Surgical complications after open and laparoscopic surgery for perforated peptic ulcer in a nationwide cohort

BACKGROUND: Surgery for perforated peptic ulcer (PPU) is associated with a risk of complications. The frequency and severity of reoperative surgery is poorly described. The aims of the present study were to characterize the frequency, procedure-associated risk and mortality associated with reoperation after surgery for PPU.

METHODS: All patients treated surgically for PPU in Denmark between 2011 and 2013 were included. Baseline and clinical data, including 90-day mortality and detailed information on reoperative surgery, were collected from the Danish Clinical Register of Emergency Surgery. Distribution frequencies of reoperation stratified by type of surgical approach (laparoscopy or open) were reported. The crude and adjusted risk associations between surgical approach and reoperation were assessed by regression analysis and reported as odds ratio (OR) with 95 per cent c.i. Sensitivity analyses were carried out.

RESULTS: A total of 726 patients were included, of whom 238 (32·8 per cent) were treated laparoscopically and 178 (24·5 per cent) had a laparoscopic procedure converted to laparotomy. Overall, 124 (17·1 per cent) of 726 patients underwent reoperation. A persistent leak was the most frequent cause (43 patients, 5·9 per cent), followed by wound dehiscence (34, 4·7 per cent). The crude risk of reoperative surgery was higher in patients who underwent laparotomy and those with procedures converted to open surgery than in patients who had laparoscopic repair: OR 1·98 (95 per cent c.i. 1·19 to 3·27) and 2·36 (1·37 to 4·08) respectively. The difference was confirmed when adjusted for age, surgical delay, co-morbidity and American Society of Anesthesiologists fitness grade. However, the intention-to-treat sensitivity analysis (laparoscopy including conversions) demonstrated no significant difference in risk. The risk of death within 90 days was greater in patients who had reoperation: crude and adjusted OR 1·53 (1·00 to 2·34) and 1·06 (0·65 to 1·72) respectively.

CONCLUSION: Reoperation was necessary in almost one in every five patients operated on for PPU. Laparoscopy was associated with lower risk of reoperation than laparotomy or a converted procedure. However, there was a risk of bias, including confounding by indication.