The clinical motivation is straightforward. Dysregulated brain metabolism is increasingly recognized as a central player in the development and progression of neurodegenerative diseases, not merely a byproduct of neuronal death. The workhorse technique for probing this metabolism in patients has been positron emission tomography, or PET, most famously with the glucose analog 18F-fluorodeoxyglucose. PET can show where the brain takes up a radioactive tracer, and reduced uptake in certain regions is a well-established signature of neurodegeneration. But PET has a fundamental blind spot: it reveals only the initial uptake of the tracer and offers no insight into its downstream metabolic fate. It cannot tell clinicians whether glucose that enters a neuron is being funneled into energy production, diverted into biosynthetic pathways, or left to accumulate. It also provides no window into dynamic metabolic fluxes and turnover, the actual rates of biochemical reactions that define how a cell is functioning.

MRS and MRSI approach the problem from an entirely different physical angle. Rather than detecting radioactive decay, these techniques exploit nuclear magnetic resonance, the same phenomenon underlying MRI, to identify specific chemical compounds based on the characteristic resonance frequencies of their atomic nuclei. The result is a non-invasive, spatially resolved, in vivo measurement of metabolite levels, achieved without any radioactive tracer. Proton MRS and MRSI, which detect hydrogen nuclei, are the most widely utilized variants because they can be implemented on standard clinical MRI scanners and require no special hardware beyond what most hospitals already possess. In neurodegenerative research, proton MRS has long been used to track markers such as N-acetyl-L-aspartate, a proxy for neuronal health, and the combined signal of glutamate and glutamine, which reflects excitatory neurotransmitter metabolism. These measurements, however, remain essentially equilibrium measurements: they report how much of each metabolite is present, not how quickly it is being made, consumed, or exchanged.

The review’s central conceptual contribution is a framework for moving beyond this equilibrium limitation toward dynamic metabolic imaging. In a dynamic experiment, a labeled substrate is introduced into the body and the scanner watches as the label is incorporated, step by step, into downstream metabolites. The rate at which labeled products appear encodes the flux through specific enzymatic pathways, such as the tricarboxylic acid cycle that powers oxidative phosphorylation in mitochondria. This is the kind of information that reveals metabolic reprogramming and metabolic bottlenecks, the mechanistic signatures that static concentration measurements cannot capture. The authors lay out the key considerations for designing such experiments, including the choice of tracer, the method of signal enhancement, the acquisition hardware, the image reconstruction and preprocessing pipeline, and the metabolic modeling needed to convert time-course data into meaningful biochemical rate constants.

Tracer selection is where the physics gets interesting. Proton MRSI can be extended to heteronuclear approaches that detect less abundant nuclei, including deuterium, carbon-13, nitrogen-15, and oxygen-17. Each of these broadens the scope of accessible metabolites and pathways. Deuterium, the heavy isotope of hydrogen, has emerged as a particularly attractive option. Deuterium-labeled glucose can be administered safely, and the resulting deuterium metabolic imaging, or DMI, tracks its conversion into products such as glutamate and lactate. The revival of deuterium MRSI, the review notes, has opened new avenues for dynamic metabolic investigation in the human brain without requiring hyperpolarization, making it one of the most practical routes to clinical translation. Because deuterium resonates at a frequency far from that of protons, the technique can be implemented with relatively modest hardware additions to existing scanners.

For metabolites present at very low concentrations, however, even heteronuclear detection struggles against the intrinsically poor sensitivity of magnetic resonance. This is where nuclear spin hyperpolarization enters the picture. Hyperpolarization techniques, including dissolution dynamic nuclear polarization and parahydrogen-based methods such as parahydrogen-induced polarization and signal amplification by reversible exchange, can transiently boost the nuclear polarization of a tracer by four to five orders of magnitude before it is injected. The result is a signal strong enough to detect low-abundance metabolites in vivo in real time, enabling hyperpolarized metabolic imaging of reactions that would otherwise be invisible. Dissolution DNP, for example, can hyperpolarize carbon-13-labeled pyruvate, allowing researchers to watch the enzyme lactate dehydrogenase convert it to lactate within seconds, a direct readout of a key step in cellular energy metabolism. The review discusses these hyperpolarization methods in detail, along with their technical demands, including the specialized dissolution equipment and rapid injection systems they require.

None of this is trivial to implement. The authors are candid about the major methodological and translational challenges standing between dynamic MRSI and routine clinical use. Signal-to-noise ratio remains the persistent adversary, since metabolic signals are orders of magnitude weaker than the water signal used in conventional MRI. Higher magnetic field strengths improve sensitivity and spectral resolution but bring their own problems, including increased specific absorption rate and greater hardware cost. Vendor hardware variability compounds the difficulty, as scanners from different manufacturers differ in gradient performance, coil configurations, and pulse sequence availability. Perhaps most consequentially, there is currently a lack of standardization across acquisition, reconstruction, preprocessing, and modeling strategies, which makes it difficult to compare results between sites or to pool data across studies. The review also points to the role of modern computational tools, including machine learning and deep learning approaches, in accelerating image reconstruction and improving the extraction of quantitative information from noisy spectroscopic data.

