How to Investigate Potential Poisonings With Horses

The investigation of potential poisonings requires a systematic approach, because most pastures and fencerows contain a number of poisonous plants. Jumping to a conclusion about the first toxic plant found often results in missing the primary cause of the problem, such as poor-quality feed or underfeeding.

1. Introduction

When a vague disease syndrome, such as alopecia, chronic itching, colic, laminitis, or sudden death occurs during the pasture period, veterinarians and owners wonder if there is some insidious weed growing in the pastures/paddocks (Fig. 1). Frantically, practitioners commonly take a multitude of samples, praying that the laboratory will give them a diagnosis. Unfortunately, the pathological findings are often non-specific, and the histories are commonly incomplete.

Only a few veterinarians have advanced training in toxins and toxic plants, and they are usually confined to teaching institutions. Private practitioners often lack the confidence in their abilities to investigate these cases, but they are ideally suited because of their extensive knowledge of agricultural practices, horse behavior, nutrition, feeding practices, and what is normal in a horse’s world. Often, the primary problem is a change in husbandry that has encouraged horses to seek anything edible. A common sense and systematic approach will commonly uncover the etiology of the problem and prevent an investigation from going in the wrong direction.

2. Materials and Methods

The first step in an investigation is to determine if the horses are looking for something to eat or if there is a point source of a toxin. Too often, the focus is on looking for a point source, and the investigation is led astray by the presence of numerous toxic plants on the property. Before searching for a point source, conduct a systematic investigation of the overall farm husbandry. The following is a guide to your investigation.

Determine the body condition score of each horse using either the 1 - 9 scale, described by Henneke et al. [1], or the 0 - 5 scale, described by Carroll and Huntington [2]. Estimates of weight will be needed for ration evaluation. The nomograms using height and body condition, described by Carroll and Huntington [2], are useful in this endeavor (Fig. 2). Refer to the OMAFRA website (www.omafra.gov.on.ca) for details.

Identify the individual horses at risk, especially those that are too thin or too fat (Fig. 3).

Figure 1. Blistering and dermatitis of the muzzle.

Determine the feeding practices, including the presence or absence of pasture and/or supplemental feeding, and the feeding schedule (e.g., access to pasture and/or feed 24 h per day or limited access).

List and sample all feed ingredients including hay, whole grains, commercial diets, vitamins, minerals, and neutraceuticals. Typically, three samples of each feed ingredient are placed into a 10-in, square, zip-lock bag and labeled. One sample remains with the owner, one sample is available for initial testing, and one sample is held in storage by the veterinarian.

Identify all feed ingredients with their names, manufacturers’ names, and batch numbers.

Record the average daily consumption of each feed ingredient for each type of horse (e.g., the average brood mare, yearling). Weigh scales capable of measuring pounds or kilograms and grams are required. Ensure that both scales have been recently calibrated.

Weigh several hay bales and determine the average daily hay consumption of representative horses.

Sample 5 - 10 bales of hay using a hay core sampler (Fig. 4). Place samples in zip-lock bags.

Figure 2. Nomogram example of a horse 14 hands high with a body condition score of 3 (using the 0 - 5 scale).

Open several bales of each type of hay (e.g., mixed, alfalfa), and identify the plants and the percentage composition of each plant. If clover is present, determine the species of clover (e.g., red, white or alsike; Fig. 5). If non-cultivated plants are present, enquire as to whether baling of hay occurred along bush lots, rivers, or creeks.

Examine the bedding. Determine the composition and source (e.g., shavings can be kiln dried, air dried or freshly cut). They can be cedar, hardwoods such as oak, or mixed. Straw can be the by-product of many crops, including wheat, barley, and mixed grain.

Walk the entire pasture, preferably without the distraction of the owner talking to you. Look for places where the horses or trees are leaning/hanging over the fence. Look for areas where the horses have browsed trees and eaten plants. Take note of all material that they are eating. All plant material is suspect.

Figure 3. Horse with a body condition score of 1 (using the 0 - 5 scale).

Figure 4. Hay core sampler attached to a portable drill. B. Serrated, tapered end of a hay core sampler.

Identify the cultivated plants and non-cultivated plants in the pasture/field. Make the following determinations:

• What plants are the horses eating or not eating?
• What is the plant quality and quantity in relation to the time of year and moisture content of the soil?
• Estimate the nutrient content of the plants. Is there rapid lush growth with high crude protein and digestible energy content or are the plants mature with seed heads, very low crude protein and digestible energy?

Figure 5. Red clover (T. pratense), white clover (T. repens), and alsike clover (T. hydridum).

Identify the primary plants on the edge of the field, and collect samples of unknown plants/trees for later identification.

Look for areas where water collects, and examine the water in stock tanks for contamination and/or evidence of blue-green algae.

Determine sources of drinking water (e.g., deep or shallow well, ponds, etc.), and determine if it has been tested recently (e.g., presence of coliforms and E. coli). Collection containers and instructions can be obtained from the regional public health department.

