# Comparison of quantitative lung ultrasound scores with automated quantitative CT: to overcome the ultrasound limitations

## TL;DR

This study compares lung ultrasound scores with CT scans to assess interstitial lung disease in systemic sclerosis patients, finding strong correlations that suggest ultrasound can evaluate deeper lung regions.

## Contribution

The study introduces a novel pleural line irregularity score that correlates with deeper lung disease volume, overcoming traditional ultrasound limitations.

## Key findings

- Quantitative LUS scores correlated strongly with CT-measured ILD extent, including surface and core volumes.
- The novel PLI score showed significant correlations with basal and mid-lung ILD changes.
- Multivariate analysis confirmed the associations between LUS scores and CT-defined ILD volumes.

## Abstract

Lung ultrasound (LUS) is emerging as a valuable tool for assessing systemic sclerosis-associated interstitial lung disease (SSc-ILD), although it traditionally explores only superficial lung regions. Building upon our preliminary findings, this study investigated correlations between quantitative LUS scores and automated quantitative computed tomography (qCT) measurement of ILD extent, including both superficial and deeper lung involvement.

Between 2021 and 2023, 82 consecutive SSc patients underwent concurrent LUS and CT scans. Total B-lines (BL) count (range 0–140) and our novel pleural line irregularity (PLI) score (range 0–28) were obtained using a 14-intercostal space scanning protocol. CT scans were analysed by automated texture analysis software, quantifying volumes of ILD, ground-glass opacities (GGO) and reticulations (RET), segmented in three levels (apices, midfields, bases) and subdivided in surface and core lung parenchyma.

Total BL count and PLI score correlated with total ILD, GGO and RET volumes (all P < 0.0001), as well as with surface and core ILD volumes (all P < 0.0001). Basal lung BLs and PLI score correlated with basal ILD, GGO, RET (all P < 0.005) and corresponding surface and core ILD volumes (all P < 0.005). Mid-lung PLI correlated also with corresponding ILD-related changes and surface and core ILD (all P < 0.005). These associations were confirmed by multivariate regression analysis.

Quantitative LUS score correlated with qCT-defined ILD extent, especially at lung bases. LUS scores (particularly the novel PLI score) were found to correlate with deeper ILD volume, suggesting potential to overcome traditional LUS limitations related to superficial lung assessment.