Background The modern wound dressing is produced to absorb fluid and protect against external contamination. The choice of which wound dressing to apply after surgery is usually based on local tradition. There are various impervious dressings on the market. Even if the wound is sterile before application, there will be subsequent recolonisation of skin microbiota. Previous studies suggest that a high bacterial load on the skin hampers wound healing and might be a risk for SSI. Aim The aim was to compare bacterial recolonisation on the shoulder under three different wound dressings, 48 h after sterile preparation of the skin as in preparation for surgery. Method In 25 healthy volunteers, a standard pre-surgical skin disinfection for a deltopectoral incision was made on the left shoulder with 0.5% chlorhexidine solution in 70% ethanol. Three different wound dressings were then placed on the shoulder, and 48 h later the skin beneath each dressing was swabbed, subsequently cultured and bacterial density analysed using viable count. Results The bacterial recolonisation under air-dry (gauze) dressing was significantly lower ( p = .0001) compared to semipermeable and occlusive wound dressings. Conclusion Choosing a less permeable wound dressing may lead to an increased bacterial load on the skin during the first 48 h after surgery.