Table 1.

Summary of the major findings of studies investigating the role of Mycobacterium tuberculosis as a trigger of PBC. The findings are presented in relation to Koch's postulates. To date, no conclusive evidence has been presented which links Mycobacterium tuberculosis with PBC.

Koch's postulates	Finding	Study
Infectious aetiology   Microorganism must be present in every case of diseased individuals	Positive staining for mycobacterial hsp65 in PBC cases has been demonstrated, but hsp65 is conserved among all mycobacterial species and it is unclear whether this staining is Mycobacterium tuberculosis-specific and characteristic for PBC patients with granulomata	[75]
Isolation   The microorganism must be isolated and grown in pure cultures	No such data exist
Disease causality   Pure cultures inoculated in healthy animals must reproduce the disease	No such data exist Only one case report of a female that developed PBC following tuberculosis infection has been published There is no epidemiological or evidence linking tuberculosis with PBC, including vaccination against tuberculosis	[76] [77–80]
Reproducibility   The microorganism must be recovered from the diseased animal	No such data exist
Other data indirectly relevant to Koch's postulates	PBC specific AMA was detected in 43% of tuberculosis patients, but in low titres, and did not show the typical indirect immunofluorescent patterns of antimitochondrial antibodies seen in PBC. These patients do not have clinical features of PBC.	[19, 21, 22, 81]