Dataset Information

Development and validation of a prediction model for tuberculous pleural effusion: a large cohort study and external validation.

ABSTRACT:

Background

Distinguishing tuberculous pleural effusion (TPE) from non-tuberculosis (TB) benign pleural effusion (BPE) remains to be a challenge in clinical practice. The aim of the present study was to develop and validate a novel nomogram for diagnosing TPE.

Methods

In this retrospective analysis, a total of 909 consecutive patients with TPE and non-TB BPE from Ningbo First Hospital were divided into the training set and the internal validation set at a ratio of 7:3, respectively. The clinical and laboratory features were collected and analyzed by logistic regression analysis. A diagnostic model incorporating selected variables was developed and was externally validated in a cohort of 110 patients from another hospital.

Results

Six variables including age, effusion lymphocyte, effusion adenosine deaminase (ADA), effusion lactatedehy drogenase (LDH), effusion LDH/effusion ADA, and serum white blood cell (WBC) were identified as valuable parameters used for developing a nomogram. The nomogram showed a good diagnostic performance in the training set. A novel scoring system was then established based on the nomogram to distinguish TPE from non-TB BPE. The scoring system showed good diagnostic performance in the training set [area under the curve (AUC) (95% confidence interval (CI)), 0.937 (0.917-0.957); sensitivity, 89.0%, and specificity, 89.5%], the internal validation set [AUC (95%CI), 0.934 (0.902-0.966); sensitivity, 88.7%, and specificity, 90.3%], and the external validation set [(AUC (95%CI), 0.941 (0.891-0.991); sensitivity, 93.6%, and specificity, 87.5%)], respectively.

Conclusions

The study developed and validated a novel scoring system based on a nomogram originated from six clinical parameters. The novel scoring system showed a good diagnostic performance in distinguishing TPE from non-TB BPE and can be conveniently used in clinical settings.

SUBMITTER: Liu Y

PROVIDER: S-EPMC9145463 | biostudies-literature | 2022 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Development and validation of a prediction model for tuberculous pleural effusion: a large cohort study and external validation.

Liu Yanqing Y Liang Zhigang Z Yuan Songbo S Wang Shanshan S Guo Fei F Peng Weidong W Yang Jing J Wu Aihua A

Respiratory research 20220527 1

<h4>Background</h4>Distinguishing tuberculous pleural effusion (TPE) from non-tuberculosis (TB) benign pleural effusion (BPE) remains to be a challenge in clinical practice. The aim of the present study was to develop and validate a novel nomogram for diagnosing TPE.<h4>Methods</h4>In this retrospective analysis, a total of 909 consecutive patients with TPE and non-TB BPE from Ningbo First Hospital were divided into the training set and the internal validation set at a ratio of 7:3, respectively ...[more]

PMID: 35624515

Similar Datasets

Project description:BackgroundTuberculous pleural effusion (TPE) is a challenging extrapulmonary manifestation of tuberculosis, with traditional diagnostic methods often involving invasive surgery and being time-consuming. While various machine learning and statistical models have been proposed for TPE diagnosis, these methods are typically limited by complexities in data processing and difficulties in feature integration. Therefore, this study aims to develop a diagnostic model for TPE using ChatGPT-4, a large language model (LLM), and compare its performance with traditional logistic regression and machine learning models. By highlighting the advantages of LLMs in handling complex clinical data, identifying interrelationships between features, and improving diagnostic accuracy, this study seeks to provide a more efficient and precise solution for the early diagnosis of TPE.MethodsWe conducted a cross-sectional study, collecting clinical data from 109 TPE and 54 non-TPE patients for analysis, selecting 73 features from over 600 initial variables. The performance of the LLM was compared with logistic regression and machine learning models (k-Nearest Neighbors, Random Forest, Support Vector Machines) using metrics like area under the curve (AUC), F1 score, sensitivity, and specificity.ResultsThe LLM showed comparable performance to machine learning models, outperforming logistic regression in sensitivity, specificity, and overall diagnostic accuracy. Key features such as adenosine deaminase (ADA) levels and monocyte percentage were effectively integrated into the model. We also developed a Python package ( https://pypi.org/project/tpeai/ ) for rapid TPE diagnosis based on clinical data.ConclusionsThe LLM-based model offers a non-surgical, accurate, and cost-effective method for early TPE diagnosis. The Python package provides a user-friendly tool for clinicians, with potential for broader use. Further validation in larger datasets is needed to optimize the model for clinical application.

