|
Treatment of nosocomial postoperative pneumonia in cancer patients: a prospective randomized study |
Raad I, Hachem R, Hanna H, Abi-Said D, Bivins C, Walsh G, Thornby J, Whimbey E, Huaringa A, Sukumaran A |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology The health interventions examined in the study were piperacillin/tazobactam (P/T) 4.5 g IV every 6 hours versus clindamicyn 900 mg plus aztreonam 2 g IV every 8 hours (CI/Az) for the treatment of nosocomial pneumonia in cancer patients. All patients were also treated with amikacin 500 mg IV every 12 hours.
Economic study type Cost-effectiveness analysis.
Study population The study population consisted of adult patients (minimum age: 18 years), who were non-neutropenic (greater than or equal to 1,000 neutrophils/mm^3) with recent onset of fever (greater than 38 degrees C) and with the presence of new pulmonary infiltrates, determined radiologically, that occurred postoperatively and more than 72 hours after admission, and that developed in the absence of fever or pulmonary infiltrates at the time of admission. Patients were excluded if allergic to penicillin or in had any of the following conditions: prior treatment with broad-spectrum antibiotics for more than 48 hours or response to antimicrobial therapy administered for nosocomial pneumonia; prior allergic reaction to any of the study drugs; pregnant or nursing an infant; or bacterial infection known to be resistant to the study antibiotics. Further exclusion criteria were reported in the paper.
Setting The setting was hospital. The economic study was carried out at the University of Texas, M.D. Anderson Cancer Centre in Houston, Texas, USA.
Dates to which data relate Data on effectiveness were gathered from April 1994 to April 1996. No price year was reported.
Source of effectiveness data The effectiveness evidence was derived from a single study.
Link between effectiveness and cost data It was not stated explicitly, but it appears that the costing was conducted prospectively.
Study sample No power calculations were performed and it was not stated whether the initial sample of patients was appropriate for the study question. The method of sample selection was not reported. A sample of 52 patients was enrolled in the study: 22 patients (mean age: 63 +/- 13 years; 77% smokers) were included in the CI/Az group and 30 patients (mean age: 60 +/- 13 years; 80% smokers) in the P/T group. There were also sub-samples for the microbial etiology of pneumonia that occurred in each group. For example, there were 12 episodes of Gram-negative bacilli for the CI/Az group and only 9 for the P/T group.
Study design This was a prospective, open, randomised, controlled trial, which was carried out in a single centre. The randomisation scheme was designed as 1.5 to 1.0 ratio for P/T versus CI/Az to balance the expected difference related to penicillin allergy. The actual method of randomisation was not reported. Patients were assessed after 48 hours of entry onto the study. For patients with nosocomial pneumonia caused by any organism other than P. aeruginosa, the amikacin was discontinued 48-72 hours after initiation. For pneumonias caused by P. aeruginosa, amikacin was continued for as long as the other study drugs were given. The length of follow-up was two weeks after discontinuation of antibiotic therapy. No loss to follow-up was reported.
Analysis of effectiveness The authors did not state whether they used an intention to treat or per-protocol analysis. All patients were accounted for in the study but there were two cross-overs and it was not stated how this was dealt with in the analysis. Study groups were shown to have been comparable at baseline in terms of age, sex, underlying malignancy, duration of fever, duration of intubation, and hospitalisation before diagnosis of pneumonia. The primary health outcomes used in the effectiveness study were the clinical response rate (defined as the complete resolution of all signs and symptoms of pneumonia), death rate during the study period, death rate during hospitalisation, and overall death rate from infection. The duration of treatment, adverse infections, occurrence of superinfections, the number of cases of nosocomial postoperative pneumonia, the cases of mixed respiratory flora and the response rate to therapy based on microbial etiology of pneumonia were reported.
Effectiveness results The effectiveness results were as follows:
There were 29 cases of nosocomial postoperative pneumonia with an identified bacterial organism.
The clinical response rate was 86% in the CI/Az group and 83% in the P/T group.
The death rate during the study period was 0 in both groups;
the death rate during hospitalisation was 4% in the CI/Az group and 13% in the P/T group;
the overall death rate from infection was 0 in the CI/Az group and 13% in the P/T group.
The duration of treatment was 8.1 +/- 3.0 days in the CI/Az group and 6.8 +/- 3.0 days in the P/T group.
The occurrence of adverse infections and superinfections was comparable in both study groups.
None of the differences in the outcomes reported above reached statistical significance.
There were 12 episodes of Gram-negative bacilli and 2 of Gram-positive cocci in the CI/Az group; and
9 episodes of Gram-negative bacilli and 6 of Gram-positive cocci in the P/T group.
Mixed respiratory flora was observed in 8 cases (100% response rate) in the CI/Az group and in 15 cases (87% response rate) in the P/T group.
Clinical conclusions The effectiveness study showed that the two treatments were similarly effective and safe.
Measure of benefits used in the economic analysis No summary benefit measure was used in the economic analysis and no statistically significant differences were found in any of the clinical outcomes used in the effectiveness study. Thus the study was categorised as a cost-minimisation analysis.
