Correlation between sputum Xpert MTB/RIF Ultra semi-quantitative load with acid-fast bacilli smear grade

The World Health Organisation has recommended the rapid, semi-quantitative Xpert MTB/RIF or Xpert MTB/RIF Ultra point-of-care testing (POCT) as the initial test in TB evaluation with the hope of replacing AFB smears (traditionally used to estimate disease burden and infectiousness) due its superior sensitivity and specificity. We examine the correlation between Xpert Ultra semi-quantitative results with AFB smear grades. Xpert Ultra POCT was performed on one ml aliquots of sputum samples of patients undergoing TB evaluation. The remaining sample was sent to the laboratory for liquefaction, centrifugation and AFB smear. A total of 3435 sputum samples underwent Xpert Ultra and AFB smear: 348 (10.1%) were positive for Mycobacterium Tuberculosis (MTB) on Xpert Ultra. Table 1 shows the relationship between AFB smear grades and Xpert Ultra semi-quantitative load. Ninety-one (26.1%) samples were smear positive. Twenty-seven (64.3%) of those with high load were smear ≥3+ and 78 (98.7%) of those with trace load were smear negative. Twenty-eight (10.4%) with low, very low and trace load were smear positive: 6 of these showing smear ≥2+. <fig><object-id>erj;64/suppl_68/PA633/F1</object-id><object-id>F1</object-id><object-id>F1</object-id><graphic></graphic></fig> Results suggest poor correlation between Xpert Ultra POCT semi-quantitative load and smear grade. Further statistical analysis is underway to verify this. Clinicians should not dismiss low/ very low/ trace Xpert Ultra results as insignificant, as 10% of these are smear positive.