ALS, carbs and ketones: what happens when the science points in two directions at once

Richard Moore
Mar 11
10 min read

When I wrote about ketogenic therapy last November, I asked whether it might be a lifeline, a landmine, or something in between. The earlier 2024 Lee paper (Lee et al., 2024) pointing toward higher-glycaemic index (GI) diets does not resolve that debate. If anything, it highlights just how metabolically complex ALS is, and why the disease may not be looking for one perfect fuel at all.

When a disease takes away movement, speech and independence, food stops being a lifestyle choice and starts looking more like medicine. For families living with ALS, that makes nutrition both emotionally charged and scientifically bewildering, especially when the science seems to point in two directions at once.

A 2024 study in Annals of Neurology found that a higher and glycaemic load (GL) diet was linked to slower disease progression in ALS patients (Lee et al., 2024). That is, broadly speaking, the “more carbohydrate” signal. Separately, researchers have been exploring whether ketones, produced when the body burns fat rather than glucose, might help struggling nerve cells find an alternative energy source (Phillips et al., 2024). That is the “less carbohydrate, more fat” signal.

At first glance, these ideas seem to belong in opposite corners. But ALS rarely behaves like a neat either-or problem, and neither does the evidence around it.

Why nutrition matters so much in ALS

ALS is not only a disease of motor neurones. In many patients, it is also a disease of weight loss and metabolic strain. The body can enter a state of hypermetabolism, burning through energy faster than it can be replaced, and this has been linked to faster decline and worse survival (He et al., 2022). Meanwhile, swallowing difficulties make eating progressively harder.

That is why maintaining or restoring nutritional status sits at the heart of ALS care. The 2024 European Academy of Neurology guidelines make this explicit, as do recent systematic reviews (Van Damme et al., 2024; de Carvalho Vilar et al., 2025). Many people with ALS are not just fighting neurodegeneration. They are also fighting a body that may be burning through fuel too quickly.

So the real question is not which camp, carbs or fat, is right. It is what the evidence can actually tell us, for whom, and how confidently.

The study that started the conversation

The Lee et al. (2024) paper drew on the ALS COSMOS cohort and asked whether baseline dietary patterns, including GI and GL, were linked to subsequent function and survival in sporadic ALS. This was a prospective observational study, not a controlled feeding trial, which matters immediately: it can identify associations, but it cannot prove that diet caused the outcomes observed.

Among 304 participants who completed the baseline diet questionnaire, higher GI and GL were associated with less functional decline, measured by ALSFRS-R, over three months. Compared with those in the lowest GI quartile, people in the higher quartiles showed roughly 1.6 to 2.0 fewer points of decline. Higher GI also showed a trend toward longer tracheostomy-free survival, but that signal did not remain statistically significant in the fully adjusted model.

That is clinically interesting. But it is not the same as showing that eating a high-GI diet slows ALS.

Why the question was worth asking

Lee and colleagues were not chasing a random idea. Their paper sits on top of a real clinical observation: weight loss in ALS is common and prognostically important, and earlier research had already suggested that higher-calorie feeding might be safe and potentially beneficial (Wills et al., 2014; Dorst et al., 2013).

A phase 2 randomised trial by Wills et al. (2014) found that hypercaloric enteral nutrition was safe and tolerable in ALS patients, and supported the case for studying nutritional interventions further. Dorst et al. (2013) found that high-calorie supplements could help stabilise weight over 12 weeks. So the idea that easier, faster-access carbohydrate might help some patients meet their energy demands is biologically plausible, not a stretch.

What the study did well

This was a serious, multicentre study. Participants came from 16 US clinical centres, diagnoses were based on accepted ALS criteria, and the researchers adjusted for a range of clinically relevant factors: age, sex, disease duration, BMI, onset site, baseline function, lung capacity, diagnostic certainty and riluzole use. They also attempted external replication, which strengthens credibility. The problem is not that the study was careless. The problem is that nutrition research in a fast-moving disease is a blunt instrument, even when the study is thoughtfully done.

Four reasons to read the findings carefully

These are not reasons to dismiss the paper. They are reasons to hold the conclusions at arm’s length until stronger evidence arrives.

