Sycamore and atypical myopathy: why not every horse on the same pasture falls ill

What happens in the muscle during atypical myopathy

Atypical myopathy is not a metabolic disorder of the horse, it is a poisoning. The trigger is hypoglycin A, an amino acid from the sycamore (Acer pseudoplatanus). In the body this gives rise to a substance that blocks several enzymes of fat breakdown, and precisely those that open short-chain, medium-chain and branched fatty acids for energy production. Muscle, which draws a large part of its energy from fat, thereby loses its main source, and muscle tissue breaks down.

Where this conversion takes place was shown only late. In a study of tissue samples from five horses that had died, the activated breakdown products were found at very high concentrations above all in the skeletal muscle, while a large part of the hypoglycin A taken in remained unchanged in the organs. The muscle is therefore not only the site of the damage, but also the site of activation. The work covers five animals and describes a state after death, not a sequence in time.

Why does it strike the muscles in particular?

Because the conversion happens there, and because the postural, respiratory and cardiac muscles work without interruption. An analysis of the blood protein pattern of 26 affected horses found signs of inflammation, of oxidative stress and of increased sugar burning at the same time, with which the body was evidently trying to replace the blocked breakdown of fat.

That analysis, however, was carried out on horses that were already ill. It describes what distinguishes a poisoned horse from a healthy one, not what would have distinguished a horse about to fall ill from its grazing companions beforehand. Precisely that question remains open to this day.

Why most horses on the same pasture stay healthy

The most striking finding of the past ten years runs counter to the idea that a sycamore by the fence is a switch that gets thrown. In a German study carried out during an outbreak, hypoglycin A was measured in 16 affected horses, in blood or in urine. The blood values of the ill horses ranged from just under 400 to over 8000 micrograms per litre, those of unexposed comparison horses stayed below 10. The clinically healthy grazing companions from the same paddocks averaged around 109 micrograms per litre: clearly above the comparison horses and clearly below the ill ones.

Two further pieces of work show the same pattern. On two Swedish holdings, on each of which one horse had fallen ill, 20 of the 22 clinically healthy grazing companions were positive in blood, and a month later, after a move to other pastures, still 9 of 20. Ten horses from unaffected holdings remained negative throughout. These figures come from a congress contribution without a full publication and should be read as a provisional, if consistent, indication.

The largest analysis so far covered 263 horses: 95 ill horses, 73 grazing companions, 19 horses with colic and 76 comparison horses. The pattern of acylcarnitines in the blood, that is, of the intermediate products of the blocked breakdown of fat, clearly separated the ill horses from the comparison horses and also from the colic horses. The grazing companions lay in between. Specialists therefore speak of a subclinical form: taken in, measurable, without signs of illness.

Why the tree by the fence is not a measure of risk

Anyone wishing to know how exposed their own pasture is runs into results that cannot be translated into a risk figure. The only Swiss work on the question examined seeds from six holdings with confirmed cases and from two holdings without cases. Hypoglycin A was found in every sample. The mean content was higher on the holdings without cases of disease than on the affected ones. A statistical comparison of the two means is not reported, and the seeds were collected in 2012, while the cases went back as far as 2007.

A Dutch study of 278 samples, by contrast, did find a difference in the expected direction: seeds from pastures with a case of disease carried on average a higher load than seeds from pastures without a case. The spread, however, was as large as the difference itself, and for leaves and shoots it did not exist at all. The specialists there note expressly that no prediction for an individual horse can be derived from a measurement taken on the tree.

An English study from 2025 examined whether the content can be read off the tree: tar spot disease, an urban or rural location, trunk circumference, seed weight, orientation of the branches. No feature examined explained the toxin content. What differed was the year. The same trees carried markedly different contents from one year to the next, and they did so in step. Across several regions, the reported case count followed the estimated seed production; that comparison is calculated at the level of regions and says nothing about a particular horse.

What was noticed on affected pastures is described as well, but is less well supported than it sounds. In the European analysis of 600 cases reported between 2006 and 2009, dead foliage, dead wood and trees on the pasture, a sloping site and permanent access to grazing went together with a case more often. That analysis was made before hypoglycin A was recognised as the cause in Europe, and its comparison group consisted of reported but subsequently rejected cases, not of healthy exposed horses.

