Diaphragmatic attenuation is a common imaging artifact that occurs during nuclear heart scans, where the diaphragm muscle sitting beneath the heart absorbs some of the gamma rays emitted by the radioactive tracer before they reach the camera. The result is a falsely reduced signal along the bottom (inferior) wall of the heart, which can look disturbingly like a real perfusion defect caused by blocked coronary arteries. This artifact has been a recognized headache in cardiac imaging for decades and remains one of the most frequent causes of false-positive results on myocardial perfusion studies, particularly in men and in people with larger body habitus.
How the Artifact Forms
During a myocardial perfusion scan, a small amount of radioactive tracer is injected into your bloodstream and taken up by heart muscle cells in proportion to blood flow. A specialized camera (called a gamma camera) then rotates around your chest, detecting the gamma rays the tracer emits. The camera uses those signals to build a three-dimensional map of blood flow through your heart. In a perfect world, the gamma rays would travel unimpeded from your heart to the camera. In the real world, they have to pass through layers of tissue, and some get absorbed along the way.
The diaphragm is a thick, dome-shaped muscle that separates the chest from the abdomen. It sits directly below the heart, and when you lie on your back for a scan, the diaphragm rises upward, pressing closer to the inferior wall of the heart. Gamma rays trying to escape from the bottom of the heart have to pass through this muscle, and a proportion of them get scattered or absorbed before reaching the detector. The camera registers fewer counts from that region, which it interprets as reduced blood flow. The physics are straightforward: more tissue between the source and the detector means more signal loss.
What makes the problem worse is that the diaphragm is not the only culprit. The stomach, liver, and intestines sit just below the diaphragm, and these organs can contribute their own interference. One research group pointed out that much of what gets labeled “diaphragmatic” attenuation is actually gastric in origin, since the stomach lies in direct contact with the inferior heart wall when you are supine, and its contents and surrounding tissue absorb additional photons.1PubMed. The way to a man’s heart is through his stomach: much ‘diaphragmatic’ attenuation is likely gastric, and effervescent granules enhance cardiac imaging Subdiaphragmatic organs can also take up the radiotracer themselves, creating hot spots of activity in the liver or gut that further distort the inferior wall signal.2PubMed Central. Subdiaphragmatic activity-related artifacts in myocardial perfusion scintigraphy
Why It Affects Some People More Than Others
The artifact does not hit everyone equally. Men are more susceptible than women because of body composition differences. Men tend to carry more abdominal tissue and have a higher, more prominent diaphragm when lying down. Women, meanwhile, are more prone to a different attenuation artifact caused by breast tissue overlying the front of the heart. These sex-based patterns have been recognized for years as major factors that reduce the accuracy of perfusion imaging.3The American Journal of Cardiology. Effect of patient obesity on the accuracy of thallium-201 myocardial perfusion imaging
Body weight matters too. In people with obesity, the diaphragm is pushed higher by increased abdominal pressure, and there is simply more soft tissue for gamma rays to traverse. A study using hybrid SPECT/CT found that patients with a BMI of 30 or above showed significantly greater amounts of inferior wall attenuation, both visually and by quantitative measurement, compared to leaner patients.4PubMed Central. The Value of Attenuation Correction in Hybrid Cardiac SPECT/CT on Inferior Wall According to Body Mass Index This is one reason why interpreting nuclear heart scans in heavier patients requires extra care.
The Clinical Problem It Creates
The real danger of diaphragmatic attenuation is that it mimics coronary artery disease. On a perfusion map, an artifact in the inferior wall looks the same as a genuine blood flow deficit caused by a narrowed right coronary artery. If the reading physician cannot distinguish between the two, the patient might be sent for an unnecessary cardiac catheterization, an invasive procedure that carries its own risks. In one study evaluating inferior wall abnormalities, even after prone imaging was used to reduce false positives, a small percentage of patients with normal coronary arteries on angiography still had findings ultimately attributed to diaphragmatic attenuation.5Arquivos Brasileiros de Cardiologia. The Benefits of Prone SPECT Myocardial Perfusion Imaging in Reducing Both Artifact Defects and Patient Radiation Exposure
The artifact also complicates things in the opposite direction. If a physician sees an inferior wall defect and dismisses it as “just attenuation,” they could miss real disease. This is the fundamental tension: being too aggressive leads to unnecessary invasive testing, while being too conservative risks overlooking genuine ischemia. A fixed defect (one that doesn’t change between stress and rest images) is somewhat more likely to be artifact, while a reversible defect (present at stress but not at rest) is more suggestive of real ischemia. But the overlap between these patterns is frustratingly common, which is why the imaging community has invested so much effort in correction techniques.