What could dynamic MRSI actually deliver for patients with neurodegenerative diseases? The authors argue that the technique provides a promising foundation for investigating brain metabolism in vivo, offering mechanistic insights into metabolic fluxes, reprogramming, and bottlenecks that could reshape biomarker development and diagnostic practice. Mitochondrial dysfunction is a recurring theme across neurodegeneration, and dynamic measurements of TCA cycle flux or oxidative phosphorylation could reveal disease mechanisms years before structural changes appear on conventional MRI. In disorders such as multiple system atrophy, progressive supranuclear palsy, and corticobasal degeneration, where distinguishing between overlapping clinical syndromes is notoriously difficult, metabolic fingerprints might aid patient stratification for clinical trials. Longitudinal disease monitoring is another compelling application: because MRSI involves no ionizing radiation, it can be repeated safely, making it well suited to tracking disease progression and treatment response over time in a way that repeated PET scans cannot easily match.

The translational path, however, requires honest acknowledgment of where the field stands. Dynamic MRSI is currently limited to a handful of research institutions with the specialized hardware, chemistry infrastructure, and analytical expertise to pull it off. Hyperpolarized tracers must be produced under good manufacturing practice conditions, and regulatory pathways under medical device regulation frameworks add further complexity. The authors emphasize that ongoing technical and methodological advances are steadily improving feasibility, biochemical specificity, and the prospects for broader implementation, but they are equally clear that standardized and harmonized acquisition, reconstruction, and preprocessing protocols are necessary before these approaches can achieve broad clinical implementation. Without such harmonization, the field risks fragmenting into incompatible local practices that cannot generate the large, comparable datasets needed for clinical validation.

The bigger picture is one of a field at an inflection point. The shift from static to dynamic MRSI, the review concludes, has great potential to advance biomarker and diagnostic tool development, patient stratification, and longitudinal disease monitoring in neurodegenerative disorders, provided that current challenges and limitations are sufficiently addressed. In an era when most experimental therapies for Alzheimer’s and Parkinson’s diseases have stumbled, partly because patients are enrolled too late in a disease process that begins decades before symptoms, a technology that can measure the metabolic fluxes of the living brain, non-invasively and repeatedly, offers exactly the kind of window that drug developers and neurologists have been seeking. The photograph is becoming a movie, and what it shows may change how neurodegeneration is understood.

Subject of Research: Dynamic magnetic resonance spectroscopic imaging of brain metabolism in neurodegenerative disorders

Article Title: From equilibrium to dynamic metabolic imaging: the evolving role of magnetic resonance spectroscopic imaging in neurodegenerative disorders

Article References: Pokotylo, M. M., Peters, J. P., Diercks, A., Mohamad, F. H., Brüggemann, N., Pravdivtsev, A. N., Hövener, J.-B., & Prasuhn, J. (2026). From equilibrium to dynamic metabolic imaging: the evolving role of magnetic resonance spectroscopic imaging in neurodegenerative disorders. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-09084-z

Image Credits: AI Generated

DOI: 10.1186/s12967-026-09084-z

Keywords: magnetic resonance spectroscopy, MRSI, hyperpolarization, deuterium metabolic imaging, neurodegenerative diseases, brain metabolism, PET, metabolic flux, Alzheimer's disease, Parkinson's disease, biomarkers, metabolic modeling

The strategy, described in the Journal of Translational Medicine, is what the investigators call reaction-based chemical shift engineering. Rather than trying to untangle overlapping resonances with ever more sophisticated pulse sequences or higher magnetic fields, the team sidestepped the problem entirely. They designed a molecular probe that reacts selectively with a chosen target — in their proof-of-concept study, the neurotransmitter norepinephrine — and converts it into a new chemical species whose protons resonate far outside the crowded endogenous window. The product of the reaction, 4-hydroxybutanal, carries an aldehyde proton that rings out at approximately 9.7 parts per million, a spectral neighborhood essentially deserted by the metabolites of living tissue and safely distant from the towering water signal that dominates every in vivo proton spectrum.

The choice of norepinephrine was no accident. This catecholamine neurotransmitter sits at the center of the brain’s arousal and reward circuitry, and disturbances in its signaling have been implicated in depression for decades — indeed, some of the most widely prescribed antidepressants, including fluoxetine, are thought to modulate noradrenergic tone among their downstream effects. Yet norepinephrine has been notoriously difficult to observe in a living brain. Optical probes cannot penetrate the skull, positron emission tomography tracers offer limited chemical specificity, and conventional spectroscopy cannot pick the molecule out of the dense thicket of overlapping signals. A probe that renders norepinephrine visible as a clean, isolated peak would therefore fill a genuine gap in both neuroscience research and translational drug development.

