Flock and Herd logo

ARCHIVE FILE

This article was published in 1963
See the original document

INSTITUTE OF INSPECTORS OF STOCK OF N.S.W. YEAR BOOK.

Changing Concepts With Coliform Diseases

K. G. JOHNSTON, B.V.Sc., Dip. Bact. (Lond.) Senior Clinical Pathologist, University of Sydney, School of Veterinary Science, Rural Veterinary Centre, Camden, N.S.W.

INTRODUCTION

Some coliforms are common inhabitants of the large bowel. They are not infrequent post-mortem invaders. From time to time attention has been focussed on the possible pathogenicity of strains of E. coli. How could these be distinguished from non-pathogenic types? Cultural and biochemical methods were not sufficient to give the degree of precision needed. For these reasons there has been a reluctance in the past to ascribe to them any aetiological role in disease.

HISTORICAL

As early as 1891 Jensen, in Denmark, thought that some strains of E. coli were potential pathogens in the calf. In the early 1920's Theobald Smith and co-workers published a series of papers in the Journal of Experimental Medicine which claimed a causative role for strains ofE. coli in the so-called "white scours" of calves. This and related work was accepted in veterinary circles but progress in the next twenty years was relatively slow (for a detailed review see Lovell, 1955).

Meanwhile P.B. White in Britain and Kauffmann in Scandinavia were working on the classification of those recognised enteric pathogens, the Salmonella group. By 1934 the International Association of Microbiologists had recommended the adoption of their terminology. Using the White-Kauffmann scheme, the Salmonellae could be typed into groups on the basis of their somatic O antigens and specifically on the basis of their flagellar H antigens.

Some time elapsed before the White-Kauffmann approach to the typing of Salmonellae was applied toE. coli. The first papers appeared in Scandinavia at about the end of World War II. At the present time about 150 O antigen groups have been characterised, 86 K (capsular or envelope, divided again into three types, L, A and B) antigens and fifty flagellar H antigens. Within any O group differentiation depends on K and H antigens. For example, Wright et al.. (1953) during the investigation of four ward outbreaks of infantile gastroenteritis, found the 055 : B6 as responsible in each outbreak but that three types, viz., 2, 6 and 7 were present at various times, each due to the admission of a single case to the ward.

The technical advances made with antigenic typing enabled the precise recognition of certain coliform types in association with various conditions. In 1947. Kauffmann reported the progress he had made and pointed out that groups 2, 4, 6, 8 and 9 were commonly found associated with appendicitis, cystitis, peritonitis and pyelitis. He further suggested that these serotypes could be important in their aetiology.

At about the same time in London, Bray (1945) found a particular coliform in association with epidemic diarrhoea of the new born. Similar observations were made in Aberdeen by Giles and Sangster (1948). The organism concerned was O 111. Soon many other reports appeared with O 111 and O 55 predominating. Later O 119 and O 26 were recognised.

There was controversy for a time. Were these serotypes a primary cause of infantile diarrhoea? Gradually the position cleared, and, on the basis of epidemiological studies, investigation of the ecology of these serotypes, and limited feeding trials, the evidence was gradually accepted. It should be understood, however, that the importance now attached to pathogenic E. coli serotypes in this respect does not exclude viral agents from a role in some neo-natal diarrhoea, especially in the calf. In infants, however, Young et al.. (1959), rarely found recognisable enteroviruses in patients with diarrhoea attributable to specific E. coli serotypes.

Concurrently with the work on infantile diarrhoea E. coli strains from calf colibacillosis came under scrutiny in Scandinavia and subsequently in other places. Certain serotypes have emerged as being important. It has also been shown that as far as calf colibacillosis is concerned, the protective effect of colostrum is due to its content and type of specific E. coli antibody. (see Briggs et al. 1961; Ingram et al.., 1956, and also the recent paper by Williams Smith, 1962, on the relationship of colostrum to calf survival in the face of neo-natal diarrhoea).

This then is the general background to our present knowledge of the role of pathogenic E. coli serotypes in diseases of man and domestic animals. It has depended on the development of techniques for detailed antigenic analysis, allowing precise identification of a strain. Many but not all of the pathogenic serotypes are haemolytic on blood agar.

The days are clearly gone when the bacteriologist casts aside all coliform cultures as "post-mortem invaders."

Present knowledge of coliform diseases is very briefly outlined below:

SWINE:

There are three conditions in pigs in which coliforms are important. These are (1) diarrhoea in baby piglets analogous to neo-natal diarrhoea in calves and infants (see Saunders et al. 1960), (2) oedema disease (see Sojka, et al. 1957), (3) enteritis in weaned pigs (see Richards and Frazer, 1961, and Gregory, 1962).

There are three groups principally concerned, viz., O 138, O 139 and O 141. Sojka et al. (1960) found O 141 predominating in enteritis and O 139 in oedema disease. This is not always so. In Ontario, Richards and Frazer found O 138 predominant in enteritis. At Camden we have never encountered oedema disease. Heavy pure growths of haemolytic E. coli have, however, been isolated from acute enteritis of 10-16 week old pigs. These were typed at Weybridge by Dr. Sojka and found to be O 141 K85 (Johnston, to be published, Aust. vet. J.).

O 138, O 189 and O 141 are almost host specific for the pig. They are infrequently encountered in other species. Their precise role in the pathogenesis of oedema disease is still the subject of investigation. Present evidence favours an enterotoxaemic hypothesis.

Oedema disease has been described in South Australia but the serotypes concerned were not determined (Cameron - Stephen 1963).

DOGS

The most frequently encountered O groups have been O2 and O6 in association with such conditions as endometritis, cystitis and pneumonia (Mansson, 1962). Coliforms are common canine urogenital pathogens and we have had haemolytic coliforms in acute haemorrhagic pleuro-pneumonia in a dog. This strain has not as yet been serotyped. CATTLE.