Samples from all sources can be sent for testing as the clinical signs and history dictate. Some samples can be retained for later submission but should be held in a freezer to prevent deterioration. Laboratory analysis could include hay (nutrient, fusarium, and ergot alkaloid mycotoxins analysis), straw and grain (mycotoxins analysis), and commercially prepared feed and minerals (ionophore contamination). You will need to know the type of test that is being used (e.g., enzyme-linked immunosorbent assay, (ELISA), high-performance liquid chromotography (HPLC), if the test has been validated for the type of sample you are submitting, and the sensitivity (lower detectable limit) of the test.

Rations can be analyzed using a ration-balancing program or by working with one of the nutritionists employed by feed companies. For commercially prepared feeds, obtain the complete nutrient analysis, including digestible energy, from the feed company.

3. Results

The investigation quickly provides you with a sense of whether the horses are fed adequately, starved, or overfed based on the feeding program and the body condition of the horses. Evaluation of the volumes of feed (hay and supplements) being fed per day and the average nutrient content of each ingredient will give an overall sense of whether or not the feeding program was within normal husbandry practices before the ration-balancing analysis. For example, the average mature horse should be consuming 2 - 2.5% of its body weight or 2 - 2.5 lb per 100 lb of body weight in dry matter each day. This means that the average 50-lb bale of hay will be fed to two 1000-lb horses within a day. Similar calculations can be made for all of the feedstuffs being fed.

The general "walk-around" will give you a sense of whether or not there is something abnormal on the horse farms. The determination of the quality and quantity of pasture will reflect the horse’s need to search for something edible. A number of poisonous plants will have been identified, and the problem may be identified at this point.

If the problem is not obvious at this point, then a more intensive analysis of feedstuffs and plant samples will be required. Additionally, toxic plants/trees must be identified. Excellent resources are available.

Figure 6. Horsetail (Equisetum arvense).

At this point, it is time to match the presenting clinical signs and pathology report with textbook descriptions of the toxic effects of identified poisonous plants.

4. Discussion

From my experience, improper or poor husbandry practices were a primary contributing factor to the disease problem in >50% of recent investigations. Too commonly, a single plant in the pasture is blamed when an underlying problem, such as insufficient high-quality feed, is the primary cause. The following key discussion points are presented with a situation or example.

As an investigator, you need to know the circumstances when a plant/tree could be a problem. Weeds are commonly present in horse pastures because of poor pasture management, and old fence bottoms are often present as well. Weeds, such as horsetail (Equisetumspp.), grow in damp, overgrazed areas of the pasture where few other plants survive (Fig. 6). Weeds in the hay are often a result of baling along side of ditches and bush lots where non-cultivated plants encroach on the cultivated land; additionally, weeds may have invaded hay fields if the fields are run down.

Horses commonly encounter horsetail in pastures, but a toxic condition is rarely reported. Horses can eat horsetail or bracken fern infrequently and in limited quantities, because it is only a small portion of their diet each day. However, if a producer was baling hay along a bush lot where bracken fern was abundant or in poorly drained fields where horsetail was prevalent, then the feeding of this hay over a period of a couple of weeks could lead to vitamin B1 deficiency in non-ruminants as a result of thiaminase being present in these plants.

Figure 7. Cereal rye straw containing ergot alkaloids.

Figure 8. Alsike clover.

Horses that are in good body condition and given free access to feed and water are less likely to go searching for other sources of nutrients. However, horses that have limited access to feed, either because of the lack of feed, an owner’s desire to control the horse’s body condition, or the presence of poor-quality feed, are more likely to go searching for something desirable to eat.

Fifty-five pregnant mares with body condition averaging 4 in the 1 - 9 system were being fed round bales (crude protein ≈9%) in limited quantities. When placed in foaling stalls and fed poor-quality hay in limited quantities, they were observed consuming large volumes of straw bedding. The crude protein and energy content of the straw was almost equal to the hay. Of the first eight mares to foal, seven had dystocia and dead foals. Unfortunately, the straw was cereal rye straw and contained ergot alkaloids (Fig. 7). The clinical syndrome was similar to fescue toxicity. Circumstances may have been different if the hay quantity was higher and the quality was better.

Horses are selective grazers but retain the tendency to browse when something succulent is available. Horses graze some areas to within one inch of the ground like a golf green, but they will leave other areas (roughs) long and defecate in them. Some toxic plants will be present in both the roughs and the greens. The key is to observe if they are being grazed or being ignored. In most cases, poisonous plants can be present in horse pastures, and the horses will not touch them unless there is nothing else to eat. When horses consume stored feed or are not given a choice, they may ingest sufficient quantities of toxic plants that would not be a problem when consumed occasionally.

A Thoroughbred racehorse had started well in her maiden races. She was shipped in from home to race. As the summer wore on, the paddocks at home became severely depleted of grass. The owner started hand grazing the mare along the side of the driveway for a few hours per day. Serial blood samples revealed increasing liver enzymes, and the mare’s performance diminished. The investigation revealed that the boulevard adjacent to the driveway had been seeded with a highway forage mix containing 40% alsike clover (Fig. 8). Cessation of hand grazing along the driveway resulted in the mare returning to her normal performance level.

