Hidden Heart Damage Revealed by PET Blood Flow Even When Standard Scans Look Normal

For decades, cardiologists have relied on a simple visual logic when reading a nuclear heart scan: if the perfusion images look evenly lit and the pumping chamber squeezes with normal force, the heart is presumed healthy. A new study from Kuwait challenges that assumption at its foundation. Researchers at Kuwait University and Kuwait Chest Disease Hospital analyzed 292 patients who underwent adenosine-stress rubidium-82 positron emission tomography, or PET, and found that patients with a documented history of cardiac disease had substantially lower absolute myocardial blood flow during stress even though their perfusion categories and ejection fractions appeared entirely normal. The finding, published in Frontiers in Nuclear Medicine, suggests that the two numbers clinicians most often trust may be quietly missing disease that a third, less commonly emphasized measurement can detect.

The technology behind the study is worth unpacking. Rubidium-82 is a radioactive tracer produced on demand from a strontium-82 generator, which means hospitals can perform quantitative cardiac PET without an on-site cyclotron. When the tracer is injected at rest and again during pharmacological stress induced by adenosine, dynamic imaging allows software to compute myocardial blood flow in absolute physiological units of milliliters per minute per gram of tissue. From rest and stress values, the myocardial flow reserve, or MFR, is derived as the ratio between them. Professional societies in nuclear cardiology now recommend incorporating these quantitative measures into routine reporting, yet in many laboratories the categorical perfusion read and the left ventricular ejection fraction remain the dominant basis for clinical decisions.

The Kuwaiti cohort is particularly interesting because of where it comes from. The Arabian Gulf carries one of the world’s heaviest burdens of diabetes and obesity; Kuwait’s age-standardized diabetes prevalence stood at 25.6 percent in 2024, and the average body mass index in this study cohort was about 31.5 kilograms per square meter, well into the obese range. Both diabetes and obesity impair the microscopic vessels of the heart and lower baseline blood flow. Existing reference values for rubidium-82 flow measurements derive almost entirely from North American and European populations, so the researchers set out to ask whether absolute flow behaves differently in a population that has been largely absent from the normative datasets.

The study’s design was deliberately conservative. From 330 consecutive scans performed between January 2018 and December 2023, the team deduplicated repeat studies and excluded scans with internally inconsistent flow values, yielding 292 source-verified patients with a mean age of 63.1 years. Every quantitative value in the analysis file was checked against the original clinical PET report. Within this cohort, 84 patients had what the researchers called normal-appearing studies: a C1 perfusion category, meaning visually normal perfusion with a flow reserve of at least 2.0, combined with a resting ejection fraction of 50 percent or higher. Of these, 38 had a documented history of cardiac disease and 46 had none. The primary analysis compared these two subgroups directly.

The results were striking. Patients with prior cardiac disease had a median global stress myocardial blood flow of 2.25 milliliters per minute per gram, compared with 2.99 in patients without cardiac history, a difference corresponding to a Cohen’s d effect size of negative 0.78, which is conventionally considered a large effect. The deficit was even larger in the territory supplied by the right coronary artery, where stress flow was 2.25 versus 3.29, an effect size of negative 1.02. Resting flow was also lower in the cardiac history group. Crucially, however, the myocardial flow reserve itself did not differ significantly between the groups, and neither did ejection fraction or left ventricular volumes. The association survived multivariable adjustment for age, sex and body mass index, with prior cardiac disease independently associated with roughly half a milliliter per minute per gram lower global stress flow.

The pattern of the deficit matters as much as its size. Because both rest and stress flow fell together while their ratio remained preserved, the signature does not match the classic picture of coronary microvascular dysfunction, in which resting flow is typically maintained but the reserve capacity to increase flow under stress is blunted. A proportional reduction in both rest and stress flow is what one would expect instead from a lower rate-pressure product, from medication effects such as beta-blockade or nitrates, or from a reduction in viable myocardial mass after infarction. The researchers were candid about this ambiguity: resting heart rate, blood pressure and medication use were not recorded in the imaging database, and the composite cardiac history variable could not be decomposed into its constituent diagnoses. They therefore explicitly inferred no mechanism.

