ETD

• DTF
• Lung Ultrasound and Diaphragmatic Ultrasound
• LUS score
• PISA index

Backround and aims: In the Covid era, the lung ultrasound, using LUS score (LUS aeration score), and diaphragmatic ultrasound, using DTF (Diaphragmatic thickening fraction), have been studied to the aim of improving the management of patients with pneumonia. the literature supports the role of these tools for diagnosis and as a predictor of severity disease, adverse outcomes and mortality. However there is little evidence of their use as predictor of intubation.
In this contest the aims of this study was to evaluated the ability of LUS score, DTF, and the PISA index (ratio between DTF an LUS) as predictors of orotracheal intubation and mechanical ventilation and to determinate the optimal cut-off value in COVID 19 pneumonia.
Methods: We retrospectively analyzed the data of 343 patients admitted to the general intensive care and internal medicine and submitted oxygen therapy with Venti-mask and CPAP before considering intubation and mechanical ventilation. LUS aeration score (LUS score) and Diaphragmatic thickening fraction (DTF) inferred by ultrasound within 12 hours of Hospital admission.
Results: 55 patients required intubation and mechanical ventilation. the univariate logistic regression models performed for orotracheal intubation, only PISA index [beta:-6,884; OR 0,001; p < 0,001]resulted statistically significant. In the sub subsequent multivariate model, PISA index reached statistical significance [Beta:- 5,758; OR 0,03; p < 0,001].The best cutoff value is 1,6 (AUC 0,94, CI 0,91-0,97, sensibility 87%, specificity 78%, p < 0,001). The diagnostic accuracy of two parameter taken alone is lower: LUS score with cut-off value > 17 (AUC 0,73; CI 0,67-0,79; sensibility 69%; specificity 68%, p < 0,001) and DTF with cut-off value < 21(AUC 0,86; CI 0,80-0,91; sensibility 91%; specificity 65%; p < 0,001). Cumulative probability for lack of intubation in patients with PISA index higher or lower than 1.6 was calculated with the Kaplan-Meier product-limit estimator. The log-rank (Mantel-Cox) test was applied to evaluate the difference in probability for lack of intubation after grouping for PISA index higher or lower than 1.6.
Conclusions: In COVID-19 Pneumonia, a PISA index score < 1,6 may represents a promising early predictor of intubation with a higher diagnostic accuracy compared with LUS or DTF.