Isoniazid (isonicotinic acid hydrazide) is a common, first-line antimycobacterial agent used to treat tuberculosis in humans and other animals.1–4 Although tuberculosis is rare in developed countries (only 9,421 cases of tuberculosis were reported in the United States in 20145), the use of isoniazid with its low margin of safety puts companion animals at risk of accidental ingestion. Human patients and dogs that accidentally ingest or are overdosed with isoniazid can develop acute life-threatening toxicosis.6–9
Isoniazid inhibits cell wall synthesis of Mycobacterium tuberculosis by inhibiting lipid and DNA synthesis. In human patients, isoniazid is rapidly and almost completely absorbed in the gastrointestinal tract following oral administration. It undergoes substantial first-pass metabolism, and peak plasma concentrations are achieved within 1 to 2 hours after administration. Isoniazid has poor protein binding (approx 10%) in plasma, has a small volume of distribution, and readily penetrates into the CNS and CSF.10,11 In humans, the elimination half-life of isoniazid is dependent on a patient's ability to acetylate; the elimination half-life is 1 to 2 hours in fast acetylators and 2 to 5 hours in slow acetylators.10–12 Dogs are considered slow acetylators because of their inherent lack of NAT2.13 Thus, in dogs, the elimination half-life of isoniazid is 2 to 5 hours, and the lack of NAT2 increases the risk of isoniazid-induced neurologic and hepatic toxicosis.13 In humans patients, approximately 50% to 70% of isoniazid is excreted unchanged or as metabolites in urine within 24 hours after administration.10,11
In both dogs and humans, clinical signs of isoniazid toxicosis are generally observed within 30 to 60 minutes after ingestion and are characterized by recurrent grand mal seizures subsequent to a decrease in GABA (an inhibitory neurotransmitter) concentration and an increase in glutamic acid concentration in the CNS.6,14,15 High toxic doses of isoniazid are associated with what is known as the classic triad, which includes seizures that are refractory to conventional treatments, lactic acidosis, and coma. Patients with this classic triad of signs of isoniazid toxicosis can die without prompt diagnosis and proper treatment.8,9,14,15
The veterinary literature contains reports of accidental6 and experimentally induced15,16 isoniazid toxicosis in dogs. In a report6 of 27 dogs with accidental isoniazid toxicosis, the amount of the drug ingested ranged from 22 to 270 mg/kg (10 to 123 mg/lb; mean ± SD, 109 ± 82 mg/kg [49.5 ± 37 mg/lb]). Clinical signs in those dogs developed within 30 minutes to 2 hours after isoniazid ingestion, with the most common sign reported being recurrent grand mal seizures followed by stupor.6 Among the dogs that developed seizures in that study,6 the lowest dose of isoniazid ingested was 37 mg/kg (17 mg/lb). In an experimental study,16 the lowest consistently lethal dose of isoniazid in dogs was 75 mg/kg (34 mg/lb), with the onset of seizures and death occurring approximately 1 to 2 hours after drug exposure. In another experimental study,15 IV administration of pyridoxine (vitamin B6) as an antidote to dogs with isoniazid toxicosis had a dose-dependent protective effect against seizures and death; dogs administered pyridoxine at a dose equivalent to the dose of isoniazid ingested (ie, 1:1 dose ratio of pyridoxine to isoniazid) survived.
Although isoniazid toxicosis has been described in dogs following accidental6 and experimental15,16 ingestion of the drug, little information is available regarding the minimum toxic dose of isoniazid in dogs, or the signalment, clinical signs, treatment, and outcome for a large number of dogs with accidental isoniazid toxicosis. The goals of the study reported here were to establish the minimum toxic dose of isoniazid in dogs, characterize the clinical signs and review outcomes for dogs following accidental ingestion of isoniazid, determine whether IV administration of pyridoxine to those dogs was protective against death, and describe various approaches used to manage dogs with isoniazid toxicosis in clinical settings.
Materials and Methods
Case selection criteria
The electronic database of the ASPCA APCC was searched to identify the records of dogs with isoniazid exposure between January 2004 and December 2014. That database contains a record of queries received by APCC personnel from both animal owners and veterinarians in the United States and Canada. Dogs were included in the study if isoniazid exposure was classified as observed or evidenced on the basis of a history of observed ingestion, missing medication, or physical evidence of tampering with medication bottles. Dogs were excluded from the study if there was evidence that other drugs in addition to isoniazid had been ingested, the medical record was incomplete, or there was a low likelihood of isoniazid exposure on the basis of information provided to and recorded by the APCC veterinary toxicologist or veterinarian.