The most instructive counter-test comes from North America. In twelve cases of the seasonal pasture myopathy seen there, box elder seeds lay on all the affected pastures, but also on 61 per cent of the comparison pastures visited, which had no cases. The tree alone therefore does not separate the two groups. What did distinguish the groups was longer grazing times, more closely grazed pastures and less supplementary feeding: observations from interviews conducted after the cases, not measures that were tested.

Which maple, which part of the plant

Not every maple carries hypoglycin A. In Dutch and Belgian samples the substance was not detectable in Norway maple (Acer platanoides) or in field maple (Acer campestre), while it was found in almost all sycamore samples. In North America a corresponding disease is linked to the box elder (Acer negundo). A Czech analysis additionally detected the substance in several ornamental maples, for which not a single case of disease has yet been described: detected does not mean disease-causing here.

Within the tree the distribution is thus uneven: seedlings carry by far the highest load, seeds markedly less, leaves and flower clusters less still. That ranking says nothing, however, about what a horse actually eats. The accessibility and palatability of the plant parts were not measured in the same work, and the specialists point this out expressly.

Two windows in the year, and a route through water

The European reporting data show an unusually sharp seasonality. Around 94 per cent of cases fall into two windows of three months each: one opens in October with the falling seeds, the other in March with the seedlings. The disease thus has two faces, an autumn one and a spring one, and the same substance underlies both. The spring window is the less familiar of the two, even though seedlings are the most heavily loaded part of the plant.

Besides seeds and seedlings, further sources come into question. In a Belgian pilot study, hypoglycin A was found in spring in every sycamore sample, including the flower clusters and rainwater collected after a wet night from moistened seedlings. An English study found the substance in water in which seeds had lain, still after 48 hours. Both observations concern plant and water samples, not horses: whether a case of disease has ever arisen from them was not examined.

Horses do distinguish between the developmental stages of the plant. In an observation of 29 pastured horses with 117 video sequences analysed, 19.1 per cent of the young plantlets at the cotyledon stage were eaten, but only 5.46 per cent of the older seedlings with their first true leaves, which contain considerably more phenolic compounds. The link between plant chemistry and behaviour is observed, not proven, and the authors themselves phrase it as a possibility. Above all, the avoidance is incomplete: a remainder is eaten.

What happens to the plants was examined in its own right. In a laboratory experiment, neither mowing nor two herbicides that were tested lowered the content of the seedlings over two weeks; after mowing it even rose temporarily. Grass that had been cut together with seedlings carried small amounts after a week, and in hay and silage the substance was still detectable after six to eight months of storage. What was measured throughout was the content of the plant, never the risk to a horse.

How often the disease ends in death

How many horses survive atypical myopathy can only be estimated from reporting networks, and the two largest analyses arrive at different figures. In Great Britain, 224 cases in total were reported between 2011 and 2015; for 189 of them the outcome was known, and 38.6 per cent of those horses survived. For 600 European reports from the years 2006 to 2009, of which 354 were classified as atypical myopathy, the survival rate was 26 per cent.

Both figures rest on voluntary reporting. The denominator is unknown: how many horses fell ill in the same regions without being reported is known to no one, and the British work itself points to possible misclassification, because some reports came directly from keepers. The gap between 26 and 38.6 per cent is therefore no evidence of an improvement: these are different periods, different networks and different inclusion criteria.

Which findings went together with death is similar in both analyses. In the British work, a single sign remained in the joint analysis: recumbency, that is, a horse that stays down. In the European analysis, sweating, refusal of feed and laboured, accelerated breathing additionally went together with death, while horses that mostly stood, showed a normal body temperature and passed droppings survived more often. In the British work, the administration of vitamins also went together with better survival: an observation on treated horses, not proof of effect.

What a blood test can and cannot do

Blood diagnostics have developed strongly within a few years. In the analysis of 263 horses, the acylcarnitine pattern separated the ill horses not only from healthy comparison horses, but also from horses with colic, that is, from another cause of an acutely and severely ill horse. One single one of these substances also proved there to be an indication of survival. This is the largest data set published on the question so far.