The Breathing Problem Within the Problem
Diaphragmatic attenuation gets more complicated once you factor in breathing. During a scan that can last several minutes, the diaphragm moves up and down with every breath. This motion blurs the boundary between the heart and the diaphragm, creating a second layer of artifact on top of the static tissue absorption. Research using mechanical phantoms that simulated breathing found that the motion artifact becomes clearly visible at about 20 millimeters of diaphragmatic excursion and severe at 30 millimeters.6Journal of Nuclear Medicine. Effect of Mechanically Simulated Diaphragmatic Respiratory Motion on Myocardial SPECT Processed With and Without Attenuation Correction The pattern of the artifact also depends on how someone breathes: if the diaphragm spends more time in its raised (exhaled) position, the inferior wall takes the worst hit; if it spends more time low (inhaled), the anterior wall suffers instead.
Here is the frustrating part: standard attenuation correction, which works well for the static tissue-absorption component, does essentially nothing for the respiratory motion component. The same phantom study confirmed that attenuation correction fully addressed the static absorption of inferior wall counts but had no visible effect on the motion-related artifact.7Journal of Nuclear Medicine. Effect of Mechanically Simulated Diaphragmatic Respiratory Motion on Myocardial SPECT Processed With and Without Attenuation Correction When subdiaphragmatic tracer activity from the liver or gut is also present, the combination of breathing motion and hot organ overlap creates a messy mixture of cold and hot artifacts that attenuation correction cannot untangle.8PubMed. Contributions of subdiaphragmatic activity, attenuation, and diaphragmatic motion to inferior wall artifact in attenuation-corrected Tc-99m myocardial perfusion SPECT
Why the Tracer You Get Matters
Not all radiotracers are equally vulnerable to attenuation. Thallium-201, one of the original tracers used for perfusion imaging, emits relatively low-energy gamma rays, which makes it especially prone to tissue absorption. Technetium-99m-based agents (like sestamibi) emit higher-energy gamma rays that pass through soft tissue more easily. Early comparisons found that switching from thallium to sestamibi could dramatically reduce inferior wall artifacts: in one small study, inferior wall perfusion defects seen with thallium appeared in only one out of thirteen cases when the same patients were imaged with sestamibi.9PubMed. 99mTc-sestamibi can improve the inferior attenuation of TL-201 myocardial spect imaging
That said, the improvement is not as large as the energy difference between the two tracers might suggest. A study in women found that the higher energy of technetium-99m reduced tissue attenuation by only about 15% compared to thallium-201.10Journal of the American College of Cardiology. Comparative Diagnostic Accuracy of Tl-201 and Tc-99m Sestamibi SPECT Imaging (Perfusion and ECG-Gated SPECT) in Detecting Coronary Artery Disease in Women Phantom experiments have also shown that extracardiac tracer activity in the gut can actually push the inferior-to-anterior count ratio upward, partially masking the attenuation deficit; this effect was more pronounced with thallium, which means thallium images sometimes look paradoxically more normal in the inferior wall when there is a lot of gut activity, muddying the picture further.11PubMed. Attenuation-corrected rest thallium-201/stress technetium 99m sestamibi myocardial SPECT in normals The takeaway is that technetium-based tracers help, but they do not eliminate the problem.
