Background: Introduction: Point-of-care ultrasound (POCUS) expediently detects the pulmonary manifestations of COVID-19 and is predictive of patient outcomes. Although lung POCUS can be used in the diagnosis and management of patients hospitalized with COVID-19, there are no standardized scanning protocols. Furthermore, it is not known how much scanning is needed to sufficiently rule-in the pulmonary complications of COVID-19.
Methods: Methods: We conducted a prospective cohort study at two hospitals from 3/2020-4/2021 to evaluate the utility of lung POCUS with COVID-19. Inclusion criteria included adult patients hospitalized for COVID-19 who received lung POCUS with a 12-zone protocol. Each image was independently interpreted by two reviewers blinded to patient information, who then met to provide a final interpretation. If no consensus could be reached, the study was excluded. Cohen Kappa values were used to calculate the level of agreement of clinical findings between a 2-zone, 4-zone, or 8-zone scanning protocol and the original 12-zone protocol. Clinical findings of interest included B-lines, consolidations, or a normal A-line pattern.
Results: Results: N=97 patients were included and there were no excluded studies. Patient characteristics are shown in Table 1. B-lines were present in n=88 patients (90.7%), consolidations in n=68 patients (70.1%), and a normal scan in n=4 patients (4.1%). B-lines on 2-zone (k=0.10 [95% CI: -0.03-0.24]), 4-zone (k=0.25 [95% CI: 0.04-0.46]), or 8-zone protocol (k=0.46 [95% CI: 0.19-0.73]) had low to modest agreement to 12-zone scans (Table 2). Consolidations on 2-zone (k=0.17 [95% CI: 0.08-0.25]), 4-zone (k=0.37 [95% CI: 0.25-0.51]), or 8-zone protocol (k=0.54 [95%CI: 0.39-0.69]) had low to modest agreement to 12-zone scans (Table 2). Normal scans on 2-zone (k=0.21 [95% CI: 0.03-0.39]), 4-zone (k=0.34 [95% CI 0.09-0.59]), or 8-zone protocols (k=0.54 [95% CI: 0.23-0.86]) had low to modest agreement to 12-zone scans (Table 2). Although only 4% of 12-zone scans were considered normal, a higher proportion of 2-zone scans (26.8%), 4-zone scans (17.5%), and 8-zone scans (10.3%) were considered normal despite originating from the same patients.
Conclusions: Simplified lung ultrasound protocols such as a 2-zone, 4-zone, or even 8-zone protocol may miss important pathologies for COVID-19, such as B-lines or consolidations. Furthermore, a patient may be misclassified as having minimal pulmonary involvement in a simplified protocol. While there is a balance between the time spent scanning and the additional information to be gained, these findings suggest that a 12-zone protocol (6 sites on each hemithorax) may be better at detecting important clinical markers of COVID-19. Additional comparisons of the number of zones sampled and patient outcomes are needed to confirm these findings.