EP913 - How Low Should You Go? Quality of Hartmann’s Procedure and Its Impact on Stoma Reversal

Purpose/Background: There is a paucity of data concerning the quality of Hartmann’s procedures performed in clinical practice. We aimed to identify the frequency of features indicating a less-than-ideal Hartmann’s procedure (retained distal sigmoid and non-transrectus stoma location) and determine downstream impact on subsequent Hartmann’s reversal.

Methods/Interventions: This is a single-center retrospective study of patients undergoing Hartmann’s reversal between 2015-2025 at a university-based academic institution. Inclusion criteria were an index Hartmann’s procedure and index diagnosis of diverticular disease. Pathologic evidence of retained sigmoid, operative time, length of stay, and other clinical variables related to the stoma reversal were recorded. Pre-reversal CT scans, when available, were reviewed. Retained sigmoid was determined if the gross pathologic length of the trimmed portion of the “rectal” stump was >=4.3cm per receiver operating characteristic curve analysis. If pre-reversal CT imaging was available, the stoma location was classified as either transrectus or non-transrectus.

Results/Outcomes: A total of 53 patients underwent reversal of an initial Hartmann’s procedure for diverticular disease. The prior Hartmann’s procedure was performed in an academic setting in 15 patients (28.3%) and in a community setting in 38 patients (71.7%). Retained sigmoid was observed in 64.2% of cases, with a significantly lower incidence if the index surgery was performed in an academic hospital compared with community (40.0% vs 73.7%; p=0.02). The mean “rectal” stump specimen length was also significantly different in these groups (4.0cm vs 8.8cm; p=0.03). Non-transrectus stoma location was noted in 17.4% of the cases with pre-reversal CT available (0% in the academic group vs 24.2% in the community group; p=0.051). Retained sigmoid was associated with a significantly longer operative time at the time of stoma reversal (401 min vs 312 min; p=0.03) on univariate analysis. Multivariate analysis demonstrated that retained sigmoid predicted 87.5 minutes longer operative time (p=0.02) and non-transrectus stoma location predicted 4.6 days longer post-operative stay p=0.004). Neither was associated with blood loss or complications.

Conclusion/Discussion: Following reversal of an initial Hartmann’s procedure for diverticular disease, retained sigmoid was identified frequently and more often following index operations performed at community hospitals. This retrospective review shows the impact of retained sigmoid and non-transrectus colostomy on those patients who then undergo Hartmann’s reversal. Considering the high incidence of less-than-ideal Hartmann’s procedures and their downstream impact, quality improvement opportunities remain in this very common operation.