Letter to the Editor: D-dimer and Presumptive Diagnosis of COVID-19.

To the Editor:

We read with interest, “D‐dimer Triage for COVID‐19” by Li et al. This observational cohort study from Wuhan, China, concludes that a D‐dimer performed at admission could be an effective and easily available diagnostic surrogate marker for coronavirus disease‐19 (COVID‐19). We are concerned that several limitations in the study's methodology affect the authors’ conclusions.

Our first concern is the use of D‐dimer as a rule‐in strategy. A negative D‐dimer is useful but when positive can be caused by multitude of conditions that are unrelated to COVID‐19. The utility of the D‐dimer requires age adjustment, which this study doesn't seem to have taken into account as the positive threshold was set as >1.0 mg for all patients.

We are also concerned about the lack of reporting of important outcomes for the population included in this study. No information is provided on the impact of the clinical management on patients with positive D‐dimer results. As well, the proportion of patients who subsequently underwent a pulmonary angiogram and the result of imaging studies are not reported. The characteristics of the study population such as age, sex, risk factors, pregnancy, comorbidities, or current medications especially anticoagulation medications are not provided. These data are necessary to be able to understand to which population the conclusions, if valid, could be applied to.

A particular statement caught our attention: “Elevated D‐dimer levels could be presumptively diagnosed as COVID‐19 and triaged as higher risk.” We feel that this conclusion is erroneously inflating the diagnostic specificity of a positive D‐dimer for COVID‐19 and we would like the authors to report a 2 × 2 table showing the positive predictive and negative predictive values with a criterion standard PCR test for COVID‐19.

Another use for D‐dimer might be as a predictor of the severity of the disease rather than its diagnostic value. However, the authors do not report all the relevant outcomes to assess whether or not a positive D‐dimer in their cohort was associated with the severity COVID‐19.

Finally, because this is a single‐center study of 749 patients, we believe that a larger number of patients and reporting of both clinical outcomes and COVID PCR results are necessary to validate the accuracy of D‐dimer value as a diagnostic tool in COVID‐19 patients.

While we appreciate the authors’ intent to establish diagnostic testing for COVID‐19–suspected patients, we feel that this study's claims are far too bold and require much more transparency of information.