For the question that owners and keepers ask, however, this helps only to a limited extent. In a Belgian case series, a clinically healthy horse on a pasture with seedlings and seeds lay higher than the mean of the affected horses. This is a single animal, whose further course is not reported. The finding is enough, though, to mark the limit: detection in the blood documents an intake, not an impending illness.

Can I have my horse tested as a precaution?

No procedure follows from the studies for that purpose. They measure exposure, and exposed horses remained healthy in the majority in all the work analysed here. No threshold value above which illness is to be expected has been defined. Whether an examination makes sense in an individual case is decided by the attending veterinary practice.

The analytical methods themselves have become considerably more sensitive. A method validated at the German federal risk assessment institute lowered the limit of quantification for hypoglycin A by roughly thirty-fold and additionally detected, in sycamore and box elder, two dipeptides that had not been described in maple since 1973. What these compounds do in the horse is open: the work is purely analytical and examined no animal.

What the view beyond the species shows

The same group of substances also makes humans ill, but differently. In India and other growing regions, an acute encephalopathy in children is linked to the toxins of the lychee aril, among them the same substance that occurs in sycamore. The mechanism is shared, the clinical picture is not: in those children, low blood sugar with involvement of the brain dominates rather than a breakdown of muscle, and it is described only where undernutrition, long fasting and the eating of large quantities of fruit come together.

Sheep take up the substance too. In a British study, hypoglycin A was detectable in 13 of 15 sheep grazing on loaded pastures, without signs of illness and without abnormal blood values, and in two of five suckling lambs as well, which points to passage into the milk. Neither rumen fluid nor equine gastric juice broke the substance down in the test tube. The authors therefore suspect a more resistant metabolism in sheep, without demonstrating it.

The chain from the tree to the sick horse, link by link: what was measured and where the inference stops

Original analysis

Limitations and uncertainty

• There is no randomised trial and no systematic review of this poisoning, and neither would be readily feasible: none of the work used here reaches the level of systematic review, meta-analysis or randomised trial. The editorial requirement to base at least half of the sources at that level cannot currently be met in this field; the causal chain is carried by case series, comparisons of exposed groups and plant analysis from several countries.
• Frequency and survival figures come from voluntary reporting networks. The denominator is unknown, mild courses are reported less often, and in the British analysis some reports came directly from keepers, which makes misclassification possible.
• The quantity at which a horse falls ill has still not been determined. Blood concentrations cannot be converted back into quantities taken in, and none of the cited work measured how much a horse actually ate.
• Why one horse falls ill and its grazing companion does not is unexplained. The protective features found in one analysis, greater age and gelding status, come from calculations on a convenience sample and were not confirmed prospectively; the blood protein pattern was measured only after the disease had broken out.
• The only Swiss survey covers eight holdings, and its seeds were collected years after the cases of disease, even though the content varies strongly precisely from year to year. The remaining plant data come from England, the Netherlands, Belgium, Czechia and Germany and cannot readily be transferred to a particular Swiss pasture.

Open questions

• Why do individual horses on the same pasture fall severely ill, while the majority of demonstrably exposed animals stay healthy?
• Do horses that later fall ill differ from their grazing companions even before intake, or does the difference arise only from the quantity taken in?
• What do the newly detected dipeptides and the methylenecyclopropylglycine compounds contribute to the clinical picture in the horse?
• Does the lower sensitivity of sheep really rest on a more resistant metabolism, or was the season that was observed simply a mild one?

Frequently asked questions

Do I have to fell the sycamore next to my pasture?

The state of the studies does not answer this question, and for a substantive reason: there is no threshold value. In an English study, no examined feature of a tree explained the toxin content of its seeds, the same trees carried markedly different contents from year to year, and in the only Swiss survey the mean content was even higher on holdings without cases of disease than on affected ones. A laboratory experiment also showed that neither mowing nor two tested herbicides lowered the content of the seedlings over two weeks. What makes sense on a particular pasture is a decision to be taken with the attending veterinary practice, not an inference from these figures.

My horse has stood under maples for years and has never been ill. Is the pasture safe, then?

No, that cannot be concluded from it. The contents vary above all between years: in an English study the same trees carried markedly different amounts from one year to the next, and across several regions the reported case count followed the estimated seed production. In Swiss samples the mean content was higher on holdings without known cases than on affected ones. A European review of the reporting data therefore states expressly that a pasture with a sycamore nearby counts as exposed even when nothing has ever happened there. A year without incident is an observation, not a property of the pasture.

