Derivation, validation and comparative performance of a simplified chest X-ray score for assessing the severity and outcome of pulmonary tuberculosis

Systematic review of prediction models for pulmonary tuberculosis treatment outcomes in adults

OBJECTIVE

To systematically review and critically evaluate prediction models developed to predict tuberculosis (TB) treatment outcomes among adults with pulmonary TB.

DESIGN

Systematic review.

DATA SOURCES

PubMed, Embase, Web of Science and Google Scholar were searched for studies published from 1 January 1995 to 9 January 2020.

STUDY SELECTION AND DATA EXTRACTION

Studies that developed a model to predict pulmonary TB treatment outcomes were included. Study screening, data extraction and quality assessment were conducted independently by two reviewers. Study quality was evaluated using the Prediction model Risk Of Bias Assessment Tool. Data were synthesised with narrative review and in tables and figures.

RESULTS

14 739 articles were identified, 536 underwent full-text review and 33 studies presenting 37 prediction models were included. Model outcomes included death (n=16, 43%), treatment failure (n=6, 16%), default (n=6, 16%) or a composite outcome (n=9, 25%). Most models (n=30, 81%) measured discrimination (median c-statistic=0.75; IQR: 0.68-0.84), and 17 (46%) reported calibration, often the Hosmer-Lemeshow test (n=13). Nineteen (51%) models were internally validated, and six (16%) were externally validated. Eighteen (54%) studies mentioned missing data, and of those, half (n=9) used complete case analysis. The most common predictors included age, sex, extrapulmonary TB, body mass index, chest X-ray results, previous TB and HIV. Risk of bias varied across studies, but all studies had high risk of bias in their analysis.

CONCLUSIONS

TB outcome prediction models are heterogeneous with disparate outcome definitions, predictors and methodology. We do not recommend applying any in clinical settings without external validation, and encourage future researchers adhere to guidelines for developing and reporting of prediction models.

TRIAL REGISTRATION

The study was registered on the international prospective register of systematic reviews PROSPERO (CRD42020155782).

Development and validation of a prognostic score during tuberculosis treatment

BACKGROUND

Death under care is a major challenge for tuberculosis (TB) treatment programs. We derived and validated a simple score to predict mortality during tuberculosis treatment in high endemicity areas.

METHODS

We used data for patients aged ≥15 years, diagnosed and treated for tuberculosis at the Yaounde Jamot Hospital between January 2012 and December 2013. Baseline characteristics associated with mortality were investigated using logistic regressions. A simple prognosis score (CABI) was constructed with regression coefficients for predictors in the final model. Internal validation used bootstrap resampling procedures. Models discrimination was assessed using c-statistics and calibration assessed via calibration plots and the Hosmer and Lemeshwow (H-L) statistics. The optimal score was based on the Youden's index.

RESULTS

A total of 2250 patients (men 57.2%) with a mean age of 35.8 years were included; among whom 213 deaths (cumulative incidence 9.5%) were recorded. Clinical form of tuberculosis (C), age (A, years), adjusted body mass index (B, BMI, kg/m²) and status for HIV (Human immunodefiency virus) infection (I) were significant predictors in the final model (p < 0.0001) which was of the form Death risk = 1/(1 + e ^{- (-1.3120 + 0.0474 ∗ age - 0.1866 ∗ BMI + 1.1637 (if smear negative TB) + 0.5418(if extra - pulmonary TB) + 1.3820(if HIV+))}). The c-statistic was 0.812 in the derivation sample and 0.808 after correction for optimism. The calibration was good [H-Lχ² = 6.44 (p = 0.60)]. The optimal absolute risk threshold was 4.8%, corresponding to a sensitivity of 81% and specificity of 67%.

CONCLUSIONS

The preliminary promising findings from this study require confirmation through independent external validation studies. If confirmed, the model derived could facilitate the stratification of TB patients for mortality risk and implementation of additional monitoring and management measures in vulnerable patients.

Assessment of treatment response in tuberculosis

Antibiotic treatment of tuberculosis has a duration of several months. There is significant variability of the host immune response and the pharmacokinetic-pharmacodynamic properties of Mycobacterium tuberculosis sub-populations at the site of disease. A limitation of sputum-based measures of treatment response may be sub-optimal detection and monitoring of Mycobacterium tuberculosis sub-populations. Potential biomarkers and surrogate endpoints should be benchmarked against hard clinical outcomes (failure/relapse/death) and may need tailoring to specific patient populations. Here, we assess the evidence supporting currently utilized and future potential host and pathogen-based models and biomarkers for monitoring treatment response in active and latent tuberculosis. Biomarkers for monitoring treatment response in extrapulmonary, pediatric and drug resistant tuberculosis are research priorities.