It is known that hospitalized, bedridden patients are particularly prone to development of deep venous thrombosis (DVT); in the absence of adequate prophylaxis, its overall incidence among in-hospital patients is 0.9%, rising up to 15% to 32% among intensive care unit (ICU) patients.1 ^, 2 However, during the last month, a remarkable increase in the diagnosis of DVT has been noticed among non-ICU hospitalized patients as well due to the numerous cases of DVT observed among patients infected by coronavirus disease 2019 (COVID-19). In the last month, the Unit of Vascular Surgery of the teaching hospital S. Matteo, in Pavia, one of the hub centers for COVID-19 in Lombardy, the Italian region most affected by the pandemic, was requested to perform 30 compression ultrasound scans of the venous system, of both lower and upper limbs, of COVID-19 non-ICU patients with signs or symptoms suggestive of DVT, and 16 of them had positive results; 6 more had no DVT but superficial thrombophlebitis or upper limb lymphedema. The most frequent localization of the thrombus was the iliac-femoral-popliteal axis, followed by the brachial-axillary veins and the calf veins (Table ). Four cases developed from a central vein catheter placed in the femoral vein. Although the numbers are too small to perform statistical analysis, the same period in 2019 could provide a term of comparison: in March 2019, the same unit performed 24 compression ultrasound scans on the whole hospital and found only 5 cases of DVT.

Currently, no literature exists about the pathogenetic mechanism of DVT in COVID-19 patients. A relationship was found between DVT and acute respiratory distress syndrome in influenza A H1N1, for which an enhanced leukocyte adhesion to the vein walls due to the production of inflammatory molecules was hypothesized, and empirical anticoagulation was proposed in all patients with severe acute respiratory distress syndrome.3 Production of procoagulant factors, such as D-dimer and tissue factor, is another possible mechanism that has been well outlined in other community-acquired pneumonias, although no relationship with DVT has been established; a similar procoagulant stimulation can also occur in COVID-19 infection.4 A third possible favoring element that could also explain the upper limb involvement is the use of a continuous positive airway pressure ventilator, which is often tied in a way that can compress the superficial or deep vessels of the upper limbs; all patients with involvement of the upper limbs, in fact, were receiving continuous positive airway pressure therapy.

Unfortunately, no data are yet available on the prognosis of patients developing DVT during a COVID-19 infection, so further research should inquire as to how the two conditions interact with each other and whether they affect the chance of recovery. However, considering these preliminary data, our institution is starting to administer anticoagulant doses of low-molecular-weight heparin in hospitalized COVID-19 patients, after monitoring of coagulation test results and kidney and liver function.

In conclusion, DVT can be considered a frequent and potentially lethal complication of COVID-19. It deserves further attention to establish incidence, mortality rate, and the opportunity of a screening program and prophylactic therapy in these patients.