Introduction

This is a first case report detailing clinical presentation, blood work changes, medical management, and diagnostic investigation of fatal oleander toxicosis in two miniature horses from the same farm.

Case Report

A 1 yr old female American miniature horse, weighing 57 kg, was evaluated by a veterinarian due to lethargy and inappetance of 12 hr duration. The horse had a heart rate of 40 beats per minute (BPM) (reference range: 28–40), but with inconsistently dropped beats on auscultation, decreased gut sounds in all quadrants, and extremities that were cool to the touch. Nasogastric tube returned a small amount of fetid stomach contents, while installation of water via the nasogastric tube did not indicate an obstruction (no net reflux). Blood results showed a mildly decreased packed cell volume (27.49%, reference range: 32–53), a low hemoglobin (9.6 g/dL; reference range: 11–19), and mean corpuscular volume (32 fL; reference range: 37–59). They also showed an elevated gamma-glutamyltransferase (45 u/L; reference range: 5–24), elevated albumin (3.8 g/dL; reference range: 2.2–3.7), and decreased globulin (2.4 g/dL; reference range: 2.7–5.0). One L of lactated ringer's solution (LRS) was administered IV, which seemed to resolve the arrhythmia and increased interest in food. The owners were instructed to monitor manure output overnight and offer frequent small feedings of hay.

The recheck examination 24 hr later showed continued inappetance, lethargy, cool extremities, decreased gut sounds in all quadrants, and generalized muscle fasciculation. The patient received another L of LRS, 2 mL of omeprazole (2 g/15 mls) per os (PO), 100 lb dose flunixin paste^a PO, 1440 mg sulfamethoaoxazole-trimethoprim PO, 3 mL compounded injectable vitamin supplement, and salt paste. Following treatment, the horse passed several mucoid fecal balls and appeared more interested in grazing.

Forty-eight hr after initial examination, the patient continued to show clinical signs, this time with mild hyperthermia (102.4°F; reference range: 99.5–101.5), moderate tachycardia (80 BPM), and the beginning of a bright pink to bluish-purple line of discoloration at the periphery of the gingiva around the incisors (a “toxic line”). Nasogastric tube retrieved fetid stomach contents, but no net reflux. The horse was given 1 mL flunixin (50 mg/mL) IV. Additionally, the patient was sedated with 0.1 mL of detomidine (10 mg/mL) IV and 0.1 mL of butorphanol IV (10 mg/mL) in preparation for abdominal ultrasound. Ultrasound revealed a fluid-filled large colon and a thickened small intestine with decreased motility. During the procedure, the horse's heart rate decreased to 36 BPM. Repeat complete blood count showed leukocytosis (19.92 × 10³/ug; reference range: 5.4–14.3) with a mild thrombocytosis (441x 10³/ul; reference range: 100-400). Recheck serum chemistries showed slight increases in creatine phosphokinase (514 u/L; reference range 120-470), gamma-glutamyltransferase (57 u/L; Normal: 5–24), and aspartate aminotransferase (AST) (371 u/L; reference range: 175–340). Additionally, elevations in blood urea nitrogen (42 mg/dL; reference range: 7–25) and creatinine (5.6 mg/dL; reference range: 0.6–2.2) were noted along with mild hyperglycemia (155 mg/dL; reference range: 65–110). At this point, the horse was referred to the local veterinary school for additional treatment and diagnostics.

The horse appeared comfortable 10 hr after presentation to the veterinary school with continued inappetance, worsening hyperglycemia (267 mg/dL), and slightly improving lactate level (5.2 mmol/L). Due to worsening hyperglycemia, the fluid rate was increased and an insulin continuous rate infusion was instituted (0.2 unit/min). One hr later, the pet was re-evaluated due to continuing tachycardia (70 BPM) and increased muscle fasciculation. Repeat abdominal ultrasound showed no abnormal findings, nor did passing of a nasogastric tube. Abdominal exploratory surgery was discussed, but rejected due to the current state of the horse as a poor anesthesia candidate. A lidocaine (102 mg/hr) and butorphanol (1.3 mg/hr) continuous rate infusion was instituted, which seemed to mildly improve the clinical appearance of the horse. Three hr later, the horse was found dead in its stall (approximately 72 hr since development of clinical signs).