Medical records review
For each dog evaluated, information extracted from the record included signalment, history, clinical signs, amount of isoniazid ingested, treatment, and outcome. The data for all dogs were recorded into a commercially available spreadsheet program.a Clinical signs were individually analyzed and classified into the following categories: seizures, CNS signs without seizures (eg, depression, coma, ataxia, blindness, vocalization, tremors, miosis, mydriasis, and anisocoria), gastrointestinal (eg, vomiting, ptyalism, and diarrhea), cardiovascular (eg, tachycardia, bradycardia, arrhythmias, and pallor), respiratory (eg, tachypnea, dyspnea, and pulmonary edema), and urogenital (eg, urinary incontinence and pigmenturia).
Investigators contacted and corresponded with pet owners and primary care veterinary staff via telephone or facsimile to obtain additional information such as more in-depth history, clarification of clinical signs, medical treatments administered, hospitalization data, and outcome. Dogs that recovered were classified as survivors, whereas those that died prior to or after hospitalization or that were euthanized while hospitalized were classified as nonsurvivors.
Statistical analysis
Descriptive statistics were generated for all demographic variables, clinical signs, isoniazid dose ingested, and outcome. Age was evaluated as both a continuous and categorical (young [≤ 1 year old] or adult [> 1 year old]) variable. Dogs that died or were euthanized were combined and analyzed together in the nonsurvivor category. The isoniazid dose ingested (reported in mg/kg) was estimated on the basis of the dog's body weight, the number of isoniazid tablets ingested, and the concentration of isoniazid in those tables. Treatment with pyridoxine was classified as yes or no regardless of the route (IV, PO, or rectal) of administration. The distribution of data for continuous variables such as age, weight, and isoniazid dose ingested was assessed for normality by use of the Shapiro-Wilk test, and all those variables were found to have nonparametric distributions. The Wilcoxon test was used to assess the respective relationships between each continuous variable and the presence of seizures (yes or no), whether the dog was treated with pyridoxine (yes or no), and outcome (survivor or nonsurvivor). The Bonferroni correction was used when multiple pairwise comparisons were indicated. The respective associations between each categorical variable and the presence of seizures and outcome were assessed by use of a Fisher exact (when the count within ≥ 1 cell was ≤ 5) or χ2 test. Individual variables that were associated (P < 0.25) with outcome were individually entered into logistic regression models for presence of seizures or outcome. Logistic fit and inverse prediction were used to estimate the probability of seizures following ingestion of various isoniazid doses.
Results
Dogs
During the evaluation period, the APCC received queries regarding isoniazid ingestion for 172 dogs, of which 35 were excluded from the study because exposure to isoniazid was considered unlikely to cause the clinical signs reported, multiple drugs or compounds in addition to isoniazid were ingested, or the medical record was incomplete. Thus, 137 dogs were evaluated in the study.
The 137 dogs evaluated included 69 (50.4%) males and 60 (43.8%) females; the sex was not recorded in the record for 8 (5.8%) dogs. Age was recorded for 133 dogs, and the median age for those dogs was 1 year (range, 0.13 to 13 years). Body weight was recorded for 135 dogs, and the median body weight for those dogs was 11 kg (24.2 lb; range, 1.22 to 45 kg [2.7 to 99 lb]). Forty-four breeds were represented in the study population, with the most common being Labrador Retriever (n = 19 [13.9%]), Chihuahua (10 [7.3%]), German Shepherd Dog (7 [5.1%]), Beagle (6 [4.4%]), Boxer (6 [4.4%]), Maltese (6 [4.4%]), Miniature Dachshund (6 [4.4%]), Border Collie (5 [3.6%]), Golden Retriever (5 [3.7%]), and mixed breed (5 [3.6%]); all other breeds were represented by ≤ 4 dogs.