The probe itself, designated FS, works through a cascade of nucleophilic substitution reactions. When it encounters norepinephrine, the catecholamine’s amine and hydroxyl groups attack the probe in sequence, triggering a chemical rearrangement that liberates 4-hydroxybutanal as the detectable product. The team did not simply assert this mechanism; they supported it with theoretical simulations that mapped the reaction pathway and predicted the chemical shift of the resulting aldehyde proton. That computational grounding matters, because the entire strategy hinges on a quantitative prediction: that the product’s resonance would land at a frequency far enough from every endogenous signal to be resolved cleanly even at the moderate field strengths available in clinical and preclinical settings.

Validation proceeded in careful steps, from the test tube to the living brain. In aqueous solutions, the probe showed favorable selectivity for norepinephrine over a panel of structurally related catecholamines and other biological amines, along with robust resistance to interference and stability across the physiological pH range — properties that are essential for any reagent intended to function in the chemically noisy environment of cells and blood. The investigators then moved to PC12 cells, a widely used model line derived from rat adrenal tissue that differentiates into neuron-like cells and secretes catecholamines. Inside these cells, the probe successfully detected endogenous norepinephrine release, demonstrating that it could function not merely against purified standards but against the genuine molecular inventory of a living cell.

The decisive test came in vivo. Working on a 7.0 Tesla magnetic resonance system — a field strength common in preclinical research and increasingly in human imaging — the team administered the probe to live rats that had been treated with fluoxetine, a pharmacological manipulation expected to elevate norepinephrine signaling in the brain. The characteristic aldehyde resonance near 9.7 parts per million appeared in the spectra of the treated animals, providing what the researchers describe as in vivo monitoring of pharmacologically elevated norepinephrine. In other words, for the first time in this experimental context, a specific neurotransmitter could be tracked in a living brain as a single, interference-free spectroscopic peak, using the same fundamental physics that powers every clinical MRI scanner.

The technical significance of this achievement is worth unpacking for readers who do not live and breathe spectroscopy. Chemical shift — the tiny displacement of a nucleus’s resonance frequency caused by its electronic environment — is the fundamental currency of magnetic resonance spectroscopy. Aldehyde protons are among the most deshielded protons in organic chemistry, which is why they resonate at such high parts-per-million values. By designing a reaction that manufactures an aldehyde in situ, the team effectively hijacked the extreme end of the chemical shift scale as a private detection channel. No pulse-sequence trickery, no spectral deconvolution, no isotope labeling with its attendant cost and regulatory burden: just a targeted chemical transformation that moves the signal out of the crowd. The approach is, in principle, generalizable — any analyte for which a selective reaction can be devised that yields a product with a distinctive chemical shift becomes a candidate for the same treatment.

The translational implications stretch across several domains. In disease diagnosis, an interference-free window would allow clinicians to quantify specific biomarkers in the brain or in tumors with a specificity that conventional proton spectroscopy cannot deliver, potentially distinguishing pathological biochemistry from normal variation. In dynamic metabolic monitoring, the ability to follow a neurotransmitter’s fluctuations over time could illuminate how psychiatric medications actually reshape brain chemistry in individual patients. In preclinical drug evaluation, the method offers pharmaceutical researchers a non-invasive readout of target engagement in animal models — a way to confirm that an experimental compound genuinely changes the neurochemical landscape it is designed to change, without sacrificing the animals at each time point. The authors frame the work as establishing a generalizable strategy for expanding the metabolic detection scope of proton MRS, and the framing seems justified by the breadth of those applications.

Caveats remain, as they always do on the road from proof of concept to bedside. The study was conducted in rats at 7 Tesla with a probe administered exogenously, and the safety, pharmacokinetics, and clearance of such probes in humans would require extensive evaluation before any clinical deployment. The sensitivity of the method — how little norepinephrine can be detected, and how quickly — will determine whether it can capture the fast, transient dynamics of neurotransmission or only slower, tonic changes. And for each new target molecule, chemists will need to design a new probe with the right selectivity, kinetics, and biocompatibility, a nontrivial task that the elegant norepinephrine example does not automatically solve for every analyte. Still, the conceptual barrier that has constrained proton spectroscopy for half a century — the crowding of all endogenous signals into a few parts per million — has now been shown to be circumventable by design rather than by brute force.

What makes this work resonate beyond its immediate field is the way it reframes a classic instrumentation problem as a chemistry problem. For decades, the response to spectral congestion in magnetic resonance has been to build bigger magnets and cleverer sequences. This team’s answer was to reach into the molecule itself and rearrange it. If reaction-based chemical shift engineering proves as adaptable as its creators suggest, the crowded 1-to-5-ppm window that has defined the limits of proton spectroscopy may no longer be the whole story — and the quiet, empty region beyond 5 parts per million could become a busy new address for the molecules of the mind.

