Introduction/Purpose: In the veterinary literature, a comprehensive assessment of pleural abnormalities on computed tomography (CT) in a large number of patients with pleural effusion is lacking. The purpose of the study was to describe pleural abnormalities on CT in dogs and cats with pleural effusion and to evaluate for potential discriminatory pleural features between different effusion types.
Methods: Patients with cytologic evaluation of pleural effusion and/or cytologic/histopathologic evaluation of the pleura that underwent pre- and post-contrast thoracic CT were included. CT examinations were reviewed by two radiologists for presence (yes, no), appearance (smooth, irregular, nodular, mass), location (costal, mediastinal, diaphragmatic, pulmonary pleura), distribution (focal, multifocal, circumferential) and severity (minimum/maximum measurements) of pleural abnormalities. Between-group comparisons were made using Fisher’s exact test for categorical variables and Wilcoxon rank sum test or Kruskal Wallis test for continuous variables. Statistical significance was set at p < 0.05.
Results: One hundred thirty-six patients with malignant (n=24), exudative (43), chylous (38), transudative (19), and hemorrhagic (12) effusion were included. Pleural thickening ≥10mm in any location (33/67 [49.3%], p<0.001), presence of mediastinal mass or mass-like thickening (25/67 [37.3%], p<0.001]), circumferential pleural thickening (20/67 [29.9%], p=0.008) and irregular costal thickening (59/67 [88.1%], p<0.001) were all significantly associated with malignant or exudative effusion compared with other types (6/69 [8.7%]; 2/69 [2.9%]; 8/69 [11.6%]; 39/69 [56.5%], respectively). Nodular pleural thickening in any location, particularly diaphragmatic nodules, was significantly (p<0.001) higher with malignancy (any location: 18/24 [75%]; diaphragm: 9/24 [37.5%]) than with non-malignant effusions (any location: 37/112 [33%]; diaphragm: 9/112 [8%]). However, mediastinal nodular thickening did not differ (p=0.267) between malignancy (14/24 [58.3%]) and exudate (19/43 [44.2%]). Mediastinal thickening involving only the pericardium was more common (p<0.001) with transudative, chylous and hemorrhagic effusion (14/69 [21.5%]) than with malignancy or exudate (1/67 [1.5%]). Pulmonary pleural thickening was significantly (p=0.013) more common with exudative and chylous effusion (42/81 [51.9%]) compared to other types (16/55 [29.1%]). Absence of any pleural thickening was significantly (p=0.001) more likely with transudative effusion (5/19 [26.3%]) than with other effusions (5/117 [4.3%]).
Discussion/Conclusion: There are some discriminatory pleural features that would allow for prioritization of differential diagnoses in patients with pleural effusion, although malignant and exudative effusions shared many common features.