Lung ultrasound findings in hospitalized COVID-19 patients in relation to venous thromboembolic events: the ECHOVID-19 study

Skaarup, Kristoffer Grundtvig; Lassen, Mats Christian Højbjerg; Espersen, Caroline; Lind, Jannie Nørgaard; Johansen, Niklas Dyrby; Sengeløv, Morten; Alhakak, Alia Saed; Nielsen, Anne Bjerg; Ravnkilde, Kirstine; Hauser, Raphael; Schöps, Liv Borum; Holt, Eva; Bundgaard, Henning; Hassager, Christian; Jabbari, Reza; Carlsen, Jørn; Kirk, Ole; Bodtger, Uffe; Lindholm, Matias Greve; Wiese, Lothar; Kristiansen, Ole Peter; Walsted, Emil Schwarz; Nielsen, Olav Wendelboe; Lindegaard, Birgitte; Tønder, Niels; Jeschke, Klaus Nielsen; Ulrik, Charlotte Suppli; Lamberts, Morten; Sivapalan, Pradeesh; Pallisgaard, Jannik; Gislason, Gunnar; Iversen, Kasper; Jensen, Jens Ulrik Stæhr; Schou, Morten; Skaarup, Søren Helbo; Platz, Elke; Biering-Sørensen, Tor

PURPOSE: Several studies have reported thromboembolic events to be common in severe COVID-19 cases. We sought to investigate the relationship between lung ultrasound (LUS) findings in hospitalized COVID-19 patients and the development of venous thromboembolic events (VTE). METHODS: A total of 203 adults were included from a COVID-19 ward in this prospective multi-center study (mean age 68.6 years, 56.7% men). All patients underwent 8-zone LUS, and all ultrasound images were analyzed off-line blinded. Several LUS findings were investigated (total number of B-lines, B-line score, and LUS-scores). RESULTS: Median time from admission to LUS examination was 4 days (IQR: 2, 8). The median number of B-lines was 12 (IQR: 8, 18), and 44 (21.7%) had a positive B-line score. During hospitalization, 17 patients developed VTE (4 deep-vein thrombosis, 15 pulmonary embolism), 12 following and 5 prior to LUS. In fully adjusted multivariable Cox models (excluding participants with VTE prior to LUS), all LUS parameters were significantly associated with VTE (total number of B-lines: HR = 1.14, 95% CI (1.03, 1.26) per 1 B-line increase), positive B-line score: HR = 9.79, 95% CI (1.87, 51.35), and LUS-score: HR = 1.51, 95% CI (1.10, 2.07), per 1-point increase). The B-line score and LUS-score remained significantly associated with VTE in sensitivity analyses. CONCLUSION: In hospitalized COVID-19 patients, pathological LUS findings were common, and the total number of B-lines, B-line score, and LUS-score were all associated with VTE. These findings indicate that the LUS examination may be useful in risk stratification and the clinical management of COVID-19. These findings should be considered hypothesis generating. CLINICALTRIALS.GOV ID: NCT04377035