Impact of preoperative antibiotic prophylaxis on the incidence of surgical site infection in a Ugandan Hospital

Background:

Surgical site infection (SSI) constitutes a major clinical problem worldwide causing significant morbidity, mortality and increased hospital costs. Literature provides little information on risk factors associated with SSI in Sub Saharan African hospitals and the implementation of pre-operative Anti Microbial Prophylaxis (AMP).

Objective:

Determine incidence of SSI, risk factors and the subsequent impact of AMP implementation on the incidence of SSI in a Ugandan referral hospital, utilizing a standardized data capturing method for SSI surveillance (Hospitals in Europe Link for Infection Control and Surveillance).

Methods:

In 2009, a nine-month, prospective surveillance study was conducted in the general surgery (including orthopedics) and gynecological wards of the St. Raphael of St. Francis Hospital, Kampala, Uganda. 1149 patients were recruited. Observation period was divided into 2 phases: pre-intervention phase (no AMP administered; combination post-operative antibiotic therapy successively given) and intervention phase (AMP administered preoperatively). Use of innovative patient follow-up schemes such as discharge questionnaires and contacting patients after discharge on mobile phones on the 15^th and 30^th post operative day. Ascertainment of SSI was made using strict definitions.

Results:

SSI occurred in 92 (9.64%) of 954 patients in the preintervention phase and in 5 (2.56%) of 195 patients in the intervention phase (p=0.003). We have estimated that 69 of the 92 preintervention SSIs could have been prevented with AMP; not having AMP pre-operatively constitutes a RR of 4,05 (CI 1,6-10,1).

Higher SSI incidence rates were observed in orthopedic surgery (34%), large bowel surgery (28%) and caesarean sections (7.2%); the rate of SSI were reduced to 0%, 5.55% and 2.5% respectively with the implementation of AMP in the intervention phase. 

Almost half (43%) of SSI were ‘captured’ by the use of the new protocol. There was complete follow up over a 30 day observation period for SSI in both phases.

Risk factors for the development of SSI in the preintervention group were found to be ASA score ≥2 (p=<0.05), wound class >3 (p=<0.05) length of stay (p=<0.000), duration of operation (p=<0.05), pre-operative hair removal (p=0.004), diabetes (p=<0.05), nutrition status poor (p=<0.05), intraoperative transfusions (p=0.05), male gender (p=0.000). After the intervention with AMP only ASA score ≥2 and wound class remained significant risk factors for SSI in the intervention group.

Conclusions: The high incidence of SSI in the preintervention group was effectively reduced with only the administration of a single preoperative dose, 30 minutes before incision. Moreover, innovative follow up methods improved the detection of postoperative SSI. Although there is obviously room for improvement, the four-fold reduction in SSI using a cheap and sustainable method is impressive.