Kinetics of mean platelet volume predicts mortality in patients with septic shock
Datasets usually provide raw data for analysis. This raw data often comes in spreadsheet form, but can be any collection of data, on which analysis can be performed.
Thrombocytopenia is well recognized as a poor prognosis sign associated with increased mortality and prolonged Intensive Care Unit (ICU) stay, particularly in septic patients. Mean platelet volume (MPV) could represent a relevant predictive marker of mortality. Here we investigated whether MPV kinetics during the first 15 days after hospital admission has a potential prognostic value for clinical outcome in septic shock.
We performed a retrospectively analysis of a cohort of 301 septic patients admitted in ICU. Three-month mortality was the primary endpoint. The prognostic value of the covariates of interest was ascertained by multidimensional analysis. We proposed a classification and regression trees analysis to predict survival probability.
MPV kinetics was significantly different between 90-day survivors and non-survivors when followed during 15 days (except on day 3). 10-day MPV >11.6fL was an independent predictive factor of 90-day mortality (Hazard Ratio (HR) 3.796, 95% Confidence Interval (CI) [1.96–7.35], p = 0.0001) in multivariate analysis. Base excess on day 4 <1.9mmol/L was also a predictive factor of mortality (HR 2.972, 95%CI [1.38–6.40], p = 0.0054.
MPV increase during the first 15 days after ICU admission in non-survivors was observed during septic shock and 10-day MPV >11.6fL was an independent predictive factor of 90-day mortality. This could be explained by the emergent response to acute platelet loss during septic shock, leading to megakaryocyte rupture to produce new but potentially immature platelets in the circulation. Therefore, continuous monitoring of MPV may be a useful parameter to stratify mortality risk in septic shock.