[Utility of molecular biology in the microbiological diagnosis of mycobacterial infections].

Species within the Mycobacterium genus are of major medical interest, since, together with environmental and opportunistic species, there are two species (Mycobacterium tuberculosis and Mycobacterium leprae) that remain an important public health challenge. Despite efforts to control tuberculosis (TB), this disease remains one of the most prominent health problems worldwide. In the last few years, mycobacteriology has experienced major technological advances. Nevertheless, the early diagnosis of mycobacterial infection and, especially of TB, is still based on microscopic examination of properly stained samples. At present, this procedure is still the simplest, fastest and most cost-effective method for preliminary diagnostic guidance. Effective control of TB is based on rapid detection of M. tuberculosis, followed by immediate implementation of the appropriate antituberculosis therapy. Because of the emergence of multidrug resistant strains, the development of rapid diagnostic methods, both for identification of M. tuberculosis and susceptibility testing, has become a pressing need. The availability of molecular epidemiology methods that are easy to implement and standardized and that would allow identification of related cases is of key importance to identify epidemic outbreaks and control the spread of TB. Despite the evident progress in the molecular diagnosis of mycobacterial infections, the available techniques are still inadequate. In this review, we describe the state of the art of the main molecular techniques for direct detection of mycobacteria in clinical samples, their identification, detection of resistance to the most important antituberculosis agents, and molecular epidemiology. In each case, we describe the advantages and limitations of current techniques. In the near future, clinical mycobacteriology will probably evolve to the universal use of genetic techniques for direct diagnosis and detection of resistance. The molecular epidemiology of TB will be performed, in its various applications, by faster and more automated techniques than those currently available.