(A) CALF. A growing list of serotypes has been incriminated in neo-natal diarrhoea and septicaemia in the calf. These include O groups 35, 78, 137, 26, 8, 119, 9, 86, 45, 15, 115, 103 and 117. In Britain, Rees (1968) found O35, O78, K80 and O137 to be associated with rapidly fatal infections. Glantz (1960) in Pennsylvania produced typical fatal diarrhoea in colostrum deprived calves by feeding O26 B6, O8 and O119.

We have encountered haemorrhagic coliform enteritis in a calf analogous to that seen in some of our pigs. (Note. The role of viral agents in calf diarrhoea is, at the present time, complex. For a discussion of the Washington State College work on this problem, see Moll, 1957).

(B) ADULTS. The groups reported have mostly been from coliform mastitis and include O78 K80, O9, 24, 86, 81, 21, 6, 8 and O2. SHEEP.

The problem at the moment has been confined to lambs up to about eight weeks of age. There may be diarrhoea comparable to coliform diarrhoea in the calf and piglet. In peracute outbreaks, there has been sudden death with massive serofibrinous effusions in serous cavities. In such outbreaks, or those which are less acute, survivors are likely to show focal localisation in joints and meninges with the appropriate clinical signs.

Roberts (1957, 1958) described such outbreaks in West Australia. The group was O78 but the K antigen was not given. Charles (1957) noted coliform meningoencephalitis in lambs (untyped).

Kater, Davis, Haughey and Hartley (1963) have described similar outbreaks in New Zealand. The serotppe was O78 K80.

Similar reports have come from Britain (Terlecki and Shaw 1959, Rees, 1958 and Hughes, 1962).

As with the pig and calf we have seen acute haemorrhagic enteritis in a ram from which haemolytic coliforms were recovered in heavy pure growth.

With regard to ruminants it might be noted that O78, K80 has an established place as a pathogen. It can hardly be regarded as a "normal" gut inhabitant.

ASSESSMENT OF COLIFORM SIGNIFICANCE

A. J. Stevens in F.A.O. Animal Health Branch Monograph No. 3 entitled "A Laboratory Handbook of Veterinary Diagnostic Procedures," Rome, 1961, lists the following factors of importance in assessing the significance of coliform isolates:

(1) Freshness of carcass: within three hours if possible but visceral invasion of gut coliforms usually not seen within 24 hours.

(2) Type of outbreak: young animals mostly; more often in intensive systems.

(3) Source of isolation: determine if present in sites other than intestine and especially higher levels of small intestine.

(4) Abundance and purity: significant growths usually heavy and relatively pure.

(5) Absence of other pathogens: exclude other possibilities although there may be co-existent condition, e.g., anaemia and nutritional deficiencies.

(6) Haemolysis on blood agar: NOT the lone criterion of pathogenicity but helpful when seen. Non-haemolytic types may be pathogens.

(7) Biochemical reactions: Some of these help, especially the Eijkman test and some fermentations.

(8) Serology: Every effort should be made to ascertain the serotype as certain serotypes are fairly constantly associated with particular diseases.

IN RETROSPECT.

It has not been the intention in this paper to give a detailed account of the diseases mentioned but rather, a brief outline of how concepts have changed and why.

In the short bibliography cited there are two brief reviews which contain general discussion of recent developments viz., Kramer (1961) and Glantz (1960)

REFERENCES:

  1. Bray, J. (1945) — J. Path. Bact., 57: 239
  2. Briggs, et al. (1951) — Brit. J. Nutrit., 5: 349, 356
  3. Cameron-Stephen, I. D. (1963) — Aust. vet. J., 39: 126
  4. Charles, G. (1957) — Aust. vet. J., 33: 329
  5. Giles, C. and Sangster, G. (1949) — J. Hyg., Camb., 46: 1
  6. Glantz, P.J. (1960) — Corn. Vet.. 50: 9
  7. Gregory, D. W. (1962) — J. Amer. vet. Med. Ass., 141: 917
  8. Hughes, L. E. (1962) — Vet. Rec., 74: 350
  9. Ingram, et al. (1956) — J. Path. Bact., 72: 561
  10. Kater, et al. (1963) — N.Z. vet. J., 2: 32
  11. Kauffmann, F. (1947) — J. Immunol., 57: 71
  12. Kramer, T. (1961) — Canad. vet. J., 2: 326
  13. Lovell, R. (1955) — Vet. Revs. Annot., 1: 1
  14. Mansson, I. (1962) — Acta Veterinaria Scandinavica, 3: 65
  15. Moll, T. (1957) — Vet. Med., 52: 115
  16. Richards, W. P. C. and Frazer, C. M. (1961) — Corn. Vet., 51: 245
  17. Rees, T. A. (1958) — )J. comp. Path. Therap., 68: 388, 399
  18. Roberts, D. S. (1957) — Aust. vet. J., 33: 43
  19. Roberts, D. S. (1958) — ibid., 34: 152
  20. Saunders, C. M. et al.., (1960) — Res. Vet. Sci., 1: 28
  21. Smith, H. Williams (1962) — J. Path. Bact., 84: 147
  22. Sojka, W. J. et al. (1957) — Vet. Rec., 69: 293
  23. Sojka, W. J. et al. (1960) — Res. Vet. Sci., 1: 1
  24. Terlecki, S. and Shaw, W. E. (1959) — Vet. Rec., 71: 181
  25. Wright, J. et al. (1953) — Am. J. Hyg., 58: 133
  26. Young, et al. (1959) — Proc. Soc. exp. Biol., N.Y., 100: 579
Site contents Copyright 2006-2026©