S
sh!+_stains
Guest
A variety of extra-intestinal manifestations of IBD have been described[65] and in 'established' IBD environments are useful distinguishing features of CD. However, the ability of TB to involve multiple extra-pulmonary sites and associated immunological phenomena are common clinical manifestations in endemic regions and may be difficult to distinguish from extra-intestinal features of IBD. TB involvement of the lower limb joints[66], skin, eye and liver may mimic extra-intestinal CD. Immunologically mediated reactive polyarthritis (Poncet's disease)[66], erythema nodosum, erythema induratum[67] and uveitis[68] can also be interpreted as manifestations of Crohn's. An association between inflammation and a hypercoagulable state is common to both Crohn's and tuberculosis. Patients with IBD are at a 3.6-fold increased risk of thromboembolic disease.[69] Similarly, patients with active TB are at risk of deep vein thrombosis.[70,71] In areas of endemic tuberculosis cautious interpretation of extra-intestinal signs of CD is required.
Fistulization is one of the clinical hallmarks of CD. However, entero-enteric, entero-cutaneous and peri-anal fistulas are all well described in intestinal TB.[63,72,73] In a South African series, 17% of peri-anal fistulas referred to a surgical department were tuberculous in origin.[74] In a series from Taiwan, 8% of ITB patients presented with anorectal disease.[75]
Crohn's disease and TB are both chronic granulomatous conditions which affect the gastrointestinal tract in a similar manner. Mycobacterium tuberculosis is the causative organism in ITB whereas the aetiology of CD is multi-factorial and includes genetic, immunological, environmental and microbial factors. It is not surprising given striking morphological similarities that they share many common immune pathways of pathogenesis, nor that corticosteroids have been used effectively in both disorders to control deleterious inflammatory reactions.[20-22] Both trigger potent adaptive TH1 cytokine responses which result in granuloma formation and are characterized by robust production of interferon-gamma (IFN-γ), IL-12 and IL-23.[23] This is necessary to contain M. tuberculosis and prevent dissemination, and the protective effect of these cytokines is best demonstrated by the predisposition to disseminated, atypical mycobacterial infections in individuals with deleterious mutations in the IL-12/IL-23/IFN-γ axis.[24] The flip side, however, is the development of intestinal disease indistinguishable from CD.
In contrast to the forceful adaptive immune responses seen, both CD and TB appear to be associated with impaired innate immunity.[25,26] Only 5-10% of patients infected with M. tuberculosis develop active tuberculosis. Furthermore, as with CD the protean manifestations of TB suggests that individual variation in host-bacterial interactions may contribute to disease phenotype and that host genetics may play a role in dictating the efficacy of innate immune responses.
Nucleotide-binding oligomerization domain-2 (NOD2) and Toll-like receptors (TLRs) may play a role in the early, inductive stages of both diseases.[27-30] NOD2 single nucleotide polymorphisms (SNPs) confer susceptibility to CD in certain populations.[27] Furthermore, NOD2, as well as TLR 2, 4 and 9 are non-redundant recognition systems for sensing the presence of M. tuberculosis and the important role of NOD2 was recently shown in mononuclear cells of individuals with CD homozygous for the 3020insC NOD 2 mutation, as well as in NOD2 knock-out mice.[31,32] Despite this, NOD2 mutations have not been shown to increase susceptibility to pulmonary TB in African populations[33,34]
However, as with CD several other genetic polymorphisms have been found to confer susceptibility to pulmonary TB. Generally the associations have been weak but have provided valuable insight into disease pathogenesis. Interestingly several candidate genes, such as the Vitamin D receptor and the SCL11A1 (formerly NRAMP1) genes, have been shown to impact on both CD and ITB.[35-38] As in the Asia-pacific region NOD2 mutations are not associated with CD in our setting.[39]
However, both TB and CD are characterised by enormous heterogeneity and it may be, as has been shown in CD, that polymorphisms predict disease phenotype.[40,41] In contrast to pulmonary TB the contribution of genetic mutations in intestinal TB has not been adequately explored and is currently being evaluated in our unit.
No discussion of CD and ITB would be complete without addressing the longstanding debate on the role of Mycobacterium avium paratuberculosis (MAP) in CD, which remains unresolved. It has long been recognized that CD does not exist in germ-free environments and that luminal bacteria are required for the development of inflammation in animal models of IBD.[42,43] The recognition of NOD2 gene mutations in enhancing susceptibility to CD has placed emphasis on the role of luminal microflora in this disorder.[44-46] It is, however, overly simplistic to assume a true infectious causality of CD given the success of anti- tumour necrosis factor-α (TNF-α) therapies, which should markedly worsen the course of an active mycobacterial infection. As such if MAP plays a role, it is likely to be a little more esoteric. One possible explanation evokes the concept of molecular mimicry, with antibodies directed against mycobacterial antigen cross reacting with intestinal components.[47,48]
End Quote
It may be overly simplistic to assume anti-TNF agents will make all mycobacterial infections worse:
