Introduction:

COVID pneumonia was associated with a high risk of thromboembolic complications particularly pulmonary embolism (PE). The D-dimers level, a marker of inflammation, is often high in COVID 19. What would be, then, its place in the diagnostic prediction of PE in COVID 19?

Aim:

To study the diagnostic performance of D-dimer to exclude PE in COVID19.

Methods:

We included 470 patients hospitalized in the pulmonary department of Charles Nicolle hospital in Tunisia, from September 2020 to September 2021, for COVID pneumonia.

We defined two groups: G1 (with PE (n=20) and G2 (without PE (n=450))

ROC curve was generated to measure the diagnostic performance and the cutoff level of D-dimer in PE.

Results:

Our 470 patients had a mean age of 60 years old, with a male predominance at 54%.

PE was statically associated with the chest pain (40% vs 3%, p<0.001), hypocapnia (32 mmHg vs 39 mmHg, p=0.001), also with a higher troponin level (139 ng/l vs 18 ng/l, p<0.001) and d-dimer level (15,347 ng/ml vs 4.104 ng/ml, p= 0.002).

ROC curve of the D-Dimers level according to the PE occurrence showed an area under the curve (AUC) of 0.842 ([0.753-0.932] 95% CI, p<0.0001).

The best D-dimer cutoff value on the ROC curve for PE diagnosis was 1690 ng/ml with a sensitivity of 88%.

Conclusions:

Our results agree with the literature in terms of higher D-dimer concentration and higher D-dimer threshold to exclude PE. Thus, the use of a higher D-dimer cut-off than the conventional one could have a better diagnostic performance of PE in COVID.