Role of transcranial Doppler in cerebrovascular disease
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.190265
Source of Support: None, Conflict of Interest: None
Keywords: Cerebrovascular ischemia; emboli monitoring; transcranial Doppler; sonothrombolysis; vasomotor reactivity
Transcranial Doppler (TCD) ultrasonography is a bedside tool that provides reliable information about cerebral blood flow in real-time in a reliable manner. Some of the important clinical uses of this imaging modality are described in this article.
TCD plays an important role in establishing the presence, location, and severity of intracranial arterial occlusion in acute stroke. It helps in evaluating the hemodynamic consequences of an acute arterial occlusion, such as flow diversion, and compensatory flow increase in other intracranial arteries with reasonable accuracy. It plays an important role in the real-time monitoring of recanalization of arteries during intravenous thrombolysis. Therefore, it aids in identifying the non-responders to intravenous thrombolysis who might benefit from urgent endovascular therapy. TCD helps in determining the pathogenic mechanisms of stroke that, in turn, may help in selecting the most appropriate secondary stroke prevention treatment. Therapeutic ultrasound or sonothrombolysis remains a widely debated application of TCD.
In patients with established stroke, TCD helps in the diagnosis of steno-occlusive disease of intracranial and/or extracranial arteries. In addition, it plays an important role in risk stratification.
TCD is the only imaging modality that can detect an embolus and help in establishing the artery-to-artery (emboli in a relevant arterial segment distal to the stenoocclusive disease) or cardioembolic (emboli detected in multiple intracranial arteries) mechanism of acute stroke. In addition, paradoxical embolization (right-to-left shunt) can be diagnosed by TCD with nearly 100% sensitivity when compared to transesophageal echocardiography.
Fast-track transcranial Doppler insonation protocol in acute cerebral ischemia
“Time is brain” for a patient with acute ischemic stroke presenting to the emergency room within the window period when thrombolysis is feasible. Accordingly, the shorter the door-to-needle time, the better the outcomes. Therefore, a fast-track method of TCD insonation has been developed to save time and increase efficiency in the emergency settings. In addition, TCD, when performed by an experienced Neurosonologist, has a diagnostic yield comparable to catheter angiography. Furthermore, TCD has been validated against computed tomography angiography (CTA) with acceptable accuracy parameters.,
TCD should be considered to be an extension of the clinical examination. Hence, the clinical stroke localization should be an important consideration when performing a fast-track TCD in acute ischemic stroke. TCD should begin in the suspected uninvolved hemisphere (normal hemisphere) to get an idea of the normal arterial waveform pattern and velocities as well as the expected quality of the temporal acoustic window. One should then insonate the affected hemispherical side, starting at a mid-middle cerebral artery (MCA) depth of 50–58 mm. In patients with a normal MCA flow, distal MCA segments are assessed first, followed by the evaluation of internal carotid artery (ICA) bifurcation. If signals for MCA flow cannot be obtained, attempt may be made for insonation from the contralateral side across the midline. This is especially useful for patients with a poor temporal acoustic window on the involved side. Ophthalmic artery (OA) flow direction is assessed next at 50–58 mm depth followed by ICA siphon assessment at 60–64 mm. Basilar artery (BA) is insonated next to determine the presence of a compensatory flow velocity increase. Furthermore, it excludes a pre-existing steno-occlusive disease of the BA. Insonation of vertebral arteries (VA), OA, and ICA on the unaffected side and the posterior cerebral arteries may be performed later in a non-urgent manner. Using this method, TCD can be performed rapidly and parallel to the neurological and National Institutes of Health Stroke Scale (NIHSS) assessment and blood checks, avoiding any delay in the door-to-needle time.
