Romulo Iii Babasa, Karl Christian Aralar, Joanah Bithao

INTRODUCTION AND OBJECTIVES

Pulmonary Tuberculosis (PTB) is one of the leading causes of morbidity in the Philippines.

Its prevalence rate is 438/100,000 population.

The key to stopping the spread of the disease is accurate diagnosis and consistency of treatment and follow-up.

Diagnosis is based on clinical findings and confirmation of infection by chest x-ray, sputum analysis, and molecular testing. This standard work-up is hampered by its cost, availability of diagnostic centers and the tediousness of testing especially in resource-limited areas.

Recently, Lung Ultrasound (LUS) has become a valuable modality in diagnosing pulmonary diseases because of its portability, accuracy and cost-effectiveness.

PTB is one of the few diseases that have not been extensively studied with regards to sonographic diagnosis.

It is our intention to determine the diagnostic accuracy of LUS in patients with suspected PTB using chest radiography, sputum microscopy and nucleic acid amplification testing as the reference standard.

If proven accurate, LUS can potentially simplify the detection of PTB in patients, facilitating quick diagnosis and dispensing of treatment.

METHODS

We conducted a cross-sectional, prospective observational study of 131 consecutive patients, 18 years old and above, with presumptive PTB in a tertiary hospital outpatient clinic from September to November 2018.

After securing consent, lung ultrasound was performed on all subjects to detect previously observed findings suggestive of PTB, namely subpleural nodules, pleural effusion, consolidation, and C-lines.

A single emergency medicine physician trained in lung ultrasound performed all the scans and a senior emergency medicine consultant with fellowship training in point-of-care ultrasound (POCUS) interpreted the clips.

All patients subsequently underwent x-ray studies of the chest, direct sputum microscopy, and nucleic acid amplification testing using a commercially available assay* to confirm presence of TB infection.

The sonographers were blinded to the results of the diagnostic work-up.

RESULTS

Out of 131 patients, 102 patients had positive LUS findings of PTB.

LUS compared to standard work-up had a specificity of 92.59% (75.71% to 99.09%), sensitivity of 55.77% (45.70% to 65.50%), and accuracy of 63.36% (54.50% to 71.60%) at 95% confidence interval (CI).

Positive and Negative LR were 7.53 (1.96 to 28.88) and 0.48 (0.38 to 0.61) respectively.

Furthermore, LUS had a PPV of 96.67% and a NPV of 25.21%.

CONCLUSION

Our study demonstrated that LUS has robust specificity and can be an acceptable, simplified diagnostic test to detect PTB infection.

This is compared to a reference standard consisting of positive chest x-ray findings, sputum microscopy, and nucleic acid amplification testing.

Follow-up studies comparing LUS to the current gold standard of TB diagnosis – the sputum culture, should be done in various stages of the disease to further validate the accuracy of LUS in PTB diagnosis