Project description:BackgroundDiagnosing tuberculous pleural effusion (TPE) in patients presenting with Lymphocyte-Predominant Exudative pleural effusion (LPE) is challenging, due to the poor clinical utility of TB culture. Adenosine deaminase (ADA) has been recommended for diagnosis, but its high cost and limited availability hinder its clinical utility. We aim to develop diagnostic prediction tools for Thai patients with LPE in scenarios where pleural fluid ADA is available but yields negative results and in situations where pleural fluid ADA is not available.MethodsTwo diagnostic prediction tools were developed using retrospective data from patients with LPE at Surin Hospital. Model 1 is for ADA-negative results, and Model 2 is for situations where pleural fluid ADA testing is unavailable. The models were derived using multivariable logistic regression and presented as two clinical scoring systems: round-up and count scoring. The score cut-point that achieves a positive predictive value (PPV) comparable to the post-test probability of a pleural fluid ADA at a cut-point of 40 U/L was used as a threshold for initiating anti-TB treatment.ResultsA total of 359 patients were eligible for analysis, with 166 diagnosed with TPE and 193 diagnosed with non-TPE. Age <40 years, fever, pleural fluid protein ≥5 g/dL, male gender, pleural fluid color, and pleural fluid ADA ≥20 U/L were identified as final predictors. Both models demonstrated excellent discriminative ability (AuROC: 0.85 to 0.89). The round-up scoring demonstrated PPV above 90% at cut-off points of 4 and 4.5, while the count scoring achieved cut-off points of 3 and 4 for Model 1 (Lex-2P2A) and Model 2 (Lex-2P-MAC), respectively.ConclusionThese diagnostic tools offer valuable assistance in differentiating between TPE and non-TPE in LPE patients with negative pleural fluid ADA (Lex-2P2A) and in settings where pleural fluid ADA testing is not available (Lex-2P-MAC). Implementing these diagnostic scores may have the potential to improve TPE diagnosis and facilitate prompt initiation of treatment.

Project description:BACKGROUNDCurrent diagnostic tools for tuberculous pleural effusion (TPE) are often inadequate, making accurate diagnosis challenging. Effective identification of TPE is critical for ensuring proper treatment and preventing tuberculosis relapse. This study explored the potential of granzyme A (GZMA) as a biomarker for TPE.METHODSPatients with TPE, malignant pleural effusion (MPE), and parapneumonic pleural effusion (PPE) were recruited into discovery and validation cohorts. The discovery cohort consisted of 200 patients with TPE and 100 patients with MPE, while the validation cohort included 167 patients with TPE, 84 patients with MPE, and 69 patients with PPE.RESULTSIn the discovery cohort, GZMA levels were significantly elevated in TPE compared with MPE, demonstrating 90% sensitivity and 91% specificity at a cutoff of 102.29 ng/mL for effectively distinguishing between the two conditions. In the validation cohort, GZMA maintained high diagnostic performance, distinguishing TPE from MPE with 87% sensitivity and 87% specificity and from PPE with 87% sensitivity and 84% specificity. Incorporating GZMA, lactate dehydrogenase (LDH), and adenosine deaminase (ADA) into a random forest model further improved diagnostic accuracy. In the discovery cohort, this model achieved 92% sensitivity and 100% specificity, and in the validation cohort, it distinguished TPE from MPE with 87% sensitivity and 94% specificity and from PPE with 87% sensitivity and 91% specificity.CONCLUSIONOverall, GZMA is a promising biomarker for diagnosing TPE, with improved accuracy when combined with LDH and ADA, providing a robust tool for timely identification and effective management of patients with TPE.FUNDINGThe study was supported by Science and Technology Project of Shenzhen (KCXFZ20211020163545004, KQTD20210811090219022, JCYJ20220818095610021, JSGG20220822095200001, JCYJ20210324094614038), Shenzhen Medical Research Funding (B2302035, A2302004), Provincial Natural Science Foundation of Guangdong (2022A1515220034), and Shenzhen Third People's Hospital Research Foundation (G2022155).

Dataset Information

Development and validation of a prediction model for tuberculous pleural effusion: a large cohort study and external validation.

Background

Methods

Results

Conclusions

Publications

Development and validation of a prediction model for tuberculous pleural effusion: a large cohort study and external validation.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets