Abstract

The objective : to identify and rank the predictors of lethal outcome in patients with tuberculosis resistance to at least rifampicin and concurrent HIV infection ( T B -R / HIV ). Subjects and methods. 130 cases T B -R / HIV notified in Rostov Region in 201 7-2 018 were included in an observational retrospective cohort study. Two cohorts were formed: cohort A consisted of 31 patients with a documented fatal outcome within 12 months after registration for treatment, cohort B included 99 patients didn't die at least during the first year after registration for treatment. At the first stage of the study, the statistical significance of differences of certain signs in patients of cohorts A and B was determined. The signs with confirmed statistical significance of differences were included in the second stage of the study, during which their sensitivity as predictors of lethal outcome was determined. For this, automated artificial neural networks were used. Results. The following can be used as predictors of lethal outcome for T B -R / HIV patients (in decreasing order of significance): homelessness (people who have no place of residence), addiction to alcohol or drugs, interruption of ant i- T B chemotherapy in the past, low body mass index, low hemoglobin level, including anemia (but not severe), low level of CD 4 T-l ymphocytes (the prognosis is especially unfavorable with less than 100 cell s/ μ l). The level of total protein may also be a potential predictor, however, the selection criteria for this indicator require further clarification.

Highlights

  • Цель исследования: выявить и ранжировать по значимости предикторы летального исхода у пациентов с туберкулезом, с устойчивостью как минимум к рифампицину, и ВИЧ-положительным статусом (ВИЧ)-инфекцией (ТБ-Р/ВИЧ)

  • Включенных в исследование, составил 40 лет (25% квартиль ‒ 34 года; 75% квартиль – 44 года); большинство пациентов (79/130 – 60,8%) были мужчинами

  • У пациентов когорты А медиана числа CD4+ лимфоцитов составила 117 (25% квартиль ‒ 27; 75% квартиль ‒ 379,5), в то время как у пациентов когорты Б ‒ 334 (25% квартиль ‒ 206, 75% квартиль – 456,5)

Read more

Summary

Subjects and methods

130 cases TB-R/HIV notified in Rostov Region in 2017-2018 were included in an observational retrospective cohort study. Two cohorts were formed: cohort A consisted of 31 patients with a documented fatal outcome within 12 months after registration for treatment, cohort B included 99 patients didn't die at least during the first year after registration for treatment. At the first stage of the study, the statistical significance of differences of certain signs in patients of cohorts A and B was determined. The signs with confirmed statistical significance of differences were included in the second stage of the study, during which their sensitivity as predictors of lethal outcome was determined.

Results
Материалы и методы
Доля корректно определенных состояний когорта Б когорта А
Результаты исследования
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call