Comparison of same day diagnostic tools including Gene Xpert and unstimulated IFN-γ for the evaluation of pleural tuberculosis: a prospective cohort study

Table 2 Per patient diagnostic accuracy of Xpert MTB/RIF, IFN-γ, and ADA for the diagnosis pleural tuberculosis

Assay	Definite-TB vs. Non TB
Assay	Sensitivity% (95% CI)	Specificity% (95% CI)	PPV% (95% CI)	NPV% (95% CI)	+LR (95% CI)	-LR (95% CI)
Xpert MTB/RIF	22.5% *^#□$£ (12.4, 37.6) 9/40	98% (89.2, 99.7) 47/48	90% (59.6, 98.3) 9/10	60.3% ^†◊●▪ (49.2, 70.4) 48/78	10.8 (1.43, 81.65)	0.79^¢°¤§¿ (0.67, 0.94)
IFN-γ	92.5%*^O (80.2, 97.5) 37/40	95.9% (86.1, 98.9) 46/48	94.9% (83.2, 98.6) 37/39	93.9%^†∞ (83.5, 97.9) 46/49	22.20 (5.70, 86.46)	0.08^¢ð (0.03, 0.23)
ADA (Clinical cut point)	79% ^∆#× (63.7, 89) 30/38	92.7% (80.6, 97.5) 38/41	91% (73.4, 95.4) 30/33	82.7% ^‡◊ (69.3, 91.0) 38/46	10.79 (3.59, 32.47)	0.23^° (0.12, 0.42)
ADA (Rule-in cut point)	55.3% ^□O×€¥ (39.8, 69.9) 21/38	95.2% (83.9, 98.7) 39/41	91.4% (73.3, 97.6) 21/23	69 · 7% ^∞⌂ (56.7, 80.1) 39/56	11.33 (2.85, 45.1)	0.47^¤ðþ (0.33, 0.67)
Xpert MTB/RIF followed by IFN-γ if MTB/RIF-negative	92.5%^$€ (80.2, 97.5) 37/40	93.8% (83.2, 97.9) 45/48	92.5% (80.2, 97.5) 37/40	93.8%^●⌂ (83.2, 97.9) 45/48	14.80 (4.93, 44.43)	0.08^§þ (0.03, 0.24)
Xpert MTB/RIF followed by ADA (clinical cut-point) if MTB/RIF-negative	81.6%^£¥ (66.6, 90.8) 31/38	92.7% (80.6, 97.5) 38/41	91.2% (77.1, 97) 31/34	84.5%^▪ (71.3, 92.3) 38/45	11.15 (3.70, 33.49)	0.20^¿ (0.10, 0.39)
p-value	*; ^#; ^$ and ^£: p < 0.0001 ^O; ^□; ^×; ^€ and ^¥: p < 0.05			^† and ^●: P < 0.0001 ^◊; ^∞; ▪ and ^⌂: p < 0.05		^¢; ^°; ^¤; ^§; ^¿; ^ð and ^þ: non overlapping CI

A positive MTB (Mycobacterium tuberculosis) fluid culture and/or positive MTB biopsy culture and/or histology in keeping with MTB infection used as a reference. PPV: positive predictive value, NPV: negative predictive value, +LR: positive likelihood ratio, -LR: negative likelihood ratio, CI: confidence interval, IFN-γ: interferon gamma, ADA: adenosine deaminase. IFN- γ cut point of 107.7 pg/ml determined by ROC. ADA clinical cut point of 30 U/L is used for clinical decision-making at Groote Schuur hospital. Rule-in ADA cut point of 48.85U/l determined by ROC.
Superscript symbols: *; ^#; ^$; ^£: ^O; ^□; ^×; ^€; ^¥; ^†; ^●; ^◊; ^∞; ▪; ^⌂; ^¢; ^°; ^¤; ^§; ^¿; ^ð and ^þwere used to indicate which groups were being compared for statistical analysis.

ISSN: 1471-2466