Performance of the PIRO Score for Predicting Mortality in Patients with Ventilator-Associated Pneumonia
Guilherme Henrique Campos Furtado, Dora E. Wiskirchen, Joseph L. Kuti, David P. Nicolau
Abstract
Guilherme Henrique Campos Furtado, Dora E. Wiskirchen, Joseph L. Kuti, David P. Nicolau
Abstract
The ventilator-associated pneumonia (VAP) PIRO score is a new scoring system based on the PIRO concept. The aim of this study was to validate the PIRO score against the Acute Physiology and Chronic Health Evaluation (APACHE) II and VAP APACHE II in an independent group of VAP patients. Areas under the receiver operating characteristic curves were compared to determine the tests' abilities to predict intensive care unit and 28-day mortality. Variables associated with intensive care unit mortality were evaluated. One hundred and forty-eight intensive care unit patients who met radiographic and clinical criteria for VAP were included. The area under the receiver operating characteristic curves for predicting intensive care unit mortality with the PIRO, APACHE II and VAP APACHE II scores were 0.605 (P=0.03), 0.631 (P=0.01) and 0.724 (P <0.0001), respectively. Areas under the receiver operating characteristic curve for predicting 28-day mortality were 0.614 (P=0.01) for PIRO, 0.633 (P=0.01) for APACHE II and 0.697 (P=0.002) for VAP APACHE II. No differences in area under the receiver operating characteristic curve between scores were found at either endpoint. Variables independently associated with intensive care unit mortality were bacteraemia (adjusted odds ratio 7.16, 95% confidence interval 1.19 to 42.98, P=0.03) and APACHE II (1.06, 1.01 to 1.11, P=0.006). VAP PIRO score was not a good predictor of intensive care unit and 28-day mortality. The low sensitivity and specificity of VAP PIRO score preclude its use clinically.
OpenAlex reports 13 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The ventilator-associated pneumonia (VAP) PIRO score is a new scoring system based on the PIRO concept. The aim of this study was to validate the PIRO score against the Acute Physiology and Chronic Health Evaluation (APACHE) II and VAP APACHE II in an independent group of VAP patients. Areas under the receiver operating characteristic curves were compared to determine the tests' abilities to predict intensive care unit and 28-day mortality. Variables associated with intensive care unit mortality were evaluated. One hundred and forty-eight intensive care unit patients who met radiographic and clinical criteria for VAP were included. The area under the receiver operating characteristic curves for predicting intensive care unit mortality with the PIRO, APACHE II and VAP APACHE II scores were 0.605 (P=0.03), 0.631 (P=0.01) and 0.724 (P <0.0001), respectively. Areas under the receiver operating characteristic curve for predicting 28-day mortality were 0.614 (P=0.01) for PIRO, 0.633 (P=0.01) for APACHE II and 0.697 (P=0.002) for VAP APACHE II. No differences in area under the receiver operating characteristic curve between scores were found at either endpoint. Variables independently associated with intensive care unit mortality were bacteraemia (adjusted odds ratio 7.16, 95% confidence interval 1.19 to 42.98, P=0.03) and APACHE II (1.06, 1.01 to 1.11, P=0.006). VAP PIRO score was not a good predictor of intensive care unit and 28-day mortality. The low sensitivity and specificity of VAP PIRO score preclude its use clinically.
Key concepts: Medicine, Receiver operating characteristic, Intensive care unit, APACHE II, Area under the curve, Ventilator-associated pneumonia, Confidence interval, Odds ratio