• Open Access

Arsenic Toxicosis in Cattle: Meta-Analysis of 156 Cases


  • The study has been submitted for presentation at the 2013 ACVIM Forum, Seattle, WA

Corresponding author: S.D. Taylor, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Purdue University, West Lafayette, IN 47907; e-mail: taylo248@purdue.edu.



Arsenic toxicosis is uncommon in cattle and successful treatment is rarely reported.


This analysis reviews all cases of acute arsenic toxicosis in cattle reported in the literature and describes cases from Purdue University that had a favorable outcome. Clinical presentation of the disease, treatments, and variables associated with survival are described.


One hundred and fifty-six cattle with arsenic toxicosis from 16 outbreaks.




The most common clinical signs were sudden death (68%), diarrhea (33%), ataxia (29%), dehydration (22%), and respiratory distress (4%). The most common clinicopathologic abnormalities included azotemia (100%), hematuria (100%), increased liver enzyme activity (86%), and increased hematocrit (60%). One percent of cattle survived and the survival time for nonsurvivors ranged from 20 hours to 21 days. None of the clinical signs or clinicopathologic findings was associated with survival. Treatment was attempted in 24% of cases and was not associated with survival (P = .055), but administration of an antidote and administration of fluids were associated with better outcome (P = .036 and P = .009, respectively). In the animals presented to Purdue University, treatment with IV fluids and sodium thiosulfate resulted in decreased blood arsenic concentrations in all animals (= .009) and a survival rate of 50%.

Conclusions and Clinical Importance

Although acute arsenic toxicosis has a poor prognosis, survival is possible if aggressive fluid therapy and antidotes are administered.


lethal dose 50%

Arsenic is a ubiquitous toxic element that is concentrated in soil and water as a result of industrial activities.[1, 2] It is found in organic, inorganic, trivalent, and pentavalent forms, and combines with many elements such as oxygen, hydrogen, sulfur, nickel, copper, and lead.[2-5] Before the 1960s, when arsenic was used extensively, acute toxicosis by exposure to wood preservative, pesticide, herbicide, fungicide, and paint was not uncommon in ruminants.[6, 7] Acute arsenic toxicosis still occurs when cattle have access to discarded materials, old outbuildings, or old dipping vat areas where the element persists alone or in combination with chromium or lead.[6] The prognosis for acute arsenic toxicosis is considered grave. Death of intoxicated animals occurred in all published reports of confirmed cases of acute arsenic toxicosis.[6-12] Presenting clinical signs often are vague and most affected animals are found dead.[3] The diagnosis usually is made postmortem, and data are limited regarding antemortem diagnosis and treatment of these cases.[3]

This analysis reviews all documented cases of confirmed acute arsenic toxicosis reported in cattle, including affected animals from Purdue University Veterinary Teaching Hospital that survived. Clinical presentation, diagnostic testing, treatment, and outcome are described.

Materials and Methods

All published case reports or case series of acute arsenic toxicosis in cattle were evaluated using searches of PubMed and Google Scholar. Key words included “cattle,” “cow,” “beef,” “bovine,” “arsenic,” “arsenite,” “arsenate,” “toxicosis,” and “poisoning.” Reports from peer-reviewed journals in English from 1941 to 2012 were considered. Environmental analyses and reports concerning chronic arsenic exposure were excluded. Medical records from Purdue University Veterinary Teaching Hospital also were searched for cases of acute arsenic toxicosis in cattle.

Cases were included if a definitive diagnosis of arsenic toxicosis was obtained by identification of toxic concentrations of arsenic in urine, blood, liver, or kidney. Reference ranges, obtained from the Diagnostic Center for Population and Animal Health at Michigan State University, are 0.05–0.17 ppm in urine, <50 ppb in blood, 0.004–0.40 ppm in liver, and 0.018–0.40 ppm in kidney. Animals in which arsenic was only detected in the lumen of the gastrointestinal tract were excluded because no reference range has been established for ingesta. Animals that were experimentally intoxicated were included.

Data collected from cases included signalment, number of animals in the herd, number of animals affected, arsenic origin, route of exposure, clinical signs, duration of clinical signs before presentation, clinicopathologic data, method of diagnosis, treatment, outcome, survival time from time of exposure to death, and necropsy findings.