Subject of Research: Reaction-based chemical shift engineering for interference-free in vivo proton magnetic resonance spectroscopy of norepinephrine

Article Title: Reaction-based chemical shift engineering unlocks interference-free detection window for in vivo proton magnetic resonance spectroscopy

Article References: Lv, G., Gao, J., Qi, M., Meng, X., Tang, H., Han, F., & Zhang, J. (2026). Reaction-based chemical shift engineering unlocks interference-free detection window for in vivo proton magnetic resonance spectroscopy. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-08931-3

Image Credits: AI Generated

DOI: 10.1186/s12967-026-08931-3

Keywords: magnetic resonance spectroscopy, chemical shift, molecular probes, norepinephrine, depression, neurotransmitters, in vivo imaging, fluoxetine, preclinical imaging, molecular imaging, brain metabolism, translational medicine

The scientific backdrop is sobering. Neurodegenerative diseases now affect more than 57 million people worldwide, with Alzheimer’s disease and related dementias representing the most prevalent forms, and effective therapies remain scarce once pathological decline has taken hold. Prevention, or at least a meaningful delay of onset, has therefore become the central strategy. Projection models cited by the researchers suggest that delaying dementia onset by just five years could save more than half a million dollars per patient in the United States alone. Physical activity stands out as the one lifestyle factor consistently included in brain health recommendations across major health organizations, with meta-analytic evidence indicating that high levels of activity reduce the risk of cognitive decline by roughly 38 percent compared with sedentary behavior. Yet most of this evidence lumps all movement together, leaving open a question the Swiss team set out to answer: which specific exercise modalities act on which dimensions of brain health, and through what mechanisms?

To probe that question, the researchers recruited 40 community-dwelling volunteers aged 64 to 81 and randomly assigned them to a resistance training group or a control group instructed to maintain their habitual lifestyle. The training program was anything but gentle slow-motion lifting. Participants completed three supervised 45-minute sessions per week, targeting the lower body through exercises for ankle dorsiflexion, plantarflexion, knee extension, and hip extension, with loads progressively increased by adding weight cuffs, plates, and even jump movements. After an initial three weeks of traditional strength work, the program alternated explosive sessions, in which participants contracted as fast as possible against resistance, with conventional slow-tempo training. Attendance was high, averaging 89 percent of prescribed sessions, and no injuries occurred, an important demonstration that high-velocity resistance work is feasible and safe even in the eighth decade of life.

The study’s most technically ambitious component involved measuring brain chemistry directly. Using ultra-high-field 7 Tesla magnetic resonance spectroscopy, the team quantified concentrations of metabolites in the left sensorimotor cortex, a region chosen because prior work in rodents and humans links it to metabolic adaptations following resistance exercise. Two markers took center stage. The first was lactate, a molecule that accumulates in the aging brain as mitochondrial function falters and metabolism shifts from efficient oxidative phosphorylation toward glycolysis. Healthy older adults show higher resting lactate than younger adults, and patients with Alzheimer’s disease display even greater concentrations, making lactate a candidate biomarker of compromised neural energy metabolism. The second marker was N-acetylaspartate, or NAA, one of the most abundant metabolites in the brain, which declines with age and falls further in dementia, serving as an index of neuronal integrity.

After twelve weeks, the spectroscopy data told a striking story. Participants in the training group reduced their resting sensorimotor lactate concentration by an average of 0.36 millimoles per liter, a statistically significant change with a moderate-to-large effect size, while the control group showed no meaningful change. The sex-adjusted between-group difference in lactate change was 0.45 millimoles per liter. The authors interpret this reduction as a possible sign of improved mitochondrial function, a return toward more efficient oxidative metabolism in a brain region where age-related metabolic decline typically advances. NAA, by contrast, did not change significantly; the training group showed a slight numerical increase while controls drifted downward, a pattern consistent with earlier work suggesting resistance training may protect neuronal integrity, but the result did not reach statistical significance in this sample, likely reflecting a smaller analyzable cohort and baseline imbalances between groups.

Sleep, the study’s second pillar, produced equally compelling results. Unlike most exercise-sleep research, which relies on wrist-worn actigraphy or questionnaires, this trial used wearable polysomnography in participants’ own homes, capturing electroencephalographic, electromyographic, and respiratory signals across a full night, with an adaptation night first to eliminate the distorting first-night effect. After the intervention, the training group improved objective sleep efficiency, the proportion of time in bed actually spent asleep, by 7.55 percentage points, while controls were essentially unchanged. Total sleep time lengthened and wakefulness after sleep onset, a parameter strongly associated with future cognitive decline in prospective studies, decreased. Within the training group, the share of the night spent in REM sleep also rose, though this change lacked the between-group statistical support needed to attribute it firmly to the intervention.

