Cardiac valve disease and low-dose dopamine agonist therapy: an artefact of reporting bias?

Abstract

Chronic low-dose cabergoline treatment for microprolactinoma may cause cardiac valve pathology, but the evidence is contradictory. We investigated whether the expectation of the echocardiographer could influence the report. Transthoracic echocardiograms from 40 patients aged 49·3 ± 9·6 (mean ± SD) years (Men:Women 7:33) on long-term cabergoline and bromocriptine therapy (duration 9·94 ± 4·5 years) were randomly assigned to two groups of echocardiographers so that each echocardiogram was reported twice. One group was told that 'the patients were control subjects' (Group A) and the other that 'the patients were on dopamine agonist therapy which is known to cause valve disease' (Group B). An observer who was blind to the group scored the reports for regurgitation at each valve (scores 0-4; max 16 per case). Mean total regurgitation score was significantly higher in Group B (1·43 ± 1·28; P = 0·014) than in Group A (0·73 ± 1·30). The difference was mainly from reporting trivial regurgitation: (mitral 16 vs 5, P = 0·005; tricuspid 17 vs 6, P = 0·007 and pulmonary 8 vs 1, P = 0·013). Mild regurgitation was uncommon (mitral 1 vs 1 and tricuspid 3 vs 6). Moderate regurgitation occurred in only one case and was associated with restriction of the leaflets consistent with the effects of cabergoline. Valve thickening was not reported in Group A, but in 9 (23%) mitral and 4 (10%) aortic valves in Group B. Long-term, low-dose dopamine agonist therapy rarely causes cardiac valve disease, but operator bias can lead to over-reporting of both valve thickening and trivial regurgitation.