Prone Imaging and Other Positioning Tricks
One of the simplest and oldest workarounds is to flip the patient over. When you lie face down (prone), the diaphragm drops away from the heart under gravity, and abdominal contents shift forward. This repositioning reduces the amount of tissue between the inferior wall and the camera, often making diaphragmatic attenuation artifacts disappear. Studies have confirmed that prone imaging lowers false-positive rates for inferior wall abnormalities.12PubMed. Diagnostic accuracy of supine and prone thallium-201 stress myocardial perfusion single-photon emission computed tomography to detect coronary artery disease in inferior wall of left ventricle
The typical approach is to acquire images in both supine and prone positions and compare them. If an inferior wall defect appears on the supine images but resolves when the patient is prone, it is almost certainly artifact. If the defect persists in both positions, it is much more likely to represent real disease. Prone imaging does have its own drawbacks: it can be uncomfortable for some patients, it introduces new artifacts along the anterior wall (since the chest is now compressed against the table), and it adds time to an already lengthy scan.
More creative positioning has also been tested. Researchers tried a forward-leaning “biker” position on a newer cadmium-zinc-telluride (CZT) camera and found it significantly reduced the number of heart segments showing attenuation artifact compared to both supine and sitting positions.13Journal of Nuclear Medicine. Assessment of Myocardial CZT SPECT Recording in a Forward-Leaning Bikerlike Position The forward lean tips the diaphragm and abdominal organs away from the heart, mimicking some of the benefit of prone positioning while keeping the patient upright.
Hardware-Based Attenuation Correction
The more technologically ambitious fix is to measure the attenuation directly and mathematically correct for it. This is done by acquiring a separate “transmission” scan of the chest, which maps out how much tissue is in the way at each angle around the body. Older systems used radioactive line sources (typically gadolinium-153) mounted on the camera; newer hybrid SPECT/CT systems use a low-dose CT scan for the same purpose. Either way, the attenuation map is fed into the image reconstruction algorithm, which adjusts the perfusion data accordingly.
Both methods work. Gadolinium-153 and low-dose CT attenuation correction produce broadly similar results in phantom testing, though head-to-head patient comparisons have shown that the two systems can sometimes create meaningful differences in the corrected images.14Journal of Nuclear Cardiology. Attenuation correction in myocardial SPECT using gadolinium-153 line sources or low-dose CT: Comparison of phantom and patient examinations One study evaluating gadolinium-153-based correction found that it raised specificity from about 69% to 84% in one patient group and from 91% to 100% in another, while also pushing sensitivity higher.15PubMed. Attenuation-corrected thallium-201 single-photon emission tomography using a gadolinium-153 moving line source: clinical value and the impact of attenuation correction on the extent and severity of perfusion abnormalities Comparing attenuation-corrected SPECT to PET (which is inherently attenuation-corrected and considered the gold standard) confirmed that the biggest improvement from correction occurred in the inferior wall, right where diaphragmatic attenuation lives.16Journal of Nuclear Medicine. Attenuation Correction of Myocardial SPECT Perfusion Images with Low-Dose CT: Evaluation of the Method by Comparison with Perfusion PET
Hybrid SPECT/CT attenuation correction also helps in heavier patients. The study mentioned earlier found that CT-based correction significantly reduced inferior wall defects in both obese and non-obese patients, with statistically significant improvement in both groups.4PubMed Central. The Value of Attenuation Correction in Hybrid Cardiac SPECT/CT on Inferior Wall According to Body Mass Index Still, attenuation correction is not perfect. If the CT and SPECT images do not line up precisely (a problem called misregistration, often caused by patient movement between the two scans), the correction can introduce new artifacts. And as noted above, correction addresses tissue absorption but not respiratory motion.
How Newer Cameras Shift the Pattern
The latest generation of gamma cameras uses CZT semiconductor detectors instead of traditional sodium iodide crystals. These cameras are faster, more sensitive, and allow new patient geometries, but they also change the shape of attenuation artifacts in ways that readers of older textbooks might not expect. A direct comparison between CZT and conventional cameras found that the CZT system shifted the attenuation artifact from the inferolateral wall to the lateral wall and produced a larger artifact area, though the artifact was shallower (less severe at any given point).17PubMed Central. Differences in attenuation pattern in myocardial SPECT between CZT and conventional gamma cameras The practical implication is that physicians experienced with conventional cameras need to recalibrate their expectations when reading CZT images. A defect in the lateral wall on a CZT scan may be artifact where the same location on a conventional scan would have raised less suspicion.