Intriguingly, the objective improvements did not translate into better subjective sleep. Neither the one-night sleep quality questionnaire completed each morning nor the Pittsburgh Sleep Quality Index covering the preceding four weeks showed significant group-by-time differences. The authors point to a plausible explanation: their participants were relatively good sleepers at baseline, and prior trials suggest resistance training primarily benefits those who start out sleeping poorly. Indeed, a correlation analysis within the training group showed that participants with worse baseline sleep quality experienced greater improvements, echoing earlier findings in older women where only poor sleepers improved. The practical implication is that strength training may be most powerful as a sleep medicine precisely for the population that needs it most, older adults whose nights are already fragmented.

Why would lifting weights lower brain lactate and consolidate sleep? The authors are careful to frame the mechanistic links as speculative, but the framework they assemble is compelling. Exercise transiently elevates lactate, which crosses the blood-brain barrier and may stimulate mitochondrial biogenesis through lactylation processes, ultimately enhancing the brain’s oxidative capacity. Meanwhile, the glymphatic system, the brain’s waste-clearance network, operates most vigorously during deep sleep, and prior work links glymphatic clearance inversely to brain lactate concentrations. Improved sleep efficiency could therefore feed forward into better metabolic housekeeping, while reduced lactate might ease the pro-inflammatory signaling and microglial activation that fragmented sleep and metabolic dysfunction are thought to promote. Notably, a supplementary analysis showed that twelve weeks of balance training, performed by a separate group in the same study, produced none of these changes in lactate or sleep efficiency, hinting that the adaptations are specific to resistance exercise rather than generic to being more active.

The study is not without limitations, and the authors enumerate them with unusual candor. Brain health was assessed through a narrow set of markers without direct measurement of neuroinflammation, glymphatic flow, or cognition itself; sleep was recorded on a single night per timepoint; simple randomization left baseline imbalances, including lower NAA in the training group; and spectral quality constraints reduced the lactate sample to twelve participants per group, with p-values hovering near the significance threshold. The researchers also acknowledge that a passive control group cannot fully separate training-specific effects from the benefits of general social activity, though the balance training comparison argues for specificity. Replication in larger, baseline-matched samples with multi-night sleep recordings and direct cognitive testing remains the essential next step.

Even with those caveats, the trial marks a meaningful advance. It moves beyond the familiar observation that active people age better and instead identifies a concrete, prescribable intervention, three sessions per week of progressive explosive resistance training, that shifts two physiological markers implicated in the path toward dementia: neural oxidative metabolism and sleep continuity. For a rapidly aging global population seeking low-cost, low-risk ways to protect the brain, the message emerging from the 7 Tesla scanner in Fribourg is unexpectedly muscular. The dumbbell rack, long the domain of athletes chasing power and physique, may deserve a permanent place in the preventive medicine toolkit for the aging brain, particularly for older adults whose nights have grown restless and whose neural engines are quietly losing their oxidative edge.

Subject of Research: Effects of resistance training on brain lactate metabolism and sleep in older adults

Article Title: Reduced sensorimotor lactate concentration and improved sleep efficiency following resistance training in older adults

Article References: Scherrer, S., Egger, S., Liu, X., Wick, A. Z., Rasch, B., Xin, L., Lauber, B., & Taube, W. (2026). Reduced sensorimotor lactate concentration and improved sleep efficiency following resistance training in older adults. GeroScience. https://doi.org/10.1007/s11357-026-02519-x

Image Credits: AI Generated

DOI: 10.1007/s11357-026-02519-x

Keywords: resistance training, aging, brain metabolism, lactate, sleep efficiency, magnetic resonance spectroscopy, N-acetylaspartate, neurodegeneration, GeroScience, older adults, polysomnography, mitochondrial function

The scale of the clinical associations is striking. Excessive intramuscular fat has been linked not only to sarcopenia in older adults but also to coronary microvascular dysfunction, where it independently predicts adverse cardiovascular outcomes. It is tightly connected to insulin resistance and type 2 diabetes, with intermuscular adipose tissue shown to directly impair the insulin sensitivity of muscle tissue. In metabolic dysfunction-associated steatotic liver disease, greater intramuscular fat correlates with disease severity, visceral adiposity and even gut dysbiosis. Worsening muscle quality is associated with increased mortality in chronic kidney disease, poorer outcomes in cirrhosis, and complications after transplantation. In oncology, myosteatosis has emerged as an independent prognostic marker across multiple malignancies, predicting survival in ways that simple muscle volume measurements cannot capture.

Part of the appeal of intramuscular fat as a biomarker lies in its sensitivity to change. While ageing may cause roughly five to ten per cent muscle loss over a decade, the fat fraction within muscle can double over the same period, explaining functional decline far better than muscle mass alone. Sedentary behaviour or immobilisation quickly increases intramuscular fat content, whereas regular physical activity helps maintain lean muscle quality. Fatty infiltration in the thigh and calf muscles of older people correlates with poorer balance and increased falls risk, and in neuromuscular disorders such as muscular dystrophies, MRI fat fraction mapping has already become an invaluable outcome measure for tracking disease progression.