For the detection of an offending lesion in acute ischemic strokes, TCD alone has sensitivity, specificity, positive, and negative predictive values of 96%, 75%, 96% and 75%, respectively. The additional advantages include detection of multiple intracranial stenosis, acute hemodynamic flow alterations and recruitment of collaterals, monitoring of the recanalization during intravenous thrombolysis; as well as real-time detection of re-occlusion (deterioration following improvement). On the other hand, the corresponding sensitivity, specificity, positive, and negative predictive values for carotid duplex sonography are 94%, 90%, 94%, and 90%, respectively. However, the combined use of TCD and carotid duplex scan in the emergency evaluation of acute ischemic stroke results in a better diagnostic yield as compared to the individual modalities alone. Chernyshev et al., used a fast-track TCD with the carotid duplex scan to identify lesions amenable to the interventional treatment (LAIT). LAIT was defined as occlusion, near-occlusion, stenosis of 50% or more, or thrombi in an artery supplying the area of ischemia. Presence of a pre-existing chronic and asymptomatic stenosis remains an important confounder for this emergency diagnostic approach, and the information obtained from clinical evaluation and other imaging modalities should be used wisely to overcome this limitation.
The diagnostic yield of TCD in acute ischemic stroke is particularly high when performed early after the symptom onset. It has a high sensitivity (>90%) and is best used for acute proximal MCA and terminal ICA occlusive disease. The high yield of TCD is mainly contributed by the observation that up to 90% of patients who are considered eligible for thrombolytic therapy and with NIHSS of >10 points show an acute intracranial occlusion., Therefore, TCD evidence of occlusion helps in confirming ischemia whereas a normal TCD could indicate a lacunar stroke or a non-stroke event such as complicated migraine. Demchuk et al., described the TCD diagnostic criteria and specific flow findings that characterized proximal intracranial occlusions.,
Transcranial Doppler monitoring during intravenous thrombolysis
Compared to contrast angiography, TCD demonstrated a very high sensitivity and specificity when persistent MCA occlusion and complete recanalization were assessed (91% sensitivity and 93% specificity). Accordingly, with a normal TCD, the chances of detecting an intracranial occlusion on contrast angiography are less than 5%. TCD is an important tool to witness the recanalization onset, duration, as well as its extent during intravenous thrombolysis. TCD is not very sensitive in the evaluation of posterior circulation steno-occlusive disease (with a sensitivity of 55–60%), especially when performed without contrast or transcranial color-coded duplex (TCCD) scans.
TCD may show no signal or minimal flow signals at and distal to the site of the intracranial occlusion and there may be beat-to-beat changes in flow signals due to the dynamic nature of the occlusion. i.e., clot dissolution, distal migration, and extension of the clot. The clinical recovery correlates well with the extent, timing, and completeness of arterial recanalization after thrombolysis. Tissue plasminogen activator (TPA) binding to the clot depends upon the actual area of the clot exposed to blood flow. TPA binding to the clot initiates recanalization by softening the clot, allowing the residual flow to improve and bringing more TPA to the site of the offending thrombus. The speed of clot lysis can be measured with real-time TCD monitoring with flow signal intensity, appearance of microembolic signals, flow pulsatility, and velocity changes. This cascading process facilitates more clot lysis with the help of arterial blood pulsations.
The beginning, speed, timing, and extent of recanalization are important determinants of stroke thrombolysis, and the following parameters are used in the assessment of recanalization:
Once the recanalization process begins, TCD can detect the attainment of the highest TIBI flow grade to determine the completion of recanalization. Arterial recanalization can be sudden or abrupt appearance of stenotic low-resistance or a normal signal; or, a stepwise (flow improvement over 1–29 min) or slow (flow improvement over 30–60 min) improvement in circulation. An example of thrombolysis-induced arterial recanalization and corresponding TCD changes are shown in [Figure 1].
The first noticeable improvement of flow in an intracranial artery occurs at a median time of 17 min after the TPA bolus  whereas the median time to completion of recanalization is 35 min. It is noted that those patients who achieve complete recanalization before the end of 1 h of TPA infusion are 3.5 times more likely to achieve a favourable outcome at 3 months. Furthermore, early recanalization is associated with early and dramatic recovery and an excellent long-term outcome. Labiche et al., noted that the clinical benefits of early (within 2 h) recanalization was sustained at 3 months after stroke. The likelihood of early complete recanalization with systemic TPA in the cases with M2 occlusion is around 44%. The corresponding rates for M1 MCA occlusion, tandem MCA–ICA occlusion, and terminal ICA occlusion are 30%, 27%, and less than 6%, respectively. Compared to early recanalization, slow flow improvement or dampened TIBI flow signals are less favourable prognostic indicators. Furthermore, partial or no recanalization is often associated with persistent or worsening neurological deficits.