Are all maple species dangerous?

No. In Dutch and Belgian samples, hypoglycin A was not detectable in Norway maple or field maple, while it was found in almost all sycamore samples. In North America the box elder is linked to a corresponding disease, and its seedlings also lay high in a Czech analysis. The same analysis additionally detected the substance in several ornamental maples, among them sugar maple and Japanese maple, for which no case of disease has been described. Detected and disease-causing are two different statements. Identifying the species belongs, in case of doubt, in competent hands.

When in the year is the risk greatest?

Around 94 per cent of the cases reported in Europe fall into two windows of three months each: one opens in October with the falling seeds, the other in March with the seedlings. The spring window is less well known, but well documented. Seedlings were by far the most heavily loaded part of the plant in a Czech analysis, and in a Belgian case series spring seedlings could be traced as the source. A Belgian pilot study additionally found the substance in spring in flower clusters and in rainwater that had run off moistened seedlings.

Can hay from a pasture with sycamore be contaminated?

In a laboratory experiment, yes. Seeds and seedlings were introduced there into hay and silage, and hypoglycin A was still detectable after six to eight months of storage. Grass that had been cut together with seedlings also carried small amounts after a week. This work examined plant material only and no horse: it shows that the substance survives conservation, not that a case of disease arises from it. Such a case following the feeding of conserved forage is not described in the work analysed here.

Is atypical myopathy really fatal almost every time?

According to the available reporting data, no. Of 189 British cases with a known outcome, 38.6 per cent survived; in the older European analysis of 354 cases it was 26 per cent. Both figures come from voluntary reports: the denominator is unknown, mild courses are reported less often, and the British work itself points to possible misclassification. The difference between the two figures is no evidence of an improvement, because periods, networks and inclusion criteria differ. What is certain is that a considerable proportion of the reported horses died and that a suspicion tolerates no delay.

