Background Malignant pleural effusions (MPEs) cause disabling dyspnoea in over 1 million people worldwide per year. Currently recruiting and recently reported randomised controlled trials (RCTs) use the visual analogue scale for dyspnoea (VASD) to assess mean daily breathlessness in patients with MPEs (ISRCTN12852177, ISRCTN4784593, ISRCTN73255764) in order to provide evidence for the optimal method of symptom palliation. The VASD consists of a 100 mm line which subjects mark at a point representing their dyspnoea intensity. Determination of the minimal important difference (MID) for the VASD in patients with MPEs is essential to interpret the results of these trials.

Methods Patients with a confirmed or suspected MPE undergoing a pleural procedure recorded their baseline VASD prior to the procedure and, 24 hours later, their post-procedure VASD and assessed their dyspnoea on a 7 point Likert scale. Age, gender, diagnosis, procedure performed and volume of fluid drained were also recorded.

Results A total of 114/123 (93%) questionnaires were returned. Mean age of respondents was 70 years and 56% were female. Commonest malignancies were breast (41%), mesothelioma (26%) and non-small cell lung cancer (18%). Procedures included: therapeutic aspiration (35%), medical thoracoscopy (27%), chest drain insertion (11%), diagnostic tap (11%) and indwelling pleural catheter insertion (8.8%). The mean decrease in VASD in patients reporting a ‘small but just worthwhile decrease’ in their dyspnoea (i.e. equivalent to the MID) was 19mm (95% CI 14–24 mm). The volume of fluid required to drain to produce a change in VASD of 19mm was 760ml. Mean decrease in VASD for the different procedures were: chest drain 41mm; IPC insertion and drainage 41mm; therapeutic aspiration 31mm; diagnostic aspiration 19mm; and LAT 24mm.

Conclusion The MID for the VASD in patients with a MPE undergoing a pleural procedure is 19mm (95% CI 14–24mm). This value should be used when interpreting the results of RCTs in patients with MPEs using the VASD as an outcome measure and to calculate the sample size for future RCTs.