Back to 2025 Abstracts
PREDICTORS OF UNDERGOING HARTMAN PROCEDURE AS AN INITIAL INTERVENTION FOR SIGMOID VOLVULUS
Monica Patten
*, Aisha Shaikh, Saad Shebrain
General Surgery, Western Michigan University Homer Stryker MD School of Medicine, Kalamazoo, MI
BACKGROUND: Sigmoid resection with end colostomy (Hartmann Procedure, HP) is a critical intervention in colorectal surgery, and is commonly performed on an emergency basis. While sigmoid volvulus frequently presents as an emergent colonic obstruction, sigmoid resection and anastomosis (SR-PA) remains the most common approach. This study aims to identify predictors associated with use of HP in patients who present with sigmoid colon volvulus.
METHODS: We performed a retrospective analysis of 12 years of data from the ACS-NSQIP database (2005-2017). Patients with ICD-9 and ICD-10 codes for sigmoid volvulus who underwent SR-PA or HP were included. We randomly divided the cohort into training and testing groups at a ratio of 70:30. We utilized patient demographic and preoperative variables found in the ACS-NSQIP universal and colon-specific Surgical Risk Calculators (ACS-NSQIP-SRC) as predictors. We performed univariate and multivariable logistic regression analyses and built a predictive model of the training group. We assessed performance of the predictive model on test data using calibration (Hosmer-Lemeshow’s goodness-of-fit statistic) and discrimination (concordance index [c-statistics]) techniques. We used SPSS for analysis. A two-sided p-value <.05 was considered statistically significant.
RESULTS: Of 5,732 patients who met inclusion criteria, 1,454 (25.4%) underwent HP. There were 2,779 (48.5%) females. Training and testing groups were similar in demographic characteristics and preoperative data. Univariate logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (CI) of undergoing HP associated with each independent variable found in ACS-NSQIP-SRC. The multivariable logistic regressions for the training and testing models were statistically significant. Eight variables were statistically significant independent predictors for undergoing HP. They included male sex (2-folds), admission from nursing home or chronic care facility (2.5-folds), dependent functional status (2.7-folds), chronic use of a steroid (2.0-fold), congestive heart failure (3.0-fold), emergency surgery (1.9-fold), high ASA class ( range from 1.5 to 9.6 folds), and sepsis (2.5-folds). The value of concordance (c) statistics (95% confidence interval) and the area under the ROC curve of predicting undergoing HP in the training cohort and testing cohort were 0.782 (95% CI, 0.765, 0.798) and 0.785 (95% CI, 0.760, 0.810), respectively. Both models did not differ, with p=.838, indicating that the performance in the test dataset provided a reasonable estimate of predictability.
CONCLUSION: Certain demographics and preoperative characteristics are associated with an increased likelihood of undergoing HP in patients with sigmoid volvulus. Patients should be aware of these predictors when discussing operative approaches.
Back to 2025 Abstracts
