Use of Radiology, D-Dimer, and Mean Platelet Volume Combination as a Prognostic Marker in Hospitalized Coronavirus Disease-19 Patients

Introduction: The search for biomarkers that could help in predicting disease prognosis in the Coronavirus Disease-2019 (COVID-19) outbreak is still high on the agenda.

Objective: To find out the efficacy of D-dimer and mean platelet volume (MPV) combination as a prognostic marker in hospitalized COVID-19 patients with bilateral infiltration.

Materials and methods: Study design: Retrospective observational cohort. Patients who were presented to our hospital between March 16, 2020 and June 07, 2020 were reviewed retrospectively. The primary outcome of the study was specified as the need for intensive care, while the secondary outcomes were duration of treatment and hospitalization. Receiver operator curve (ROC) analyzes were carried out to assess the efficacy of D-dimer and MPV parameters as prognostic markers.

Results: Between the mentioned dates, 575 of 1,564 patients were found to be compatible with COVID-19, and the number of patients who were included in the study was 306. The number of patients who developed the need for intensive care was 40 (13.1%). For serum D-dimer levels in assessing the need for intensive care, the area under the curve (AUC) was found to be 0.707 (95% CI: 0.620-0.794). The AUC for MPV was 0.694 (95% CI: 0.585-0.803), when D-dimer was ≥1.0 mg/L. When patients with a D-dimer level of ≥1.0 mg/L were divided into two groups considering the MPV cut-off value as 8.1, the rate of intensive care transport was found to be significantly higher in patients with an MPV of ≥8.1 fL compared to those with an MPV of <8.1 fL (32.6 vs. 16.0%, p = 0.043). For the prognostic efficacy of the combination of D-dimer ≥ 1.0 mg/L and MPV ≥ 8.1 fL in determining the need for intensive care, following values were determined: sensitivity: 57.7%, specificity: 70.8%, positive predictive value (PPV): 32.0%, negative predictive value (NPV): 84.0%, and accuracy: 63.0%. When D-dimer was ≥1.0, the median duration of treatment in MPV <8.1 and ≥8.1 groups was 5.0 [interquartile range (IQR): 5.0-10.0] days for both groups (p = 0.64). The median length of hospital stay (LOS) was 7.0 (IQR: 5.0-10.5) days in the MPV <8.1 group, while it was 8.5 (IQR: 5.0-16.3) days in the MPV ≥ 8.1 group (p = 0.17).

Conclusion: In COVID-19 patients with a serum D-dimer level of at least 1.0 mg/L and radiological bilateral infiltration at hospitalization, if the MPV value is ≥8.1, we could predict the need for intensive care with moderate efficacy and a relatively high negative predictive value. However, no correlation could be found between this combined marker and the duration of treatment and the LOS.