Comparison of Diversion Strategies for Management of Acute Complicated Diverticulitis in a US Nationwide Cohort

Yas Sanaiha,Peyman Benharash,Esteban Aguayo,Formosa Chen,Joseph Hadaya

doi:10.1001/jamanetworkopen.2021.30674

Abstract

Diverticulitis of the colon is an increasingly prevalent disease with significant implications for patient quality of life and health system resource expenditure. Although several randomized clinical trials and meta-analyses of Hartman procedure (HP) and primary anastomosis and proximal diversion (PAPD) have found surgical equipoise, questions regarding the relative performance of these treatments when applied broadly remain. To examine use of and outcomes after urgent sigmoid colectomy with end colostomy (ie, HP) vs PAPD in management of complicated diverticulitis. This retrospective cross-sectional study was a multicenter, population-based examination of inpatient hospitalizations, not including long-term rehabilitation facilities, using data from the 2014 to 2017 Nationwide Readmissions Database. It was performed from November 2020 to January 2021. Included patients were adults admitted with acute diverticulitis requiring HP or PAPD within 48 hours of admission. Undergoing HP vs PAPD. Inverse probability treatment analysis was used to compare outcomes, including index mortality, composite complications (ie, neurologic, infectious, and cardiovascular complications), length of stay, and readmissions within 90 days. During the study period, an estimated 1 072 395 adults (615 954 [57.4%] women; median [IQR] age, 64 [52-76] years) required nonelective hospitalization for acute colonic diverticulutus. A total of 34 126 patients required diversion, with 32 326 patients (94.7%) undergoing HP and 1800 patients (5.3%) undergoing PAPD within 48 hours of admission. Patients undergoing PAPD had a decreased median (IQR) age (60 [51-70] years vs 65 [54-74] years; P < .001) and decreased rates of end organ dysfunction (520 patients [28.9%] vs 11 514 patients [35.6%]; P < .001). In inverse probability treatment weight analysis, the odds of mortality (adjusted odds ratio [aOR], 0.63; 95% CI, 0.32-1.40), complications (aOR, 0.86; 95% CI, 0.66-1.13), and nonhome discharge (aOR 1.15; 95% CI, 0.83-1.60) were similar for PAPD compared with HP. Among 1772 patients who underwent PAPD and survived index hospitalization, there was an increased incidence of 90-day readmission compared with 30 851 patients who underwent HP and survived index hospitalization (393 patients [22.2%] vs 4384 patients [14.2%]; P < .001) with increased hazard of ostomy reversal (hazard ratio, 1.46; 95% CI, 1.08-1.99). This study found that the use of PAPD was associated with comparable index hospitalization outcomes vs use of HP, while readmission rate and ostomy risk were statistically significantly increased among patients who underwent PAPD compared with patients who underwent HP. These findings suggest that further delineation of criteria for appropriate application of PAPD in the urgent setting are warranted.

Highlights

IntroductionWith an increasing burden in industrialized nations, diverticulitis has been found to be associated with 1.5 million days of inpatient care and more than 2 billion dollars in annual hospitalization costs across the US. The urgent surgical management of acute perforated diverticulitis has significantly evolved, and the 2-stage operation known as Hartmann procedure (HP) is currently the most common approach. More recently, use of primary anastomosis with or without diverting loop ileostomy (PAPD) has been suggested to be safe in the treatment of patients with gross peritoneal contamination.5,6To date, several randomized clinical trials and meta-analyses have found comparable mortality and complication rates between primary anastomosis and proximal diversion (PAPD) and HP for Hinchey stage III to IV perforated diverticulitis, with reports of increased rates of stoma reversal in the PAPD group. In an attempt to evaluate the relative merits of PAPD at the national level, Gawlick et al used National Surgical Quality Improvement Program (NSQIP) data and reported similar outcomes for the 2 diversion strategies
In inverse probability treatment weight analysis, the odds of mortality, complications, and nonhome discharge were similar for primary anastomosis and proximal diversion (PAPD) compared with Hartmann procedure (HP)
Among 1772 patients who underwent PAPD and survived index hospitalization, there was an increased incidence of 90-day readmission compared with 30 851 patients who underwent HP and survived index hospitalization (393 patients [22.2%] vs 4384 patients [14.2%]; P < .001) with increased hazard of ostomy reversal

Summary

Introduction

With an increasing burden in industrialized nations, diverticulitis has been found to be associated with 1.5 million days of inpatient care and more than 2 billion dollars in annual hospitalization costs across the US. The urgent surgical management of acute perforated diverticulitis has significantly evolved, and the 2-stage operation known as Hartmann procedure (HP) is currently the most common approach. More recently, use of primary anastomosis with or without diverting loop ileostomy (PAPD) has been suggested to be safe in the treatment of patients with gross peritoneal contamination.5,6To date, several randomized clinical trials and meta-analyses have found comparable mortality and complication rates between PAPD and HP for Hinchey stage III to IV perforated diverticulitis, with reports of increased rates of stoma reversal in the PAPD group. In an attempt to evaluate the relative merits of PAPD at the national level, Gawlick et al used National Surgical Quality Improvement Program (NSQIP) data and reported similar outcomes for the 2 diversion strategies. The urgent surgical management of acute perforated diverticulitis has significantly evolved, and the 2-stage operation known as Hartmann procedure (HP) is currently the most common approach.. Several randomized clinical trials and meta-analyses have found comparable mortality and complication rates between PAPD and HP for Hinchey stage III to IV perforated diverticulitis, with reports of increased rates of stoma reversal in the PAPD group.. In an attempt to evaluate the relative merits of PAPD at the national level, Gawlick et al used National Surgical Quality Improvement Program (NSQIP) data and reported similar outcomes for the 2 diversion strategies. Goldstone et al examined statewide data and found increased mortality among patients undergoing PAPD, concluding that recommendations regarding primary anastomosis in the urgent setting may need reevaluation

Methods

Results

Discussion

Conclusion