A second, exploratory comparison underscored the power of quantitative flow measurement. When the 46 truly normal patients were compared with the remaining 246 classified as abnormal, absolute stress flow was the strongest single discriminator, with global stress flow achieving an area under the receiver operating characteristic curve of about 0.85 and flow reserve 0.81, both substantially outperforming ejection fraction, which reached only 0.74. The authors were careful to note that because the abnormal categories are defined in part by perfusion and flow reserve, this separation is partly circular and does not represent diagnostic accuracy. Still, the magnitude of the separation, and the fact that flow parameters beat ejection fraction significantly in paired statistical testing, reinforces the argument that quantitative reporting adds genuine information beyond the conventional read.

The study also illuminated a diagnostic gray zone that has long puzzled cardiologists. Forty-five patients showed the C3 pattern: normal perfusion on visual reading but reduced global flow reserve. This pattern can reflect either genuine microvascular dysfunction with clean epicardial arteries, or balanced three-vessel epicardial disease that lowers flow uniformly without creating a regional defect, and PET alone cannot distinguish the two. In this cohort, only four C3 patients underwent coronary angiography, and all four had obstructive epicardial disease, including two with balanced triple-vessel disease. Yet resting echocardiograms, available in 24 of the C3 patients, told a more nuanced story: seven showed regional wall-motion abnormalities suggesting prior infarction, while ten had explicitly preserved regional function, the pattern expected in true microvascular disease. The C3 category, in other words, appears to be a genuine mixture of both phenotypes in this population.

The limitations of the work are as instructive as its findings. Referral indication was almost perfectly confounded with cardiac history: 84 percent of the prior-disease group was referred for revascularization assessment, versus 7 percent of those without, and patients undergoing revascularization assessment are more likely to be on flow-lowering medications. The perfusion categories were assigned during routine clinical reporting without blinding, the subgroups were small, and no outcome data were collected, so the study cannot say whether reduced absolute stress flow predicts heart attacks or death. The authors frame their conclusions as hypothesis-generating and call for prospective, outcome-linked studies with recorded hemodynamics and medication data.

Even with those caveats, the message for cardiovascular medicine is resonant. A heart that pumps normally and lights up evenly on a perfusion scan may still be running on a reduced blood supply, and a single absolute number, stress myocardial blood flow, can expose that hidden deficit in a population where metabolic disease makes such deficits especially likely. As quantitative PET becomes more widely available and professional guidance increasingly endorses flow reporting, studies like this one from the Arabian Gulf suggest that the era of reading the heart by eye alone is drawing to a close, and that the numbers behind the images may hold the key to disease that has been hiding in plain sight.

Subject of Research: Quantitative rubidium-82 PET measurement of absolute myocardial blood flow in patients with prior cardiac disease and normal-appearing perfusion studies in an Arabian Gulf cohort

Article Title: Absolute stress myocardial blood flow is reduced in patients with prior cardiac disease despite normal perfusion and ejection fraction on rubidium-82 PET: an Arabian Gulf cohort

Article References: Alenezi, A., Garashi, M., Ghadenfer, L., Panchadar, S., Biswas, G., & Ali, K. (2026). Absolute stress myocardial blood flow is reduced in patients with prior cardiac disease despite normal perfusion and ejection fraction on rubidium-82 PET: an Arabian Gulf cohort. Frontiers in Nuclear Medicine, 6, Article 1957482. https://doi.org/10.3389/fnume.2026.1957482

Image Credits: AI Generated

DOI: 10.3389/fnume.2026.1957482

Keywords: rubidium-82 PET, myocardial blood flow, myocardial flow reserve, coronary microvascular dysfunction, cardiac PET, ejection fraction, myocardial perfusion imaging, Arabian Gulf, diabetes, coronary artery disease, nuclear cardiology, Kuwait