Sources

1. Bochnia M, Ziegler J, Sander J, Uhlig A. Hypoglycin A Content in Blood and Urine Discriminates Horses with Atypical Myopathy from Clinically Normal Horses Grazing on the Same Pasture. PLoS One, 2015 (Case-control study | Horse)DOI 10.1371/journal.pone.0136785
   In 16 horses that fell ill during an outbreak, hypoglycin A and its breakdown product in blood and urine were far above the values of unexposed horses, while the clinically healthy grazing companions from the same pastures lay in between at around 109 micrograms per litre: the burden is a gradient, not a switch.
2. Renaud B, Kruse CJ, Francois AC, Cesarini C. Large-scale study of blood markers in equine atypical myopathy reveals subclinical poisoning and advances in diagnostic and prognostic criteria. Environmental Toxicology and Pharmacology, 2024 (Case-control study | Horse)DOI 10.1016/j.etap.2024.104515
   In the largest analysis so far, with 263 horses, the acylcarnitine pattern separated ill horses from comparison horses and from colic horses, while the apparently healthy grazing companions lay in between: a subclinical form is thereby documented, and a single acylcarnitine served at the same time as an indication of survival.
3. Gonzalez-Medina S, Hyde C, Chang YM, Piercy RJ. Assessment of tree-associated atypical myopathy risk factors in Acer pseudoplatanus (sycamore) seeds and leaves. Equine Veterinary Journal, 2025 (Other | Horse)DOI 10.1111/evj.14475
   No examined tree feature explained the hypoglycin A content of the seeds; what differed was the year, with neighbouring trees varying in step, the regional case count following the estimated seed production, and water in which seeds had lain still testing positive after 48 hours.
4. Unger L, Nicholson A, Jewitt EM, Gerber V. Hypoglycin A concentrations in seeds of Acer pseudoplatanus trees growing on atypical myopathy-affected and control pastures. Journal of Veterinary Internal Medicine, 2014 (Case-control study | Horse)DOI 10.1111/jvim.12367
   On six Swiss holdings with cases of disease and two comparison holdings, every sycamore seed sample contained hypoglycin A, with an enormous spread and a mean content that was higher on the holdings without cases; a statistical comparison of the means is not reported.
5. Westermann CM, van Leeuwen R, van Raamsdonk LWD, Mol HGJ. Hypoglycin A Concentrations in Maple Tree Species in the Netherlands and the Occurrence of Atypical Myopathy in Horses. Journal of Veterinary Internal Medicine, 2016 (Case-control study | Horse)DOI 10.1111/jvim.13927
   In 278 samples hypoglycin A was not detectable in Norway maple or field maple, but was present in almost all sycamore samples; seeds from pastures with a case carried a higher load on average, yet the spread rules out any prediction for an individual horse, according to the authors.
6. Baise E, Habyarimana JA, Amory H, Boemer F. Samaras and seedlings of Acer pseudoplatanus are potential sources of hypoglycin A intoxication in atypical myopathy without necessarily inducing clinical signs. Equine Veterinary Journal, 2016 (Case series | Horse)DOI 10.1111/evj.12499
   Spring seedlings, and not only autumn samaras, contained hypoglycin A, and one clinically healthy horse lay above the mean of the affected animals in blood: detection in the blood documents an intake, not an impending illness.
7. Gröndahl G, Berglund A, Skidell J, Bondesson U. Clinical Research Abstracts of the British Equine Veterinary Association Congress 2015. Equine Veterinary Journal, 2015 (Cross-sectional study | Horse)DOI 10.1111/evj.12486_49
   On two Swedish holdings with one affected horse each, 20 of the 22 clinically healthy grazing companions were positive in blood and, a month later after a change of pasture, still 9 of 20, while ten horses from unaffected holdings remained negative; the figures come from a congress volume without a full publication.
8. Gonzalez-Medina S, Ireland JL, Piercy RJ, Newton JR. Equine atypical myopathy in the UK: Epidemiological characteristics of cases reported from 2011 to 2015 and factors associated with survival. Equine Veterinary Journal, 2017 (Case series | Horse)DOI 10.1111/evj.12694
   Of 189 reported British cases with a known outcome, 38.6 per cent survived; in the joint analysis only recumbency remained as an unfavourable sign, and the administration of vitamins went together with better survival on a purely observational basis.
9. van Galen G, Saegerman C, Marcillaud Pitel C, Patarin F. European outbreaks of atypical myopathy in grazing horses (2006-2009): determination of indicators for risk and prognostic factors. Equine Veterinary Journal, 2012 (Case-control study | Horse)DOI 10.1111/j.2042-3306.2012.00555.x
   Of 600 cases reported across Europe, 354 were classified as atypical myopathy, and 26 per cent survived; the pasture features and clinical signs named are associations from a reporting network, collected before hypoglycin A was recognised as the cause in Europe.
10. Gonzalez-Medina S, Montesso F, Chang YM, Hyde C. Atypical myopathy-associated hypoglycin A toxin remains in sycamore seedlings despite mowing, herbicidal spraying or storage in hay and silage. Equine Veterinary Journal, 2019 (Laboratory study | Species-independent)DOI 10.1111/evj.13070
    Neither mowing nor the tested herbicides lowered the hypoglycin A content of the seedlings within two weeks, after mowing it rose temporarily, grass cut together with the seedlings picked up a small load, and in hay and silage the substance was still detectable after six to eight months.
