|
Procalcitonin-guided algorithm to reduce length of antibiotic therapy in patients with severe sepsis and septic shock |
Hohn A, Schroeder S, Gehrt A, Bernhardt K, Bein B, Wegscheider K, Hochreiter M |
|
|
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. CRD summary This study evaluated the impact of an algorithm, based on procalcitonin, to reduce the length of antibiotic therapy, for adults with severe sepsis and septic shock, in a surgical intensive care unit. The authors concluded that the algorithm decreased antibiotic use, without compromising clinical or economic outcomes, but further research was necessary. The study was well reported, but there was no appropriate comparator, few patients, and the cost analysis was very limited. The validity of the results is unclear. Type of economic evaluation Cost-effectiveness analysis Study objective This study evaluated the impact of an algorithm, based on procalcitonin, to reduce the length of antibiotic therapy, for adults with severe sepsis and septic shock, in a surgical intensive care unit (ICU). Interventions The intervention consisted of monitoring procalcitonin in daily blood samples. Antibiotics were discontinued if the clinical signs of infection improved, and the procalcitonin level decreased to one nanogram per mL or less, or if it was over one, but had dropped to 25 to 35% of its initial level, within three days. The changes in outcomes in the five years after the introduction of the algorithm were compared. These changes over time were assumed to result from a progressive improvement in the implementation of the algorithm. Methods Analytical approach:The analysis was based on a retrospective review of records from an ICU database, from 2005 to 2009. There were 141 records with complete data; the number of patients varied from 38 in 2005, to 18 in 2009. The study perspective was not explicitly stated. Effectiveness data:The primary clinical outcomes were antibiotic days in the ICU, the ICU reinfection rate, mortality within 28 days, the length of stay in the ICU, and ventilation hours. All these outcomes were from the ICU database records. The differences over time were tested for statistical significance, using analysis of covariance (ANCOVA) and logistic regression, to adjust for differences in patient characteristics. These characteristics were measured for each year of the review, and included gender, age, Simplified Acute Physiology Score (SAPS) II, Acute Physiology and Chronic Health Evaluation (APACHE) II score, and German diagnosis-related group cost weight. Monetary benefit and utility valuations:Not relevant. Measure of benefit:The primary measures of benefit were the clinical outcomes; antibiotic days in the ICU, the ICU reinfection rate, 28-day mortality, length of stay in the ICU, and ventilation hours. Cost data:The resource use was calculated for ICU days, and hours of ventilation. The costs only included antibiotic use, which was estimated using the hospital's wholesale prices, without administration costs. All costs were reported in Euros (EUR). Analysis of uncertainty:The results were reported as means with 95% confidence intervals or standard deviations. The differences over time were assessed for statistical significance. Results The mean annual changes were: a decrease of 2.7 ICU days (95% CI -4.1 to -1.3); a decrease of 42.0 ventilation hours (95% CI -72.6 to -11.4); a decrease of 1.0 antibiotic day (95% CI -1.9 to -0.2); a decrease of 35.1% in ICU reinfection (95% CI -53.9 to -8.5); and a decrease of 22.4% in 28-day mortality (95% CI -44.3 to -8.1). The annual costs of antibiotic use decreased on average by EUR 14.3 per patient per year (95% CI -55.7 to 27.1). Of the controlled covariates, the German diagnosis-related group cost weight affected all outcomes (p<0.001), and age had a significant impact on 28-day mortality (p<0.001). The APACHE II score, SAPS II, and gender did not significantly affect the primary outcomes. Authors' conclusions The authors concluded that the procalcitonin algorithm was associated with decreased antibiotic use, without compromising clinical or economic outcomes, but further research was necessary. CRD commentary Interventions:The intervention was clearly stated and sufficiently described. The study compared the outcomes in the years after the algorithm was introduced. There was no direct comparator; instead the changes were assumed to represent the progressive improvement in the implementation of the algorithm. The authors stated that there were numerous procalcitonin algorithms available, and further research was needed to determine if specific algorithms were required for specific patient groups, rather than a general algorithm, and if there was a difference in efficacy between algorithms. Effectiveness/benefits:The effectiveness outcomes were clearly reported and appropriate analyses were undertaken to balance for differences in patient characteristics over the years. As there was no control group, it was unclear if the differences in outcomes were a result of the intervention or unknown other factors. Adherence was not monitored, so it was unclear how well the intervention was implemented. The study sample was small. The authors indicated that an important outcome that was not measured was the incidence of antibiotic-resistant bacteria. They theorised that with the procalcitonin algorithm, antibiotic resistance might be reduced, but further research was necessary to evaluate this hypothesis. Costs:The costs were presented from the perspective of the hospital, but they only included the costs of the antibiotics. This selection was too narrow for firm economic conclusions. The key missing costs were the implementation of the algorithm, and primary and secondary care. The authors acknowledged this limitation, and recommended further research to reach conclusions on the cost-effectiveness of procalcitonin algorithms. Analysis and results:The analyses were clearly presented, with sufficient detail, and appropriate analytic methods were used. The authors reported a thorough discussion of studies comparable with their work, which added to the evidence on procalcitonin algorithms for antibiotic use. They appropriately discussed their study limitations – in addition to those already mentioned, they indicated that the characteristics of their hospital and hospital database could have led to a reduced number of sepsis patients being identified, and that a larger number of patients would increase the study validity. Concluding remarks:The study was well reported, but there were several limitations. There was no appropriate comparator, the sample size was small, some relevant outcomes were not measured, and the cost analysis was very limited. The validity of the results is unclear. Funding Funding received from BRAHMS, manufacturer of a procalcitonin test. Bibliographic details Hohn A, Schroeder S, Gehrt A, Bernhardt K, Bein B, Wegscheider K, Hochreiter M. Procalcitonin-guided algorithm to reduce length of antibiotic therapy in patients with severe sepsis and septic shock. BMC Infectious Diseases 2013; 13: 158 Indexing Status Subject indexing assigned by NLM MeSH Adult; Aged; Aged, 80 and over; Algorithms; Anti-Bacterial Agents /therapeutic use; Biomarkers /blood; Calcitonin /blood; Drug Monitoring /methods; Drug Utilization /statistics & Female; Germany; Humans; Intensive Care Units; Male; Middle Aged; Protein Precursors /blood; Retrospective Studies; Sepsis /diagnosis /drug therapy; Treatment Outcome; numerical data AccessionNumber 22013017436 Date bibliographic record published 14/06/2013 Date abstract record published 14/11/2013 |
|
|
|