Macroscopic necropsy findings were nonspecific and included moderate nematodiasis (large strongyles) with diffuse mucoid material accumulation in jejunum and extensive subcutaneous edema around ventral midline. No apparent plant material was observed. Microscopic findings included mild-moderate, multifocal cholangitis and peri-cholangitis with rare hepatic necrosis. Moderate, locally extensive random lymphoplamacytic and neutrophilic hepatitis, as well as moderate, multifocal neutrophilic alveolitis, were present indicating possible sepsis. The only cardiac change was a mild, focal tract of fibrous tissue surrounded by hemorrhage and edema; this finding was attributed to the final agonal event.

Seven days after the first horse developed signs, a 2 yr old female American miniature horse, weighing 77 kg, was examined at the same farm due to lethargy and anorexia for 6 hr. The horse had a normal heart rate (40 BPM) but with intermittent dropped beats, slightly pale mucous membranes, and diffuse muscle fasciculation. A complete blood count was within normal limits, though serum chemistries showed mild hyperglycemia (119 mg/dL), mild elevations in creatinine (2.8 mg/dL), AST (365 u/L), albumin (4.1 g/dL; reference range: 2.2–3.7), and total billirubin (4.1 mg/dL; reference range: 2.2–3.7). Due to the resemblance of clinical signs to the first horse, a toxin (e.g., cardiac glycoside containing plants, organophosphate insecticides) was suspected and the pet was given 4 mL of atropine (0.54 mg/mL) IV, followed by 8 mL of atropine intramuscular. Additionally, 2 L of LRS, 0.5 mL flunixin IV, 6 mL of methocarbamol IV (100 mg/mL), 1440 mg sulfamethaoxazole-trimethoprim PO, 1 oz of di-tri-octahedral smectite^b was administered (in lieu of activated charcoal, which was not available to the attending veterinarian) with instructions for the owner to repeat later in the evening.

Follow-up examination 12 hr later showed further deterioration of condition. The horse was hypothermic (97.1°F); tachycardic (80 BPM) with intermittent dropped beats; had purple, dry mucous membranes and a prolonged capillary refill time (>2 sec). Decreased gut sounds were found in all quadrants with palpably cold extremities. The patient was depressed but responsive. She was given 2 L of LRS. At this point, she developed diffuse muscle fasciculation. The patient was given 1 mL dexamethasone (4 mg/mL) IV, 3 mL methocarbamol (100 mg/mL) IV, and 0.1 mL butorphanol (10 mg/ml) IV. Repeat bloodwork showed a significant elevation in creatine phosphokinase (2085 u/L) with a worsening in the hyperglycemia (226 mg/dL) and AST (447 u/L). Additionally, the horse developed mild-moderate azotemia (blood urea nitrogen: 37 mg/dL; creatinine: 5.9 mg/dL) and a profoundly elevated lactate level (8.3 mmol/L). Due to her worsening conditions and changes in blood values, the patient was humanely euthanized.

Based on the types of clinical signs present with cardiovascular system involvement, toxicity from a cardiotoxic agent was suspected. Samples were collected for toxicological analysis and the body submitted for complete necropsy. A walk through the farm showed several oleander plants and some loose dry leaves on the ground, though no witnessed exposure had occurred.

Necropsy results on the second horse showed mild enteritis, colitis, and presence of whipworms. Gastrointestinal contents submitted to a veterinary diagnostic laboratory confirmed presence of oleandrin using qualitative liquid chromatography-mass spectrometry. The concentration of oleandrin was not quantified in this case since any level in the gastrointestinal tract is considered diagnostic. A concurrent premortem serum sample submitted to a local human hospital laboratory confirmed via immunoassay^c the presence of digoxin at 1.70 ng/mL. No medication containing digoxin was administered to the horse. Thus, the digoxin found in the serum of this horse was likely a result of cross reaction with the cardiac glycosides found in oleander. A diagnosis of oleander toxicosis was made based on presence of oleander plant on the property, types of clinical signs (anorexia, lethargy, cardiac arrhythmias) and detection of cardiac glycosides in the gastrointestinal contents and serum of one horse.

