Introduction and objectives Chest X-ray (CXR), CT and Ultrasound (US) are commonly used to evaluate the size of pleural effusions. Accurate description of size is important in the communication of findings and urgency of intervention. With currently no standardised measurement system, significant variation in description of size by CXR, CT or US exists. The use of terms ‘small, moderate, and large’ is common, with no consensus on the limits of these sizes.

This study looked at correlation between qualitative description of effusion size by different imaging modalities and volume of effusion recorded following aspiration.

Methods This was a retrospective analysis of patients referred for pleural tap and/or drain after CXR and/or CT. CXR/CT reports were collected from PACS, US reports from the local US database, accessed by at least two US-trained Respiratory physicians.

Effusion size was estimated by the recognised method of counting intercostal spaces (ICS) from costophrenic angle (small- localised to 1 ICS, medium 2–3 ICS, large ≥4 ICS). Effusion size reported was compared to actual volume of fluid drained (till ‘dry’ or ‘safe aspiration’). For the purpose of this study, effusions <500 mL were characterised as small, 500–1000 mL moderate and >1000 mL large. Correlation was analysed using Spearman’s correlation.

Results 312 patients were referred April 2014–December 2015. 133 patients were excluded due to insufficient data, 179 patients’ data analysed. US pleural effusion size estimation correlated most closely with actual volume of fluid drained (r = 0.833, N = 179, P < 0.0001) vs. CXR (r = 0.548, N = 129, P < 0.001) and CT (r = 0.489, N = 107, P < 0.001). The error rate in size estimation was 41% (53/129) for CXR, 57% (61/107) for CT and 16% (28/179) for US. In particular, 29% (31/107) patients with ‘small’ tapped effusions were reported to be ‘medium/large’ effusions by CT scan. CT most commonly overestimated fluid present; whilst US tended to underestimate the few cases where it was inaccurate.

Conclusions This study demonstrates that US may be the most accurate modality when assessing the size of pleural effusions. CT imaging may over represent the volume of fluid present. Where imaging reports guide further management, reliability and consistency is essential to avoid unnecessary/urgent intervention and patient anxiety.