Yet the term myosteatosis conceals a biological complexity that the review works hard to untangle. At the microscopic level, intramyocellular lipids are stored as droplets within muscle fibres, serving as a readily mobilised energy reserve. When accumulation exceeds the storage capacity of fibres — a state known as lipotoxicity — insulin signalling is impaired and inflammatory pathways are activated. Crucially, other lipid species such as diacylglycerols and ceramides are metabolically bioactive, suggesting that lipid composition, not merely quantity, matters when interpreting the biomarker. Extramyocellular lipids occupy the space between fibres and fascicles, while intermuscular adipose tissue lies between muscle groups beneath the deep fascia. Each compartment carries different biological meaning, and the relative contribution of each to a measured fat signal depends on the imaging modality and its resolution.

That measurement challenge sits at the heart of the field. Magnetic resonance imaging offers a spectrum of approaches, from subjective visual grading systems such as the Goutallier scale, still widely used in orthopaedic surgical planning, to quantitative Dixon-based techniques that exploit the chemical shift between water and fat protons. Multi-echo Dixon sequences correct for field inhomogeneities, relaxation effects and spectral complexity, yielding the proton density fat fraction — a standardised measure validated against spectroscopy and increasingly applied in clinical trials. Magnetic resonance spectroscopy remains the gold standard for biochemical specificity, uniquely distinguishing intramyocellular from extramyocellular lipids, and phosphorus spectroscopy can probe mitochondrial energetics through high-energy phosphate metabolism. However, spectroscopy suffers from low spatial resolution, voxel placement variability and labour-intensive acquisition, confining it largely to mechanistic research.

Computed tomography offers a pragmatic alternative. It infers fat content from tissue radiographic density expressed in Hounsfield units, with lean muscle typically measuring around 50 to 60 HU and fat-infiltrated muscle appearing progressively less dense. Standardised assessment at the third lumbar vertebral level, using thresholds such as mean density below 41 HU for individuals with a body mass index under 25, has proven effective in predicting cancer survival and mortality in kidney and liver disease. Artificial intelligence, particularly convolutional neural networks, now enables fully automated segmentation of muscle and adipose tissue from routine clinical scans, reducing observer variability and opening the door to large-scale population studies. But CT carries ionising radiation, cannot separate lipid compartments, and is confounded by fluid retention and glycogen variation, which can lower muscle density independently of fat.

Ultrasound rounds out the toolkit as a portable, inexpensive, radiation-free option. Increased echogenicity indicates higher fat or fibrous tissue content, and quantitative echo intensity correlates with MRI-estimated intramuscular fat and with muscle performance. Ultrasound also captures muscle architecture — fascicle length and pennation angle — in a single bedside examination. Its weaknesses are reproducibility and specificity: measurements are sensitive to operator technique, probe pressure, transducer frequency and machine settings, and ultrasound cannot distinguish fat from fibrosis or resolve individual lipid compartments. The SARCUS consensus has defined anatomical landmarks for 39 muscles to improve acquisition consistency, and a cadaver-validated three-point grading system for sarcopenia has shown excellent reliability, but inter-device variability remains a barrier to multicentre trials.

Perhaps the most provocative finding in the review is that ‘normal’ intramuscular fat is anything but uniform. Even in healthy individuals, different muscles harbour inherently different fat content: the gluteus maximus typically contains around fifteen per cent fat in young men and twenty per cent in young women, while neighbouring hip abductor muscles hold only eight to ten per cent. Ageing amplifies these differences dramatically — quadriceps muscles in older men contain four to five times more fat than those of young men, despite only modest differences in muscle cross-sectional area. Sex differences appear partly mediated by conditioning: recreational cyclists of both sexes showed equivalent gluteal fat of roughly twelve per cent once body size and activity were accounted for. Ethnicity adds further complexity, with South Asian children carrying higher intramyocellular lipid than their peers, and African American and Hispanic adults accumulating more intramuscular fat for a given body size — patterns that may partly explain population differences in insulin resistance and that render European-derived reference values non-universal.

The athlete’s paradox crystallises why context is everything. Endurance athletes can carry high intramyocellular lipid yet remain exquisitely insulin sensitive, because their lipid droplets sit within oxidative fibres, are smaller, and are closely apposed to mitochondria, supporting rapid turnover. The same lipid levels in sedentary individuals signal insulin resistance and frailty. This U-shaped relationship — high fat in insulin-sensitive athletes, low fat in insulin-sensitive non-athletes, high fat again in insulin-resistant states — means intramuscular fat cannot be treated as a uniform risk factor. The review suggests composite indices, such as a muscle quality index integrating fat with strength or power measures, and points to emerging mass spectrometry imaging studies that map lipid species and their microregional organisation within muscle histology.