Discussion

Nerium oleander (common oleander), a common ornamental shrub native to Africa, Europe, the Mediterranean, and North America, is used in landscaping due to its ability to withstand drought and insect damage.¹ Fresh oleander is bitter and rarely consumed. Most cases of toxicity in animals result due to feed contamination of the more palatable dried plant material. In this particular case, although several oleander plants were present on the farm, exposure was not witnessed. The miniature horses most likely picked up some of the dried plant material from the ground. The concentration of oleander toxins in the dried plant is higher compared to green plant and toxins remain active in the dried plant.¹ Based on some experimental evidence, ingestion of even small amounts of plant material may be enough to cause toxicity. Some species require as little as 0.005% of body weight in dry leaves to result in fatalities.^3,4 The weight of the two miniature horses discussed here was 57 and 77 kg respectively. This means each horse needed to consume approximately 0.28 and 0.38 kg of dried plant material, which was possible based on the amount of dried leaves that were seen on the ground on the farm. The toxins present in oleander (often up to 0.5%) are steroidal glycosidic cardenolides and pentacyclic terpenoids found in all parts of the plant, with the highest concentration in the roots, seeds, and flowers.¹ The concentration of these toxins increases during flowering and flower color variations reflect varying concentration of toxins with the red flower variant showing higher concentration of cardenolides than the pink (which contains more than the white).²

Cardiotoxicity of oleander results from interference of cardiac glycosides with Na⁺-K⁺-ATPase pump in the cardiac cells resulting in increased intracellular Na⁺ concentrations and disruption of the Na⁺/Ca²⁺ exchange mechanism.^3,4 This leads to elevations in intracellular Ca⁺ with positive inotropic effects, increase in the resting membrane potential, cellular depolarization, and altered myocardial automaticity.^3,4 Additionally, effects on the Na⁺,K⁺-ATPase pump result in hyperkalemia that further contributes to altered cardiac function.^3,4

The renal effects seen in oleander toxicity could be due to direct glycoside action on ATPase in the renal tubules or as a result of decreased perfusion secondary to reduced cardiac output.^3,4 Gastrointestinal effects can result due to direct irritant effects of the toxins or through reduced profusion (e.g., ischemic mucosa, necrosis).^3,4

Toxicoses from oleander have been reported in humans, dogs, chickens, fish, cattle, and equines.^5–9 To the authors knowledge, no previous case reports exist for oleander toxicosis in miniature horses. Whether miniature horses are inherently more sensitive or whether their smaller size compared to other equine make them more susceptible to oleander toxicity is not known. Most of these toxicity cases have been reported due to ingestion of plant materials either by feed contamination or by direct ingestion secondary to inquisitive behavior.⁵ Following ingestion, clinical signs of toxicosis can be seen within 2 hr or delayed up to 48 hr depending on type of material ingested (earlier for seeds, delayed for roots) and the level of mastication of that material (ruminants, equine species showed signs more rapidly than monogastric species).^1,3,4,9 The onset time of clinical signs in the two horses could not be determined because exposure time was not known. The most commonly reported clinical signs in equines are nonspecific initially and may include lethargy with either diarrhea or ileus.^6–9 Other signs may include muscle fasciculation, ataxia, weakness, profuse diarrhea, abdominal pain, and seizures along with a variable degree of bradycardia, including first or second degree heart block.^9,11–13 Progression of signs is generally relative to a worsening cardiac output, resulting in hypothermia, recumbency, venricular tachyarrhythmias, and dypsnea.^9,11–13 Death usually occurs within 12–36 hr after the development of signs.^9,11–13 The first horse in the current case report died approximately 72 hr after the start of clinical signs. Aggressive supportive care may have delayed the time of death in this case. Death in the first horse could be due to the development of fatal cardiac arrhythmias, though this was not confirmed since the horse was unattended at the time of death. The remaining clinical signs in the current case were similar to previously reported equine cases with a combination of GI, renal, and cardiac effects. Another distinctive finding in this case was that the cardiac signs resolved early in treatment, with a persistence of the renal signs.

Common blood chemistry changes associated with oleander toxicosis include azotemia, hyperglycemia, elevated creatine phosphokinase, elevated lactate levels, and hyperkalemia. In humans, profound hyperkalemia indicates significant toxicity.⁹ Presence of persistent hyperglycemia and hyperkalemia in humans are considered a poor prognostic indicators of cardiac glycoside toxicosis. The importance of this in veterinary patients has not been determined, although in the first horse in this case, the hyperglycemia was refractory to treatment and increased as the horse's clinical appearance declined.

