A Blood and Biochemical Indicator-Based Prognostic Model Predicting Latent Tuberculosis Infection: A Retrospective Study.

BACKGROUND AND OBJECTIVES

Abnormal blood and biochemical indicators could increase the risk of infectious diseases. However, the association between blood together with biochemical indicators and latent tuberculosis infection (LTBI) has not been well confirmed.

MATERIALS AND METHODS

Our aim was to assess the role of blood and biochemical indicators in the risk of LTBI. We enrolled 965 freshmen who were originating from tuberculosis key areas of a college in Nanjing. We used logistic regression models, restricted cubic spline (RCS), and nomograms to evaluate the association between blood and biochemical indicators and LTBI. In addition, calibration curves were performed to evaluate the quality of the model.

RESULTS

Among these 965 participants, 311 were diagnosed as LTBI according to TST. Multivariate models showed that the population with an eosinophils percentage around <0.5% (OR: 2.82, 95% CI: 1.39-5.74, = 0.004) and 0.5-5% (OR: 2.78, 95% CI: 1.07-7.23, = 0.036) were positively associated with LTBI. Elevated uric acid levels (OR: 1.01, 95% CI: 1.00-1.02, = 0.047) were significantly associated with LTBI. In addition, participants with a history of tuberculosis exposure (OR: 3.26, 95% CI: 1.39-7.66) and a history of tuberculosis (OR: 10.92, 95% CI: 1.24-96.08) were also positively correlated with LTBI.

CONCLUSIONS

Eosinophils percentage and uric acid are associated with LTBIs. Participants who have tuberculosis exposure history and tuberculosis history are the critical target population.