First, who made it into the analysis? Completing a detailed diet questionnaire requires stamina, organisational capacity and a certain degree of preserved function. The 304 people who did so may systematically differ from those who were more unwell, more cognitively affected, or more dysphagic. The sample was relevant, but not necessarily a clean mirror of the full ALS population.

Second, diet in ALS is shaped by the disease itself. Because no one was assigned to eat a particular way, the researchers compared people who happened to eat differently at baseline. But swallowing difficulties, fatigue, appetite changes and food texture needs can all push dietary patterns toward softer, quicker, higher-GI foods, regardless of whether GI itself is the biologically important factor. When higher GI tracks with slower decline, it may reflect the value of easily accessible carbohydrate. Or it may be a proxy for lower disease burden or better support. Covariate adjustment helps, but cannot eliminate residual confounding that was never fully measured.

Third, measuring diet precisely in ALS is hard. The study used an 85-item food frequency questionnaire to estimate usual intake over the preceding six months. This is standard in large observational research, but it gives you a broad sketch rather than the precise measure that would have been provided with diet diary based research, which is the gold standard model for research involving dietary analysis. In ALS, eating patterns can change quickly as symptoms evolve, so a six-month retrospective estimate may miss meaningful instability.

Fourth, the timeline was short and attrition matters. The clearest signal was for functional change over just three months. In ALS, three months matters. But short windows are more vulnerable to noise and temporary baseline imbalances. The survival signal, meanwhile, was a trend rather than a clear statistically significant finding. And in a disease where the sickest patients often drop out of datasets first, missing data are rarely random.

What the paper actually showed, and what it did not

A fair reading is this: higher baseline GI and GL were associated with slower short-term functional decline in this cohort.

What it did not show is that prescribing a high-GI diet will slow ALS. The findings are hypothesis-strengthening, not treatment-settling.

That distinction matters, because an intriguing association can quickly become a headline that outruns the data, and in a disease like ALS, false hope is not a kindness.

Does the wider ALS literature support the carbohydrate side?

The strongest interventional support comes from Wills et al. (2014), but that trial was small and involved enterally fed patients, so it was not the final word on macronutrient composition. In terms of weight gain, Dorst et al. (2022) compared fat-rich and carbohydrate-rich supplements and found that all three high-calorie formulations tested could increase body weight, with tolerability differences between them. However, weight gain is one part of a much larger picture. A systematic review and meta-analysis by Zhu et al. (2023) concluded that high-calorie therapy appears effective for increasing weight and BMI with few side effects, but did not show significant superiority for functional outcomes or survival. Prado et al. (2023) reached a similarly cautious conclusion.

The wider takeaway is not that carbohydrates win. It is that energy support matters, while the best macronutrient balance remains unsettled.

What about the ketogenic approach?

The ketogenic argument rests on a different logic: if neuronal energy handling is impaired in ALS, perhaps ketones can provide an alternative fuel while also benefiting mitochondrial function and reducing oxidative stress. That broader framework has been developed across several neurodegenerative conditions and is articulated in the Phillips et al. (2024) “metabolic icebergs” paper. But a compelling mechanism is not the same as direct ALS treatment evidence.

In ALS specifically, the main human paper most often cited is a 2024 case study of a time-restricted ketogenic diet in one man with bulbar-onset ALS (Phillips et al., 2024). The report is genuinely interesting. The authors describe it as the first documented case of ALS managed with this kind of combined protocol, with improvements or stabilisation across several domains. They also note that weight loss was only slowed rather than reversed, and that LDL and total cholesterol rose substantially. The authors are explicit that it is a case study, which means it cannot establish efficacy or how widely the results might apply.

The ketogenic literature contributes a plausible biological hypothesis in ALS, but it does not yet provide a reliable clinical answer.

Can both stories be partly true?

Annoyingly, yes. And this is worth sitting with.

The carbohydrate story says: in a hypermetabolic, weight-losing disease, easier and faster energy delivery may help some patients maintain function. The ketogenic story says: when the cellular energy machinery is compromised, ketones might help some tissues work around that bottleneck. These ideas only look incompatible if we assume ALS is one single metabolic problem. It probably is not.

