Exploring the Spectrum of Clinical and Radiological Findings in Active Tuberculosis

Technical Summary: Exploring the Spectrum of Clinical and Radiological Findings in Active Tuberculosis

Problem Statement
Clinical diagnosis of pulmonary tuberculosis (PTB) is frequently hindered by symptom overlap with other respiratory conditions such as pneumonia and COPD, leading to delays and misdiagnoses. While low bacillary loads may evade microbiological detection, the interpretation of clinical and radiographic findings often suffers from gaps, particularly in resource-constrained, high-burden settings. Although the World Health Organization (WHO) advocates for systematic screening using symptoms, chest radiography, and rapid molecular tests, the specific clinical and radiographic patterns of confirmed TB and their relationship to disease progression and diagnostic decision-making remain insufficiently characterized in geographically underserved regions. This study addresses these evidence deficits by systematically describing the clinical symptoms and chest radiographic abnormalities among patients with confirmed TB in southwestern Ethiopia.

Methodology
The study employed a cross-sectional design conducted between January 2, 2018, and June 26, 2020, across three administrative zones in southwestern Ethiopia: Bench-Sheko, Kaffa, and Sheka. A multistage sampling technique was used to select high-risk zones and healthcare facilities equipped with TB diagnostic modalities (specifically Xpert MTB/RIF).

• Participants: A total of 410 confirmed TB cases were analyzed, comprising 328 (80%) PTB cases, 79 (19.3%) extrapulmonary TB (EPTB) cases, and 3 (0.73%) cases involving both forms. The cohort was predominantly newly diagnosed (99.3%) with a mean age of 27.65 years and 55% male representation.
• Data Collection: Data were gathered via structured face-to-face interviews by healthcare professionals and review of medical records. Diagnostic confirmation relied on smear microscopy, GeneXpert Ultra, chest X-rays, and, for EPTB, lesion biopsy or fine-needle aspiration cytology (FNAC).
• Analysis: Descriptive and inferential statistics were performed using Stata 14/16. Categorical variables were analyzed using Pearson chi-square tests and logistic regression to identify factors associated with PTB. Standardized odds ratios (SORs) were derived from multiple logistic regression analyses to assess clinical predictors.

Key Results
The study identified a high prevalence of the classic "TB tetrad" (prolonged cough, night sweats, fever, and appetite loss) among participants.

• Clinical Findings:

  • Systemic Symptoms: Night sweats were reported in 93.17%, fatigue in 90%, appetite loss in 89%, and weight loss in 88%. Fever was common, with 87% presenting with temperatures <39°C and only 5.4% with high fever (>40°C).
  • Respiratory Symptoms: Cough was reported in 94.2% of cases, with 77% being prolonged (>3 weeks) and 91% productive. Sputum character varied, with mucoid sputum being the most common (74.9%).
  • Predictors: Statistical analysis revealed that chest abnormalities (OR=6.87), prolonged cough (OR=6.4), mucoid sputum (OR=14.6), and mucopurulent sputum (OR=56.4) were strong predictors of active PTB. Conversely, lymphadenopathy showed an inverse association (OR=0.16). Appetite loss, pleural effusion, and fibrosis showed weak associations. Notably, high fever (>40°C) was a less likely predictor of PTB (OR=0.19).

• Radiological Findings:

  • Chest X-ray abnormalities were detected in 16.3% of the total cohort. Specific findings included pleural effusion (11.2%), cavitary formations (13%), and enlarged lymph nodes (14%).
  • A significant portion of the data (approx. 70-75% for specific features) was unrecorded or missing, reflecting limitations in diagnostic infrastructure and data completeness.
  • Among the recorded abnormalities, pleural effusion (OR=2.41) and fibrosis (OR=2.31) showed weak associations with PTB, while the presence of any chest abnormality was a strong predictor.

Significance and Claims
The paper claims that the integration of clinical assessments with radiographic evaluations offers a promising approach to improving disease detection in high-burden settings where microbiological confirmation may be limited. The study highlights that while the classic tetrad of symptoms is prevalent, specific sputum characteristics (mucoid/mucopurulent) and the duration of cough are statistically significant predictors of PTB.

The authors emphasize the diagnostic value of radiology, noting that chest abnormalities significantly increase the odds of PTB. However, the study modestly concludes that while these findings support the WHO's four-symptom screening strategy, the interpretation of radiographic features (such as lymphadenopathy) requires caution, as some findings (like enlarged lymph nodes) may be inversely associated with PTB in this specific cohort. The research aims to provide context-specific evidence to support more comprehensive clinical evaluations and earlier recognition of disease in resource-constrained environments.

Limitations
The authors acknowledge several limitations, including the imbalance in sample sizes between PTB and EPTB groups, the unequal distribution of participants across study sites (heavily weighted toward Bench-Sheko), and disruptions in data collection in the Sheka zone due to ethnic conflict and insecurity. These factors may have introduced selection and information biases, potentially leading to an underreporting of the actual TB burden in some sites. Additionally, the study notes that multidrug-resistant (MDR) TB was not fully addressed in this specific publication.