PET imaging reveals septic hearts switch to glucose when pumping fails

When septic shock drives the heart into failure, the struggling muscle appears to change its fuel. A small prospective imaging study from Université de Sherbrooke in Quebec reports that patients whose hearts lost contractile function during septic shock showed markedly higher glucose uptake in the myocardium than septic shock patients whose hearts kept pumping normally, a pattern the investigators say has not been clearly documented in humans before.

The findings, published on 10 July 2026 in a Springer Nature collection on critical care, come from a pilot observational cohort study that used three positron emission tomography tracers in sequence to measure fatty acid metabolism, glucose handling, and oxygen consumption in the same beating heart. Twenty-four patients admitted to intensive care or acute coronary care underwent the imaging within roughly two days of admission, and the researchers argue the results lay the groundwork for a larger multicenter effort.

Sepsis-induced myocardial dysfunction, abbreviated SIMD, is a common and dangerous complication of septic shock. It is defined in this study as a new drop in left ventricular ejection fraction below 45 percent, or a fall of at least 20 percentage points from a prior echocardiogram, in a patient requiring vaso-inotropic support. Previous work suggests the condition affects roughly 30 to 60 percent of septic shock patients within the first 48 to 72 hours, and reduced contractility clusters with higher mortality. Yet the metabolic picture inside the septic heart has remained contested, with most evidence drawn from animal and cell models describing a shutdown of both fatty acid and glucose use, and a single invasive human study dating back nearly four decades.

To resolve that ambiguity, the team recruited patients within 48 hours of admission for circulatory shock or acute heart failure and assigned them to three groups of eight: septic shock with myocardial dysfunction, septic shock without dysfunction, and acute heart failure with reduced ejection fraction but no sepsis. Bedside echocardiography performed within 24 hours before the scan determined group allocation. The researchers deliberately avoided metabolic-interfering drugs such as propofol, etomidate, insulin, and dextrose infusions before imaging, pausing them for defined intervals so the tracer signals would reflect the patients’ real-time physiology rather than treatment effects.

Each patient received three tracers in a single session on a Siemens Biograph Vision 600 PET/CT scanner. Carbon-11 acetate traced myocardial blood flow and oxygen consumption, carbon-11 palmitate traced fatty acid uptake, oxidation, and esterification, and fluorine-18 fluorodeoxyglucose traced glucose uptake. Kinetic modeling converted the dynamic scans into quantitative rates, and the analyst segmenting the images was blinded to group assignment. The imaging protocol took about 90 minutes, a burden the authors acknowledge constrained recruitment.

The central result was stark. Median myocardial glucose uptake was 19.1 micromoles per 100 grams per minute in septic shock patients with myocardial dysfunction, compared with 1.3 in septic shock patients without dysfunction, a difference the authors report as statistically significant with a p value of 0.0071. Glucose uptake was not elevated in the non-septic heart failure group. Within the two septic groups, higher glucose uptake tracked with more severe declines in ejection fraction, with a Spearman correlation of minus 0.6756 across 16 patients. Notably, the association held even without correcting for blood glucose levels, and glucose uptake did not correlate with circulating glucose, insulin, or the vasoactive-inotropic score.

Fatty acid handling told a complementary story. Fatty acid uptake was lower in the dysfunctional septic hearts than in the well-functioning septic hearts, and among septic patients lower fatty acid uptake was associated with lower ejection fraction. Meanwhile, a larger share of the fatty acid entering the myocardium was being esterified, meaning routed into storage lipids rather than burned for energy, and this esterified fraction tended to rise as ejection fraction fell. The authors note that esterification and lipid accumulation have been linked to lipotoxicity in prior work, though this study cannot establish that mechanism directly.

Efficiency measurements added a functional dimension. Myocardial oxygen consumption was elevated in both septic groups, and myocardial external efficiency, the ratio of external cardiac work to total energy expenditure, tended to be lower in the dysfunctional group at a median of 10.3 percent versus 17.2 percent, a difference that did not reach statistical significance in this small sample. Across all septic patients, lower efficiency correlated with lower ejection fraction. The authors interpret this as an imbalance between cardiac work and energy consumption, and they caution that increased glucose uptake does not prove increased glucose oxidation, because the FDG tracer cannot distinguish whether glucose is oxidized, fermented to lactate, stored as glycogen, or diverted into lipid synthesis.

The study’s limitations are considerable and the authors state them plainly. It was a single-center pilot with eight patients per group, chosen pragmatically given patient instability and scanner logistics, and the analyses were explicitly exploratory and hypothesis-generating. Sepsis-induced myocardial dysfunction was defined solely by ejection fraction, a load-dependent measure, and loading conditions differed between septic and non-septic patients. There were no healthy controls, glycemia during imaging was imperfectly controlled, and the single time-point design precludes any conclusion about how the metabolic pattern evolves over the course of illness. The authors also cannot rule out a contribution from catecholamine support to the glucose uptake signal.

Even with those caveats, the team frames the work as a feasibility success and a challenge to prevailing assumptions. The prevailing preclinical narrative holds that the septic heart shuts down both fatty acid and glucose metabolism as a protective response; these human data instead suggest that when contractility fails, the heart reaches for glucose while shunting fatty acids toward storage. A planned follow-up multicenter study intends to replace FDG with carbon-11 glucose to determine whether the extra glucose is usefully oxidized or harmfully stored, and to track how myocardial metabolism changes over time. Until then, the authors suggest, the findings support interest in therapies targeting myocardial energetics in septic shock, and in hemodynamic drugs designed with mitochondrial properties in mind.

Subject of Research: Cardiac glucose and fatty acid metabolism in sepsis-induced myocardial dysfunction measured by three-tracer PET imaging

Article Title: Increased cardiac glucose uptake in sepsis-induced myocardial dysfunction: a three-tracer positron emission tomography imaging observational cohort study

Article References: Lesur, O., Molnar, T., Chagnon, F., Farand, P., & Croteau, É. (2026). Increased cardiac glucose uptake in sepsis-induced myocardial dysfunction: a three-tracer positron emission tomography imaging observational cohort study. Critical Care: Sepsis and Severe Infection, 1(1), Article 9. https://doi.org/10.1186/s44541-026-00015-5

Image Credits: AI Generated

DOI: 10.1186/s44541-026-00015-5

Keywords: septic shock, sepsis-induced myocardial dysfunction, PET imaging, cardiac metabolism, glucose uptake, fatty acid oxidation, myocardial efficiency, left ventricular ejection fraction, FDG, critical care, echocardiography, pilot study