Microscopic giant platelets versus automated mean platelet volume for severity assessment in COVID-19

Vollständiger Abstract

Worum geht es in dieser Arbeit?

COVID-19-associated coagulopathy (immunothrombosis) contributes to mortality. Automated hematology analyzers are widely used for risk stratification but may not capture morphological platelet abnormalities. We examined whether peripheral blood smear (PBS) findings, particularly giant platelets, are associated with severe COVID-19 beyond automated mean platelet volume (MPV) in hospitalized patients. In this retrospective cohort study, 96 patients with confirmed COVID-19 were classified at hospital admission as mild/moderate (n = 60) or severe (n = 36). Admission PBS samples were independently reviewed by 2 blinded observers. Multivariable logistic regression was used to assess factors associated with severe disease, and ROC analysis compared the discriminatory performance of microscopic giant platelet detection with automated MPV. Giant platelet presence was associated with severe disease after adjustment for age, sex, and diabetes mellitus (OR: 2.86, 95% CI: 1.02–8.07, P = .047); older age and diabetes mellitus were also associated with severity, whereas female sex was inversely associated. The corresponding unadjusted association did not reach conventional statistical significance, and the adjusted estimate had a wide confidence interval. In an exploratory model additionally including MPV, giant platelet presence remained associated with severe disease (OR: 3.88, 95% CI: 1.25–12.01, P = .019), whereas MPV was not associated with severity in the same model ( P = .114). Giant platelet detection showed weak stand-alone discrimination (AUC: 0.597, 95% CI: 0.497–0.694), and its discriminatory performance did not differ significantly from that of MPV (AUC: 0.465, 95% CI: 0.349–0.584; DeLong P = .058). In this cohort, giant platelet presence was associated with severe COVID-19 at admission after adjustment for selected baseline covariates; however, the imprecision of the estimate, nonsignificant unadjusted comparison, and weak discriminatory performance indicate that this finding should be considered exploratory. These results do not establish that PBS review improves clinical severity assessment, and external validation in larger prospective cohorts is required.