Predictors for Pulmonary Tuberculosis Outcome and Adverse Events in an Italian Referral Hospital: A Nine-Year Retrospective Study (2013-2021)

Background: The COVID-19 pandemic has undone years of progress in providing essential TB services and controlling the TB burden. Italy, a low TB burden country, has an incidence of 7.1 cases per 100,000 people. To control the TB spreading in Italy is critical to investigate the characteristics of patients with the worst outcomes and the highest risk of adverse events related to antituberculosis therapy. Therefore, we conducted a large retrospective study in TB patients admitted to the Clinic of Infectious Diseases University of Bari, Italy, in order to describe the clinical presentation and the factors associated with adverse events and outcomes.

Methods: We retrospectively evaluated the patients admitted to the Clinic of Infectious Diseases from January 2013 to 15 December 2021. We stratified our cohort into two groups: <65 years of age and ≥65 years in order to assess any differences between the two groups. Two logistic regression models were implemented considering the dependent variables as: (I) the adverse events; and (II) the unsuccessful treatments.

Results: In total, 206 consecutive patients [60% (n = 124) M, median age 39 years, range 16-92] were diagnosed and admitted with TB at Clinic of Infectious Diseases. Of the whole sample, 151 (74%) were <65 years and 55 (26%) were ≥65. Statistically significant differences between the two groups were detected (p-value < 0.05) for nationality (p-value = 0.01), previous contact with TB patient (p-value = 0.00), type of TB (p-value = 0.00), unsuccessful treatment (p-value = 0.00), length of hospitalization (p-value = 0.02) and diagnostic delay (p-value = 0.01). Adverse events related to TB drug regimen were reported in 24% (n = 49). Age < 65 years (O.R. = 3.91; 95% CI 1.72-4.21), non-Italian nationality (O.R. = 4.45; 95% CI 2.22-4.98.), homeless (O.R. = 3.23; 95% CI 2.58-4.54), presence of respiratory symptoms (O.R. = 1.23; 95% CI 1.10-1.90), diagnostic delay (O.R = 2.55; 95% CI 1.98-3.77) resulted associated with unsuccessful treatment outcome (death, failure or lost to follow up). Finally, age < 65 years (O.R. = 1.73; 95% CI 1.31-2.49), presence of pulmonary TB (O.R. = 1.15; 95% CI 1.02-1.35), length of hospitalization (O.R. = 1.82; 95% CI 1.35-2.57) and TB culture positive (O.R. = 1.35; 95% CI 1.12-1.82) were associated with adverse events in our populations.

Conclusions: The pharmacological approach alone seems insufficient to treat and cure a disease whose ethiopathogenesis is not only due to the Mycobacterium tuberculosis, but also to the poverty or the social fragility. Our data suggest that young foreigners, the homeless, and the people with low social and economic status are at higher risk of an unfavorable outcome in low incidence TB countries. Targeted actions to support this highly vulnerable population both in terms of outcome and occurrence of adverse events are needed.