Guide to the Application of Genotyping to Tuberculosis Prevention and Control

Developing a Tuberculosis Genotyping Program

Establishing Procedures for Requesting RFLP Analysis

The two PCR-based genotyping methods (spoligotyping and MIRU analysis) provide sufficient discriminatory power for most TB control purposes. IS6110-based RFLP analysis can provide additional discriminatory power, but usually it is not needed. Because RFLP analysis is labor-intensive, its use should be limited to situations where additional information that can be provided only by RFLP will guide further action.

RFLP analysis is indicated only for isolates that have already been found to have matching PCR genotypes. One purpose of genotyping is to determine whether isolates are genetically distinct and therefore unlikely to be involved in the same chain of recent transmission. If any genotyping test shows that two isolates have different genotypes, this provides strong evidence that the two isolates are genetically distinct. Additional testing of genetically distinct isolates will not change the conclusion that they are genetically distinct. Therefore, if two isolates have different spoligotypes or different MIRU types, the results of the RFLP analysis will rarely be helpful. In the following discussion of when to request an RFLP analysis, it is assumed that isolates have already been tested by spoligotyping and MIRU analysis and that both tests resulted in matches.

TB programs should request RFLP analysis usually for isolates from patients who are involved in ongoing investigations of outbreaks, because it is important to know with as much accuracy as possible whether the isolates represent the same strain of M. tuberculosis. During outbreak investigations and during contact investigations of multiple cases with isolates that match by the two PCR tests, RFLP results help to establish which patients belong to the outbreak and which patients do not. Either result helps to focus the outbreak investigation on those patients involved in the same chain of transmission.

Some false-positive culture investigations should include RFLP analysis, but others do not require it. The determination of when to request RFLP analysis during a false-positive culture investigation rests on whether there was a preexisting suspicion of a problem. In circumstances where the treating physician or the clinical laboratory suspects a false-positive culture, a match by spoligotype and MIRU analysis is sufficient to confirm that suspicion. On the other hand, if the PCR genotyping results showing a possible false-positive culture were unexpected by the laboratory and the clinicians, RFLP analysis should be requested to confirm the PCR results.

When patients who are clustered by the two PCR tests are thought to be involved in the same chain of recent transmission and previous contact investigations established known epidemiologic links between them, chances are very good that RFLP results will confirm the genetic identity of the isolates. Therefore, in this case, RFLP analysis is not indicated. In contrast, for similar patients for whom only possible epidemiologic links have been found (i.e., PCR test results match and patients could be involved in recent transmission), RFLP analysis should be considered. Here, the presence of possible epidemiologic links raises the question of whether the patients represent the same chain of recent transmission. If RFLP results show that the isolates are different, the TB program will decide that no cluster investigation is needed because the RFLP results indicated that no cluster existed. If the RFLP pattern shows a match, the TB program will have added motivation to conduct a cluster investigation.

A final situation that could warrant a request for RFLP analysis occurs when certain strains of M. tuberculosis match by spoligotyping and MIRU analysis but can be differentiated only by RFLP. This situation may become increasingly important as we gain more information about the distribution of isolates that match by spoligotyping and MIRU analysis. The genotyping laboratories will be a good source of information about the need for conducting RFLP analysis in this situation.