Isoniazid dose ingested
The amount of isoniazid ingested was recorded for 130 of the 137 dogs. The median amount of isoniazid ingested was 300 mg (range, 50 to 8,400 mg). The most frequently reported source of isoniazid ingested was 300-mg tablets, with dogs ingesting a mean of one 300-mg tablet (range, 0.5 to 30 tablets). The estimated median isoniazid dose ingested on the basis of body weight was 59.5 mg/kg (27.0 mg/lb; range, 2.0 to 1,061.9 mg/kg [0.91 to 482.7 mg/lb]). Of the 137 dogs, ingestion of isoniazid was observed in 28 (20.4%) and strongly suspected but not actually observed in the remaining 109 (79.6%).
Clinical signs
One hundred thirty-four of 137 (97.8%) dogs developed clinical signs of isoniazid toxicosis. Clinical signs were not observed but were inferred or evidenced in the remaining 3 (2.2%) dogs; the estimated doses of isoniazid ingested by those 3 dogs were 2, 6.3, and 8.8 mg/kg (0.9, 2.9, and 4 mg/lb). The most commonly reported clinical sign was seizures. Of the 132 dogs for which the seizure status was known, 104 (78.8%) had seizures. The lowest estimated dose of isoniazid ingested by a dog that subsequently developed seizures was 19.8 mg/kg (9.0 mg/lb). Dogs that developed seizures had a lower median body weight (P = 0.021) and ingested a higher dose of isoniazid (P < 0.001) on a milligram-per-kilogram basis than dogs that did not develop seizures. Thirty-seven of the 104 (36%) dogs that developed seizures subsequent to isoniazid ingestion died or were euthanized (ie, were classified as nonsurvivors), whereas all 28 dogs that did not develop seizures were classified as survivors. Results of the logistic fit and inverse prediction model indicated that an isoniazid dose of 22.2 mg/kg (10.1 mg/lb; 95% CI, 1.6 to 32.3 mg/kg [0.73 to 14.7 mg/lb]) had a 50% probability of causing seizures and a dose of 48.6 mg/kg (22.1 mg/lb; 95% CI, 38.5 to 70.2 mg/kg [17.5 to 31.9 mg/lb]) had an 80% probability of causing seizures (Figure 1).
Results of a logistic fit and inverse prediction model for the effect of dose of isoniazid ingested on the probability of a dog developing seizures. Data used for the model were obtained from ASPCA APCC records for 137 dogs that inadvertently ingested isoniazid between January 2004 and December 2014. Brackets represent the model-predicted 95% CIs.
Citation: Journal of the American Veterinary Medical Association 251, 6; 10.2460/javma.251.6.689
A total of 288 clinical signs were reported for the 134 dogs that developed clinical signs following isoniazid ingestion. The most common clinical sign category recorded was seizures (n = 104 [36%]) followed by CNS signs without seizures (94 [33%]), which included ataxia (33), altered mentation (17), tremors (12), blindness (8), vocalization (8), agitation (5), coma (4), aggression (3), anisocoria (3), and mydriasis (1). The next most commonly reported clinical sign category was gastrointestinal abnormalities (n = 41 [14%]), which included vomiting (24), ptyalism (15), and diarrhea (2). Other clinical signs recorded included hyperthermia (n = 33 [11%]), cardiovascular abnormalities (19 [7%]), urogenital abnormalities (4 [1%]), and aspiration pneumonia (1 [0.3%]). All dogs (n = 6) that developed blindness and survived regained their vision.
Pyridoxine treatment and outcome
Information regarding both pyridoxine treatment and outcome was available for 66 dogs, of which pyridoxine was administered to 35. Pyridoxine was administered IV to 28 dogs, orally to 6 dogs, and rectally to 2 dogs (1 dog was administered pyridoxine by both the IV and rectal route). All dogs that were treated with pyridoxine had seizures, and pyridoxine administration was significantly (P < 0.001) associated with dogs being classified as survivors. The majority (33/35 [94%]) of dogs that developed seizures and received pyridoxine were classified as survivors. The 2 dogs with seizures that were treated with pyridoxine and classified as nonsurvivors were euthanized. One dog was euthanized after 4 days of pyridoxine treatment because it remained in a comatose state, and the other was euthanized for unknown reasons. Of the 31 dogs that did not receive pyridoxine, 26 (84%) had seizures and 5 (16%) did not. Of the 26 dogs that had seizures and were not treated with pyridoxine, 11 (42%) were classified as survivors and 15 (58%) were classified as nonsurvivors. All 5 dogs that did not have seizures and were not treated with pyridoxine were classified as survivors. Dogs that received pyridoxine IV were approximately 29 times (OR, 28.7; 95% CI, 6.5 to 207.8; P < 0.001) as likely to survive as dogs that were not treated with pyridoxine by the IV route.
