3 results
4298 Prediction models for pulmonary tuberculosis treatment outcomes: a systematic review
- Lauren Saag Peetluk, Felipe Ridolfi, Valeria Rolla, Timothy Sterling
-
- Journal:
- Journal of Clinical and Translational Science / Volume 4 / Issue s1 / June 2020
- Published online by Cambridge University Press:
- 29 July 2020, p. 34
-
- Article
-
- You have access Access
- Open access
- Export citation
-
OBJECTIVES/GOALS: Many clinical prediction models have been developed to guide tuberculosis (TB) treatment, but their results and methods have not been formally evaluated. We aimed to identify and synthesize existing models for predicting TB treatment outcomes, including bias and applicability assessment. METHODS/STUDY POPULATION: Our review will adhere to methods that developed specifically for systematic reviews of prediction model studies. We will search PubMed, Embase, Web of Science, and Google Scholar (first 200 citations) to identify studies that internally and/or externally validate a model for TB treatment outcomes (defined as one or multiple of cure, treatment completion, death, treatment failure, relapse, default, and lost to follow-up). Study screening, data extraction, and bias assessment will be conducted independently by two reviewers with a third party to resolve discrepancies. Study quality will be assessed using the Prediction model Risk Of Bias Assessment Tool (PROBAST). RESULTS/ANTICIPATED RESULTS: Our search strategy yielded 6,242 articles in PubMed, 10,585 in Embase, 10,511 in Web of Science, and 200 from Google Scholar, totaling 27,538 articles. After de-duplication, 14,029 articles remain. After screening titles, abstracts, and full-text, we will extract data from relevant studies, including publication details, study characteristics, methods, and results. Data will be summarized with narrative review and in detailed tables with descriptive statistics. We anticipate finding disparate outcome definitions, contrasting predictors across models, and high risk of bias in methods. Meta-analysis of performance measures for model validation studies will be performed if possible. DISCUSSION/SIGNIFICANCE OF IMPACT: TB outcome prediction models are important but existing ones have not been rigorously evaluated. This systematic review will synthesize TB outcome prediction models and serve as guidance to future studies that aim to use or develop TB outcome prediction models.
3166 Association between HIV and early weight loss and the impact on subsequent treatment outcomes among patients with tuberculosis
- Lauren A Saag, Peter F. Rebeiro, Marcelo Cordeiro-Santos, Afranio Kritski, Bruno B. Andrade, Betina Durovni, Solange Calvacante, Megan Turner, Marina C. Figueiredo, Valeria C. Rolla, Timothy R. Sterling
-
- Journal:
- Journal of Clinical and Translational Science / Volume 3 / Issue s1 / March 2019
- Published online by Cambridge University Press:
- 26 March 2019, p. 34
-
- Article
-
- You have access Access
- Open access
- Export citation
-
OBJECTIVES/SPECIFIC AIMS: Previous research suggests that weight loss during early TB treatment (first two months of anti-TB therapy) is a predictor of poor tuberculosis (TB) treatment outcomes among HIV-negative populations, but the relationship has not been well studied in the context of HIV. We examined the association between HIV and weight change during the first two months of anti-tuberculosis treatment, and also assessed the effects of HIV and early weight change on tuberculosis (TB) treatment outcomes. METHODS/STUDY POPULATION: Adults with culture-confirmed, drug-susceptible, pulmonary TB, regardless of HIV status, were enrolled into the Regional Prospective Observational Research for Tuberculosis (RePORT)-Brazil cohort and followed on standard anti-TB therapy. For the primary analysis, we compared weight change in persons living with HIV (PLWH) and HIV-negative patients between baseline and two months using multivariable bootstrapped quantile regression and modified Poisson regression. For secondary analysis, we examined the separate effects of HIV and weight change on poor TB treatment outcome (treatment failure, TB recurrence, or death) using Cox proportional hazards regression. RESULTS/ANTICIPATED RESULTS: Among 323 participants, 45 (14%) were HIV-positive. On average, PLWH lost 0.7% (interquartile range (IQR): −5.1%, 4.4%) of their baseline body weight between baseline and two months; those without HIV gained 3.5% (IQR: 0.8%, 6.7%). After adjusting for age, sex, and baseline BMI, PLWH lost 4.1% (95% confidence interval (CI): −6.5%, −1.6%) more weight during the first two months of anti-TB treatment than HIV-negative individuals. HIV infection was associated with weight loss ≥5% (adjusted odds ratio = 9.3; 95% CI: 4.2-20.6). Regarding the secondary analysis, 14 patients had a poor TB treatment outcome: 2 treatment failures, 4 cases of recurrent TB, and 8 deaths. PLWH and patients who lost ≥5% weight had significantly increased risk of poor TB treatment outcome with hazard ratios of 8.77 (95% CI: 2.96-25.94) and 4.09 (95% CI: 1.11-15.14), respectively. DISCUSSION/SIGNIFICANCE OF IMPACT: Our results suggest that HIV is associated with weight loss during early TB treatment, and both HIV and early weight loss were associated with poor treatment outcome. Future research should examine the potential etiologies of these findings and identify the types of interventions that would best promote weight gain during TB treatment, especially among PLWH, in order to prevent poor TB treatment outcomes.
