Random blinded rechecking of sputum acid-fast bacilli smear using fluorescence microscopy: 8 years' experience

Area of Science:

• Microbiology
• Laboratory Diagnostics
• Quality Assurance

Background:

• The Hong Kong TB Reference Laboratory processes approximately 400 sputum smears daily using fluorescence microscopy (FM).
• High-volume laboratories require robust quality assurance methods for accurate tuberculosis diagnosis.

Purpose of the Study:

• To evaluate the efficacy of blinded rechecking as a quality assurance strategy for fluorescence microscopy (FM) in a high-throughput setting.
• To identify potential challenges and areas for improvement in the blinded rechecking process.

Main Methods:

• A proportion of sputum smears (2.5-5%) were randomly selected, relabeled, and re-examined by a separate technician.
• Smears were restained and re-examined, with discrepancies resolved by a supervisor.
• Data collected from 2003 to 2010.

Main Results:

• Low false-negative (LFN) error rates were generally within acceptable limits (0.10-0.27%), indicating high sensitivity.
• However, LFN errors were more frequent among the rechecking technicians (0.28-0.62%).
• Low false-positive (LFP) cases (0.13-0.75%) were observed, though some were confirmed by mycobacterial culture.

Conclusions:

• Blinded rechecking serves as a valuable quality assurance tool in high-throughput laboratories utilizing FM for tuberculosis detection.
• Addressing issues related to restaining, such as increased background fluorescence, is crucial to minimize LFP errors.
• Further optimization is needed before widespread field application of blinded rechecking, especially in settings without routine mycobacterial cultures.