Analysis of platelet-activating factors in severe sepsis by flow cytometry and its correlation with clinical sepsis scoring system: A pilot study.

Background: Sepsis is a major global healthcare concern. Platelets and leucocytes play a key role in sepsis. Whole blood flow cytometry (FCM) is a powerful new technique for the assessment of leucocyte and platelet parameters and their functional state. In the present study, we have used FCM to examine platelet and leucocyte functions and parameters in sepsis patients.

Methods: Prospective, non-interventional cohorts of all adult patients with sepsis and history of intensive care unit stay for more than 24 h at mixed surgical- medical ICU were evaluated. The Simplified Acute Physiology Score-3 (SAPS 3) admission score was obtained, and whole blood FCM analysis of select platelets and leukocyte parameters were performed using a combination of monoclonal antibodies in a predefined panel. We also evaluated the correlation between these parameters and the severity of the illness, based on SAPS 3 admission score.

Results: Total leucocyte count (TLC) was statistically and significantly different between all the study samples, but platelet count was not. SAPS 3 acted as the best discriminant between the study groups. With a cut-off score of 55.5, SAPS 3 score predicted hospital mortality with a sensitivity of 82.8% and a specificity of 83.9%, with an area under receiver operating curves (AUROC) of 0.888 (95% CI = 0.807-0.969, p < 0.000). Parameters for CD62P, platelet-leucocyte aggregates (PLAs) and CD11b showed statistically significant differences between the patients and healthy volunteers. CD62P expression was positively correlated to PLA variables in severe sepsis patients. The median fluorescence intensity was found to be more informative than mean fluorescence intensity. New "62P adhesion index (62P AI)" and "PLA adhesion index" are proposed and is likely to be more informative.

Conclusion: SAPS 3 score was the most robust of the parameters evaluated. Our study suggest the idea that the incorporation of platelet and leucocyte activation parameters, rather than mere static counts, will add the existing prognostic model though we could not conclusively prove the same in this study.