Administering cefonicid (a second-generation cephalosporin antibiotic with broad-spectrum activity against gram-positive cocci and gram-negative rods) in a single dose of 1g intravenously 60 minutes prior to the initiation of the surgical procedure for prevention of postoperative infectious wound complications following axillary lymph node dissection (ALND) in patients with either breast carcinoma or melanoma.

Primary prevention.

The aim of the study was to assess the cost-effectiveness of administering cefonicid versus placebo for prevention of postoperative infectious wound complications following ALND in patients with either breast carcinoma or melanoma. The strategy of using placebo (normal saline) was regarded as the comparator.

Cost-effectiveness analysis.

Patients with either breast carcinoma or melanoma who underwent ALND. The exclusion criteria were as follows: a history of allergy to cephalosporin, aspirin use within 5 days of surgery, age less than 18 years, recent (within 7 days of the procedure) administration of antibiotics either orally or parenterally, recent infection, current pregnancy or breast-feeding, creatinine level greater than 2.0 mg/dL, bilirubin level greater than 1.5 mg/dL, concomitant isolated limb perfusion, inflammatory breast cancer, immediate breast reconstruction, clinically significant wound complication from prior surgery within 4 weeks of planned operation, medical condition requiring antibiotic prophylaxis, diabetes mellitus, and steroid use. These exclusion criteria were chosen to minimise potentially confounding variables and to assure a population at low risk for postoperative infections.

Hospital. The economic study was carried out in the USA.

Not given.

The evidence for the final clinical outcomes was derived from a single study.

Costing appears to have been undertaken prospectively on the same patient sample as that used in the effectiveness analysis.

Power calculations were used to determine the sample size. To detect a significant difference in the postoperative infection rate from 11% (based on historical experience and on the literature) to 3% at a level of significance of 0.05 with a power of 80%, about 100 patients were required in each study group; retrospectively the power of the study was 0.644). The study sample consisted of 200 consecutive patients (about 15% of patients who underwent ALND during a 63-month time period) enrolled in the study by 16 surgeons. After excluding 22 patients with protocol violations, the placebo group had 90 patients with a mean age of 53.6 (range: 27-86) years versus 88 in the intervention (cefonicid) group with a mean age of 49.6 (range: 20-82) years.

This was a prospective, randomised, double-blind, placebo-controlled trial, carried out in a single centre. Loco-regional signs of infection were monitored for 4 weeks postoperatively. All patients with postoperative wound infections were followed until the infection resolved and then at routine intervals depending on the patient's disease process and requirement for further therapy. The loss to follow-up (excluded because of protocol violations) was 10 patients in the placebo group and 12 patients in the intervention group. A research nurse contacted the patient and referring physicians for wound follow-up for 4 weeks after surgery. Patients were stratified by use of preoperative chemotherapy, whether an open biopsy had been performed within 7 days prior to the procedure, and type of surgical procedure to assure balance of potential risk factors for infection. A computer-generated block randomisation table was used.

The principle used in the analysis of effectiveness was treatment completers only. The primary end point of the study was the development of a postoperative infectious wound complication. The rate of severe infections requiring hospitalisation and length of hospital stay were also reported. The patient groups were comparable in terms of demographic features (age and gender), prior chemotherapy, and type of operation performed.

No patient experienced a complication related to the antibiotic administration. A trend was observed toward a decrease in the incidence of postoperative wound infections (13% in the placebo group versus 6% in the intervention group, p=0.080). There was also a significant reduction in infections severe enough to require hospitalisation (8% in the placebo group versus 1% in the intervention group, p=0.033). It was reported that the characteristics of the patients having developed postoperative wound infections were similar between the study groups. Length of hospital stay was 5.9 days in the placebo group and 3 days in the intervention group (NS).

Although this study did not demonstrate a statistically significant reduction in the overall incidence of infections, a significant reduction in severe postoperative wound infections was observed following the administration of a single dose of cefonicid prior to axillary lymph node dissection.

No summary benefit measure was identified in the economic analysis, and only separate clinical outcomes were reported.

Costs were not discounted due to the short time frame of the cost analysis. Some quantities were reported separately from the costs. Cost items were not reported separately. The cost analysis covered the costs of antibiotic, outpatient treatment (for a 7-day course of oral antibiotic), and hospitalisation. The perspective adopted in the cost analysis was not explicitly specified. Charge data were used instead of true costs. The price year was not given.

Not included.

US dollars ($).

A statistical comparison was made between the study groups in terms of costs.

Not conducted.

Not applicable.

The intervention (cefonicid) was associated with a cost of $48.80 per patient, significantly less than the $364.87 per patient for the placebo group.

Costs and benefits were not combined since the intervention was the dominant strategy (with equal or better clinical outcomes, and significantly lower costs).

The authors demonstrated a trend toward fewer overall infections and significantly fewer severe infections in patients given prophylactic antibiotics, which translated into a decrease in the cost of treatment for infectious complications.

The authors chose placebo as the comparator for the intervention drug thus allowing the active value of the treatment to be evaluated.

The effectiveness results are likely to be internally valid in view of the randomised design of the study. However, the fact that the effectiveness analysis was based on treatment completers only rather than intention to treat may have introduced some bias into the study. The study groups were found to be comparable in terms of demographic and prognostic features. The patient sample was chosen to represent a low risk population. It was noted that a single dose of antibiotic is not usually associated with the development of resistant micro-organisms.

The authors did not derive a measure of health benefit. The study may therefore be regarded as a cost-consequences analysis.

Some quantities were reported separately from the costs and adequate details of methods of cost estimation were given. Using charge data as opposed to true costs may have adversely affected the internal and external validity of the cost results. The effects of different procedures on indirect costs (productivity loss) were not evaluated. Cost results may not be generalisable to other settings or countries. The price year was not given.

The authors' conclusions appear to be justified given the uncertainties in the data. The issue of generalisability to other settings or countries was not addressed, although appropriate comparisons were made with other studies. Whether the study sample was representative of the study population was addressed by pointing out that the study sample was meant to represent a low-risk population.

The study findings support antibiotic prophylaxis for patients undergoing ALND.

Supported in part by a grant from SmithKline Beecham Laboratories.

This record was compiled by CRD commissioned reviewers according to a set of guidelines developed in collaboration with a group of leading health economists.