Some patients may primarily need more total energy. Some may do better with easy carbohydrate access. Some may have disease features that make a fat-based fuel strategy more plausible. The field does not yet have the phenotyping tools or comparative trials needed to sort these groups cleanly (He et al., 2022; de Carvalho Vilar et al., 2025). That is not a reason for despair. It is an honest account of where the science currently stands.

Are low-GI foods suddenly the villains?

No, and this is one of the easiest ways to misread the Lee et al. paper. GI tells you how quickly a carbohydrate raises blood glucose. It does not capture total nutritional quality, and it certainly does not tell you everything that matters in ALS: calorie density, ease of swallowing, protein, micronutrients, food texture and the practical realities of mealtimes when disease is advanced.

A higher-GI pattern in this study may reflect the value of easier-access carbohydrate in some patients. It may also reflect the kinds of foods people can still tolerate as disease progresses. That is very different from recommending that everyone with ALS eat more refined carbohydrate or ignore overall diet quality.

Where the evidence lands today

The most defensible conclusion is not that carbohydrate has defeated fat, or that ketones have exposed the limits of glucose. It is that ALS nutrition is still a fuel problem without a settled fuel answer.

The Lee et al. paper is thoughtful, clinically relevant and worth serious attention. It makes a genuine case for further intervention trials and strengthens the plausibility of higher-GI dietary patterns in ALS care. But because of its observational design, dietary measurement method, short primary window, and the realities of attrition, it does not support a strong causal claim (Lee et al., 2024). The ketogenic literature offers a coherent biological rationale and some intriguing early signals, but ALS-specific human evidence remains too limited for confident practice recommendations (Phillips et al., 2024).

So the most evidence-based position is also the least glamorous one: protect nutritional status, avoid unnecessary weight loss, and tailor the fuel strategy to the person rather than to an ideology (Van Damme et al., 2024; de Carvalho Vilar et al., 2025).

The lesson worth keeping

When a carbohydrate paper looks promising, it does not mean a ketogenic approach must be wrong. And when ketone research offers an intriguing signal, it does not mean carbohydrate thinking was misguided. ALS keeps returning us to the same humbling truth: biology is rarely loyal to a single narrative.

Sometimes the body needs quick fuel. Sometimes it may need a different kind of fuel. Often it simply needs enough fuel, in a form the person can tolerate, maintain, and live with.

That is less dramatic than a miracle diet.

It is also much closer to the evidence.

Looking for personalised support?

At You Nutrition Clinic, we support people living with ALS and other neurodegenerative conditions through a functional medicine and nutritional therapy approach that is evidence-informed, practical and individualised. We look beyond one-size-fits-all advice to consider nutritional status, symptoms, metabolic stress, gut health, biochemistry and the realities of daily life, helping you or your loved one navigate complex decisions with clarity and compassion.

If you would like personalised support, you can learn more about our team and approach at You Nutrition Clinic. Contact us admin@younutritionclinic.com for a free introductory chat and we’ll talk through what’s relevant for your situation.

🧩 Connect with us

For research updates, practical tips, and ongoing inspiration, follow us on Instagram:

👉 @nutritionandthebrain

Stay curious. Stay hopeful. Support your brain. 🧠

Disclaimer: This article is for informational and educational purposes only and does not constitute medical advice. Always consult with a qualified, registered medical doctor (MD) for diagnosis and treatment decisions.

References

de Carvalho Vilar, M. D., Coutinho, K. M. D., Vale, S. H. L., Dourado Junior, M. E. T., de Medeiros, G. C. B. S., Piuvezam, G., Brandão-Neto, J., & Leite-Lais, L. (2025). Evidence-based nutritional recommendations for maintaining or restoring nutritional status in patients with amyotrophic lateral sclerosis. Nutrients, 17(5), 782. https://doi.org/10.3390/nu17050782

Dorst, J., Cypionka, J., & Ludolph, A. C. (2013). High-caloric food supplements in the treatment of amyotrophic lateral sclerosis: A prospective interventional study. Amyotrophic Lateral Sclerosis and Frontotemporal Degeneration, 14(7–8), 533–536. https://doi.org/10.3109/21678421.2013.823999

