The enigma of Pictou cattle disease

The first report of a fatal disease of cattle of unknown etiology in Pictou county, Nova Scotia, appeared in 1880 (1), although the disease had existed for a number of years previously. By September 1881, 1396 animals had been lost, 203 of which had died during the current season (2). Duncan McEachran, principal of the Montreal Veterinary College and an inspector in the Canada Department of Agriculture since 1879 (3), was appointed by the Minister of Agriculture to investigate the disease, and his initial findings were reported in the following year (4). The clinical signs included depression, emaciation, reduced yield of bitter tasting milk, diarrhea, and terminal nervous signs. Death generally occurred within 1 mo of the onset of clinical signs. On postmortem examination the carcass was emaciated with an excess of peritoneal fluid, the abomasum and intestines showed sub-mucosal edema with ulceration, and the liver was firm and pale.

The cause of the disease was undetermined, and remained an enigma for over 20 y, although from the outset there was a strong local belief that it was associated with the presence of the weed Senecio jacobœa (ragwort) on pastures in the area (Figure 1). However, for many years, during which McEachran was the lead investigator, the disease was regarded as contagious, and the control measures applied were appropriate for the control of a contagious disease. It is puzzling that, in spite of investigations by experienced and distinguished pathologists, including William Osler, Wyatt Johnston, and George Adami, it took so long for the cause of the disease to be identified.

Figure 1.

Pasture containing Senecio jacobœa (ragwort) showing the typical yellow flowers. Image kindly provided by Dr. Jeff Caswell, Department of Pathobiology, University of Guelph.

This short paper will examine the cases for contagion or ragwort poisoning as the cause of Pictou cattle disease, and describe the final outcome of the etiological enigma. The reasons for the prolonged delay in determining the cause of the disease will be discussed in the context of the microbiological revolution in the later part of the 19th century.

The case for contagion

While McEachran’s preliminary report (2) failed to determine the nature and cause of the disease, he nevertheless concluded that it was contagious, and recommended that it be controlled by the slaughter and burial or burning of affected animals, by the quarantine of contact cattle, and disinfection of the premises. The owners of affected stock would be compensated for their losses. To enable these recommendations to be implemented they were embodied in an Order in Council under the Animal Contagious Diseases Act of 1869 on May 25, 1882 (4), and they remained the basis for the control of the disease over the next 20 y. This action was taken despite a report by William Osler in the same year which concluded that it was doubtful that the disease was contagious (5). Osler, assisted by William McEachran, a veterinary inspector in the Department of Agriculture, attempted to transmit the disease by inoculation of a calf and a sheep with peritoneal fluid from an affected cow, with negative results. Similarly, a heifer placed in a shed contaminated with excreta from an infected cow also failed to develop the disease. Osler commented that the pathology and epidemiology of Pictou cattle disease resemble to some extent the intestinal form of anthrax, but he failed to find bacilli in the mesenteric blood vessels and lymph nodes, although bacilli were plentiful in the intestinal contents.

McEachran claimed initial success for the policy of control by slaughter and quarantine, since the annual number of deaths declined from 92 in 1882, when the measures were introduced, to 18 in 1886 (6). However, a modest resurgence in the number of cases began in 1888, and the quarantine policy continued to be enforced by William Jakeman, the Canada Department of Agriculture veterinary officer in Halifax (7). Jakeman also conducted further unsuccessful attempts to transmit the disease by inoculation and cohabitation. In the following year the Minister of Agriculture, impressed by Jakeman’s findings, commented that since the disease was not contagious it could not be dealt with under the Animal Diseases Act, and should be the responsibility of the local authorities and the provincial government (8). However, the Minister’s comment appeared to fall upon deaf ears since McEachran, appointed Chief Inspector in 1892, continued to pursue the policy of slaughter and quarantine, apparently to no avail since a record 363 cattle affected with the disease were slaughtered in 1892 (9).

In 1891, an experimental station for the investigation of Pictou cattle disease was established at Stellarton, Nova Scotia, on a farm adjoining the town, and Wyatt Johnston, a medical pathologist at McGill University, was appointed to participate in the investigation (10). Laboratory facilities were provided for Johnston at the Pictou Academy, and he reported his findings in the following year (11). Johnston conducted 34 autopsies on affected animals, and reported, for the first time, marked cirrhosis of the liver in advanced cases, an important finding which Osler had not observed in his earlier investigations. Bacteriological examination of the tissues yielded negative results. In addition, Johnston made several attempts to transmit the disease experimentally. These included the feeding of feces or milk from affected cattle to healthy animals, and the inoculation of blood, serum, or ascitic fluid intraperitoneally or intravenously into healthy calves — all with negative results. While Johnston concluded that there was no evidence that the disease was contagious, he nevertheless recommended continuation of the policy of control by slaughtering and burying diseased animals, with compensation of their owners.