Cases were grouped by outcome (discharged alive or not) and compared. Statistical analysis was performed by commercially available statistical software (Statistica1). Normality of data was assessed by a Shapiro–Wilk test. Normally distributed data were reported as mean ± SD and compared by a Student's t-test. Nonnormally distributed data were reported as median and range and compared by the Mann–Whitney U-test. Because of the low number of survivors, categorical data were compared with a Fisher's exact test. Kaplan–Meier log-rank survival analysis was used to calculate the median survival time. For all comparisons, P < .05 was considered significant.


One hundred and fifty-six animals met the inclusion criteria. From the literature, 13 reports describing 16 outbreaks of acute arsenic toxicosis in cattle were retrieved.[1, 3, 6-16] In these reports, 1,122 animals were exposed to arsenic, 151 of which demonstrated clinical signs and had a confirmed diagnosis of acute arsenic toxicosis. The Purdue University Veterinary Teaching Hospital medical records search added a herd of 13 animals with 5 confirmed cases. Complete data were not available for all individual cases.

Beef breeds represented 96% of cases (n = 146) and included mixed breed, Angus, Hereford, Angus-Hereford, and Shorthorn. Dairy breeds represented 4% of cases (n = 6). The breed was not reported in 4 cases. Of the 44 cases for which sex was recorded, 95% were female (n = 42) and 5% were male (n = 2). The age at the time of diagnosis ranged from 6 weeks to 6 years with a median of 18 months. The median number of animals exposed in a herd was 39 (range, 11–280) and the median number of animals showing clinical signs per herd was 8 (range, 3–101). No epidemiologic data were associated with survival.

The most common clinical signs were sudden death (68%, n = 106), diarrhea (33%, n = 52), ataxia (29%, n = 46), dehydration (22%, n = 34), respiratory distress (4%, n = 7), decreased milk production (3%, n = 5), and increased salivation (1%, n = 2). Surprisingly, the presence of dehydration based on clinical signs was positively associated with survival (P = .046).

Partial or complete clinicopathologic data were available in 7 cases. The most common abnormalities included azotemia (100%, n = 7), increased aspartate aminotransferase activity (86%, n = 6), hyperbilirubinemia (83%, n = 5), increased creatine kinase activity (57%, n = 4), increased alkaline phosphatase activity (40%, n = 2), hypocalcemia (33%, n = 2), acidemia (33%, n = 2), and increased γ-glutamyl transferase activity (33%, n = 2). Increased hematocrit was noted in 60% of cases (n = 3). Hematuria was reported in all cases in which a urinalysis was performed (n = 5). None of the clinicopathologic data was associated with survival.

Method of diagnosis was reported in 36 cases. Overall, the most common method utilized a silver diethyldithiocarbamate colorimetric procedure (75%, n = 27). In more recent cases, inductively coupled plasma atomic emission spectrometry was used (17%, n = 6). In 23 cases, arsenic concentrations in urine, blood, liver, or kidney were available. Urine was analyzed in 4 cases and was diagnostic in all with a median concentration of 26.8 ppm (range, 3.7–45 ppm; reference range, 0.05–0.17 ppm). Blood samples were obtained in 5 cases and were diagnostic in only 4 cases, with a median concentration of 677 ppb (range, 267–747 ppb; reference range, <50 ppb in blood). Arsenic was measured at necropsy in 15 liver samples, all of which yielded toxic concentrations of arsenic (median, 8.5; range, 3.2–350 ppm; reference range, 0.004–0.40 ppm). Fourteen kidney samples obtained at necropsy were tested for arsenic concentrations, all of which contained toxic concentrations of arsenic (median, 11.2 ppm; range, 1.44–44 ppm; reference range, 0.018–0.40 ppm). Blood and urine arsenic concentrations were not associated with survival.

In 79% of cases described in 2 reports (n = 124), arsenic toxicosis resulted from experimental administration. In 1 report, animals were treated with an arsenic-based pour-on insecticide; and in another report, arsenic was administered through an orogastric tube as 0.4% solution in tap water. In the other cases (21%, n = 32), arsenic toxicosis resulted from accidental environmental exposure. Among these 32 animals, the main sources for ingestion of arsenic included ashes (34%, n = 11), water (28%, n = 9), pesticides (13%, n = 4), fertilizers (13%, n = 4), and 1 case each of paint and sheep dip ingestion. In 2 cases, the source of arsenic was not reported. Accidental arsenic exposure was associated with a better outcome (P = .038) compared with those experimentally intoxicated. In 34% of cases (n = 54, including 24 experimental cases), arsenic was ingested by cattle and in 65% of cases (n = 100, all experimental cases), arsenic toxicosis occurred by transcutaneous absorption. In 2 cases, the route of exposure was not documented. The route of exposure was not associated with survival.

In 24% of cases (n = 37, including 20 experimental cases), treatment was attempted. The most common treatments included administration of an antidote (87% of cases, n = 30, including 20 experimental cases) and administration of oral fluids, IV fluids, or both (41% of cases, n = 15, all natural exposure). Other treatments included antibiotics, laxatives, anti-inflammatory drugs, and probiotics. Interestingly, treatment attempts were not associated with increased survival (P = 0.055). The most commonly administered antidote was sodium thiosulfate (20–40 mg/kg IV q8h and 80 mg/kg PO q24h; 47%, n = 14). Other antidotes administered in 13% of cases each (n = 4) included cysteine (200 mg/kg IV q8h), thioctic acid (50 mg/kg IV q8h), dimercaptopropanol (3 mg/kg IM q4–12h), and a combination of thioctic acid and dimercaptopropanol. There was no association between the type of antidote used and survival, but administration of an antidote was associated with better outcome (P = .036). The most commonly administered type of IV fluid was lactated Ringer's solution at an initial rate of 5 mL/kg/h (27%, n = 4). Other types of fluids included hypertonic saline and 50% dextrose. Oral fluids were given at a rate of 30–40 mL/kg q12h and consisted of tap water with or without electrolyte supplementation. Administration of IV or oral fluids was associated with a better outcome (P = .009). In the 4 animals presented alive to Purdue University Veterinary Teaching Hospital, treatment with oral and IV fluids and sodium thiosulfate resulted in a significant decrease in blood arsenic concentration in all animals (P = .009), 50% of which survived.

Of 156 cases, 1% (n = 2) survived and completely recovered from arsenic toxicosis. Both animals were successfully inseminated after recovery and calved normally the following year. Time of exposure was available in 132 animals. The median survival time was 20 hours (range, 20 hours to 21 days). Two cases had prolonged survival times of 10 and 21 days, respectively. Complications experienced by these 2 animals included, in 1 cow, marked normocytic normochromic anemia because of hemolysis, which resulted in severe icterus (Fig 1) and in another cow, persistent diarrhea associated with melena and intestinal sloughing. Cattle that received an antidote or fluids had a significantly prolonged survival time (P < .0001 and P < .0001, respectively).

Figure 1.

Severe icterus in a cow 12 days after exposure to arsenic, (A) whole body, (B) eye, (C) vulva.

Necropsy was performed in 19% of cases (n = 30) and the lesions were located in the gastrointestinal tract and in the urinary system. Gastrointestinal lesions were described as resembling salmonellosis with dark and putrid intestinal fluid with varying degrees of erosion and submucosal edema of the esophagus, abomasum, jejunum, and rectal folds. Urinary lesions consisted of renal tubular necrosis, suppurative pyelonephritis, and mucosal petechiae of the urinary bladder.


Acute arsenic toxicosis is a rare, sporadic condition in cattle. As observed in this study, ashes from treated or painted wood are the most common environmental source, with ingestion of ashes resulting in disease. The burning of treated lumber results in oxidation and increased concentrations of arsenic.[2, 4, 5, 8, 17-19] Arsenic has 3 oxidative stages, and toxicosis is associated with either the trivalent (arsenite) or pentavalent (arsenate) form. Trivalent arsenic inactivates intracellular sulfhydryl-containing compounds (lipoic acid, alpha-keto oxidases) and therefore inhibits oxidative phosphorylation.[9, 10] The pentavalent form of arsenic induces formation of 1-arseno-3-phosphoglycerate (instead of a 1,3-diphosphoglycerate), which cannot release phosphate for ATP formation.[9, 10] Pentavalent arsenic is reduced to trivalent arsenic in the rumen.[10] Toxic ranges of arsenic after ingestion in cattle are wide with a lethal dose 50% (LD50) of 1–25 mg/kg for trivalent arsenic and a LD50 of 30–100 mg/kg for pentavalent arsenic.[2-5, 10]