Outcome was available for 87 of the 137 (63.5%) dogs. The majority (61/87 [70%]) of dogs were classified as survivors, and of the 26 (30%) dogs that were classified as nonsurvivors, 18 (69%) died and 8 (31%) were euthanized. The median body weight of dogs that were classified as survivors (14.6 kg [32.1 lb]; range, 1.8 to 37.6 kg [4.0 to 82.7 lb]) was significantly (P = 0.006) greater than that of dogs that were classified as nonsurvivors (4.5 kg [9.9 lb]; range, 1.2 to 45 kg [2.6 to 99 lb]). Survival was also significantly associated with the dose of isoniazid ingested (P = 0.036) and treatment with pyridoxine (P < 0.001). The median dose of isoniazid ingested by dogs that were classified as survivors (50.4 mg/kg [22.9 mg/lb]; range, 2.0 to 666.6 mg/kg [0.91 to 303.0 mg/lb]) was significantly less than that ingested by dogs classified as nonsurvivors (132.7 mg/kg [60.3 mg/lb]; range, 29.4 to 1,061.9 mg/kg [13.4 to 482.7 mg/lb]). The lowest dose of isoniazid ingested by a dog that subsequently died was 29.4 mg/kg. Among the dogs that were not treated with pyridoxine, the median dose of isoniazid ingested by dogs that were classified as survivors (19.8 mg/kg) was significantly (P = 0.002) less than that ingested by dogs that were classified as nonsurvivors (132.7 mg/kg). Additionally, among the dogs classified as survivors, the median dose of isoniazid ingested by dogs that were treated with pyridoxine (66.4 mg/kg [30.2 mg/lb]) was significantly (P = 0.002) greater than that of dogs that were not treated with pyridoxine. Age (P = 0.55) and sex (P = 0.32) were not significantly associated with outcome.
Discussion
The use of isoniazid for the treatment of human patients poses a risk for inadvertent exposure of dogs to the drug. For dogs that weigh < 10 kg (22 lb), ingestion of one 300-mg tablet of isoniazid will cause severe clinical signs and potentially death. Most dogs that were inadvertently exposed to isoniazid in the present study had clinical signs attributable to CNS hyperexcitability similar to those previously described for dogs6,16 and human patients7–9 with acute isoniazid toxicosis. Although body weight and the amount of isoniazid ingested were significantly associated with survival in the present study, those criteria should not be used as the sole predictors of outcome. In the present study, the development of seizures was a significant predictor of outcome; 36 of 104 (35%) dogs that developed seizures died or were euthanized, whereas all 28 dogs that did not develop seizures survived. Moreover, dogs that developed seizures and were treated with pyridoxine IV were 29 times as likely to survive as were dogs that developed seizures and were not treated with pyridoxine. That finding supported the results of other studies15,16 in which IV administration of pyridoxine was an effective antidote for isoniazid toxicosis in clinically affected dogs and rats.