Deep Learning Without the CT
Many nuclear cardiology labs, especially smaller or older facilities, do not have hybrid SPECT/CT systems. They use standalone SPECT cameras with no way to acquire a CT attenuation map. This is where deep learning has made surprising progress. Researchers have trained neural networks to look at uncorrected SPECT images and predict what the attenuation-corrected versions would look like, essentially generating a virtual correction without any CT hardware at all.
A feasibility study showed that this approach substantially reduced attenuation artifacts, bringing average segmental errors from about 6% down to under 1% compared to actual CT-based correction. The deep-learning-corrected images correlated much more closely with the CT-corrected reference than the uncorrected images did.18Journal of Nuclear Medicine. Direct Attenuation Correction Using Deep Learning for Cardiac SPECT: A Feasibility Study The authors noted that performance was not perfectly consistent across all patients, likely because different people have different amounts of attenuation and different tracer uptake patterns.
Since that initial work, the approach has been validated more rigorously. A multicenter study involving thousands of patients developed a model called DeepAC and tested it on external populations from dozens of sites. The deep-learning correction improved diagnostic accuracy for obstructive coronary artery disease compared to uncorrected images, and its performance was statistically similar to actual CT-based attenuation correction.19Journal of Nuclear Medicine. Deep Learning–Based Attenuation Correction Improves Diagnostic Accuracy of Cardiac SPECT A subsequent large validation confirmed that the model’s accuracy held up across a broad, real-world patient population.20PubMed Central. General Purpose Deep Learning Attenuation Correction Improves Diagnostic Accuracy of SPECT MPI: A Multicenter Study The correction runs in under a second, requires no additional hardware or radiation dose, and could be deployed as a software upgrade to existing cameras. For labs that cannot afford hybrid SPECT/CT systems, this is a meaningful advance.
What Patients Can Do Before the Scan
Some of the interference in the inferior wall comes not from the diaphragm itself but from radiotracer that accumulates in the liver, stomach, and intestines after injection. When these organs glow brightly on the scan, their signal spills into the adjacent heart wall and corrupts the image. Labs have long asked patients to eat a fatty snack after tracer injection to trigger gallbladder contraction, which helps flush tracer out of the liver and into the gut, where it moves further from the heart. This works, but it adds logistical hassle and is not always practical.
An alternative that has gained traction is simply drinking water. A randomized study found that carbonated water given before adenosine SPECT significantly reduced extracardiac activity interfering with the heart images.21PubMed. A randomized study of the effect of carbonated water prior to myocardial SPECT Another study comparing water to fatty meals found the two approaches performed similarly, with water being easier and cheaper to administer routinely.22Egyptian Journal of Radiology and Nuclear Medicine. The effect of fatty meal and water on interfering extracardiac activity in myocardial perfusion single-photon emission computed tomography If you are scheduled for a nuclear heart scan and the lab offers you water or a snack partway through, this is the reason. It is not about comfort; it is about image quality.
When Artifacts Lead to Repeat Scans
Subdiaphragmatic artifacts do not just affect diagnosis; they affect workflow. When a scan is compromised by intense liver or gut activity overlapping the heart, the reading physician often cannot make a confident call. The result is a repeat acquisition, which means more time on the camera table for the patient and reduced availability of the scanner for the next patient in line. In busy nuclear medicine departments, this is a real operational problem.2PubMed Central. Subdiaphragmatic activity-related artifacts in myocardial perfusion scintigraphy Strategies that reduce these artifacts at the front end, whether through patient preparation, positioning, or software correction, save time and resources beyond just improving diagnostic accuracy. For patients, fewer repeat acquisitions also means less total radiation exposure, which is a secondary but welcome benefit.