Computerized Tomography-Based Quantification of Regional Volume in Murine Lung Specimens

Objectives Alterations in regional lung morphology are indicative of pathological changes in the lung. Accurate estimation of morphology-related metrics, particularly volume, is critical in understanding the pathophysiology and progression of disease. Current techniques for lung volume measurement include morphometry based on histologic sectioning and micro-computerized tomography (micro-CT). Advantages of micro-CT as a modality include 3-dimensional (3D) structure preservation, time efficiency and automation. Limitations of current micro-CT imaging techniques for lung volume determination include inefficient procedures for sample preparation and reduced contrast at high resolution to study lung morphology reliably. We describe a simple and efficient ex vivo lung preparation technique, adapted from a previously reported method used to evaluate lung fibrosis, to quantify regional lung volumes of wild type (WT) and Claudin18 knockout (KO) mice with lung enlargement using high-resolution (0.7 µm) micro-CT scanning.

Methods Lungs from three wild type (WT) and Claudin18 KO mice (age 4 months) were insufflated with 4% paraformaldehyde at 20 cm H2O pressure overnight. Samples were dehydrated through an ethanol gradient and incubated in hexamethyldisilazane (HMDS) before air-drying. Lungs were scanned at an isotropic resolution of 10 µm at 45kVp, 200mAs with a Scanco µCT50 scanner. Small sections (~2 mm3) cut from distal lung were scanned at an isotropic resolution of 0.7 µm with an energy focused X-ray spectrum. Raw CT data were processed and visualized using AMIRA (FEI) software. 3D segmentation of lung tissue and airspace was performed by exploiting the enhanced radiodensity difference between tissue and air achieved by our method. Total lung and regional volumes (i.e., conducting airways, alveolar space and parenchyma of the small section) were automatically quantified by AMIRA.

Results Consistent with literature values, WT total lung volume (Vlung) and conducting airway volume (Vairway) were 0.62 ± 0.02 ml and 0.110 ± 0.00 ml, respectively. Alveolar airspace fraction (Falv) and parenchymal fraction (Fpar) of the small lung section were 0.79 ± 0.00 and 0.21 ± 0.00, respectively. Alveolar airspace volume (Valv) and parenchymal volume (Vpar), calculated as Falv x (Vlung-Vairway) and Fpar x (Vlung-Vairway), were 0.40 ± 0.01 ml and 0.11 ± 0.01 ml, respectively. Claudin18 KO mice lungs showed increased total lung volume (1.05 ± 0.08 ml, p<0.05) with increased parenchymal volumes (0.56 ± 0.07 ml, p<0.05). WT and Claudin 18 KO total lung volumes were consistent with results obtained by volume displacement showing ~2-fold increases in lung volume in Claudin 18 KO mice.

Conclusions The lung preparation method described preserves lung 3D structure for micro-CT scanning while accentuating tissue-air contrast within the lung. The use of an energy focused X-ray spectrum reduces imaging artifacts, allowing improved visualization of alveolar microstructure and conducting airways at high resolution and facilitating quantification of lung volume changes in different lung compartments. This platform shows promise for accurate quantification of lung volumes in both normal and diseased lungs.