Isoniazid ingestion results in CNS hyperexcitability owing to a decrease in GABA concentrations subsequent to pyridoxine inhibition and enhanced elimination. Isoniazid affects pyridoxine metabolism. It binds with P5P, the active form of pyridoxine, to form isonicotinilhydrazide, which is readily excreted in the urine and leads to pyridoxine depletion. Isoniazid derivatives also inhibit pyridoxal phosphokinase, which is an important enzyme in the metabolism of pyridoxine to P5P. Pyridoxal-5-phosphate is an important cofactor of glutamic acid decarboxylase, which is responsible for the synthesis of GABA from glutamic acid.14 Therefore, depletion of P5P, whether because of enhanced excretion or inhibited production, decreases GABA synthesis. Overall, the toxic mechanisms of isoniazid and its derivatives cause a decrease in GABA concentration and uninhibited CNS excitation, which can culminate in life-threatening seizures. In addition to causing CNS signs, isoniazid and its major metabolites also have a role in other clinical syndromes in both dogs and humans such as a decrease in lactate metabolism, mild self-limited hepatotoxicosis, and rhabdomyolysis.15,17–20 Metabolic acidosis and hyperlactatemia are common sequelae of isoniazid toxicosis and are thought to be caused by both seizure activity and indirect inhibition of the conversion of lactate to pyruvate by the liver.14,18
In the present study, seizures and death were reported in dogs that ingested isoniazid at estimated doses as low as 19.8 and 29.4 mg/kg, respectively. Given the severity of clinical signs, high likelihood of seizures, and poor prognosis associated with seizure-like signs, dogs with known or suspected isoniazid ingestion at doses ≥ 20 mg/kg (9.1 mg/lb) or that have substantial clinical signs of toxicosis should be immediately treated with pyridoxine, the antidote. Pyridoxine can be acquired from human hospitals or pharmacies. For dogs with isoniazid toxicosis, pyridoxine should be diluted to a 5% to 10% solution and administered IV over 30 to 60 minutes at a dose equal to the amount of isoniazid ingested on a gram-per-gram basis6,8,16,21–23 or, if the amount of isoniazid ingested is unknown, beginning at a dose of 70 mg/kg (32 mg/lb).6,14 In the present study, clinicians frequently reported that seizures in dogs with isoniazid toxicosis ceased immediately after initiation of IV administration of pyridoxine. Numerous human patients with isoniazid toxicosis have remained in a prolonged comatose state following pyridoxine administration, which was reversed by the administration of an additional 3 to 5 g of pyridoxine.24 Because the injectable formulation of pyridoxine may not be readily available to veterinarians, rectal administration of crushed pyridoxine tablets has been anecdotally reported to be of some benefit for the treatment of dogs with isoniazid toxicosis. However, pyridoxine is slowly absorbed by the rectal mucosa, and rectal administration should not be substituted for IV administration unless the IV formula cannot be obtained.
Pyridoxine has a wide margin of safety, but rare adverse effects, such as peripheral neuropathy, associated with excessive pyridoxine administration have been reported in humans, rats, and dogs.25–27 In dogs, oral administration of pyridoxine at a dose of 200 mg/kg/d (91 mg/lb/d) for 40 days has been associated with ataxia and muscle weakness.26 However, dogs with acute isoniazid toxicosis do not require long-term treatment with pyridoxine, and the risks of pyridoxine-induced adverse effects in such dogs are generally considered minimal.
In addition to pyridoxine, dogs with acute isoniazid toxicosis should undergo appropriate decontamination, supportive care, and monitoring. Decontamination via emesis should only be performed within 1 hour after isoniazid ingestion or for patients in which isoniazid ingestion is known or strongly suspected but that have not yet developed clinical signs of toxicosis. Emesis should not be performed in patients with clinical signs because of the risk for aspiration of gastric contents and possible induction of seizure activity. Gastric lavage with the patient anesthetized and its airway protected is another option for decontamination. Oral administration of activated charcoal with a cathartic reduces absorption of isoniazid in humans, rats, and rabbits if given within 1 hour after ingestion of the drug28,29 and might have a similar effect in dogs. Decontamination or oral administration of activated charcoal was performed for very few dogs evaluated in the present study because most of the dogs had impaired neurologic function and the risks associated with such treatment were considered too great. Supportive care may include fluid therapy, antiemetics (to prevent secondary aspiration pneumonia), thermoregulatory measures, and nursing care. Intravenous fluid therapy is imperative for dogs with clinical signs of isoniazid toxicosis because they are often hyperthermic, have CNS dysfunction which precludes oral fluid intake, and are hemodynamically unstable at the time of diagnosis and commonly develop metabolic acidosis and hyperlactatemia. The purpose of IV fluid therapy is to prevent dehydration, treat hypotension, enhance perfusion, and potentially facilitate urinary elimination of isoniazid. Fluid therapy should be tailored to each patient on the basis of hemodynamic variables such as blood pressure, lactate concentration, pulse quality, capillary refill time, and heart rate. Monitoring of critically ill patients with isoniazid toxicosis should include frequent assessment of blood pressure, ECG, and other cardiac, neurologic, and respiratory variables. Blood gas analysis should be performed as necessary to assess the patient's acid-base status, oxygenation, and ventilatory function.