Patients with persisting proximal occlusions have only about 10% chance of complete recovery at 3 months. These are the subset of patients who might benefit from an early endovascular intervention and clot retrieval with or without intra-arterial TPA (as seen in MR CLEAN trial) following administration of systemic TPA.
Arterial reocclusion after partial or complete recanalization could be the clinical equivalent of deterioration following improvement (DFI), observed in 13% of patients in the the National Institute of Neurological Disorders (NINDS) study. Alexandrov et al., found that almost one-third of their patients with early recanalization experienced reocclusion within 2 h of intravenous thrombolysis. They found that two-thirds of patients with DFI had an early reocclusion, indicating that DFI is a clinical surrogate for an early arterial reocclusion. Interestingly, patients with an early reocclusion had better long-term outcomes than those with no recanalization.
Therapeutic transcranial Doppler
Systemic thrombolysis has a low rate of complete arterial recanalization in acute intracranial arterial occlusions. del Zoppo et al., showed that partial or full recanalization after systemic thrombolysis with duteplase occurred only in 26% of the patients. Furthermore, the observation that almost 50% of acute ischemic stroke patients are left with either moderate or severe disability despite systemic thrombolysis led to the search for supportive therapeutic approaches that could enhance the TPA-induced recanalization rates. Continuous exposure of the clot to TCD during thrombolysis is one such therapeutic option. Experiments involving cadaver skull demonstrated that 1-MHz TCD exposure of the clot worked synergistically with TPA and helped to recanalize 90% of the clots. Ultrasound waves create a small pressure gradient, allowing more TPA to bind to the clot to allow quicker recanalization without the untoward side-effects of the strong mechanical vibrations, as seen with KHz ultrasound.
Therapeutic TCD works by reversible disaggregation of fibrin threads, leading to the formation of small cavities in the thrombus, allowing for more transport of TPA and facilitating an increase in residual flow. Despite large attenuation of ultrasound signals by the skull bone, at least 10% of this energy is transmitted to the clot site, which is believed to be sufficient to enhance the TPA activity.
The preliminary observation of better recanalization rates with continuous TCD monitoring during intravenous thrombolysis led to a phase II randomized controlled CLOTBUST trial, where 126 intravenous TPA patients were randomized to continuous TCD monitoring or placebo (63 patients in each arm). Complete recanalization or dramatic recovery within 2-h after the administration of the TPA bolus occurred in 49% in the TCD group compared to 30% in the control group (P = 0.03). Only 4.8% of patients developed symptomatic intracerebral hemorrhage, demonstrating a therapeutic benefit without any increased risk. The CLOTBUST trial also showed that early recanalization led to a trend toward better clinical recovery. We demonstrated similar feasibility and efficacy of ultrasound-assisted thrombolysis in our local Asian patient population, as well in a meta-analysis of previous studies related to sonothrombolysis. Similarly, Eggers et al., noticed that recanalization was more frequent while conducting sonothrombolysis with TCCD ultrasonography as compared to the controls. In a systematic review of randomized controlled trials involving sonothrombolysis (where both TCD and TCCD were used to augment thrombolysis), sonothrombolysis was found to be safe and effective in increasing the rates of recanalization. However, a recent phase 3 multi-centre, randomized controlled double blind trial for emergent revascularization in acute ischemic stroke, called the CLOTBUST-ER, was stopped due to lack of efficacy. The final results and reasons for the futility of the procedure are awaited. Perhaps, more randomized controlled trials with better designs are needed to prove the benefit of sonothrombolysis.
The major limitation with TCD is that it is operator-dependant and requires adequate skill, experience, and knowledge of arterial anatomy. Another important limitation is an insufficient temporal acoustic window, observed in approximately 10–15% patients. Interestingly, African-Americans and Asians (considered to have a higher prevalence of intracranial stenosis) have unfavourable temporal bone characteristics, which are responsible for the insufficient acoustic windows. However, incorporation of power motion-mode (PMD) in the recent TCD systems enables multigated flow information, thus allowing simultaneous insonation of multiple vessels. PMD-TCD reduces operator dependence and improves the detection of the best acoustic window.