2300 Association between source case cavitation on chest radiograph and QuantiFERON-TB Gold In-Tube conversion among close contacts of active tuberculosis cases in Brazil
- Lauren A. Saag, Marcelo Cordeiro-Santos, Afranio Kritski, Bruno Andrade, Solange Cavalcante, Betina Durovni, Megan Turner, Marina Figueiredo, Valeria Rolla, Timothy Sterling
-
- Journal:
- Journal of Clinical and Translational Science / Volume 2 / Issue S1 / June 2018
- Published online by Cambridge University Press:
- 21 November 2018, p. 4
-
- Article
-
- You have access Access
- Open access
- Export citation
-
OBJECTIVES/SPECIFIC AIMS: QuantiFERON-TB Gold In-Tube (QFT) conversion from negative to positive, is regarded as a marker of recent latent tuberculosis infection and may be predictive of incident active tuberculosis (TB) disease. However, it remains unclear how conversion is influenced by individual and environmental factors, including the infectiousness of the source case to whom the contact was exposed. We aimed to examine the effect of infectiousness of TB in the source case, as measured by presence of cavitation on chest X-ray, on the incidence of QFT conversion among close contacts of the pulmonary TB index case, after adjusting for potential confounding by contact and source case characteristics. METHODS/STUDY POPULATION: The Regional Prospective Observational Research for Tuberculosis (RePORT)-Brazil is an ongoing prospective cohort study that enrolls close contacts of culture-confirmed pulmonary TB patients and follows them for 24 months for development of active TB. Demographic, clinical, and diagnostic information are obtained at baseline and during follow-up at clinical visits and by telephone. QFT testing is performed at baseline and repeated after 6 months if the baseline QFT is negative. A positive IFN-γ value is defined as >0.35 IU/mL, as recommended by the manufacturer and the CDC, and QFT conversion is defined as a negative QFT at baseline followed by a positive QFT at 6 months. RESULTS/ANTICIPATED RESULTS: Among 260 enrolled contacts with nonpositive baseline QFT results and 6 months of follow-up, 198 (76%) were retested with QFT 6 months after enrollment. Of those retested, 26 (13%) converted to positive. Presence of any cavitation in the source case, based on chest radiography, was significantly associated with QFT-conversion (ORunadjusted=2.4, 95% CI: 1.0–5.7). Additional univariate analyses revealed that QFT conversion was associated with black and brown race (compared with white race) of the contact, current smoking and current alcohol use in the source case. After adjusting for potential confounders (age, sex, and race of the contact and current smoking of the source case), the association between source case cavitation and QFT conversion remained (ORadjusted=2.5 95% CI: 1.0–6.2). As of December 6, 2017, none of the QFT-retested contacts had developed active TB, with a median follow-up of 12.3 months (IQR: 7.1–13.1). We anticipate that ongoing enrollment and follow-up may yield cases of active TB; future analyses will provide greater precision for examining predictors of QFT-conversion and its association with incident TB. DISCUSSION/SIGNIFICANCE OF IMPACT: Our preliminary results agree with published literature suggesting the infectiousness of TB in the index case is a predictor of incident LTBI. Along with recent LTBI, immune suppression, HIV co-infection, and type 2 diabetes are considered risk factors for progression to active TB disease. Because only a small proportion of persons progress from LTBI to active TB disease, it is not appropriate to treat all persons with LTBI. Thus, more research is needed to identify groups at highest risk for QFT-conversion and incident TB disease, so these groups can be targeted for TB prevention, interventions, and facilitate a decline in TB incidence and mortality.