In dogs, a hallmark of isoniazid toxicosis is seizures that are refractory to traditional anticonvulsants. Isoniazid-induced seizures are difficult to control because most veterinary hospitals do not keep pyridoxine on hand, and affected patients may need to be administered a constant rate infusion of propofol or otherwise anesthetized to control seizures until pyridoxine can be obtained. Traditional anticonvulsants such as diazepam, midazolam, levetiracetam, and phenobarbital should be used in conjunction with pyridoxine because they synergistically improve GABA function in the CNS.15 Although those convulsants are readily available to veterinarians, they must be administered in conjunction with pyridoxine because they will not prevent or ameliorate isoniazid-induced seizures when administered alone. Pyridoxine is the antidote for isoniazid toxicosis and is necessary for the successful management and treatment of severely affected patients. Results of 1 study15 indicate IV administration of diazepam (1 mg/kg [0.45 mg/lb]) in combination with pyridoxine at an amount equal to that of isoniazid ingested provides synergistic protection against seizures and death, whereas administration of pyridoxine alone prevents death but does not eliminate seizures. In that study,15 IV administration of diazepam alone failed to prevent or stop isoniazid-induced seizures at all doses evaluated but did provide some protection against death at a dose of 6 mg/kg (2.7 mg/lb); however, that dose was associated with severe CNS depression (eg, respiratory depression and coma). Many dogs evaluated in the present study were administered a benzodiazepine followed by a barbiturate (typically phenobarbital) to control seizures with little success. When pyridoxine was not readily available, the use of propofol or isoflurane to induce and maintain anesthesia was anecdotally successful in reducing or resolving isoniazid-induced seizures in some dogs.
The ASPCA APCC records indicated that many dogs with isoniazid toxicosis evaluated in the present study had abnormally high alanine aminotransferase and alkaline phosphatase activities indicative of hepatopathy, findings that were similar to those of other studies involving dogs6 and human patients8 with isoniazid toxicosis. It is believed that isoniazid induces oxidative stress, mitochondria dysfunction, and apoptosis, which result in hepatopathy.19 The hepatotoxicity of isoniazid for any species or individual is dependent on the activity of NAT2 and cytochrome P450 2E1, the 2 primary enzymes responsible for metabolizing the drug.19 Dogs inherently lack NAT2, which is responsible for the metabolism of isoniazid to acetyl isoniazid and the acetylation of toxic acetyl hydrazine to nontoxic diacetyl hydrazine. Glutathione depletion is believed to be involved in the development of oxidative stress, and administration of glutathione precursors such as N-acetylcysteine and other hepatoprotectants, such as S-adenosylmethionine, may be indicated for patients with severe hepatopathy. For most patients, isoniazid-induced hepatopathy is mild and self-limiting and typically resolves within 6 to 7 days after ingestion of the drug.19 Regardless, liver enzyme activities should be serially monitored in all patients with isoniazid toxicosis until they are within reference limits. For the dogs with abnormally increased activities of hepatic enzymes evaluated in the present study, no clinically relevant sequelae were reported, and some received supportive treatment with S-adenosylmethionine, the clinical benefit of which was unknown.
Some patients with isoniazid toxicosis may benefit from hemodialysis. Although hemodialysis has been used to successfully treat isoniazid toxicosis in human patients,30,31 to our knowledge, the use of hemodialysis for the treatment of isoniazid toxicosis in veterinary patients has not been described. However, routine use of hemodialysis for the treatment of patients with isoniazid toxicosis may not be necessary owing to the short half-life of the drug and the availability of an effective antidote (pyridoxine).
Patients with isoniazid toxicosis should also be monitored for rhabdomyolysis, a condition that, albeit infrequent, has been reported in both humans and dogs following isoniazid ingestion.17,19 A 6-month-old Rottweiler developed acute kidney injury subsequent to rhabdomyolysis after a prolonged period of isoniazid-induced seizures.17 The authors of that report17 postulated that the release of myoglobin subsequent to prolonged seizure activity resulted in acute renal tubular necrosis and injury. Seizure-induced rhabdomyolysis, although rare, can cause acute kidney injury and should be on the rule-out list for any patient that has a substantial increase in creatine kinase activity and myoglobinuria following prolonged seizure activity.
We had only limited access to the medical records for most dogs evaluated in the present study. Therefore, it was not possible to determine why some dogs that ingested high doses of isoniazid survived without being treated with pyridoxine, whereas dogs that ingested lower doses of isoniazid did not survive regardless of whether they received pyridoxine. It is possible that the time between isoniazid ingestion and examination by a veterinarian or contact with the APCC differed among dogs. The duration from isoniazid ingestion to veterinary examination was not routinely recorded for most dogs and was not evaluated in the present study. Isoniazid ingestion was not observed for 109 of the 137 (79.6%) dogs evaluated in the present study, and it is likely that the onset of clinical signs of toxicosis was also unobserved in those dogs. The owners of many dogs evaluated in this study found their pet seizing; therefore, the duration of that seizure activity was unknown, and many dogs died before they could receive medical treatment. It seems safe to assume that the longer patients have clinical signs of isoniazid toxicosis, especially seizures, before treatment, the more likely they are to die regardless of the dose of the drug ingested or treatment administered.
Limitations of the present study included its retrospective nature, a small sample size, and the lack of complete medical records for all cases. Because of the retrospective nature of the study, information regarding diagnostic test results, treatments administered, and hospitalization data (eg, physiologic variables) was not uniformly available for all cases. The retrospective nature of the study also precluded implementation of a standardized treatment protocol for each case. Despite those limitations, results indicated that survival of patients with isoniazid toxicosis was positively associated with body weight and treatment with pyridoxine and negatively associated with the dose of isoniazid ingested and the presence of seizures. The findings of this study also supported the current recommendation to treat dogs with isoniazid toxicosis with a 5% to 10% solution of pyridoxine at a dose equivalent to the amount of isoniazid ingested on a gram-per-kilogram basis, IV, over 30 to 60 minutes.
Isoniazid is an antimycobacterial agent commonly used to treat tuberculosis in human patients and can cause acute toxicosis in dogs following inadvertent ingestion. Many veterinarians are unfamiliar with isoniazid; therefore, rapid diagnosis of toxicosis, consultation with the ASPCA APCC, and prompt treatment of affected dogs with pyridoxine and other necessary supportive care are imperative for achieving a successful outcome.
Acknowledgments
The authors declare that there were no conflicts of interest.
Footnotes
Microsoft Excel, Microsoft Corp, Redmond, Wash.
ABBREVIATIONS
| APCC | Animal Poison Control Center |
| ASPCA | American Society for the Prevention of Cruelty to Animals |
| CI | Confidence interval |
| GABA | γ-Aminobutyric acid |
| NAT2 | N-acetlytransferase 2 |
| P5P | Pyridoxal-5-phosphate |
References
1. Castagnino RD, Barsallo J, Tabacchi L. Isoniazid and bovine tuberculosis Rev Invest Pecu 1973; 2:161–167.
2. Wolf RH, Gibson SV, Watson EA, et al. Multidrug chemotherapy of tuberculosis in rhesus monkeys? Lab Anim Sci 1988; 38:25–33.
3. van Zyl L, du Plessis J, Viljoen J. Cutaneous tuberculosis overview and current treatment regimens Tuberculosis (Edinb) 2015; 95:629–638.
4. Parekh MJ, Schluger NW. Treatment of latent tuberculosis infection Ther Adv Respir Dis 2013; 7:351–356.
5. CDC. Tuberculosis (TB). Available at: www.cdc.gov/tb/publications/factsheets/statistics/tbtrends.htm. Accessed Jan 1, 2016.
6. Villar D, Knight MK, Holding J, et al. Treatment of acute isoniazid overdose in dogs? Vet Hum Toxicol 1995; 37:473–477.
7. Tibussek D, Mayatepek E, Felix D, et al. Status epilepticus due to attempted suicide with isoniazid? Eur J Pediatr 2006; 165:136–137.
8. Topcu I, Tentur EA, Kefi A, et al. Seizures, metabolic acidosis and coma resulting from acute isoniazid intoxication? Anaesth Intensive Care 2005; 33:518–520.
9. Temmerman W, Dhondt A, Vandewoude K. Acute isoniazid intoxication: seizures, acidosis and coma Acta Clin Belg 1999; 54:211–216.
