Robotic Transcranial Doppler Improves PFO Detection After Stroke

The use of a robotic-assisted device to perform transcranial Doppler ultrasound greatly improved the detection of right-to-left cardiac shunts in patients with presumed embolic strokes compared with the current standard of care — transthoracic echocardiography, in a new study.

And being far easier to perform than regular transcranial Doppler ultrasound, it’s hoped that use of the robotic device will enable many more patients to undergo the more sensitive transcranial screening modality and increase the number of shunts identified.

“I believe robot-assisted transcranial Doppler ultrasound can fill the gap between the rate of shunt detection we get with current standard of care, namely transthoracic echocardiography, and what a 30-plus year published history tells us we could be achieving with transcranial Doppler as the front-line screen,” lead author Mark Rubin, MD, told theheart.org | Medscape Cardiology.  

Rubin, who is assistant professor of neurology at University of Tennessee Health Science Center, Memphis, presented results of the BUBL study at last week’s International Stroke Conference (ISC) 2022, where they were greeted with applause from the floor.

Rubin explained that patients with suspected embolic stroke are routinely screened for right to left shunts, particularly those in the heart such as patent foramen ovale (PFO), that allow blood to flow from the right chamber to the left chamber and can lead to clots from the venous system accessing the arterial system, and then traveling to the brain and causing an ischemic stroke.

The current standard of care in screening for such shunts is the use of transthoracic echocardiography (TTE), a widely available and easy to perform, non-invasive procedure. “But we have known for decades that TTE does not pick up these shunts very well. With a sensitivity of only around 45% for PFO, it identifies less than half of PFOs when they’re actually present,” Rubin noted.  

The more sensitive transesophageal echocardiography (TEE) gives much better results, and is the “gold standard” for PFO diagnosis, but it is an invasive and unpleasant procedure with the ultrasound probe being passed down the throat, and the patient needing to be sedated, so it’s not appropriate for everyone, he noted.

Transcranial Doppler ultrasound (TCD) also gives excellent results, with a sensitivity of about 96% for detecting PFO, but this procedure is relatively difficult to perform, requiring a great deal of skill in placing the probes in the right position and interpreting the signal, Rubin said. TCD has been around for decades, but expertise remains scarce and stroke centers offering the diagnostic similarly rare,” he added.

“The robotic-assisted transcranial Doppler device may give us the opportunity to achieve the sensitivity of TCD without needing scarce expert operators. This should improve accessibility to this diagnostic technique,” he said.

“With such technology we can make significant strides into more accurate diagnoses on the cause of stroke, which should lead to better preventive treatments in those found to have right-to-left shunts,” he added.  

For the BUBL study, the robotic TCD technique was compared with the standard TTE in 129 patients who had a diagnosis of presumed embolic stroke or transient ischemic attack (TIA), with all patients undergoing both procedures.  

The robotic TCD device resembles a giant pair of headphones containing the ultrasound probes, which are attached to a frame. In the study, it was operated by a healthcare professional without TCD skills. Each ultrasound probe independently scans the temporal area autonomously — with angling and positive pressure against the scalp akin to a sonographer — to find and optimize bilateral middle cerebral artery signals, Rubin explained.

The primary endpoint was the detection of a right-to-left shunt. This occurred in 82 of the 129 patients (63.6%) with the robotic TCD device, but in only 27 patients (20.9%) when TTE was used. This gives an absolute difference of 42.6% (95% CI, 28.6% – 56.7%; P <.001 which rubin described as>

However, he said he was not surprised by these results.

“In my experience with transcranial Doppler, I find shunts in patients every day that have not been seen with transthoracic echo,” he commented.

He noted that a previous meta-analysis has suggested a similar difference between TCD and transthoracic echo, but the current study provides prospectively collected data produced in a clinical trial setting and is therefore more reliable. 

“What I hope comes from this is that more stroke patients will be able to undergo transcranial Doppler, which is a far superior screening technique for identifying right-to-left shunts,” Rubin said.

He explained that finding a right-to-left shunt in stroke patients is particularly important as it can direct treatment strategies to reduce future risk for recurrent strokes. For example, certain patients with a large PFO — especially those that are young with no other stroke risk factors — may benefit from PFO closure in addition to medicine for stroke risk reduction. If the doctors never see the PFO, they cannot make that recommendation, and a patient may be left with a less-than-best stroke risk reduction strategy.

In this study, the robotic-assisted TCD detected three times as many large shunts that were considered “intervenable” compared with transthoracic echo, identifying these shunts in 35 patients (27%) compared to just 13 (10%) with TTE.

“Of the 35 patients with potentially intervenable shunts detected with robotic transcranial Doppler, TTE was completely negative in 18 of them, and only suggested a small shunt in the others. So, plainly stated, the standard of care (TTE) missed or underestimated two-thirds of the large shunts detected by TCD, and completely missed half of the patients with large shunts,” Rubin reported.

Study Should “Dramatically Change” Practice

Commenting on the study for theheart.org | Medscape Cardiology, Patrick Lyden, MD, professor of physiology and neuroscience and neurology, University of Southern California, Los Angeles, said: “Most clinicians hesitate to use transcranial Doppler given the need for specialized technical expertise to obtain a reliable result. This study showed that a robotic transcranial Doppler device — which can be applied by any cardiac non-invasive lab technician — provides reliable and rigorous data.”

He added: “This result will dramatically change the typical evaluation of patients with suspected PFO: in place of an invasive transesophageal echo that requires anesthesia and a cardiologist, most patients can have a non-invasive, robotic-guided transcranial Doppler and get the same diagnostic benefit.”

Lyden also pointed out that the cost of TCD is typically one tenth that of TEE, although he said the cost of the robotic guided TCD “is not clear.”

A representative of the company that makes the robotic assisted device, NovaSignal, says the cost of the equipment is approximately $250,000, but “understanding the importance of the technology, we work with each hospital to meet their unique needs.”

The company adds that it currently has “over 45 commercial solutions deployed across 25 centers with 3-4 times growth expected year over year.”

The study was supported by NovaSignal, the company which makes the robotic device. Rubin reports acting as a consultant for the NovaSignal.

International Stroke Conference 2022: Abstract LBS19. Presented February 11, 2022.

For more from theheart.org | Medscape Cardiology, join us on Twitter and Facebook