Clinical Correlations of Doppler Microembolic Signals in Patients With Prosthetic Cardiac Valves

Background—The clinical relevance of Doppler microembolic signals (MES) in patients with prosthetic cardiac valves was evaluated by merging and statistically reanalyzing patient data from four research institutions (Departments of Neurology, Universities of Aachen, Halle, and Münster, Germany; Department of Medicine and Therapeutics, University of Glasgow, Scotland, and Department of Cardiothoracic Surgery, Western Infirmary, Glasgow, Scotland). Methods—Transcranial Doppler monitoring for MES was performed over the middle cerebral arteries for 30 to 60 minutes per patient. Prevalence of neurological complications was evaluated with a standard neurological questionnaire in patients carrying the valve implant longer than 3 months (n5369). Results—Significant differences in MES prevalence and counts were noted among the 580 patients depending on valve type (presented with medians and [95% confidence intervals]): St Jude Medical, n5200, 72%, 4 [3 to 6]; Björk Shiley Monostrut, n599, 92%, 133 [93 to 181]; Medtronic Hall, n580, 47%, 1 [2 to 5]; ATS, n561, 52%, 3 [2 to 5]; Tecna, n538, 71%, 2 [1 to 4]; Carbomedics, n537, 81%, 8 [5 to 13]; Carpentier-Edwards supraannular, n554, 39%, 1 [0 to 3]; Sorin biological, n511, 9%, 0 [0 to 0]. No relation between MES counts and valve size, international normalized ratio, patients’ age, cardiac rhythm, or implant duration was noted. No significant differences in MES counts or prevalence (22 [3 to 68] versus 5 [3 to 6] and 63% versus 69%, both P..05), in valve duration, valve position, valve type, patients’ age, sex, cardiac rhythm, or international normalized ratio were evident between neurologically symptomatic (n542) and asymptomatic patients. Conclusions—MES in patients with prosthetic cardiac valves depend on the type and, in certain valve types, the position of the valve implant and possess no direct clinical significance. (Stroke. 1998;29:140-143.)