In spite of the continued control program, there was only a modest decline in the number of cattle slaughtered — 125 in 1893, 105 in 1894, and 81 in 1895 (12). George Adami of McGill University was appointed in July 1894 to investigate the disease in collaboration with George Townsend, the veterinary inspector at New Glasgow. In November of the same year Adami reported that his preliminary findings suggested that Pictou cattle disease was due to a bacterial infection although the bacillus which he isolated, while fatal on inoculation into rabbits, failed to reproduce the disease in 2 sheep and 4 cattle (13). In spite of the inconclusive nature of these results, McEachran stated that Adami had discovered the microorganism which was the cause of Pictou cattle disease, and that his findings were supportive of the departmental policy for the control of the disease (12). However, losses from the disease in the field continued unabated, and Adami pursued his research on a possible relationship between bacterial infection and cirrhosis of the liver in both cattle and humans. Adami submitted a final report in 1901 (14), in which he identified the bacteria which he had isolated from the cirrhotic livers and abdominal lymph nodes of cases of Pictou cattle disease as colon bacilli normally present in the large intestine. He postulated that the bacilli had invaded the liver from the intestinal tract following inflammation of the latter by some other irritant. Although the bacilli were pathogenic for rabbits, guinea pigs, and mice, they failed to induce cirrhosis of the liver, and therefore did not fulfil Koch’s postulates. McEachran was obliged to abandon his initial enthusiasm for the significance of the bacteria isolated from Pictou cattle disease, and he concluded in 1900, when 149 diseased animals were slaughtered, that efforts to determine the cause had failed (15). His last report was in 1901 since he retired in the following year, when he was replaced as Chief Veterinary Inspector by John Rutherford, who in 1904 became the first Veterinary Director General (3).

Rutherford’s appointment brought a more open mind to the problem of Pictou cattle disease, and in his first report he questioned the wisdom of the current control measures (16). He pointed out the lack of evidence of contagion, and considered the compensation of owners of animals affected with the disease to be unjustified. W.H. Pethick, a veterinary inspector from Prince Edward Island who had completed training in pathology at McGill University was enlisted to collect postmortem specimens in the field. The specimens were then forwarded to Charles Higgins, the first full-time veterinary pathologist in the Department of Agriculture (3) for further examination in the Biological Laboratory in Ottawa in the hope that Higgins might be successful where others had failed in determining the cause of the disease. In 1903 Higgins confirmed that cirrhosis of the liver was present in 8 suspected cases of Pictou cattle disease (17), and that, like Adami he had cultivated Bacillus coli from the livers and mesenteric lymph glands. He concluded that there was no infectious cause of the disease, and the case for contagion was finally abandoned in 1904 when Pethick reported his negative attempts to transmit the disease in contaminated stables and by direct contact (18). In 1905, following negative attempts to transmit the infection by inoculating blood or ascitic fluid into 10 cattle, Rutherford recommended that Pictou cattle disease be removed from the list of those controlled under the Animal Contagious Diseases Act (19).

The case for ragwort poisoning

Although McEachran indicated in his first report in 1881 (2) that the people of the district believed that Pictou cattle disease appeared after the introduction to the area of the weed Senecio jacobœa or ragwort, known locally as “stinking willy,” the possibility of plant poisoning as the cause of the disease was not taken seriously for at least 20 y. McEachran stated that the Senecio family was not poisonous, and that the more intelligent of the local population failed to see a connection between ragwort and the disease. The water supply was suspected as a cause, but this was discounted in a report by Professor George Lawson of Dalhousie College in Halifax (20) on soil, water, and herbage samples collected in the area, although Lawson did note that the distribution of ragwort was to some extent coincident with prevalence of the disease. The only early experimental study on ragwort was conducted by Osler (5) in conjunction with William McEachran, the local veterinary inspector. They fed the chopped weed in a bran mash to a steer and a heifer for a period of just over 2 mo, with negative results, and Osler concluded that this very limited experiment disposed of the popular notion that the disease was due to ragwort. Osler’s experiment was referred to in Adami’s 1894 report (13) as being most conclusive, and Adami also noted that the distribution of the weed failed to correspond exactly with the occurrence of the disease, and that ragwort was present in other areas of Canada in the absence of the disease. While not specifically implicating ragwort, Johnston postulated poisoning or improper feeding as a predisposing cause of the disease (11). Shortly before he retired, McEachran emphatically reiterated that the belief that the disease was due to eating ragwort was entirely erroneous (15).

