It is highly likely that the inflammatory bowel diseases are all infections, but the question remains as to the identity of the causative organism or organisms. Accepting that they are infections is more important than knowing the identity of the organisms themselves, or knowing the precise genetic makeup of persons whose response to such infections will be to develop these diseases. Their aetiology is likely to be bacterial rather than viral, and for a number of reasons these bacteria are unlikely to be present in the tissues in their conventional form. The very slow progression of these diseases in most cases, their very slow response to antibacterial agents, and the difficulty experienced in demonstrating them in the tissues all suggest that the pathology of the disease is an excessive response to a small number of very slowly growing organisms for which the term "slow bacteria" seems appropriate 1.
There are a number of diseases for which such a slow bacterial aetiology seems probable. These include sarcoidosis, rheumatoid arthritis and psoriasis as well as inflammatory bowel disease, for all of which an increase in agalactosyl immunoglobulin levels seems to be a marker 2. This aberrant form of IgG is also raised in tuberculosis and at certain times in leprosy. It appears to be both a marker for these infections and for the type of T-lymphocyte mediated tissue damage accompanied by an acute phase response that typifies the immunpathological response constituting the lesions 3.
An individual propensity for developing disease in response to infectious challenge with these organisms is likely to depend on his or her genetic makeup and environmental experience. Thus the epidemiological approach to confirm that Crohn's disease or ulcerative colitis are infections has not been easy. However, it must be remembered that neither tuberculosis nor leprosy were generally recognized as infections until the causative organisms were demonstrated.
The demonstration that leprosy was a disease with a wide spectrum of immunopathology, with the multibacillary lepromatous form at one end and the paucibacillary tuberculoid form at the other, has had a profound effect on conceptual frameworks for helping to understand disease 4. It has both helped in explaining how infection with a single bacterial species can give rise to different syndromes of disease due to different immunological responses, and hindered in obscuring spectra based on other aspects. Although much less well known, tropical ulcerative mycobacteriosis, Buruli ulcer, or Mycobacterium ulcerans infection also have a disease spectrum, but one through which each sufferer passes, from the multibacillary immunosuppressive tissue necrotic phase to the paucibacillary granulomatous fibrotic healing phase 5. In the inflammatory bowel diseases the spectrum may be due to locally different expressions of immune response to the same infectious agent taking place in different tissues. Thus the lesions of intestinal tuberculosis in man are often different when they occur in the terminal ileum from when they occur in the colon. The majority of ileal tuberculosis is granulomatous, whereas colonic tuberculosis is often ulcerative with little or no granuloma formation 6.
This is not yet the time to argue for or against a particular view, or particular results, but to record observations and await developments. I have been fortunate in my laboratory to work with Professor Lennard Jones, Dr Burnham. Dr White, Dr Visuvanathan and Mr McIntyre over a period extending from 1978 until now. During this time we have carried out 2 blinded surveys of bacteriological isolation of unusual bacteria from lymph nodes draining bowel affected with Crohn's disease, ulcerative colitis or unrelated diseases. The results obtained in the two surveys have been essentially the same 7,8,9 and see the table).
|1978 series||1988 series|
|Crohn's disease||42/76 (55%)|
P < 0.00001
P < 0.002
|3/41(7%)P < 0.00001||3/17 (18%)P = 0.002|
|Ulcerative colitis||14/27 (52%)||11/15 (73 %)|
All positive cultures were obtained on Robertson's cooked meat medium, after 3 months to 18 months at 32oC.
Although we still do not have a definitive identity for the organisms isolated, a solid body of information about them is building up. The organisms are pleomorphic, variably acid-fast, possess traces of tuberculostearic acid and mycolic acid, but lack lipoarabinomannan in their cell walls. Related to mycobacteria, but perhaps better thought of as near-mycobacteria at this stage, their genetic identity is not yet fully confirmed. Producing positive dot blots with some mycobacterial gene probes, and the majority of them appearing homologous, their final identity may come from the sequencing that is currently in progress. A number of these isolates have been passaged through nude mice by Drs McDermott and Banerjee, and slowly growing conventional mycobacteria have been isolated from their tissues, bringing to mind the studies of Mitchell and Rees 10.
Although under ideal conditions the currently available chemotherapy with 3 or 4 effective drugs is capable of successfully curing both leprosy and tuberculosis with few failures, worldwide perspectives show results less impressive. In leprosy, reactions continue do destroy nerves and faces long after effective antibacterial treatment has begun, or even been completed. In tuberculosis, costs of drugs, their administration and overcoming patient non-compliance, compounded by increasing levels of bacterial drug resistance, especially to rifampicin and isoniazid, and now, devastatingly, by HIV infection, have led to a breakdown in control. The need for shorter and more effective treatments which might overcome drug resistance has led to a return to ideas of treatments based on stimulation of the immune system to destroy bacteria. Such an approach is especially suitable for mycobacterial diseases, and to those caused by the putative slow bacteria for the following reasons.
The principle of immunotherapy is to induce recognition of even resting bacteria by their surface antigens, and to promote their killing by immune mechanisms. This requires both the enhancement of surface antigen recognition, and the replacement of T cell mediated tissue destruction by the true mechanism of protective immunity in which invading bacteria are killed. This latter should be accompanied by a return to normal levels of circulating agalactosyl immunoglobulin.