Dorst, J., Doenz, J., Kandler, K., Dreyhaupt, J., Tumani, H., Witzel, S., Schuster, J., & Ludolph, A. C. (2022). Fat-rich versus carbohydrate-rich nutrition in ALS: A randomised controlled study. Journal of Neurology, Neurosurgery & Psychiatry, 93(3), 298–302. https://doi.org/10.1136/jnnp-2021-328331

He, J., Fu, J., Zhao, W., Ren, C., Liu, P., Chen, L., Zhang, X., Cui, F., Sun, G., Wang, X., Cui, B., Cui, L., & Fan, D. (2022). Hypermetabolism associated with worse prognosis of amyotrophic lateral sclerosis. Journal of Neurology, 269(3), 1447–1455. https://doi.org/10.1007/s00415-021-10716-1

Lee, I., Mitsumoto, H., Lee, S., Kasarskis, E. J., Rosenbaum, M., Factor-Litvak, P., & Nieves, J. W. (2024). Higher glycemic index and glycemic load diet is associated with slower disease progression in amyotrophic lateral sclerosis. Annals of Neurology, 95(2), 217–229. https://doi.org/10.1002/ana.26825

Ludolph, A. C., Dorst, J., Dreyhaupt, J., Witzel, S., Petri, S., Hermann, A., Meyer, T., Zeller, D., Winterholler, M., Winkler, A. S., Grehl, T., Grosskreutz, J., Danek, A., Wolz, M., Dupuis, L., & the LIPCAL-ALS Study Group. (2020). Effect of high-caloric nutrition on survival in amyotrophic lateral sclerosis. Annals of Neurology, 87(2), 206–216. https://doi.org/10.1002/ana.25661

Phillips, M. C. L., Johnston, S. E., Simpson, P., Chang, D. K., Mather, D., & Dick, R. J. (2024). Time-restricted ketogenic diet in amyotrophic lateral sclerosis: A case study. Frontiers in Neurology, 14, 1329541. https://doi.org/10.3389/fneur.2023.1329541

Phillips, M. C. L., O’Connell, G. C., Romero-Ferrando, B., Mungall, A., & Budgeon, C. A. (2024). Neurodegenerative disorders, metabolic icebergs, and mitohormesis. Molecular Neurodegeneration, 19, 72. https://doi.org/10.1186/s40035-024-00435-8

Prado, M. B., Pedro, K. M., & Adiao, K. J. B. (2023). Efficacy, safety and tolerability of high caloric diet in amyotrophic lateral sclerosis patients: A systematic review and meta-analysis. Revue Neurologique, 179(10), 1008–1019. https://doi.org/10.1016/j.neurol.2023.01.731

Van Damme, P., Al-Chalabi, A., Andersen, P. M., Chiò, A., Couratier, P., de Carvalho, M., Hardiman, O., Kuźma-Kozakiewicz, M., Ludolph, A., McDermott, C. J., et al. (2024). European Academy of Neurology guideline on the management of amyotrophic lateral sclerosis in collaboration with European Reference Network for Neuromuscular Diseases (ERN EURO-NMD). European Journal of Neurology, 31(5), e16264. https://doi.org/10.1111/ene.16264

Wills, A.-M., Hubbard, J., Macklin, E. A., Glass, J., Tandan, R., Simpson, E. P., Brooks, B., Gelinas, D. F., Mitsumoto, H., Mozaffar, T., Sanders, D. B., Ravits, J., Bedlack, R. S., Cudkowicz, M. E., & the MDA Clinical Research Network. (2014). Hypercaloric enteral nutrition in patients with amyotrophic lateral sclerosis: A randomised, double-blind, placebo-controlled phase 2 trial. The Lancet, 383(9934), 2065–2072. https://doi.org/10.1016/S0140-6736(14)60222-1

Zhu, Q., Xu, D., Huang, H., Li, D., Yang, D., Zhou, J., & Zhao, Y. (2023). The safety and effectiveness of high-calorie therapy for treating amyotrophic lateral sclerosis: A systematic review and meta-analysis. Journal of Neurology, 270(10), 4729–4743. https://doi.org/10.1007/s00415-023-11838-4