A tentative diagnosis of oleander poisoning is made based on history or evidence of exposure (plant material in the feces or GI tract) and the types of clinical signs present.^3,4,9 Presence of cardiac glycosides in the blood of the affected animal can confirm diagnosis by using the fluorescence polarization immunoassay. A newer immunoassay test, using rabbit polyclonal antibiodies against digoxin^c, is considered more sensitive, providing results in a shorter period of time.^4,9 For quantitative results, high performance liquid chromatography-mass spectrometry is necessary and available in some veterinary diagnostic laboratories. ^3,4,9 While the presence of toxins in body fluids can help diagnose oleander toxicity, this does not help determine prognosis due to variability in the toxic properties of each toxin.^3,4,9 In the current case, oleander poisoning was confirmed in the second horse by determining the presence of oleandrin in the GI contents and cardiac glycosides in the serum. The diagnosis in the first horse was presumptive due to similarity of clinical signs and blood work changes to second horse. The common differential diagnoses list in this case was ruled out by the absence of other cardiac glycoside containing plants and chemicals, such as foxglove, azaleas, laurel, squill, and ionophores, while the necropsy findings were not consistent with myocarditis or endotoxemia.

Treatment of oleander toxicosis includes prompt decontamination within 2 hr of exposure in asymptomatic animal using activated charcoal (1–2 g/kg PO) with a stomach tube mixed with a cathartic like sodium sulfate or magnesium sulfate (250 g/horse).¹⁰ Oleander toxins undergo enterohepatic recirculation; therefore a second dose of activated charcoal may be beneficial 8 hr after the first dose to prevent reabsorption.^4,14 All patients with observed ingestion should be monitored closely for first 12 hr with serial cardiac evaluations hourly.

In symptomatic patients, administration of IV fluids, preferably using non-calcium containing solutions, should be started immediately for maintaining perfusion and cardiovascular support, though fluid diuresis does not increase cardiac glycoside elimination.⁹

Bradyarrhythmia including heart block can be treated with atropine (0.01–0.02 mg/kg IV). Check serum potassium prior to using atropine as hyperkalemia is frequently contributory to bradyarrhythmias.^4,11 Tachyarrhythmia are more frequently encountered, with ventricular premature contractions and ventricular tachycardia being the most common cardiac rhythm alterations.^{3,4,9,11,12,13} Corrections of these arrhythmia can be accomplished using either phenytoin (10–22 mg/kg PO q 12 hr) or lidocaine (0.5 mg/kg IV slowly; then 20–50ug/kg/min). Recent evidence in both humans and animals suggests that phenytoin may be more successful in treating ventricular tachyarrhythmia caused by cardiac glycosides than lidocaine.^9,15,16 Arrhythmias once developed can require up to 7 days of treatment for resolution, with some cases showing permanent disruption to the electroconductivity of the myocardium.^9,13,16 Additional supportive care may include use of gastroprotectants, diuretics, nutritional support, and priobiotics.

For persistent cardiac abnormalities or profound hyperkalemia, the use of digoxin-specific Fab fragments^d should be considered. This agent directly binds digoxin and other cadioactive sterols, thereby inactivating them.^4,9 No studies are available to determine effective dosing in equines. In companion animals, administration of 1–2 vials of digoxin-specific Fab fragments IV has been recommended as an initial treatment, with additional vials administered as dictated by clinical signs.^8,17 In humans, the amount of digoxin-specific Fab fragments administered is determined by measuring serum digoxin concentrations, which is not readily available to veterinary practioners.⁹ In humans, where the digoxin level cannot be determined, 10–20 vials are recommended as an empirical dose.^9,4 Administration of excessive digoxin-specific Fab fragments that do not bind digoxin is not expected to be detrimental. For most cases in equines, this can be cost prohibitive with each vial costing approximately $1400 (personal communication, 2013). Animals with renal disease should be monitored for relapse of cardiac signs because prolonged excretion times could result in disassociation between the glycoside and the antibody.⁹

Conclusion

One miniature horse in this case died and the other was euthanized despite aggressive treatment. Though this is the first case report of fatal oleander toxicosis in miniature horses, several case reports show favorable outcomes for horses that receive prompt treatment after exposure. Potentially, miniature horses may be at increased risk for more severe signs due to their smaller body size relative to other equids. More research is needed to determine their sensitivity to oleander toxins.