OBJECTIVE: To investigate clinically significant differences between vancomycin and cefuroxime for perioperative infection prophylaxis in coronary artery bypass surgery.

DESIGN: A total of 884 patients were randomized prospectively to receive either cefuroxime (444) or van-comycin (440) and were assessed for infectious complications during hospitalization and 1 month postoperatively.

SETTING: A university hospital.

RESULTS: The overall immediate surgical-site infection rate was 3.2% in the cefuroxime group and 3.5% in the vancomycin group (difference, −0.3; 95% confidence interval, −2.6-2.1).

CONCLUSIONS: The data suggest that vancomycin has no clinically significant advantages over cephalosporin in terms of antimicrobial prophylaxis. We suggest that cefuroxime (or first-generation cephalosporins, which were not studied here) is a good choice for infection prophylaxis in connection with coronary artery bypass surgery in institutions without methicillin-resistant Staphylococcus aureus problems. In addition to the increasing vancomycin-resistant enterococci problem, the easier administration and usually lower price of cefuroxime make it preferable to vancomycin.

OBJECTIVE: To identify preoperative and perioperative features that may lead to a risk of surgical-site infection (SSI) after coronary artery bypass surgery.

DESIGN: 884 patients who underwent coronary artery bypass grafting in 1992 and 1993 were studied. The associations between 23 preoperative and perioperative features and the presence of SSI at the donor site or in the chest area were evaluated by univariate analysis followed by multivariate logistic regression analysis.

SETTING: A university hospital.

RESULTS: 172 patients (19.5%) either had an SSI recorded before discharge or had received antibiotics prescribed for a suspected SSI during the 1-month surveillance period after discharge. Multivariate logistic regression analysis showed an extreme body mass index (BMI; P=.015), female gender (P=.023), and chronic obstructive pulmonary disease (COPD; P=.030) to be independent risk features for SSI. The donor site was infected in 136 patients (15.4%), an event for which female gender (P=.003) was the only independent risk feature. Forty-seven patients (5.3%) had an SSI of the chest area, with diabetes (P=.003) and extreme BMI (P=.010) as independent risk features.

CONCLUSION: Extreme BMI, female gender, and COPD are highly significant independent predictors of the development of SSI. Female gender is a risk feature specifically for SSI at the donor site, whereas diabetes and extreme BMI predict it in the chest area.

OBJECTIVE: Evaluate vancomycin prescribing patterns in a tertiary-care hospital before and after interventions to decrease vancomycin utilization.

DESIGN: Before/after analysis of interventions to limit vancomycin use.

SETTING: 420-bed academic tertiary-care center.

INTERVENTIONS: Educational efforts began August 10, 1994, and involved lectures to medical house staff followed by mailings to all physicians and posting of guidelines for vancomycin use on hospital information systems. Active interventions began November 15, 1994, and included automatic stop orders for vancomycin at 72 hours, alerts attached to the medical record, and, for 2 weeks only, computer alerts to physicians following each vancomycin order. Parenteral vancomycin use was estimated from the hospital pharmacy database of all medication orders. Records of a random sample of 344 patients receiving van-comycin between May 1, 1994, and April 30, 1995, were reviewed for an indication meeting published guidelines.

RESULTS: Vancomycin prescribing decreased by 22% following interventions, from 8.5 to 6.8 courses per 100 discharges (P<.05). The estimated proportion of van-comycin ordered for an indication meeting published guidelines was 36.6% overall, with no significant change following interventions. However, during the 2 weeks that computer alerts were in place, 60% of vancomycin use was for an approved indication.

CONCLUSIONS: Parenteral vancomycin prescribing decreased significantly following interventions, but the majority of orders still were not for an indication meeting published guidelines. Further improvement in the appropriateness of vancomycin prescribing potentially could be accomplished by more aggressive interventions, such as computer alerts, or by targeting specific aspects of prescribingpatterns.

OBJECTIVE: To determine the total and attributable costs of surgical-wound infections in a Canadian teaching hospital.

DESIGN: Retrospective incidence series study with chart review and examination of resource utilization attributable to wound infection. The charts of inpatients with wound infections were examined using the Appropriateness Evaluation Protocol (AEP), a validated chart review instrument designed to determine appropriateness of care, modified for wound infections.

SETTING: A university referral center in Canada.

PATIENTS: Medical records were abstracted from patients with wound infections who underwent an inpatient clean or clean-contaminated procedure during 1991.

MEASUREMENTS: During the wound-infection treatment period, the hospital costs associated with providing care were tabulated for all inpatient days and for outpatient and emergency visits. Costs taken into account included nursing salary and benefits, nonphysician professional services, operating room time, laboratory, pharmacy, supplies, ancillary tests, and hotel costs.

RESULTS: We identified 108 wound infections. Twenty-two patients required 28 surgical procedures related to a wound infection. Inpatient days totalled 1,116, costing $394,337. Fifty-five emergency and 42 clinic visits occurred, costing $27,193. By applying the AEP to the inpatient days, 833 days, or 10.2 days per case, were directly attributable to the wound infection. The hospital costs for inpatient care attributable to wound infections were $321,533 in total, or $3,937 per infection. Costs were distributed as follows: nursing, 51%; hotel, 14%; pharmacy, 10%; laboratory, 9%; emergency and outpatient clinic, 6%; professional services, 5%; operating room, 3%; and ancillary tests, 2%.

CONCLUSIONS: Wound infections contribute markedly to extra days of hospitalization and related costs. The AEP method is applied easily to determine attributable days of care and costs of wound infections, which are necessary to calculate the cost-benefit of infection control programs.

Among surgical intensive-care units (ICUs), we assessed differences in risk-adjusted nosocomial infection rates between cardiothoracic (CT) and general surgery ICUs, using National Nosocomial Infection Surveillance data from 1987 to 1995. Device-associated rates and average length of stay were significantly lower in CT ICUs. Comparisons of risk-adjusted nosocomial infection rates among CT ICUs should be made separately from rates from general surgery ICUs.

Sixteen percent of hospital room surfaces remained colonized by vancomycin-resistant enterococci (VRE) after routine terminal disinfection. Disinfection with a new "bucket method" resulted in uniformly negative cultures. Conventional cleaning took an average of 2.8 disinfections to eradicate VRE from a hospital room, while only one cleaning was required with the bucket method.

This is the second in a two-part series discussing and illustrating the application of statistical process control (SPC) in hospital epidemiology. The basic philosophical and theoretical foundations of statistical quality control and their relation to epidemiology are emphasized in order to expand the mutual understanding and cross-fertilization between these two disciplines. Part I provided an overview of the philosophy and general approach of SPC, illustrated common types of control charts, and provided references for further information or statistical formulae. Part II now discusses alternate possible SPC approaches, statistical properties of control charts, chart-design issues and optimal control limit widths, some common misunderstandings, and more advanced issues. The focus of both articles is mostly nonmathematical, emphasizing important concepts and practical examples rather than academic theory and exhaustive calculations.