RESUMEN
PURPOSE: Results of a study incorporating real-world results into a predictive model to assess the cost-effectiveness of procalcitonin (PCT)-guided antibiotic use in intensive care unit patients with sepsis are reported. METHODS: A single-center, retrospective cross-sectional study was conducted to determine whether reductions in antibiotic therapy duration and other care improvements resulting from PCT testing and use of an associated treatment pathway offset the costs of PCT testing. Selected base-case cost outcomes in adults with sepsis admitted to a medical intensive care unit (MICU) were assessed in preintervention and postintervention cohorts using a decision analytic model. Cost-minimization and cost-utility analyses were performed from the hospital perspective with a 1-year time horizon. Secondary and univariate sensitivity analyses tested a variety of clinically relevant scenarios and the robustness of the model. RESULTS: Base-case modeling predicted that use of a PCT-guided treatment algorithm would results in hospital cost savings of $45 per patient and result in a gain of 0.0001 quality-adjusted life-year. After exclusion of patients in the postintervention cohort for PCT test ordering outside of institutional guidelines, the mean inpatient antibiotic therapy duration was significantly reduced in the postintervention group relative to the preintervention group (6.2 days versus 4.9 days, p = 0.04) after adjustment for patient sex and age, Charlson Comorbidity Index score, study period, vasopressor use, and ventilator use. Total annual hospital cost savings of $4,840 were predicted. CONCLUSION: Real-world implementation of PCT-guided antibiotic use may have improved patients' quality of life while decreasing hospital costs in MICU patients with undifferentiated sepsis.