Can Doppler time domain analysis of microembolic signals discriminate between gaseous and solid microemboli in patients with left ventricular assist device?

OBJECTIVE

Microembolic signals (MES) can be detected by transcranial Doppler sonography (TCD). To identify gaseous microemboli the inhalation of oxygen is an established method in patients with prosthetic heart valves. Time domain analysis of sample volume length (SVL) and of frequency modulation showed promising results in the discrimination between solid and gaseous microemboli. We investigated whether these time domain analyses allow the discrimination of different types of microemboli in patients with the non-pulsatile DeBakey left ventricular assist device (LVAD).

MATERIALS AND METHODS

Repeated unilateral detection of MES was performed by TCD in 20 patients supported with DeBakey LVAD. Each monitoring session consisted of 20 minutes without and 20 minutes with inhalation of 100% oxygen (6 l/min). A total of 500 MES, detected with (n=250) or without (n=250) the supply of oxygen, were randomly chosen for offline analysis. The SVL (in cm) was calculated by duration and velocity of the MES measured in the time domain mode. Additionally, frequency modulation of MES was classified into three main types: Without modulation (type I), with gradual changes (type II) and with rapid changes (type III).

RESULTS

With oxygen supply, both prevalence (26.4% versus 36.2%, p<0.01) and mean counts of MES per hour (49+/-293 versus 108+/-550, p<0.001) significantly declined compared with the MES load while breathing room air. There was no significant difference in the SVL of MES under oxygen (0.85+/-0.38 cm) compared with those without oxygen delivery (0.92+/-0.37 cm, p=0.6). Furthermore, no significant differences were noted for the MES frequency modulation types in time domain analysis with regard to oxygen supply.

CONCLUSIONS

The reduction of MES under oxygen delivery confirms the gaseous nature in a substantial number of circulating microemboli produced by the DeBakey LVAD. However, SVL and frequency modulation of MES did not appear to provide valuable information regarding the structural nature of the underlying microembolic material.