What Is a TCD Bubble Study and How Does It Work?

A TCD bubble study is a non-invasive ultrasound test that checks whether blood is crossing abnormally from the right side of your heart to the left, typically through a small hole called a patent foramen ovale (PFO). The technique works by injecting agitated saline, which creates tiny microbubbles, into an arm vein while an ultrasound probe on your temple monitors blood flow in a major brain artery. If any bubbles appear on the brain-side recording, it means they made it through a shunt that should not be letting them pass. Doctors order the test most often after an unexplained stroke in a younger person, but it also shows up in migraine workups, diving medicine, and follow-up after heart procedures.

How the Test Works

You lie down, and a sonographer places a small ultrasound probe against one or both temples, angled to pick up blood flow in the middle cerebral artery. A nurse or technician then connects two syringes to an IV in your arm: one filled with sterile saline (sometimes mixed with a small amount of your own blood) and the other with air. The two syringes are rapidly pushed back and forth through a stopcock to whip up a froth of microbubbles, then the mixture is injected quickly into the vein. Under normal circumstances, those bubbles travel to the right side of the heart and get filtered out by the lungs. They never reach the brain. But if there is an opening, such as a PFO, some bubbles slip through to the left side of the heart and travel up into the brain’s blood supply, where the Doppler probe picks them up as characteristic high-pitched “hits” on the audio signal and bright spikes on the recording screen.

The test is usually repeated several times. At least one injection happens while you are resting quietly, and at least one happens while you perform a Valsalva maneuver, which is bearing down as if straining during a bowel movement. The Valsalva temporarily raises pressure on the right side of the heart, which can push open a PFO flap that stays shut under normal resting conditions. Detection rates rise substantially with the Valsalva: one study found that a right-to-left shunt was detected in only about 36% of patients at rest but in 99% during the maneuver, compared with about 83% detected by a different echo technique during the same maneuver.¹

Why Timing the Valsalva Matters

Getting the Valsalva right is one of the trickiest parts of the test, and small differences in timing can change the result. One well-known protocol comparison found that the most accurate approach was to start the Valsalva five seconds after the contrast injection began, holding the strain for about five seconds.² A separate study compared injecting the contrast before versus during the Valsalva and found that mid-Valsalva injection produced higher detection rates and shorter times to the first bubble appearing on the recording.³ In practice, many labs now use a standardized protocol, but if you have had a negative test and your doctor still suspects a shunt, the timing of the Valsalva is one of the first things worth revisiting.

What the Contrast Is Made Of

The simplest version of the contrast is plain agitated saline mixed with air. But adding a small amount of the patient’s own blood to the saline makes the microbubbles more numerous and longer-lasting. An in vitro and in vivo study found that saline mixed with about 10% blood produced roughly eight times more microbubbles immediately after agitation than plain agitated saline, and those bubbles stayed stable for longer and maintained a more consistent size.⁴ A clinical comparison confirmed that adding blood led to higher detection rates: shunts were identified in about 54% of patients with the blood-saline mix, versus about 43% with saline alone.⁵ Labs that want the highest sensitivity tend to use the blood-saline mixture, though both approaches are considered acceptable.

Accuracy Compared With Other Tests

The reference standard for finding a PFO has traditionally been transesophageal echocardiography (TEE), where a probe is passed down the throat to get a close-up view of the heart’s septum. TEE can physically see the hole and watch bubbles cross it in real time. TCD cannot see the hole at all; it can only tell you that bubbles reached the brain. Despite that limitation, TCD performs remarkably well as a screening tool. A meta-analysis pooling data from prospective studies found that TCD had a weighted sensitivity of 97% and specificity of 93% for detecting right-to-left shunts, compared with TEE.⁶

A study that performed TCD and transthoracic echo (TTE, the standard chest-wall echo) simultaneously found that TCD consistently and accurately identified right-to-left shunts in all confirmed cases, while TTE missed several of them.⁷ That kind of head-to-head comparison is why many neurologists prefer TCD as the initial screening step. Where the test falls a little short is specificity: because it picks up any right-to-left passage of bubbles, it cannot distinguish a cardiac hole from a shunt located somewhere else in the body. One early comparison found that TCD with the Valsalva maneuver and coughing identified all shunts but had a specificity of only about 65% compared with TEE.⁸ A more recent study combining TCD with carotid ultrasound in young stroke patients reported sensitivity of 98% and a negative predictive value of about 97% for findings that would matter on TEE, though specificity was only around 48%.⁹ In practical terms, that means a negative TCD bubble study is very reassuring, but a positive one usually needs confirmation with another test.