The prevalence of asymptomatic intracranial stenosis has been reported to be between 5.9–24.5%. It is more common than extracranial stenosis in Hispanics and Africans. Intracranial atherosclerotic disease is a major cause of ischemic stroke, being responsible for up to 10% of transient ischemic attacks and strokes in Caucasians and a higher percentage of strokes in Asians (China 33–50%, Korea 28–60%, Thailand and Singapore 47–48%).
The current data suggests that the prognosis becomes worse with increasing degree of stenosis and with increasing number of affected vessels. It has been claimed by some to be the most common cause of stroke world-wide. The presence of intracranial atherosclerosis is associated with an increased risk of recurrent stroke ranging from 10 to 50% per year.
Intracranial stenoocclusive disease causes focal velocity increase, collateral flow patterns, and a post-stenotic decrease in velocity. Focal velocity increase remains the main diagnostic criterion for the diagnosis of moderate intracranial stenosis. At present, multiple diagnostic criteria are employed for the diagnosis of moderate intracranial stenosis [Table 1]. However, the increasing length of stenosis and multiplicity of stenotic lesions influence the flow velocities adversely. Accordingly, in such lesions, the flow velocity can decrease paradoxically. In the patients with multiple tandem lesions and diffuse intracranial stenotic disease, TCD shows a low mean flow velocity in the presence of a high pulsatility index (PI), suggestive of diffuse intracranial atherosclerosis [Figure 2]. Involvement of multiple intracranial arteries tends to produce TCD findings of a diffuse intracranial disease.
TCD can reliably detect stenosis and/or occlusion of the proximal M1 segment of MCA, ICA siphon, intracranial vertebral artery, proximal basilar artery, and proximal (P1) segment of the posterior cerebral artery. TCD has a higher sensitivity, specificity, and positive and negative predictive value in detection of anterior circulation lesions when compared to the posterior circulation ones. Perhaps, this occurs because of a more reliable anatomy in the anterior circulation and wider anatomical variations in the vertebro-basilar anatomy.
TCD is more reliable in ruling out intracranial stenosis. The “Stroke Outcomes and Neuroimaging of Intracranial Atherosclerosis” (SONIA) trial aimed to define the sensitivity, specificity, and positive (PPV) and negative predictive value (NPV) of TCD/magnetic resonance angiography for the identification of 50 to 99% intracranial stenosis in the intracranial ICA, MCA, VA, and BA. The following cut-offs for the mean flow velocities (MFV) were used for the identification of ≥50% stenosis: MFV MCA > 100 cm/s, TICA/ACA > 90 cm/s, VA/BA/PCA > 80 cm/s. The accuracy parameters of TCD (SONIA MFV cutoffs) against catheter angiography for ≥50% stenosis were: sensitivity 89%, specificity 99%, PPV 93%, NPV 98%), and overall accuracy 97%. Use of stenotic-to-prestenotic ratio of >2 increased the PPV further. The optimal mean flow velocity (MFV) cut-off for the detection of ≥70% stenosis was 128 cm/s (sensitivity 78%, specificity 96%) in the anterior circulation and 119 cm/s (sensitivity 100% and specificity 99%) in the posterior circulation. Another study by Wong et al., proposed an elaborate scoring system incorporating velocity scale (score 0–6 for peak velocity), hemodynamic scale (score 0–5 for diffuse or focal velocity increase; score 0–6 for differences between bilateral MCA; and score 17 for blunted flow), and spectrum scale (score 0–2 for normal spectrum, turbulence, and musical murmurs). This new method was more reliable and demonstrated better PPV than the traditional criteria. We suggest that a neurosonology laboratory should use one or a combination of the established diagnostic criteria and validate their findings against any form of angiographic study (preferably a contrast study) to determine the most reliable criteria applicable to their laboratory.
In conclusion, TCD provides important information and serves as a rapid and reliable screening tool for the detection of an arterial occlusion, for monitoring of arterial patency during thrombolysis, as well as for the assessment of cerebral hemodynamic alterations in patients with acute ischemic stroke. In patients with established intracranial stenosis, TCD aids in the diagnosis as well as the assessment of its severity. Emboli monitoring by TCD in patients with acute cerebral ischemia as well as chronic intracranial stenosis is an important tool for risk stratification as well as optimization of anti-thrombotic therapy.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest
[Figure 1], [Figure 2]