The presence of many weeds can be reduced in paddocks with pasture rotation, fertilization, and occasional clipping; however, sometimes over-management can be equally dangerous.

A group of four horses was grazed on a 5-acre parcel of land that had been seeded down the previous year. The land was divided into paddocks, and the horses were rotated from one paddock to the next. The land was fertilized heavily with chicken manure from the brother-in-law’s poultry operation. In October, the paddocks contained a lush thick stand of clover, 10 in high, with a crude protein content of 28% on a dry matter basis and an equally high digestible energy content. The previous year a forage seed mix containing grass and clover had been planted. The drought conditions of that spring resulted in clover being the principle plant species in the pasture. A number of horses developed photosensitization-type lesions (Fig. 9), especially of the forehead region, and their liver enzymes were increased. More than 70% of the pasture was a cultivar of Trifolium repens (huia). It is postulated that the lush growth and predominance of white clover resulted in skin changes, as described as trifoliosis (dew poisoning), or a more acute form of the hepatogenous photosensitization seen with al-sike-clover poisoning. The skin lesions improved rapidly with stall confinement and the feeding of grass hay. Still unresolved is why the horses did not exhibit laminitis considering the extremely high carbohydrate intake.

Figure 9. Photosensitization-type lesions with damage to the full thickness of the dermis and peeling of the skin.

Figure 10. Yew (Taxus spp.).

Figure 11. Red maple (Acer rubrum) present in a horse pasture. Red maple is distinguishable by its serrated leaf margins; the margins of the centre lobe are almost parallel to the midvein, and the underside of the leaf is silver grey. The keys are red.

Figure 12. St. John’s Wort (Hypericum spp.).

5. Plant Dynamics

There are some plants, toxic conditions, or circumstances where toxins may be expected, and all practitioners need to be aware of and be able to recognize these on sight. These will vary from one geographic area to another. The acute toxins include the ionophores (monensin, lasalocid) found in or cross-contaminated from poultry or cattle feeds, botulism (Clostridium botulinum) that is common when feeding haylage, yew (Taxus spp.; Fig. 10), red maple (Acer rubrum; Fig. 11), poison hemlock (Conium maculatum), oleander (Nerium oleander), and cantharidiasis (blister beetle). The common and more chronic toxins include, but are not limited to, horsetail (Equisetumspp.), bracken fern (Pteridium spp.), alsike clover (Trifolium hybridum), and St. John’s wort (Hypericum spp.; Fig. 12).

There are some poisonous plants that are commonly present in pastures but are rarely consumed because of plant characteristics. These again will vary geographically but include non-verticillateleaved species of milkweed (Asclepias spp.), which are fibrous and unpalatable and some members of the nightshade family (Solanum; e.g., S. dulcamara) that are readily found in fencerows in Ontario [3].

There are some plants that are toxic in one state but are non-toxic in another. Buttercups (Ranunculus spp.), in the fresh state, contain a volatile toxin, causing a local dermatitis on the muzzle much like solar dermatitis. However, within a few days of the plant being cut and incorporated into hay or being frozen, the toxin dissipates. Horses are often seen eating buttercups in the pasture a few days after a hard frost. The reverse is true for sorghum, sudan grass, and their hybrids; cyanide intoxication is associated with the plants when they are frozen.

Owners are often keen to eradicate all potentially poisonous plants from their property. Although desirable, one needs to be careful when this is done. Sometimes a plant that was previously unpalatable or undesirable is eaten after application of an herbicide. The potential toxicity of the plant needs to be considered as well as the withdrawal time to prevent toxicity from the herbicide.

In a study of 200 cases of sudden and unexplained death in horses and ponies > 1 yr old, no cause of death was found in roughly one-third of the horses. The remaining died from cardiovascular, gastrointestinal, or respiratory lesions [4]. In many investigations, the primary cause is often missed, because practitioners rely on the laboratory and pathologists to give them the diagnosis. Practitioners need to take charge of the situation and complete a thorough farm investigation. An open mind and common sense will reduce the number of undiagnosed or unresolved cases and result in an understanding of the primary inciting cause.

Appendix: Toxic Plant Resources

1. Burrows GE, Tyril RJ. Toxic plants of North America. Ames, IA: Iowa State Press, 2001.
2. Kershaw L. Ontario wildflowers. Edmonton, Canada: Lone Pine Publishing, 2002. - Available from amazon.com -
3. Kingsbury JM. Poisonous plants of the United States and Canada. Englewood Cliffs, NJ: Prentice-Hall, 1964. - Available from amazon.com -
4. McKnight AP, Walter RG. A guide to plant poisoning of animals in North America. Jackson, Wyoming: Teton, NewMedia, 2001.
5. Ontario Ministry of Agriculture, Food and Rural Affairs. Ontario weeds. Guelph, Canada: Ontario Ministry of Agriculture, Food and Rural Affairs, 1998.
6. The Canadian Poisonous Plants Information System. Available online at http://www.cbif.gc.ca/pls/pp/poison?p_x=px. Accessed on August 31, 2007.