Diagnostic performance of fluorescent light-emitting diode microscopy for tuberculous lymphadenitis in a high-burden setting.

ABSTRACT

OBJECTIVE: Diagnosis of tuberculous lymphadenitis using fine-needle aspiration cytology is a simple and safe but low-specificity method, whereas conventional smear microscopy has variable sensitivity due to low bacterial load. We evaluated the diagnostic performance of fluorescent light-emitting diode (LED) microscopy on routinely collected fine-needle aspirates from tuberculous lymphadenitis presumptive cases. METHODS: Fine-needle aspirates were collected from patients clinically suspected of having tuberculous lymphadenitis as part of routine diagnosis. Smear preparation was performed from the aspirate and processed for cytology, conventional Ziehl-Neelsen and LED microscopy. The remaining aspirate was processed for culture on Lowenstein-Jensen media. Capilia TB-Neo test was used to differentiate M. tuberculosis complex from non-tuberculous mycobacteria. RESULT: A total of 144 tuberculous lymphadenitis presumptive cases were included. 66.7% (96/144) were positive for M. tuberculosis complex on culture. Only one isolate was identified as non-tuberculous mycobacteria. The detection rates of Ziehl-Neelsen and LED microscopy were 18.8% (27/144) and 34% (49/144), respectively. As compared to culture, sensitivity was 25.0% [95% CI 16.3-33.7] for Ziehl-Neelsen microscopy and 45.8% [95% CI 35.9-55.8] for LED microscopy. The specificity was 93.8% [95% CI 86.9-100] for Ziehl-Neelsen microscopy and 89.6% [95% CI 80.9-98.2] for LED microscopy. LED microscopy showed a statistically significant increase in sensitivity and similar specificity compared to Ziehl-Neelsen microscopy. Mean reading time of positive slides was 2.62 min/slide for Ziehl-Neelsen and 1.60 min/slide for LED microscopy. Cytology showed sensitivity of 82.3% and specificity of 54.2%. LED microscopy detected TB bacilli in 33.3% of cases cytologically classified as suppurative abscess. CONCLUSION: The LED microscopy for tuberculous lymphadenitis had significantly higher sensitivity and shorter screening time than Ziehl-Neelsen microscopy. Use of LED microscopy among cases classified as suppurative abscess on fine-needle aspirate cytology improves evidence-based diagnosis of presumptive tuberculous lymphadenitis cases. Moreover, LED microscopy could be considered as an alternative approach in settings where fine-needle aspirate cytology is impractical.