Background Pleural effusion size may determine patient management and is routinely reported using thoracic computed tomography (CT) and ultrasound scans. We aimed to compare agreement between these two modalities.

Methods Between August 2015 and January 2016, patients referred through the pleural service with pleural effusions were included if both CT and ultrasound scans had been performed ≤10 days apart with no intervening pleural procedure. Consultant radiologists reported effusions on CT as small, moderate or large. Thoracic ultrasound scans were performed by the pleural team (level 1 Royal College of Radiologists practitioners). The height of the effusion (in rib spaces) and maximal medial depth (in cm) were recorded. Small, moderate and large effusions by ultrasound were defined as ≤1, 2–3 and ≥4 rib spaces respectively. CT and ultrasound categories were compared using the Stuart-Maxwell Test.

Results In 51 patients the median age was 70 (interquartile range [IQR] 59–80) years and 26 (51%) were male. Right effusions were dominant in 29 (56%), and 48 (94%) had a CT chest or CT pulmonary angiograms. CT scans were reported by dedicated thoracic radiologists in 13 (26%) patients. CT scan reported size was available in 37 (72%) cases, and of these, 6 (16%), 11 (30%) and 20 (54%) were defined as small, medium and large respectively. For small, moderate and large effusions by ultrasound, the median height in rib spaces was 1.5 (IQR 1–2), 3 (IQR 2–3) and 3.5 (IQR 2.75–4) respectively. Height in rib spaces (r = 0.537, p < 0.001) and depth of effusion (r = 0.365, p = 0.02) correlated with CT reported size. However overall classification of ultrasound size was not associated with CT size (p = 0.04). In 15 (40%) effusions, the size classification differed.

Conclusions Our findings indicate that pleural effusions size determined by CT and ultrasound differed. Revised simple methodology to estimate effusion size should be sought, and may help to refine patient pathways of care.