Encouragingly, intramuscular fat is modifiable. A single bout of cycling depletes intramyocellular lipid by roughly twenty-four per cent in exercised muscles. Twelve months of moderate physical activity in older adults prevented the thigh muscle fat gain seen in sedentary controls, who accumulated around eighteen per cent more fat while the active group gained only two per cent — and the exercisers avoided the strength loss their sedentary peers suffered. Meta-analyses of randomised trials show combined aerobic and resistance training reduces fat infiltration most robustly when paired with weight loss. Caloric restriction and bariatric surgery consistently reduce ectopic fat depots, though rapid weight loss risks lean mass loss unless protected by resistance training and adequate protein. Most recently, glucagon-like peptide-1 receptor agonists such as liraglutide and semaglutide have shown modest but significant reductions in intramuscular fat, though concerns persist because these drugs also reduce absolute muscle mass, raising sarcopenic obesity risks in older or frail patients. Before intramuscular fat can join bone density and liver fat as a routine clinical measure, the field needs standardised protocols, multi-ethnic longitudinal cohorts to define age-, sex- and ethnicity-specific reference ranges, and consensus on what constitutes a clinically meaningful change. If those hurdles fall, clinicians may one day stratify risk across sarcopenia, cirrhosis and cancer cachexia — and track whether an intervention genuinely improves muscle quality rather than merely mass.

Subject of Research: Intramuscular fat accumulation (myosteatosis) as a biomarker of skeletal muscle health and chronic disease

Article Title: Intramuscular Fat as a Biomarker of Muscle Health: Strengths, Limitations and Open Challenges

Article References: Spence, C., Belzunce, M. A., Di Laura, A., Henckel, J., & Hart, A. J. (2026). Intramuscular Fat as a Biomarker of Muscle Health: Strengths, Limitations and Open Challenges. Journal of Cachexia, Sarcopenia and Muscle, 17(5), Article e70389. https://doi.org/10.1002/jcsm.70389

Image Credits: AI Generated

DOI: 10.1002/jcsm.70389

Keywords: intramuscular fat, myosteatosis, sarcopenia, muscle quality, Dixon MRI, magnetic resonance spectroscopy, computed tomography, ultrasound, insulin resistance, athlete's paradox, GLP-1 receptor agonists, biomarker standardisation

The technique behind the findings deserves attention in its own right. Unlike conventional magnetic resonance imaging, which maps anatomy, proton magnetic resonance spectroscopy, often abbreviated as ¹H-MRS, measures the chemical composition of living brain tissue. Each metabolite resonates at a characteristic frequency, allowing researchers to quantify molecules such as NAA, creatine, choline and glutamate without a single invasive procedure. NAA occupies a special place in this molecular catalogue. It is produced almost exclusively within neurons and their mitochondria, and decades of research have tied its concentration to neuronal density, mitochondrial metabolism and overall cellular integrity. When NAA drops in a region of the brain, it is generally read as a warning sign that neurons there are struggling, whether through loss of cells, metabolic impairment or both. Conditions as varied as multiple sclerosis, Alzheimer’s disease and diabetes all leave measurable NAA fingerprints.

To assemble the new analysis, the researchers followed the PRISMA reporting guidelines, systematically searching PubMed, Scopus and Embase for studies that had examined the relationship between body mass index and regional brain NAA. After screening, 23 eligible studies made the cut. The team then ran separate random-effects meta-analyses for two kinds of comparisons: categorical contrasts between groups of normal weight and groups with overweight or obesity, and continuous analyses that treated body mass index as a sliding scale correlated against NAA concentrations. They also conducted supplementary pooled analyses across brain regions to generate a brain-wide estimate, and they modeled the results using both lower-bound and upper-bound assumptions to test how robust the overall pattern was to analytical choices.

The headline result concerns the frontal region. Across categorical comparisons, people with higher body mass showed significantly lower frontal NAA, with a standardized effect size of Hedges’ g equal to minus 0.48 and a 95 percent confidence interval running from minus 0.93 to minus 0.03, a statistically significant difference. The continuous analyses told the same story: across all included studies, rising body mass index correlated with falling frontal NAA, with a pooled correlation coefficient of minus 0.218 and a confidence interval of minus 0.323 to minus 0.108, significant at p less than 0.001. In practical terms, the more body mass a person carried, the lower the neuronal health marker measured in the very regions responsible for planning, self-control and higher-order cognition.

Two other regions produced notable signals. In the categorical analysis, the right hippocampal region, a structure central to memory formation and famously vulnerable to metabolic and neurodegenerative disease, showed a significant NAA reduction in people with higher body weight, with Hedges’ g equal to minus 0.51 and a confidence interval of minus 0.84 to minus 0.19. In the continuous analyses, the parietal region emerged as significant, with a correlation of minus 0.256 and a confidence interval of minus 0.463 to minus 0.022. The hippocampal finding echoes earlier work suggesting that overweight is associated with reduced hippocampal NAA, and it dovetails with a broader literature linking obesity to smaller hippocampal volumes and elevated dementia risk.