11. Aboling S, Scharmann F, Bunzel D. Equine atypical myopathy: consumption of sycamore maple seedlings (Acer pseudoplatanus) by pastured horses is driven by seedling maturity and might be associated with phenolic compounds. The Veterinary Record, 2020 (Cross-sectional study | Horse)DOI 10.1136/vr.105736
    In 117 video sequences of 29 pastured horses, 19.1 per cent of the young cotyledon-stage plantlets were eaten compared with 5.46 per cent of the older, more phenol-rich seedlings; the link with phenol content is described by the authors themselves as no more than possible, and the avoidance remains incomplete.
12. Novotna T, Jahn P, Samonilova E, Kabesova M. Hypoglycin A in Acer genus plants. Toxicon, 2023 (Laboratory study | Species-independent)DOI 10.1016/j.toxicon.2023.107271
    The content depends above all on the part of the plant, with seedlings far ahead of seeds, leaves and flower clusters; hypoglycin A was additionally found in several ornamental maples for which no case of disease has been described, and accessibility and palatability were not measured.
13. Votion DM, Habyarimana JA, Scippo ML, Richard EA. Potential new sources of hypoglycin A poisoning for equids kept at pasture in spring: a field pilot study. The Veterinary Record, 2019 (Laboratory study | Species-independent)DOI 10.1136/vr.104424
    In spring, hypoglycin A was detectable in every sycamore sample, including the flower clusters and the rainwater from moistened seedlings, while Norway maple and field maple remained negative; this is a pilot study in a single area, with no examination of horses.
14. Votion DM, Francois AC, Kruse C, Renaud B. Answers to the Frequently Asked Questions Regarding Horse Feeding and Management Practices to Reduce the Risk of Atypical Myopathy. Animals, 2020 (Other | Horse)DOI 10.3390/ani10020365
    From the European surveillance data, 94 per cent of cases fall into two three-month windows from October and from March, and the authors state that any pasture with a sycamore nearby counts as at risk even if a case has never occurred there; the practical advice it contains is expert opinion, not measures that have been tested.
15. Sander J, Terhardt M, Janzen N, Renaud B. Tissue Specific Distribution and Activation of Sapindaceae Toxins in Horses Suffering from Atypical Myopathy. Animals, 2023 (Case series | Horse)DOI 10.3390/ani13152410
    In tissue samples from five horses that had died, the activation of the plant toxins took place above all in the skeletal muscle, where the active breakdown products accumulated strongly, while a large part of the hypoglycin A remained unchanged in the organs.
16. Kruse CJ, Dieu M, Renaud B, Francois AC. New Pathophysiological Insights from Serum Proteome Profiling in Equine Atypical Myopathy. ACS Omega, 2024 (Case-control study | Horse)DOI 10.1021/acsomega.3c06647
    The blood protein pattern of 26 affected horses differed from that of exposed but healthy grazing companions and pointed to inflammation, oxidative stress and a fat metabolism compensated by increased sugar burning; measurement took place only after the disease had broken out.
17. El-Khatib AH, Engel AM, Weigel S. Co-Occurrence of Hypoglycin A and Hypoglycin B in Sycamore and Box Elder Maple Proved by LC-MS/MS and LC-HR-MS. Toxins, 2022 (Laboratory study | Species-independent)DOI 10.3390/toxins14090608
    A method validated at the German federal risk assessment institute lowered the limit of quantification for hypoglycin A by roughly thirty-fold and detected two dipeptides in sycamore and box elder for the first time since 1973, whose significance for the animal remains open.
18. Valberg SJ, Sponseller BT, Hegeman AD, Earing J. Seasonal pasture myopathy/atypical myopathy in North America associated with ingestion of hypoglycin A within seeds of the box elder tree. Equine Veterinary Journal, 2013 (Case-control study | Horse)DOI 10.1111/j.2042-3306.2012.00684.x
    The North American seasonal pasture myopathy is linked to hypoglycin A from box elder seeds, yet seeds also lay on 61 per cent of the comparison pastures without cases: what separated the groups was longer grazing times, closely grazed pastures and less supplementary feeding.
19. Gonzalez-Medina S, Bevin W, Alzola-Domingo R, Chang YM. Hypoglycin A absorption in sheep without concurrent clinical or biochemical evidence of disease. Journal of Veterinary Internal Medicine, 2021 (Cohort study | Multiple species)DOI 10.1111/jvim.16077
    Sheep on loaded pastures carried hypoglycin A in their blood, without signs of illness and without abnormal blood values, and neither rumen fluid nor equine gastric juice broke the substance down in the test tube; a more resistant metabolism in sheep remains a supposition of the authors.
20. Sarkar A, Datta D, Datta SK, John TJ. Acute encephalopathy in children in Muzaffarpur, India: a review of aetiopathogenesis. Transactions of the Royal Society of Tropical Medicine and Hygiene, 2020 (Other | Human)DOI 10.1093/trstmh/traa036
    The same group of substances from the lychee aril blocks fatty acid breakdown in children, but presents there as hypoglycaemic encephalopathy rather than a breakdown of muscle, and is described only where undernutrition, long fasting and large amounts eaten come together.