Theoretically emboli, both gaseous and solid, would be expected to appear in the Doppler spectrum as short duration high intensity signals. Experimental studies have confirmed the sensitivity of the technique. Embolic signals have been detected in patients with a variety of embolic sources including prosthetic cardiac valves and carotid stenosis. In the latter group they appear to be clinically relevant correlating with time since symptoms, the degree of stenosis and plaque ulceration, and being reduced or abolished after carotid endarterectomy. However early studies have given very different frequencies of embolic signals. This may partly reflect different patient populations but also emphasizes the importance of technical factors. These are discussed in this review. A number of automated systems have been developed to detect embolic signals but none yet developed have sufficient sensitivity and specificity for routine use.
Background: The embolic signal obtained by TCD varies in sound, pitch and duration according to the size of the embolus and the materials of which it is composed. We tried to differentiate the micro-embolic signal from various embolic sources by wave analysis and also to clarify its clinical significance. Methods and Results: We recorded micro-embolic signals using TCD in 10 patients with various embolic conditions; 3 patients on a left ventricular assistance system (LVAS), 3 patients after replacement of the mitral valve with a prosthesis (MV), 2 patients with non-valvular atrial fibrillation (NVAF), and 2 patients with high-grade carotid stenosis (CS) . Acoustic power (AP), frequency, duration and background blood velocity (Vbg) were measured by a microcomputer system. The average duration of the embolic signal was longer in patients with LVAS and MV than in those with NVAF and CS. All patients who showed a correlation between Vbg and all parameters (AP, frequency, duration) were asymptomatic. Conclusion: The duration of micro-embolic signals, which reflects the size and components of the embolus, differs largely according to the embolic source. The correlation of background velocity with three parameters of micro-embolic signals (AP, duration, frequency) which appear to reflect the homogeneity of micro-emboli, may be a parameter indicating stroke risk.
High intensity transient signals (HITS) were detected in patients with a high risk of stroke. The signals were detected in the middle cerebral artery for 15-30 min by a TC2020 or Multi-Dop DX4. HITS were never detected in young healthy volunteers (n=20) . The HITS count in patients with cerebral infarction (16-20/30 min, n-42) was significantly higher than that in patients without cerebrovascular disease (0.5±0.7/30 min, n17) (p<0.01) . In the cerebral infarction group, the HITS count in patients without antiplatelet (20±29/30 min, n=20) was significantly higher than that in patients with antiplatelet (9±6/30 min, n=22) (p<0.01) . HITS were frequently detected in patients fitted with mechanical valves (St. Jude Medical valve) (14.3±34.5/15 min, n=51), and the count was significantly higher than that in open heart surgery patients without mechanical valves (0.05+0.08/15 min, n=11) (p<0.01) . In the mechanical valve replacement group, the HITS count in patients with cerebral infarction after surgery (88±66/30 min, n=43) was significantly higher than that in patients without a historyof stroke (7±7, n-8) (p<0.01) . The HITS intensity in the mechanical valve replacement group (15.1+4.3 dB, n=31) was significantly greater than that in the cerebral infarction group (10.8+1.1dB, n-32) (p<0.05) . In the mechanical valve replacement group, the HITS intensity in patients with post-surgical cerebral infarction (18±3.5 dB, n8) was significantly greater than that in patients without a history of stroke (14.5±5.6 dB, n=23) (p<0.0001) . During veno-arterial extracorporeal membrane oxygenation (V-A ECMO), HITS were frequently detected (124±60/15 min, n=6), and the HITS count was negatively correlated with activated coagulation time (ACT) . HITS may therefore reflect the presence of microemboli and indicate the risk of stroke or the state of blood coagulation.
The embolic signals detected by transcranial Doppler ultrasonography (TCD) are characterized as high-intensity transient signals (HITS) . Two types of TCD devices, conventional and embolus-oriented, are currently used. HITS demonstrated by an embolus-oriented TCD device correspond almost specifically to embolic signals. However, the characteristics of HITS in conventional TCD and their specificity to embolic signals remain to be studied. In this study, 10 ischemic patients were examined using a conventional TCD device. Six of these patients showed HITS. In all cases, these were partly unidirectional and partly bidirectional with a unilateral tendency, and the HITS were accompanied by chirping sounds. Since the chirping sounds had the typical characteristics of embolus origin which we had revealed experimentally, the HITS were interpreted to originate from moving emboli. We conclude that HITS with chirping sounds can be regarded as embolic signals in conventional TCD, irrespective of whether they are unidirectional or bidirectional with a unilateral tendency.
To evaluate intracranial pressure by transcranial Doppler ultrasound (TCD), the mean flow velocity (MFV) and Gosling's pulsatility index (PI) of the middle cerebral artery (MCA) and basilar artery (BA) were measured before and after glycerol administration, and also regional cerebral blood flow (rCBF) was investigated simultaneously. Ten patients with brain tumors and 5 normal volunteers were selected. After glycerol administration in the patient group, MFV and rCBF showed a significant increase of more than 27% and 16% respectively, in the hemispheres ipsilateral and contralateral to the tumor (p<0.05), compared with the corresponding values in the control group. Inversely, the mean PI decreased to 16% on both sides. Moreover, changes in MFV and PI in addition to those induced by glycerol were closely correlated with those of rCBF in all patients. In conclusion, these results suggest that changes in MFV and PI revealed by TCD under glycerol administration might predict intracranial pressure, especially in patients with intracranial hypertension.