To investigate the effect of preoperative initiation of low molecular weight heparin as prophylaxis for deep venous thrombosis in patients at risk of developing surgical-site infections after knee arthroplasty.

Case–control study nested in a cohort. The incidence of surgical-site infection in the cohort was calculated. With the use of data extracted from medical histories and after adjustment for other risk factors, the effect of preoperative heparinization on the risk of incisional and prosthetic infection among case-patients and control-patients (1:3 ratio) was assessed.

Orthopedic department in a tertiary-care referral hospital.

A cohort of 160 consecutive patients who had received prosthetic knee implants between October 1, 2001, and November 30, 2003.

Eighteen patients with surgical-site infections were identified, yielding an incidence of incisional and prosthetic infection of 6.9 (95% confidence interval [CI95], 3.5 to 12.0) and 4.4 (CI95,1.8 to 8.8) cases per 100 patients undergoing surgery, respectively. Surgical-site infection was associated with preoperative use of low molecular weight heparin (odds ratio [OR], 6.2 after adjustment for medical and surgical factors; CI95, 1.5 to 23). Prosthetic infection was strongly associated with preoperative use of prophylaxis (OR, undetermined [100% exposure in case-patients vs 35% exposure in control-patients]; P = .002), but incisional surgical-site infection was not.

The use of low molecular weight heparins immediately before knee arthroplasty as prophylaxis for deep venous thrombosis should be questioned because of probable increased risk of prosthetic infection.

To describe the use of laminar airflow, body exhaust, and ultraviolet lights during total knee replacement (TKR) in four U.S. states.

Survey of healthcare facilities.

Hospitals in Illinois, North Carolina, Ohio, and Tennessee that performed TKR during 2000 as identified by Medicare claims data.

Hospitals responding to a mailed questionnaire.

Two hundred ninety-five (73%) of 405 eligible hospitals that performed 18,374 primary and revision TKR procedures responded to the questionnaire. Among responding hospitals, 30% reported regular use (for > 75% of procedures) of laminar airflow, 42% reported regular use of body exhaust, and 5% reported regular use of ultraviolet lights. Among hospitals providing complete data, 150 (58%) performing 66% of procedures reported regular use of at least one of these techniques. On regression analyses, laminar airflow was used more often by hospitals with a TKR volume greater than 25 procedures per year (odds ratio [OR], 2.0; 95% confidence interval [CI95], 1.1-3.7) and orthopedic residency programs (OR, 2.8; CI95,1.3-6.3), but its use was not significantly related to hospital setting or ownership status.

Although these clean air practices are not recommended by any U.S. governmental or professional organization, they are used in nearly two-thirds of TKR procedures. Better information about their impact on current practice and more explicit guidelines may aid decisions about the use of these resource-intensive infection control practices.

To review the evidence evaluating perioperative intranasal mupirocin for the prevention of surgical-site infections according to type of surgical procedure.

Systematic review and meta-analysis of published clinical trials.

Studies included were either randomized clinical trial or prospective trials at a single institution that measured outcomes both before and after an institution-wide intervention (before-after trial). In all studies, intervention and control groups differed only by the use of perioperative intranasal mupirocin in the intervention group.

Patients undergoing general or nongeneral surgery (eg, cardiothoracic surgery, orthopedic surgery, and neurosurgery).

Risk of surgical-site infection following perioperative intranasal mupirocin versus usual care.

Three randomized and four before-after trials met the inclusion criteria. No reduction in surgical-site infection rate was seen in randomized general surgery trials (summary estimates: 8.4% in the mupirocin group and 8.1% in the control group; relative risk [RR], 1.04; 95% confidence interval [CI95], 0.81 to 1.33). In nongeneral surgery, the use of mupirocin was associated with a reduction in surgical-site infection in randomized trials (summary estimates: 6.0% in the mupirocin group and 7.6% in the control group; RR, 0.80; CI95, 0.58 to 1.10) and in before-after trials (summary estimates: 1.7% in the mupirocin group and 4.1% in the control group; RR, 0.40; CI95, 0.29 to 0.56).

Perioperative intranasal mupirocin appears to decrease the incidence of surgical-site infection when used as prophylaxis in nongeneral surgery. Given its low risk and low cost, use of perioperative intranasal mupirocin should be considered in these settings.

To determine whether certain hair removal policies are better than others to prevent surgical-site infections in patients undergoing clean surgery.

Publications were retrieved by a systematic search of Medline, the Cochrane Library, and EMBASE up to February 2005. Additionally, the reference lists of all identified trials were examined. All randomized trials, quasi-randomized trials, and systematic reviews or meta-analyses of randomized or quasi-randomized trials comparing hair removal policies in clean surgery were selected. Trials involving patients undergoing cranial neurosurgery were excluded. Two reviewers independently assessed trial quality and extracted data. Disagreements were resolved by discussion with a third reviewer. Data from the original publications were used to calculate the relative risk or risk difference of surgical-site infection. Data for similar outcomes were combined in the analysis, where appropriate, with the use of a random effects model.

Four trials were included in the review. No eligible systematic review or meta-analysis of randomized or quasi-randomized trials was found. The quality of the trials and how they were reported were generally unsatisfactory. Evidence regarding whether preoperative hair removal has any effect was inconclusive. When hair removal was considered necessary, evidence about the best time for removal was inconclusive. There was some evidence that hair removal by clipper is superior to removal by razor.

Because of insufficient evidence as a basis for recommendations, the practical consequences for ward management were essential when the Dutch Working Party on Infection Prevention formulated its recommendations for hair removal policies. Large randomized, controlled trials are needed to determine the optimal policy for preoperative hair removal.