The use of computer techniques to optimize the diagnosis of tuberculosis in HIV-infected patients at the secondary diseases stage

ВACKGROUND: The course of tuberculosis in HIV-infected patients has no significant clinical and diagnostic differences from damage to organs and systems of other etiology. The growth of new cases of HIV/TB co-infection necessitates a revision of diagnostic approaches, their improvement in order to increase the efficiency of examination and management of HIV-infected patients, depending on the stage and indicators of the immune status. AIM: Of the study was to determine by the method of complex statistical processing the main clinical and laboratory-instrumental criteria for improving the diagnosis of tuberculosis in HIV-infected patients at the stage of secondary diseases. MATERIALS AND METHODS: The study design was retrospective. The object of the research was the case histories of 113 patients with HIV infection at the stage of secondary diseases (classification by V.I. Pokrovsky), of parametric and nonparametric statistics, computer analysis of images with the gradient program proposed by Dr. Sci. A.N. Gerasimov, to assess the possibility of using micro- and macro-preparations of tissues and organs of patients with HIV infection. RESULTS: Using the method of correlation adaptometry, it was found that there are no significant differences in the clinical course of HIV-infected patients with tuberculosis of various localization, and with lesions of the respiratory organs of other etiology. The use of multivariate probability models made it possible to identify significant diagnostic risk factors for lethal outcome ― 66.7% of patients with further lethal outcome complained of chest pain during breathing (p=0.004), and ESR was significantly accelerated in patients with a lethal outcome in the hospital ― 771.99 (p=0.019). CONCLUSION: The multicomplex instrumental and laboratory examination makes it possible to diagnose tuberculosis of various localization at the initial stage of development. The use of computer techniques optimizes and unifies the diagnostic search in patients with HIV infection and determines the timely treatment tactics.