10. Weber WW, Hein DW. Clinical pharmacokinetics of isoniazid Clin Pharmacokinet 1979; 4:401–422.
11. Isoniazid [package insert] New York: Eon Labs Inc, 2016.
12. Ellard GA. Variations between individuals and populations in the acetylation of isoniazid and its significance for the treatment of pulmonary tuberculosis Clin Pharmacol Ther 1976; 19:610–625.
13. Trepanier LA, Ray K, Winand NJ, et al. Cytosolic arylamine N-acetyltransferase (NAT) deficiency in the dog and other canids due to an absence of NAT genes? Biochem Pharmacol 1997; 54:73–80.
14. Lheureux P, Penaloza A, Gris M. Pyridoxine in clinical toxicology: a review Eur J Emerg Med 2005; 12:78–85.
15. Chin L, Sievers ML, Laird HE, et al. Evaluation of diazepam and pyridoxine as antidotes to isoniazid intoxication in rats and dogs? Toxicol Appl Pharmacol 1978; 45:713–722.
16. Chin L, Sievers ML, Herrier RN, et al. Potentiation of pyridoxine by depressants and anticonvulsants in the treatment of acute isoniazid intoxication in dogs? Toxicol Appl Pharmacol 1981; 58:504–509.
17. Haburjak JJ, Spangler WL. Isoniazid-induced seizures with secondary rhabdomyolysis and associated acute renal failure in a dog J Small Anim Pract 2002; 43:182–186.
18. Chin L, Sievers ML, Herrier RN, et al. Convulsions as the etiology of lactic acidosis in acute isoniazid toxicity in dogs? Toxicol Appl Pharmacol 1979; 49:377–384.
19. Hassan HM, Guo H, Yousef BA, et al. Hepatotoxicity mechanisms of isoniazid: a mini-review? J Appl Toxicol 2015; 35:1427–1432.
20. Panganiban LR, Makalinao IR, Corte-Maramba NP. Rhabdomyolysis in isoniazid poisoning J Toxicol Clin Toxicol 2001; 39:143–151.
21. Brown A, Mallett M, Fiser D, et al. Acute isoniazid intoxication: reversal of CNS symptoms with large doses of pyridoxine? Pediatr Pharmacol (New York) 1984; 4:199–202.
22. Orlowski JP, Paganini EP, Pippenger CE. Treatment of a potentially lethal dose of isoniazid ingestion Ann Emerg Med 1988; 17:73–76.
23. Wason S, Lacouture PG, Lovejoy FH Jr. Single high-dose pyridoxine treatment for isoniazid overdose JAMA 1981; 246:1102–1104.
24. Brent J, Vo N, Kulig K, et al. Reversal of prolonged isoniazid-induced coma by pyridoxine? Arch Intern Med 1990; 150:1751–1753.
25. Schaumburg H, Kaplan J, Windebank A, et al. Sensory neuropathy from pyridoxine abuse—a new megavitamin syndrome? N Engl J Med 1983; 309:445–448.
26. Phillips WE, Mills JH, Carbonneau SM, et al. Subacute toxicity of pyridoxine hydrochloride in the Beagle dog? Toxicol Appl Pharmacol 1978; 44:323–333.
27. Krinke G, Schaumburg HH, Spencer PS, et al. Pyridoxine megavitaminosis produces degeneration of peripheral sensory neurons (sensory neuronopathy) in the dog? Neurotoxicology 1981; 2:13–24.
28. Siefkin AD, Albertson TE, Corbett MG. Isoniazid overdose: pharmacokinetics and effects of oral charcoal in treatment Hum Toxicol 1987; 6:497–501.
29. Ofoefule SI, Onuoha LC, Okonta MJ, et al. Effect of activated charcoal on isoniazid absorption in rabbits? Boll Chim Farm 2001; 140:183–186.
30. Cash JM, Zawada ET Jr. Isoniazid overdose Successful treatment with pyridoxine and hemodialysis. West J Med 1991; 155:644–646.
31. Skinner K, Saiao A, Mostafa A, et al. Isoniazid poisoning: pharmacokinetics and effect of hemodialysis in a massive ingestion? Hemodial Int 2015; 19:E37–E40.
