Background: In most patients, severity of valvular aortic stenosis can be accurately assessed non-invasively by echocardiography. However, retrograde catheterisation of the aortic valve is often undertaken. This procedure has a potential risk of neurological complications, with an unknown incidence of clinically silent embolism. We aimed to establish the frequency of clinically apparent and silent cerebral embolism after this procedure.
Methods: We prospectively randomised 152 consecutive patients with valvular aortic stenosis at a German university hospital to receive either cardiac catheterisation with (n=101) or without (n=51) passage through the aortic valve. Patients underwent cranial MRI and neurological assessment within 48 h before and after the procedure to assess cerebral embolism. Controls were 32 patients without valvular aortic stenosis who underwent coronary angiography and laevocardiography.
Findings: 22 of 101 patients (22%) who underwent retrograde catheterisation of the aortic valve had focal diffusion-imaging abnormalities in a pattern consistent with acute cerebral embolic events after the procedure; three of these patients (3%) had clinically apparent neurological deficits. By contrast, none of the patients without passage of the valve, or any of the controls, had evidence of cerebral embolism as assessed by MRI.
Interpretation: Patients with valvular aortic stenosis who undergo retrograde catheterisation of the aortic valve have a substantial risk of clinically apparent cerebral embolism, and frequently have silent ischaemic brain lesions. Patients should be informed about these risks, and this procedure should be used only in patients with unclear echocardiographical findings when additional information is necessary for clinical management.