Silent Embolism in Diagnostic Cerebral Angiography and Neuro-interventional Procedures

abstract & commentary

Source: Bendszus M, et al. Silent embolism in diagnostic cerebral angiography and neurointerventional procedures: A prospective study. Lancet 1999;354:1594-1597.

Bendszus and colleagues used clinical criteria complemented by MRI scanning with diffusion weighted imaging (DWI) to evaluate the safety of diagnostic and interventional cerebral angiography. One hundred consecutive patients (66 diagnostic, 34 interventional) were evaluated. DWI imaging showed 42 new lesions in 23 patients (17 diagnostic, 6 interventional), with no clinical correlates. The frequency of DWI lesions was significantly higher in patients with vascular risk factors (44%) than in those without (13%). Abnormal DWI was also associated with procedural factors such as difficulty in probing vessels, increases in quantity of contrast medium, fluoroscopy time, and number of catheters used.

DWI was more frequently abnormal in diagnostic (26%) than in interventional cases (18%). DWI lesions were confirmed by standard T2 and FLAIR imaging in all patients (n = 8) who had follow-up imaging.


These data suggest that a high rate of asymptomatic microemboli may be detected by DWI imaging. Patients with underlying vasculopathy were at highest risk, presumably due to an increased burden of atheroma in the aorta and great vessels. Patients undergoing interventional procedures were at lower risk, possibly due to a lesser amount of vasculopathy in these patients. The routine use of heparin in interventional cases might also have been protective (Debrun GM, et al. Am J Roentgenol 1982;139:139-142).

Are these DWI abnormalities clinically important and should the data call into question the safety of diagnostic angiography? Although nearly a quarter of the patients in this series had positive DWIs, none had clinical sequelae. Similarly, data from Transcranial Doppler studies in cardiac catheterization suggest that while microemboli (possibly composed of saline or contrast bubbles, tiny thrombi, or flecks of cholesterol) are common, they do not produce acute neurological effects (Bladin CF, et al. Stroke 1998;29:2367-2370). However, more sensitive neuropsychiatric batteries performed in follow-up of cardiac surgery patients have shown subtle effects correlating with microembolic signals on TCD (Braekken SK, et al. J Neurol Neurosurg Psych 1998;65: 573-575). —azs