In this study we found that MDW, a measure of monocyte activation that can be routinely reported on the CBC differential, has potential to be used as an ED-based screening test for respiratory viral infection. Overall accuracy (AUC) of MDW for diagnosis of both SARS-CoV-2 and influenza infection was very good; high sensitivities and negative predictive values were also achieved for both viruses using the previously established and currently reported cut-off for MDW. The identification of such a measure embedded within a laboratory panel that is routinely ordered in acute care settings is a major advancement 31,32. It affords the opportunity for the development and implementation of broad-based screening programs to enable rapid identification and isolation of patients with contagious viral infections, even for cases that are unsuspected. Negative likelihood ratios achieved by MDW also suggest it could be used to preserve more expensive molecular diagnostics under conditions where testing resources are limited, as has been the case for SARS-CoV-2 18,19. Importantly, the diagnostic performance of MDW was unique among all leukocyte parameters examined, further substantiating its independent value for decision-making in acute care environments.
Multiple recent studies have illustrated that MDW is a useful biomarker for early detection of sepsis and several have found that MDW levels increase in parallel with sepsis severity 1–3. In this study, we found that MDW is also elevated in the setting of SARS-CoV-2 and influenza infection, and that MDW elevation in this context is independent of illness severity; virally infected patients well enough to be discharged home had MDW values similar to those who required hospitalization or ICU admission. In our cohort, MDW did track linearly with disease severity as reported by others 1–3, but only in patients without SARS-CoV-2 or influenza infection. These findings suggest that infections with SARS-CoV-2 or influenza may be confounders when MDW is utilized as a screening tool for sepsis; an elevated MDW could be due viral infection with a less severe prognosis. In practice, elevated MDW should likely prompt consideration of or investigation for SARS-CoV-2 or influenza infection prior to or in parallel with activation of protocols for sepsis diagnosis and treatment.
While a small number of other studies have investigated the association between MDW and SARS-CoV-2 infection 33,34, our study is the largest to compare MDW values in patients with and without SARS-CoV-2 infections. Our findings are supported by the previous reports which suggest that MDW is elevated in patients with SARS-CoV-2 infection compared to those not infected with this virus 33,34 with mean differences between the two groups ranging from 1.8U to 7.2U. In a pooled analysis of three available studies, Lippi et al. determined that the weighted mean difference in the MDW for patients with SARS-CoV-2 compared to patients without this virus was 3.95U (95% CI 1.41U – 6.49U)34. In our study, this difference was 4.2U. Collectively, these data strongly suggest MDW is a reliable marker of SARS-CoV-2 infection.
Our study is also one of very few (and the largest to date) to have investigated MDW for the diagnosis of influenza. In the only other currently published report on MDW and influenza, Lin et al. evaluated the diagnostic performance of MDW in 174 ED patients with suspected respiratory infection who were also subjected to molecular viral testing and hospitalized after their ED encounter. Nine patients in this population tested positive for SARS-CoV-2 and 24 for influenza; MDW was differentially elevated in patients with either viral infection. In this small study, discriminatory performance of MDW was moderate for both SARS-CoV-2 and influenza, with reported AUCs of 0.70 and 0.63, respectively 33. Our study, performed in a much larger and less selected population, achieved much better diagnostic performance. The AUC of 0.83 for both SARS-CoV-2 and influenza observed here, along with the binary classification measures shown in Table 2, suggest MDW may be an effective pragmatic, ED population-wide screening test for these viral infections.
Importantly, the diagnostic performance of MDW for SARS-CoV-2 and influenza infection reported here is comparable to or better than that of many currently approved, more expensive, and less routinely available tests. While RT-PCR is the gold standard for diagnosis of these viral infections, several other platforms play important roles in their management 19,35. For example, SARS-CoV-2 antigen-based tests are routinely used in outpatient and urgent care settings and have been proposed as a useful adjunct to RT-PCR in the ED; the sensitivity of MDW for SARS-CoV-2 infection exceeds that of most antigen-based SARS-CoV-2 testing platforms 36–38. In our study, MDW was also more sensitive for influenza infection than several CLIA-waivered rapid influenza detection tests currently used in outpatient and ED settings 39,40. Our finding that this level of diagnostic performance can be achieved using a component of the CBC differential, a panel that is routinely ordered regardless of clinical suspicion for viral infection, is unique and potentially practice changing.