Clinical signs of acute arsenic toxicosis in cattle are caused by direct inflammation of gastrointestinal mucosa and vasculature, and by damage to organs with high ATP requirements.[9, 10] Therefore, the most common clinical signs include diarrhea and ataxia as they involve organs with a high metabolic rate. Clinicopathologic findings, such as azotemia and increased liver enzyme activities, are also consistent with damage to organs with high-energy requirements. Interestingly, no clinicopathologic factor was associated with survival, suggesting that the severity of organ insult was not associated with a poor prognosis. A possible explanation of this finding is that only 7 animals had complete or partial clinicopathologic testing and only 2 animals survived, making results difficult to interpret. Another explanation is that animals with clinicopathologic data available were hospitalized and received IV fluids that likely decreased renal injury. In 1 report, an association between severity of clinical signs and hematocrit was suggested.[7] This study did not support such an association, as increased hematocrit was inconsistent and not related to outcome or survival time.

Determination of arsenic concentration in blood and urine was used as an antemortem diagnostic tool. Interestingly, there was no association between blood or urine arsenic concentration and specific clinical signs or outcome. This lack of association has been previously suggested.[7] In 1 case that presented at the Purdue University Veterinary Teaching Hospital, blood arsenic concentration was below the toxic range, despite the identification of toxic arsenic concentrations in urine, kidney, and liver. Therefore, it appears that blood arsenic concentrations can be highly variable and account, to some extent, for the absence of correlation among blood arsenic concentration, clinical signs, and outcome. It also indicates that blood may not be the sample of choice for diagnosis of arsenic toxicosis. In this study, despite the low number of cases, it appears that urine is a more sensitive antemortem sample.

There was a significant increase in survival time and survival rate when an antidote and fluids were administered. Surprisingly, the type of antidote administered did not have a significant effect on survival rate or survival time. This finding was previously demonstrated by Hatch.[7] In that study, 5 different antidotes were administered to animals with acute arsenic toxicosis. There was no difference in the course of disease, and all animals died within 96 hours. In the same study, when treated animals were compared with nontreated controls, there was no difference between groups, suggesting that any antidote did not improve the prognosis. In contrast, this study supports use of an antidote, as an improved survival rate and a longer survival time were found for treated animals. A possible explanation of this discrepancy is that concurrent fluid therapy may have affected the outcome in animals treated with an antidote, but only 4 of 37 treated animals received both an antidote and fluid treatment. Other possible explanations include variation in the dosage of antidote administered or survival time bias because a large number of animals that did not receive an antidote suffered from sudden death. The cases with a favorable outcome were treated with sodium thiosulfate at higher dosages than the regimen described in other reports (40 mg/kg IV q8h and 80 mg/kg PO q24h instead of 20–40 mg/kg IV q8h).

Administration of oral or IV fluids was highly associated with survival. Maintenance of renal perfusion may have led to a more favorable outcome in these animals. This finding has been previously suggested and is further supported by this analysis, which shows that administration of fluids is of paramount importance in the treatment of acute arsenic toxicosis.[7] Although dehydration was positively associated with survival in this study, this likely represents dehydration as a confounder of the beneficial effect of fluid treatment. It is likely that the confounding effect is attributable to the fact that animals that received fluid treatment were classified as dehydrated. Therefore, it is not considered clinically relevant that dehydration is predictive of a positive outcome.

In summary, acute arsenic toxicosis has a poor prognosis as described in previous reports. Arsenic interferes with energy metabolism, resulting in injury to the gastrointestinal tract, central nervous system, liver, and kidneys. This may result in sudden death, diarrhea, and ataxia. The current meta-analysis, including recent cases diagnosed and treated at Purdue University Veterinary Teaching Hospital, suggests that survival of severe arsenic toxicosis in cattle may be possible if an antidote and aggressive fluid treatment are administered. It also serves to remind clinicians and producers that arsenic is still present in the environment and that steps should be taken to minimize exposure of cattle and other animals.


The authors thank C. Hagan and L. Lum for technical assistance, as well as the Diagnostic Center for Population and Animal Health at Michigan State University for toxicology testing.

Conflict of Interest Declaration: The authors disclose no conflict of interest.


  1. 1

    Statistica 11.0; StatSoft, Inc, Tulsa, OK