OBJECTIVE Transthoracic contrast echocardiography (TTCE) is recommended for screening of pulmonary arteriovenous malformations (PAVMs) in hereditary haemorrhagic telangiectasia. Shunt quantification is used to find treatable PAVMs. So far, there has been no study investigating the reproducibility of this diagnostic test. Therefore, this study aimed to describe inter-observer and inter-injection variability of TTCE. METHODS We conducted a prospective single centre study. We included all consecutive persons screened for presence of PAVMs in association with hereditary haemorrhagic telangiectasia in 2015. The videos of two contrast injections per patient were divided and reviewed by two cardiologists blinded for patient data. Pulmonary right-to-left shunts were graded using a three-grade scale. Inter-observer and inter-injection agreement was calculated with κ statistics for the presence and grade of pulmonary right-to-left shunts. RESULTS We included 107 persons (accounting for 214 injections) (49.5% male, mean age 45.0 ± 16.6 years). A pulmonary right-to-left shunt was present in 136 (63.6%) and 131 (61.2%) injections for observer 1 and 2, respectively. Inter-injection agreement for the presence of pulmonary right-to-left shunts was 0.96 (95% confidence interval (CI) 0.9-1.0) and 0.98 (95% CI 0.94-1.00) for observer 1 and 2, respectively. Inter-injection agreement for pulmonary right-to-left shunt grade was 0.96 (95% CI 0.93-0.99) and 0.95 (95% CI 0.92-0.98) respectively. There was disagreement in right-to-left shunt grade between the contrast injections in 11 patients (10.3%). Inter-observer variability for presence and grade of the pulmonary right-to-left shunt was 0.95 (95% CI 0.91-0.99) and 0.97 (95% CI 0.95-0.99) respectively. CONCLUSIONS TTCE has an excellent inter-injection and inter-observer agreement for both the presence and grade of pulmonary right-to-left shunts.