| Popis: |
Background. Triage of patients with COVID-19 pneumonia is very time-sensitive. It allows effective rearrangement of hospital resources and can ultimately result in saving the life of the patient. Challenges of using CT scan in the overwhelmed healthcare system require the search for additional methods to differentiate severe COVID-19 pneumonia. We are proposing a new lung ultrasound (LUS) scoring protocols with good diagnostic characteristics to determine the severity of COVID-19 pneumonia which could be used along with CT scan during triage in emergency settings Materials and methods. We retrospectively collected data from 161 hospitalized patients with confirmed COVID-19 pneumonia, who underwent both CT scan and LUS within 24 hours of admission. Three consecutive LUS protocols, including two originated by the authors (LUS NMHC), have been tested to find the most reliable for the accurate COVID-19 pneumonia diagnosis (in terms of correlation with chest CT scan). We also tested the ability of LUS to differentiate between disease severity, as found on CT scans, as well as between different outcomes. Results. Both 16-zone LUS NMHC and 12-zone LUS NMHC protocols showed good ability to differentiate between moderate (50% on CT scan) and severe lung damage (>50% on CT scan). AUC for the ROC curves were almost identical with 0.83 (95% CI, 0.75-0.90) and 0.81 (95% CI 0.73-0.88) for 16-zone LUS NMHC and 12-zone LUS NMHC protocols, respectively. The 16-zone LUS NMHC had optimal cutoff score of 20 with a sensitivity of 67% and specificity of 82%, while 12-zone LUS NMHC provided optimal cutoff score of 15 with the same sensitivity but lower specificity of only 73%. Scores from neither 16-zone or 12-zone LUS NMHC protocols were predictive of the patient outcome. Conclusion. Newly developed 16-zone LUS NMHC and 12-zone LUS NMHC protocols for patients with COVID-19 pneumonia proved to be feasible and had strong correlation with CT-findings. The 16-zone LUS NMHC protocol was more efficient in the triage of patients who had more than 50% of lung damage on CT. Both protocols could be useful in emergency settings and in the healthcare facilities with limited or no access to CT. Trial registration: retrospectively registered |
