ABSTRACT: It is still a matter of debate whether delayed primary closure (DPC) of dirty abdominal incisions reduces surgical site infections (SSIs) compared with primary closure (PC). Our objective was to determine whether DPC of dirty abdominal incisions reduces SSIs.
A controlled randomized study was conducted at an academic tertiary care 1,500-bed university hospital in Western India involving 81 consecutive patients with dirty abdominal incisions. Only 77 patients (DPC = 37, PC = 40) were evaluable because of the deaths of four patients. A total of 52 patients had peptic or typhoid perforations, whereas the rest had appendicular perforations/abscesses, penetrating or blunt abdominal injuries with gastrointestinal perforation, or intra-peritoneal abscesses. Patients were randomized to have their surgical incisions (skin and subcutaneous tissue) either closed primarily (PC) or left open with saline-soaked gauze dressings for DPC on the 3(rd) postoperative day or later if the incision conditions were inappropriate for closure. The main outcome measure was the incidence of postoperative SSI.
In the entire series, SSI developed after incision closure in 23% of the patients. Infections were significantly more common in the PC group (42.5% vs. 2.7% for DPC; p = 0.0000375). There also were significantly more cases of abdominal dehiscence in the PC group (DPC 1 [2.7%] vs. PC 10 [25%]; p = 0.005). The mean complete incision healing (CIH) time and length of hospital stay (LOS) were longer after PC (18.52 days) than DPC (13.86 days), resulting in a significant difference in the end point of healing and LOS (p = 0.0207). Short-term cosmetic results for PC incisions were significantly inferior to those for DPC (p = 0.03349).
Delayed primary closure is a sound incision management technique that should be utilized for dirty abdominal incisions. It significantly lowers the rate of superficial SSI as well as fascial dehiscence and reduces the mean CIH time and hospitalization. The short-term cosmetic appearance is superior.
Surgical Infections 05/2009; 10(2):129-36. · 1.80 Impact Factor