How Results Are Graded

When bubbles do appear on the recording, the number of microembolic signals determines the grade of the shunt. Various grading scales exist, but most divide results into categories ranging from no bubbles (negative) through a small shower (a handful of signals) up to a “curtain” pattern where the bubbles are so numerous that individual hits can no longer be counted. A higher grade generally corresponds to a larger or more functionally open shunt. One single-center study in stroke patients found that patients with a bubble-study grade above grade I had dramatically higher odds of a confirmed PFO on further testing.¹⁰

Grading matters clinically because it influences treatment decisions. A massive curtain-pattern shunt in a young stroke patient strengthens the case for PFO closure, while a handful of bubbles at rest with no increase during the Valsalva may point to a shunt too small to have caused anything. However, one study comparing TCD grading with TEE descriptions found that when TEE reports used formal quantitative language, agreement with TCD was perfect, but when TEE reports relied on informal descriptions, agreement dropped sharply.¹¹ The practical takeaway is that TCD provides a reproducible numerical grade, which can be more consistent than the subjective impressions sometimes recorded during echo.

When the Bubbles Did Not Come Through the Heart

A positive TCD bubble study means bubbles crossed from the venous side to the arterial side somewhere in the body, but that somewhere is not always the heart. In a study of 502 ischemic stroke patients who underwent TCD with bubble study, about 48% had a positive result. Of those positive cases, roughly 59% had a confirmed cardiac shunt on echocardiography. But about a quarter had echocardiograms that showed no cardiac shunt at all. When investigators looked further into the discrepant cases, they found explanations that included a pulmonary arteriovenous malformation, an arteriovenous fistula with thrombus, a brain AVM, and shunting related to malignancy.¹² Pulmonary AVMs are the most commonly discussed extracardiac cause. These are abnormal connections between arteries and veins in the lung that let blood (and bubbles) bypass the lung’s filtering capillary bed. One clue that a shunt is pulmonary rather than cardiac is the timing: bubbles from a PFO typically appear within the first few heartbeats after injection, while bubbles from a pulmonary AVM tend to arrive several beats later because they have to travel through the lung’s blood vessels first. If the timing points toward an extracardiac source, a chest CT or pulmonary angiogram may be ordered.

Safety

The test is considered very safe. The microbubbles are tiny, dissolve quickly, and are cleared by the lungs in healthy circulation. A large international multicenter study specifically designed to measure the complication rate of TCD bubble studies found zero ischemic cerebrovascular events across their entire cohort, with a 95% confidence interval that the true risk was no higher than 0.6%.¹³ That safety profile held regardless of whether patients had underlying heart abnormalities. For most patients, the biggest discomfort is the IV placement and the brief awkwardness of performing the Valsalva maneuver on command.

There is one notable side effect that occurs in a small fraction of patients: migraine with aura can be triggered by the test itself. In a study of 159 migraine-with-aura patients undergoing contrast TCD, about 7.5% experienced a typical aura attack during or immediately after the procedure. Every one of those attacks occurred in patients who had a right-to-left shunt, and most occurred in patients with large shunts.¹⁴ The auras resolved without lasting harm, but if you have a history of migraine with aura, it is worth knowing that the test itself could bring one on.

The Migraine Connection

Researchers have been exploring whether PFO contributes to migraine with aura for years. The logic is that small venous emboli or vasoactive chemicals that would normally be filtered by the lungs might slip through a PFO and trigger cortical spreading depression, the wave of nerve activity behind aura. TCD bubble studies have been used extensively in this research as a screening tool. In one study of 135 migraine-with-aura patients, about 65% had a positive TCD bubble test, though the study found no significant difference in the clinical features of the migraine itself between those with and without a shunt.¹⁵ The high prevalence of positive tests in migraine patients compared with the general population (where PFO rates are typically estimated around 25%) keeps the research question alive, even though closure trials for migraine have had mixed results. If you are seeing a neurologist about frequent migraine with aura, you may be offered a TCD bubble study as part of the evaluation, particularly if there is also a concern about stroke risk.

