Association between platelet parameters and mortality in coronavirus disease 2019: Retrospective cohort study

Abstract

Background: Thrombocytopenia has been implicated in patients infected with severe acute respiratory syndrome coronavirus 2, while the association of platelet count and changes with subsequent mortality remains unclear.

Methods: The clinical and laboratory data of 383 patients with the definite outcome by March 1, 2020 in the Central Hospital of Wuhan were reviewed. The association between platelet parameters and mortality risk was estimated by utilizing Cox proportional hazard regression models.

Results: Among the 383 patients, 334 (87.2%) were discharged and survived, and 49 (12.8%) died. Thrombocytopenia at admission was associated with mortality of almost three times as high as that for those without thrombocytopenia (P < 0.05). Cox regression analyses revealed that platelet count was an independent risk factor associated with in-hospital mortality in a dose-dependent manner. An increment of per 50 × 10⁹/L in platelets was associated with a 40% decrease in mortality (hazard ratio: 0.60, 95%CI: 0.43, 0.84). Dynamic changes of platelets were also closely related to death during hospitalization.

Conclusions: Baseline platelet levels and changes were associated with subsequent mortality. Monitoring platelets during hospitalization may be important in the prognosis of patients with coronavirus disease in 2019.

Keywords: Cohort study; coronavirus disease 2019; mortality; platelet count; severe acute respiratory syndrome coronavirus 2.

Figure 1.

Timeline charts illustrate the platelet levels in 383 patients with COVID-19 (49 non-survivors and 334 survivors) on day 1, day 3, day 7, and day 14 after admission Data are represented as median and 95% confidence interval. The dash lines in black show the upper normal limit of platelets, and the dash line in red shows the lower normal limit of platelets. Generalized linear-mixed models examined the differences in the platelets between non-survivor and survivor groups over time.* P < .05 for non-survivor vs. survivor.

Figure 2.

The smoothing spline presenting the association between baseline platelet count and APACHE II score was generated by utilizing the generalized additive model.

Figure 3.

The nonlinear relationship of (a) platelets at admission and (b) platelet changes with mortality. The smoothing splines were generated utilizing generalized additive model and adjusted for age, sex, baseline comorbidities (including chronic obstructive pulmonary disease, hypertension, diabetes, chronic kidney disease, cardiovascular disease, and cerebrovascular disease), glucocorticoid therapy, intravenous immunoglobulin, blood lactate, C-reactive protein, and lymphocyte count. The red line indicates the risk of mortality and the blue dot line indicates 95% confidence intervals.

Figure 4.

Kaplan–Meier curves stratified by platelet count at admission and platelet changes in 7 days.