Perhaps the most consequential result came from the supplementary pooled analyses that collapsed findings across all brain regions. When the team aggregated continuous results using both the lower-bound model, which assumed the most negative associations, and the upper-bound model, which assumed the least negative or most positive ones, the overall negative relationship between body mass index and NAA remained significant in both cases. That kind of consistency across bounding assumptions is a strong robustness check. The categorical pooled analyses, by contrast, retained significance only within the lower-bound model, suggesting that while the overall brain-wide association is convincing in the continuous framework, the strength of group-based contrasts is more sensitive to how studies are combined. The authors conclude that higher body mass index is associated with lower neural NAA most consistently in frontal regions, but that the pooled results point toward more widespread NAA-obesity associations across the brain.

These neurometabolic findings fit into a much larger and increasingly urgent scientific conversation. Neuroimaging studies over the past two decades have documented reduced gray matter volume, thinner frontal cortex and diminished white matter integrity in people with obesity, and large collaborative efforts such as the ENIGMA consortium have confirmed structural brain differences associated with body mass across thousands of participants. Functional imaging has added its own evidence, including reports of reduced prefrontal metabolic activity and decreased prefrontal blood flow in otherwise healthy adults with elevated body mass. Cognitive research has repeatedly linked higher body mass index to executive dysfunction, and longitudinal cohorts have connected midlife obesity to accelerated cognitive decline. What the new meta-analysis adds is a molecular bridge: a measurable chemical correlate of neuronal health that moves the field one step closer to understanding how excess body weight might translate into neural vulnerability.

The mechanistic possibilities are numerous and not mutually exclusive. Mitochondrial dysfunction is a leading candidate, since NAA is synthesized in neuronal mitochondria and its levels track the energy state of cells. Chronic low-grade inflammation, a hallmark of obesity, is known to damage neurons and has been linked to altered brain metabolites in other contexts. Insulin resistance offers another plausible pathway; studies combining spectroscopy with positron emission tomography have found that insulin-stimulated brain glucose uptake correlates with brain metabolites in severe obesity, and reduced insulin sensitivity has been tied to altered neurotransmitter levels. Vascular factors, dyslipidemia and oxidative stress have each been implicated in early neurochemical changes associated with metabolic syndrome. The new meta-analysis cannot adjudicate among these mechanisms, but by establishing a robust regional pattern, it gives future mechanistic studies a concrete target.

There are important caveats. Nearly all of the included studies are cross-sectional, meaning they capture a snapshot in time and cannot establish whether higher body mass causes NAA depletion, whether metabolic brain changes contribute to weight gain, or whether both reflect a third factor. Spectroscopy also faces a technical subtlety specific to this research question: subcutaneous and other fat deposits can influence measured metabolite signal strengths, and methodological work has examined whether superficial fat affects metabolite concentrations determined with water referencing. The authors of the meta-analysis were clearly aware of these measurement challenges, and the consistency of the frontal findings across both categorical and continuous frameworks suggests the signal is not easily explained away. Still, heterogeneity across studies in voxel placement, sequence parameters and quantification methods remains an inherent limitation of pooling spectroscopy data.

Why does this matter beyond the laboratory? If reduced NAA marks early, potentially reversible neuronal metabolic stress, then spectroscopy could serve as a sensitive biomarker for tracking brain health in obesity research, complementing structural imaging and cognitive testing. Intriguingly, one prior study found that intragastric balloon therapy in patients with morbid obesity led to normalization of brain spectroscopic markers associated with diabetes, raising the possibility that metabolic interventions might restore neurometabolic profiles. The Dalhousie team argues that their findings support the utility of proton magnetic resonance spectroscopy as a complementary neuroimaging modality within obesity-related brain research. As obesity rates continue to climb worldwide and the search intensifies for early indicators of brain vulnerability, a non-invasive window into neuronal chemistry may prove to be exactly the tool the field has been waiting for.

Subject of Research: The association between body mass index and regional brain N-acetylaspartate concentrations measured with proton magnetic resonance spectroscopy

Article Title: Body mass index and regional brain N-acetylaspartate: a systematic review and meta-analysis of ¹H-MRS studies

Article References: MacGillivray, C., Wu, L., Fraiha-Pegado, J., McWhinney, S., & Hajek, T. (2026). Body mass index and regional brain N-acetylaspartate: a systematic review and meta-analysis of ¹H-MRS studies. International Journal of Obesity. https://doi.org/10.1038/s41366-026-02214-z

Image Credits: AI Generated

DOI: 10.1038/s41366-026-02214-z

Keywords: obesity, body mass index, N-acetylaspartate, magnetic resonance spectroscopy, neurometabolism, frontal lobe, hippocampus, neuronal integrity, meta-analysis, brain health, mitochondrial metabolism, neuroimaging