Intestinal tuberculosis or Crohn's disease: Illusion or delusion or allusion.

Tuberculosis (TB) has reveled in its ability at clever disguise, cunning deceit, and masquerading as any disease beyond any logic. Interestingly, these illusory presentations of TB have been more commonly and inexplicably linked to specific anatomical sites like the intestine, the central nervous system, or the eye compared to its native habitat, the lungs. Robert Koch isolated tubercle bacilli and formulated Koch's postulates based on this, yet it is ironic that Koch's bacillus at extrapulmonary sites is defiant and resists all attempts made to fulfil its own creator's postulates. However, what Koch's favorite bacillus in the intestine had not fathomed was that the 20th century would bring in a conquistador that would be the very imitation of itself. Crohn's disease (CD) was found, and it imitated the great masquerade of TB so closely that it baffled caregivers. So much so that, despite the knowledge of its illusory state, physicians and the diagnosticians remained deluded and hoped for an allusion to end their diagnostic misery.

North America and Europe, ‘the Western world,’ found TB to be one of the major killers in the 18th and early 19th centuries, where it spread explosively due to urban human crowding, a consequence of the Industrial Revolution. This ‘white plague epidemic’ followed the Industrial Revolution, spreading to Eastern Europe, South Americas, East Asia, and Southeast Asia (‘the developing world’) in the 20th century. While the incidence and mortality of TB waned in the Western world, even before the advent of antitubercular therapy (possibly due to improving hygiene and sanitation), most of the developing world is still shackled with a high disease burden of TB. The disease process of extrapulmonary TB (EPTB) starts from the pulmonary focus, from where the bacterium disseminates to other areas, which could possibly account for the low bacterial load (paucibacillary) in extrapulmonary sites. The paucibacillary nature of EPTB makes it difficult to diagnose, more so if the Koch's postulates are followed, which has led to alternative diagnostic criteria depending upon the individual sites. The problem is further compounded at the sites that are affected by phenotypically similar but etiologically different disease processes, with the intestine (CD), thoracic cavity (sarcoidosis), and eyes (Eale's disease) primarily being such sites. , The epidemic of these immune‐mediated disorders, such as inflammatory bowel disease (IBD), has followed the decline of TB in the West, marked by a rising incidence in the 20th century that has stabilized over the last decade, while the developing world is following this pattern and is at the epidemiologic crossroads with a declining or stabilizing burden of TB , and a rising incidence of IBD and other such disorders. , , , Hence, the nontubercular etiologies at these extrapulmonary sites have largely replaced TB in the West, but the TB‐endemic regions continue to be affected by both the etiologies, creating a diagnostic challenge for clinicians in these regions.

Over the last two decades, there have been several efforts by gastroenterologists to decipher the dilemma of CD and intestinal TB (ITB), and considerable progress has been made toward dissecting this diagnostic conundrum, but a perfect solution still remains. , , The problem persists primarily due to two reasons: poor sensitivity of definite diagnostic tests for ITB (because of paucibacillary disease) and nonspecific diagnostic criteria for CD. The microbiological (positive stain for acid‐fast bacillus or culture for mycobacterium TB or positive Gene‐Xpert) or pathological tests (caseation granuloma) do have 100% specificity but are limited by poor sensitivity. , , The present study by Israrahmed et al. is another laudable effort in this direction. The authors propose a computed tomography (CT)‐based diagnostic algorithm for differentiating CD from ITB based on their findings on CT enteroclysis/enterography in a cohort of 61 patients with ITB and 24 patients with CD. Homogenous bowel wall thickening and confluent bowel involvement were significantly more common in ITB, while stratified bowel wall thickening with mucosal hyperenhancement, skip lesions, and comb sign were significantly more common in CD. Furthermore, stratified bowel wall enhancement with an intervening layer of fat was specific to CD, and necrotic lymph nodes were specific to ITB. The algorithm starts with the CT findings and incorporates endoscopy, histology and microbiology, and a therapeutic anti‐tubercular therapy (ATT) trial when these tests are inconclusive. Although the study echoes previously reported findings, , it provides a useful direction. Although the addition of radiology in this and the previous studies (necrotic lymph nodes on CT abdomen or evidence of active pulmonary TB) has increased the sensitivity, a large diagnostic gap still remains, and a similar situation exists for other etiologies such as sarcoidosis.