The case for ragwort poisoning as the cause of Pictou cattle disease began to receive credence following reports from New Zealand of a similar disease in the Winton district in Southland (14). Although this disease affected horses more frequently than cattle, it resembled the Pictou disease clinically and pathologically, and ragwort was prevalent in the affected area. Initially, J.A. Gilruth, the Chief Veterinary Officer for New Zealand denied a relationship between the weed and the Winton disease, but was later strongly of the opinion that ragwort was the cause of the disease (21). The earliest evidence for the possible toxicity of ragwort came from experiments conducted by Higgins in Ottawa (22). He fed some hay from Pictou to rabbits and guinea pigs, and found that the animals gradually lost weight, and that the guinea pigs showed cirrhosis of the liver and ulceration of the small intestine indistinguishable from the lesions found in cattle.

Higgins’ findings stimulated Rutherford to investigate further the role of ragwort in the etiology of Pictou cattle disease. He established an experimental station at Antigonish, where he planned extensive trials conducted by W.H. Pethick with pathological support from Higgins in Ottawa. Pethick fed 16 cattle, which were kept in a newly constructed stable, with local hay containing liberal amounts of ragwort. Of these, 15 died from Pictou cattle disease and the 16th was found to be suffering from hepatic cirrhosis when it was slaughtered (19). The postmortem findings were confirmed by Higgins, whose report included micrographs of the liver lesions (18). Unfortunately these were not reproduced in the printed report, and the original manuscript copy is no longer available, but the typical hepatic changes associated with pyrrolizidine alkaloid poisoning are illustrated in Figure 2. Thus the enigma surrounding the etiology of the disease was solved by these pioneering experimental studies, which clearly established the toxicity of Senecio jabobœa, within 3 y of Rutherford’s appointment as Chief Veterinary Inspector.

Figure 2.

Portal fibroplasia and bile duct proliferation with impaired regeneration of hepatocytes (megalocytosis) due to pyrollizidine alkaloid poisoning. From Hayes MA. Pathophysiology of the Liver. In: Dunlop RH, Mathieu C-H, eds. Veterinary Pathophysiology. Ames, Iowa: Backwell Publ, 2005:11;394 (Reproduced with permission of John Wiley & Sons Inc.). Image kindly provided by Dr. Tony Hayes, Department of Pathobiology, University of Guelph.

The final experiments conducted by Pethick, under Rutherford’s direction, were to investigate the feasibility of using sheep to rid the infested pastures of ragwort. Sheep were fed ragwort contaminated hay and exposed to the growing weed on pastures, without ill effects (23), although some deaths occurred in pastured sheep after the first frosts in the late fall (24). The latter finding led Pethick to recommend that sheep should be removed from ragwort infested land before the first frosts. He further recommended that in addition to keeping sheep on the infested pastures, there should be improved tillage of cultivated land and scything of the weed on the headlands, fence corners, and roadsides. Finally Pethick delivered a series of 54 lectures on the eradication of ragwort at public meetings in the affected areas of Nova Scotia and Prince Edward Island in 1908 and 1909 (24), and Rutherford reported in 1911 that losses from Pictou cattle disease had become almost unknown (25).

Discussion

Although fiercely defended by McEachren, and supported by his distinguished colleagues from McGill University, the policy of control of Pictou cattle disease by slaughter, quarantine, disinfection, and compensation proved to be a costly and futile endeavor. The compensation payments ranged from $15 000 to $60 000 annually for over 20 y (22), and additional costs were associated with quarantine and disinfection, including the salaries of the inspectors who supervised these procedures. In a brief discussion of the history of Pictou cattle disease, Dukes (26) ascribed the erroneous assumption of an infectious cause to the concurrent microbiological revolution, in which bacteria were identified as the cause of numerous human and animal diseases. In commenting on the discovery by Frank Schofield that a hemorrhagic disease of cattle, previously attributed to a bacterial infection, was associated with the ingestion of moldy sweet clover, Saunders (27) coined the term “bacteriomania” for “a trendy condition which attributes all disease to infectious organisms.” McEachran’s control program for Pictou cattle disease provides an excellent example of this condition, while not detracting from his pioneering work on the control of infectious diseases such as hog cholera, glanders, and bovine tuberculosis (3), his establishment of experimental stations in Stellarton and Outremont, the forerunners of the Biological Laboratory in Ottawa, or his contributions to comparative medicine at the Montreal Veterinary College, which he founded in 1866 (28).

Fortified by the reports of ragwort poisoning in New Zealand, and the results of his own well planned experimental studies, Rutherford solved the etiological enigma of Pictou cattle disease shortly after he succeeded McEachran as the lead investigator, and subsequently established effective control measures for the disease. This was an auspicious start to his career as Veterinary Director General, which included major contributions to disease eradication, the founding of the Canadian federal meat inspection service (29), and the culmination of his long-standing efforts to improve the academic standards at the Ontario Veterinary College (30). CVJ