Studies over many years have shown that suspensions of a killed, selected, strain of the harmless environmental saprophyte Mycobacterium vaccae, has both the required properties 12. A single intradermal injection of this suspension, used as an adjunct to antibacterial chemotherapy is proving very effective in established mycobacterial disease. Thus it doubles the rate of removal of leprosy bacilli in newly diagnosed multibacillary leprosy, and it increased the rate of disappearance of tubercle bacilli from the sputum of patients with pulmonary tuberculosis in an on-going study in Nigeria 13. In a controlled and blinded trial in The Gambia, immunotherapy with M. vaccae significantly reduced mortality over an 18 month follow-up period in patients who were considered to have complied with chemotherapy 14. In this study the level of agalactosyl IgG fell significantly faster in the recipients of the immunotherapy than in the recipients of placebo. Open studies of its use in the retreatment of treament failures due to drug resistance in tuberculosis have shown immunotherapy with M. vaccae to lead to substained disappearance of bacilli from the sputum, resolution of signs and relief of symptoms in patients in India, Iran, Romania, Vietnam and the United Kingdom.
Currently being tried in a number of diseases of uncertain ætiology, a number of dramatic, but still anecdotal, successes have been recorded. In psoriasis immunotherapy with M. vaccae has recently been found to be very effective in a group of Indian patients 15, and similar results have been achieved in on-going studies in Argentina, Iran and the United Kingdom. Although no studies have started yet in Crohn's disease, with its excessive granuloma formation and increased level of agalactosyl IgG, this is exactly the type of disease that might be expected to respond to such treatment. At the moment more than 10,000 people have received injections of killed M. vaccae as part of programmes of immunoprophylaxis or immunotherapy, with no more serious side effects than those accompanying BCG vaccination of a British school child.
It is difficult to see how much attempted immunotherapy could be harmful, and the potential benefit for the patient is enormous. Data obtained from patients with rheumatoid arthritis suggest that immunosuppressive drugs may be prejudicial to immunotherapeutic success, and therefore it might be better employ M. vaccae during a period of remission, rather than at a time when symptomatic suppression is required. An initial open study could be simple with a randomized half of a group of patients with quiescent Crohn's disease being give the injection. Follow-up might record relapses, assessments of symptomatic well-being, and the level of circulating agalactosyl immunoglobulin.
|1||Rook GAW, Stanford JL. autoimmunity or sloW bacterial infections? Immunology Today 1992 (in press).|
|2||Dude R, Rook GAW, Steele J, Brealey R, Dwek R, Rademacher T, Lennard-Jones J. Agalactosyl IgG in inflammatory bowel disease: correlation with C-reactive protein. Gut 1990;31:431-434.|
|3||Rook GAW, Al-Attiyah R. Cytokines and the Koch Phenomenon. Tubercle 1991;72:13-20.|
|4||Ridley DS, Jopling WH. Classification of leprosy according to immunity. A five group system. International Journal of Leprosy 1966;34:255-273.|
|5||Stanford JL, Rutt MSR, Phillips I, Revill WDL. Antibiotic treatment in Mycobacterium ulcerans infection. Ain Shams Medical Journal 1974;25 :suppl .5;258-261.|
|6||Vij JC. Personal communication.|
|7||Burnham WR, Lennard-Jones JE, Stanford JL, Bird RJ. Mycobacteria as a possible cause of inflammatory bowel disease. Lancet l978;2:693-696.|
|8||Stanford JL, Dourmashkin R, McIntyre G. An investigation into the role of mycobacteria in Crohn's disease. In "Inflammatory bowel disease - frontiers in aetiology". 1978. Proceedings of the 8th BSG/SK & F International Workshop 24-28.|
|9||Visuvanathan S. The characterization of pleomorphic acid fast organisms isolated from patients with inflammatory bowel disease. MD thesis, University of London. 1990.|
|10||Mitchell DN, Ress RJW. Agent transmissible from Crohn's disease tissue. Lancet 1970;2:1680171.|
|11||Stanford JL. Koch's phenomenon, can it be corrected? Tubercle 1991;72: in press.|
|12||Stanford jI, Rook GAW, Bahr GM, Dowlati Y, Ganapati R, Ohazi Saidi K, Lucas S, Ramu G, Torres P, Ho Monli Ly, Anstey N. Mycobacterium vaccae in immunoporphylaxis and immunotherapy of leprosy and tuberculosis.Vaccine 1990;8:525-530.|
|13||Onyebujoh P, Rook GAW. Mycobacterium vaccae immunotherapy. Lancet 1991;338:letter p.1534.|
|14||Corrah PT, Byass P, Stanford JL, Thomas V, Bouchier V, Whittle HC. The effects of prios vaccination with BCG and of immunotherapy on the outcome of treatment for pulmonary tuberculosis (in preparation). 1992.|
|15||Ramu G, Prema GD, Balakrishnan S, Shanker Narayan NP, Stanford JL. A preliminary report on the immunotherapy of psoriasis. Indian Medical Gazette 1990;124:381-382.|