Hence, the most important step toward bridging this gap should involve a holistic effort (across involved specialties) targeting the root, that is, its paucibacillary nature. Xpert/MTB‐RIF (Xpert/mycobacterium tuberculosis‐rifampicin) was an important addition, but its poor sensitivity for EPTB specimens was an impediment to the resolution of the situation. The cornerstone for global TB control involves the development of novel, rapid, and accurate diagnostic techniques for all forms of TB, and a global laboratory initiative (GLI) has been launched in this regard. , These efforts should also benefit the diagnostic uncertainties existing between EPTB and respective site‐specific immune‐mediated etiologies. Another approach would involve the development of biomarkers for accurate diagnosis of EPTB, and FOXP3 T‐regulatory cell enumeration in peripheral blood has revealed reasonable diagnostic accuracy in differentiating CD from ITB. It has also been prospectively validated at the same center but requires further validation at different centers and across other sites.

Pushing our boundaries of understanding through this maze have been the development of machine learning‐based diagnostic algorithms incorporating multiple features (as very few are exclusive to CD or ITB, and those that are have very poor diagnostic sensitivity) on a platform through which these algorithms can be effortlessly applied. The Bayesian meta‐analytical model by Limsrivilai et al. incorporated clinical, endoscopic, and histological features into a predictive model with good diagnostic accuracy to differentiate CD from ITB. Furthermore, a nomogram based on seven parameters, including age, transverse ulcer, rectum involvement, skipped small bowel involvement, target sign, comb sign, and a QuantiFERON gold assay (for model 1) or Mantoux test (for model 2), also revealed good sensitivity and specificity for differentiating CD from ITB.

Before we develop a perfect strategy for upfront diagnosis of EPTB or for upfront differentiation of CD from ITB, we need to rely on a therapeutic ATT trial. In addition to the concern of ATT‐induced hepatotoxicity and steroid‐induced flare of TB, the ATT trial has recently been associated with complicated disease course in patients with an eventual diagnosis of CD. The etiologies for this ATT‐induced complication have been identified as ATT‐induced diagnostic delay or the profibrotic effects of ATT (independent of the diagnostic delay) in two different studies from India. , Having realized this additional and significant side effect of ATT, more vigilance and early decisions are required in patients on a therapeutic ATT trial. Hence, the follow‐up algorithm proposed earlier and endorsed by guidelines , requires a revisit with early (at 2–3 months of ATT) objective evaluation and possible shortening of ATT duration in patients with a higher suspicion of CD.

The TB–CD tale is reminiscent of Star Wars trilogy. The latest episode is “The Return of the Jedi.” First, TB existed in the intestine; then, CD was found at the same site; and now, the 21st century brings a new permutation that is “TB on CD.” This concern, specific to TB‐endemic regions (or the regions where this diagnostic dilemma exists), regards the identification of definite evidence of TB in a patient with definite CD. The increasing disease burden of IBD and improving economy in the developing world have led to increased use of antitumor necrosis factor (TNF) therapy in IBD. TB reactivation is a major concern with anti‐TNF use, and the local TB burden has been identified as a major risk factor for this association, with the highest rates of reactivation being reported from high TB burden countries such as India. , TB that develops after anti‐TNF therapy can be extrapulmonary in up to 50% and can also involve the intestine. Therefore, the clinicians in these areas need to be aware of this emerging perplexity of “TB on CD,” which has further enhanced the diagnostic confusion with regard to “TB or CD”.

The emergence of IBD (and other immune‐mediated disorders) and persistence of TB in the developing world continue to perplex the clinicians. Although this illusion of diagnostic certainty (or uncertainty) delays the correct therapeutic approach, the progress is steady, and a joint effort toward global TB control through improved diagnostic techniques and the advent of big data features would certainly bring about a definite solution. As previously stated, the new religion is Dataism. Dataism is a word first used by David Brooks in the New York Times in 2013, underlining the emerging importance of big data. He felt it would greatly influence human thinking as “in a world of increasing complexity, relying on data could reduce cognitive biases and illuminate patterns of behavior we haven't yet noticed.” This new religion was propelled to the masses by Yuval Noah Harari is his landmark book “Homo Deus: A Brief History of Tomorrow” when he suggested “Forget about listening to ourselves. In the age of data, algorithms have the answer.” And this could very well be the gospel truth or the allusion for ending this deluded illusion of TB or CD.

Declaration of conflict of interest

None.