Back to Annual Meeting Program

Introduction: Primary anastomosis with or without proximal diversion is increasingly applied to pts requiring urgent colectomy for complicated disease of the sigmoid colon. Conversely, the Hartmann procedure (HP) is now often restricted to patients who are unstable or otherwise ill suited to primary anastomosis. As such, pts who are evaluated for Hartmann takedown often have formidable comorbities and considerable judgment is often required in pt selection. We sought to define the complication rate of Hartmann takedown in this setting.

Methods: A prospective complication database was searched for consecutive adult patients undergoing colostomy takedown with colorectal anastomosis (HP) at an academic teaching hospital from 1/1/02 to 12/31/10. Demographics, BMI, ASA classification, interval between Hartmann procedure and subsequent takedown, surgical indication, surgeon volume and specialty, length of stay and complications were recorded. Fisher's exact test was used to identify risk factors for postoperative complications.

Results: 104 pts underwent Hartmann reversal by 16 different surgeons; 7 of these surgeons did 4 or fewer procedures during the study period. 39 pts had their original Hartmann procedure done elsewhere; 38 of these reversals were done by a colorectal surgeon. During the same time period, 334 patients underwent a Hartmann procedure at our institution. 77/104 pts (74%) had their HP for complicated diverticulitis; anastomotic leak was the second most common indication. The median age was 61 years (31-84 yrs) and the interval from Hartmann procedure to reversal ranged from 87-1489 days. Only 8 pts (7.7%) had an ASA of 1 and at least 30 patients required a concomitant ventral hernia repair. 30 pts (29%) had complications and 12 (11%) had two or more complications (Table 1). There were two deaths, four anastomotic leaks, and seven patients had inadvertent enterotomies. Only ASA status predicted postop complications (p=.01)

Conclusions: Hartmann takedown is a morbid operation with a substantial risk of inadvertent enterotomy and serious complications. Excluding cases referred from elsewhere, there were more than fivefold the number of Hartmann procedures than takedowns performed during the study period. This suggests that Hartmann procedures are largely restricted to patients who are poor candidates for takedown and that their colostomy is highly likely to be permanent.

Table 1: Complications (n=30 pts)

SSI	7
Inadvertent enterotomy	7
Blood transfusions	6
Anastomotic leak	4
Arrhythmia	4
Urinary	3
MI	2
Death	2
Other	14