logistic regression. Results Of 499,433 patients who underwent colectomy in the emergency setting for acute diverticulitis, 271,288 (54.3%) had a HP and 228,145 (45.7%) had a PA. Median age was 61 years (IQR: 50–73), 53% were women, and 70.5% were white. The proportion of HP slightly increased over the study period—HP comprised 52.6% of included cases in 1993–98 and 55.2% of cases in 2014–2018 ( p = 0.017). Advanced age (reference = 18–44 years; 45–54 years: OR 1.16, 95% CI 1.10–1.22; 55–64 years: OR 1.26, 95% CI 1.20–1.33; 65–74 years: OR 1.33, 95% CI 1.25–1.42; ≥ 75 years: OR 1.51, 95% CI 1.41–1.62), complicated diverticulitis (OR 1.41, 95% CI 1.36–1.46), and severity of illness (reference = minor; moderate: OR 1.46, 95% CI 1.38–1.54; major/extreme: OR 3.43, 95% CI 3.25–3.63) were associated with increased odds of HP. Conclusions Over a 26-year period, HP has remained the most performed procedure in the emergency setting for diverticulitis. Future work should focus on knowledge translation with a possible change in practice as more randomized controlled trials provide support for PA.