Screening Divers for Decompression Risk

Scuba divers have a specific reason to care about right-to-left shunts. During ascent, dissolved nitrogen in the blood forms tiny gas bubbles. Normally these venous gas emboli travel to the lungs and are harmlessly exhaled. But if a diver has a PFO, those nitrogen bubbles can cross to the arterial side and reach the brain or spinal cord, causing neurological decompression sickness. A study examining risk stratification in divers found that those with a right-to-left shunt had a higher risk of neurological decompression sickness, and that the risk climbed with the shunt’s grade. The authors concluded that TCD is an appropriate screening method for divers.¹⁶ Many dive medicine specialists now recommend TCD bubble testing for divers who have experienced unexplained neurological symptoms after a dive, and some recommend it proactively for professional divers or those who plan to do deep or decompression-stop diving. Finding a large shunt does not necessarily mean someone must stop diving, but it typically leads to a conversation about conservative dive profiles and whether PFO closure is warranted.

Use in Children

TCD bubble studies are not just for adults. Children can have unexplained strokes too, and PFO is occasionally the culprit. The non-invasive nature of TCD makes it especially appealing in pediatric patients, who may not tolerate TEE (which requires sedation and a probe down the throat). One group evaluating PFO in children with unexplained ischemic events reported that contrast TCD with the Valsalva maneuver was a sensitive and non-invasive method for PFO detection, and found it particularly suitable for younger patients.¹⁷ The main practical challenge is coaching a child through the Valsalva maneuver, which requires enough cooperation to bear down effectively on cue. Some labs use playful instructions, like asking the child to blow up a balloon or push as if inflating something, to get an adequate strain.

Monitoring After PFO Closure

If a PFO is closed with a catheter-delivered device, doctors need a way to check whether the closure worked and whether any residual shunt remains. TCD bubble studies are the go-to follow-up tool. The test is typically repeated at around three months after the procedure. One study comparing two different closure devices found that severe residual shunt was present in 13% of patients who received one device type and 7% of those who received another.¹⁸ A separate study of a third device type found severe residual Valsalva shunt at three months in about 7% of all patients, with higher rates in patients who had more severe baseline shunting at rest before the closure.¹⁹ Because TCD gives a numerical grade, it is straightforward to compare pre-closure and post-closure results. A persistent large shunt after closure might prompt a repeat procedure or a switch to a different device.

The Temporal Window Problem

One limitation that occasionally comes up is the temporal bone window. The ultrasound beam has to pass through a thin area of the skull near the temple, and in some people this window is too thick or the angle is too unfavorable to get a usable signal. Rates of inadequate temporal windows vary by population, but they tend to be higher in older women and in certain ethnic groups. When the temporal window is insufficient, some labs attempt to monitor through alternative sites like the eye socket (using the ophthalmic artery) or the back of the skull (using the vertebral arteries through the foramen magnum). These alternatives can salvage a test that would otherwise be uninterpretable, though the standard middle cerebral artery approach remains the most validated.

Advantages Over Echo-Based Bubble Studies

You might wonder why not just do the bubble test with a regular echocardiogram, since many people are already getting an echo for other reasons. Transthoracic echo with agitated saline (TTE bubble study) is widely available and does not require any special equipment beyond a standard echo machine. The reason TCD is often preferred comes down to sensitivity. TCD monitors every bubble that reaches the brain circulation, which means it picks up shunts that are too small or too intermittent to be visualized on a chest-wall echo image. As a recent literature review put it, contrast TCD is a non-invasive, inexpensive, and accurate tool for the detection and grading of right-to-left shunts after stroke.²⁰ In many stroke centers, the standard workflow is to screen with TCD first: if it is negative, a PFO is effectively ruled out and the patient avoids the discomfort of TEE. If it is positive, TEE or transesophageal echocardiography may follow to visualize the anatomy before any treatment decision is made.

Artificial Intelligence and the Future of Bubble Counting

One of the more tedious parts of a TCD bubble study is counting the microembolic signals. At low grades, it is straightforward, but at higher grades the signals pile up and overlap, making manual counting slow and subjective. Researchers are now applying machine learning algorithms to automate this process. A recent review described how AI-based systems can detect microembolic signals with improved consistency, better rejection of artifacts that mimic real signals, and more advanced interpretation of the recordings.²¹ Robotic-assisted TCD devices are also being developed that can automatically find the temporal window and maintain probe position, which could reduce the operator dependence that currently limits the test’s availability in smaller hospitals. If these technologies mature, TCD bubble studies could become something that any emergency department can run around the clock without needing a dedicated neurosonography technician on site.