Alberto Fogagnolo1, Fabio S Taccone2, Giulia Benetto3, Federico Franchi4, Sabino Scolletta4, Antonella Cotoia5, Iryna Kozhevnikova3, Carlo A Volta3, Savino Spadaro3. 1. Unit of Anesthesia and Intensive Care, Department of Morphology, Surgery and Experimental Medicine, Sant'Anna University Hospital, University of Ferrara, Ferrara, Italy - alberto.fogagnolo@gmail.com. 2. Department of Intensive Care, Erasme Hospital, Université Libre de Bruxelles, Brussels, Belgium. 3. Unit of Anesthesia and Intensive Care, Department of Morphology, Surgery and Experimental Medicine, Sant'Anna University Hospital, University of Ferrara, Ferrara, Italy. 4. Department of Anesthesia and Intensive Care, University of Siena, Siena, Italy. 5. Department of Anesthesiology and Intensive Care, University of Foggia, Foggia, Italy.
Abstract
BACKGROUND: Thrombocytopenia is associated with worse outcomes in critically ill patients. The clinical relevance of other platelets indices is less studied. We investigated the ability of the platelets distribution width (PDW) and the mean platelet volume (MPV) to predict mortality in critically ill patients. We hypothesized that the prognostic values of PDW and MPV could be different in septic and non-septic patients. METHODS: We prospectively analyzed patients with an expected ICU length of stay ≥48 hours. Repeated measurements of PDW and MPV were considered (on ICU admission and up to day 5 thereafter). The primary outcome was to investigate the ability of PDW and MPV to predict 90-day mortality in septic and non-septic patients. RESULTS: We included in the study 234 patients of which 31% patients were septic. 90-day mortality was 39% in septic and 27% non-septic patients. PDW and MPV values on admission were 12.5±2.5% and 10.7±1.1 fL, respectively. The AUROC of PDW values on admission to predict 90-day mortality in septic patients was 0.813, being higher than those in non-septic patients (0.550, P<0.001). Similarly, the AUROC for MPV in septic patients was higher than non-septic patients (0.55, P<0.001). The combined analysis of platelets morphological indices and lactate improved the predictive accuracy (PDW and lactate AUROC=0.870; MPV and lactate AUROC=0.867). CONCLUSIONS: Platelet morphological indices are independent predictor of 90-day mortality in septic patients but not in non-septic patients. A combined analysis of platelets morphological indices and lactate in septic patients resulted in improved prediction of mortality.
BACKGROUND:Thrombocytopenia is associated with worse outcomes in critically illpatients. The clinical relevance of other platelets indices is less studied. We investigated the ability of the platelets distribution width (PDW) and the mean platelet volume (MPV) to predict mortality in critically illpatients. We hypothesized that the prognostic values of PDW and MPV could be different in septic and non-septic patients. METHODS: We prospectively analyzed patients with an expected ICU length of stay ≥48 hours. Repeated measurements of PDW and MPV were considered (on ICU admission and up to day 5 thereafter). The primary outcome was to investigate the ability of PDW and MPV to predict 90-day mortality in septic and non-septic patients. RESULTS: We included in the study 234 patients of which 31% patients were septic. 90-day mortality was 39% in septic and 27% non-septic patients. PDW and MPV values on admission were 12.5±2.5% and 10.7±1.1 fL, respectively. The AUROC of PDW values on admission to predict 90-day mortality in septic patients was 0.813, being higher than those in non-septic patients (0.550, P<0.001). Similarly, the AUROC for MPV in septic patients was higher than non-septic patients (0.55, P<0.001). The combined analysis of platelets morphological indices and lactate improved the predictive accuracy (PDW and lactate AUROC=0.870; MPV and lactate AUROC=0.867). CONCLUSIONS: Platelet morphological indices are independent predictor of 90-day mortality in septic patients but not in non-septic patients. A combined analysis of platelets morphological indices and lactate in septic patients resulted in improved prediction of mortality.