- Three randomised controlled trials met inclusion and exclusion criteria.
- No significant reduction in wound infections with the use of occlusive compared to non-occlusive surgical dressings was observed in any of the three trials. In general there was a low incidence of infection, so large trials would be required to detect a significant difference.
- The three trials involved patients who had undergone elective plastic surgery, 'clean' abdominal surgery, or heart surgery. In the largest trial Wikblad & Anderson (1995) compared the effectiveness of three types of wound dressings in patients undergoing elective coronary bypass or valve replacement surgery. Seventy-seven of the patients were randomised to an occlusive hydrocolloid dressing (DuoDerm), 92 patients to a conventional absorbent dressing and 81 patients to a semi-occlusive dressing. The incidence of wound infections requiring antibiotics among the 216 patients reported on was 5% (n=11) and were not significantly less likely in the group of patients with occlusive dressings.
- Two of the three trials did not use intention-to-treat analysis of outcome data for patients randomised and then withdrawn from the study. However, this fact is unlikely to alter the negative results for the benefits of occlusive dressings in reducing wound infections.