Direct costs Costs were incurred over a short period, thus discounting was not relevant. Unit costs were not reported separately from quantities of resources. Only the daily cost of treatment (drug expenses and drug administration) was included in the economic evaluation. The cost/resource boundary adopted in the study was not stated, but it appears to have been that of the authors' institution. The source of cost data was the authors' institution. The method used for collection of resource use data was not explicitly stated, however, it was likely to have been conducted prospectively over the period April 1994-April 1996 and on the same patient sample as that used in the effectiveness analysis. No price year was reported.
Statistical analysis of costs Costs were treated deterministically.
Indirect Costs Indirect costs were not included.
Sensitivity analysis No sensitivity analyses were conducted.
Estimated benefits used in the economic analysis Please refer to the effectiveness results reported above.
Cost results The daily cost of P/T at a dose of 4.5 g IV once every 6 hours was $73.86.
The daily cost of CI/Az at a dose of 2 g IV once every 8 hours was $99.15.
The authors also reported that the cost of Az alone given at a dose of 2 g IV every 8 hours cost $82.05.
Synthesis of costs and benefits Costs and benefits were not combined as a cost-minimisation analysis was performed.
Authors' conclusions The authors concluded that both P/T and CI/Az administered in combination with an aminoglycoside were highly effective and safe treatments for patients with nosocomial pneumonia. The therapy based on P/T had a slight economic advantage in terms of lower daily drug costs.
CRD COMMENTARY - Selection of comparators The authors stated that CI/Az and P/T were selected as comparators as both represented widely used therapies for the treatment of nosocomial pneumonia and no comparison between CI/Az and P/T had been undertaken in previous studies. There appear to be several alternative comparators such as Aztreonam alone or ceftazidine, which may treat a narrower spectrum of microbial etiology of pneumonia. You, as a user of this database, should decide whether they represent valid comparators in your own setting.
Validity of estimate of measure of effectiveness The analysis of effectiveness was based on a randomised controlled trial, which was appropriate for the study question. Study groups were comparable at baseline, the study sample appears to have been representative of the study population and follow-up was complete. However, the analysis presented several drawbacks. First, the major threat to the internal validity of the study was the lack of power calculations and the very small sample size, which did not allow the study to be powered enough to detect statistically significant differences between the two groups in terms of clinical outcomes. The authors commented that an impractically large number of patients would need to be studied. Second, the methods of sample selection and randomisation were not reported. Third, no blind assessment of the outcome was performed, thus assessment bias may have played a role. Further, the authors did not state how they dealt with the 2 cross-overs in the analysis.
Validity of estimate of measure of benefit No summary benefit measure was used in the economic analysis and no statistically significant differences were found in any of the clinical outcomes used in the effectiveness study. The analysis was therefore categorised as a cost-minimisation study (see validity of effectiveness comments above).
Validity of estimate of costs The perspective adopted in the study was not explicitly stated, although it appears to have been that of the hospital where the study was conducted. Only daily drug costs were included in the analysis and few details of the economic evaluation were reported. In order to complete a full economic evaluation of the two interventions all related costs should be compared, such as hospitalisation costs. It appears that the economic analysis represented a secondary objective of the present study as details on the cost analysis were reported incidentally in the 'discussion' session of the paper. Unit costs and quantities of resources used were not reported separately and no price year was given, thus limiting the possibility of replicating the study in other settings. It was not clear whether resource consumption was estimated alongside the clinical trial. Costs were treated deterministically and no sensitivity analyses were performed.
Other issues The findings of this study were consistent with those from other published studies quoted by the authors. They did not address the issue of the generalisability of the study results to other settings. No sensitivity analyses were performed, thus further limiting the external validity of the analysis. The study enrolled high-risk patients and this was reflected in the conclusions of the analysis.
Implications of the study The study suggested that the two treatments for nosocomial pneumonia presented similar efficacy and safety profiles, although daily drug costs were lower with the P/T therapy than with the CI/Az therapy. The authors highlighted that a systematic approach which started with a combination therapy including an aminoglycoside for the first 48 hours, based on the re-evaluation every 48 hours, and then resulting in a switch to oral antibiotics for patients responding to the therapy, would further improve the outcomes in high-risk cancer patients.
Source of funding Supported by a grant from Bristol-Myers Squibb and Wyeth-Ayerst.
Bibliographic details Raad I, Hachem R, Hanna H, Abi-Said D, Bivins C, Walsh G, Thornby J, Whimbey E, Huaringa A, Sukumaran A. Treatment of nosocomial postoperative pneumonia in cancer patients: a prospective randomized study. Annals of Surgical Oncology 2001; 8(2): 179-186 Indexing Status Subject indexing assigned by NLM MeSH Aged; Amikacin /administration & Aztreonam /administration & Chi-Square Distribution; Clindamycin /administration & Cross Infection /drug therapy; Drug Therapy, Combination /therapeutic use; Humans; Middle Aged; Neoplasms /complications; Opportunistic Infections /drug therapy; Penicillanic Acid /administration & Piperacillin /administration & Pneumonia, Bacterial /drug therapy; Postoperative Complications /drug therapy; Prognosis; Prospective Studies; derivatives; dosage; dosage; dosage; dosage; dosage /analogs & AccessionNumber 22001000627 Date bibliographic record published 31/07/2003 Date abstract record published 31/07/2003 |
|
|
|