Skip to main content

Table 2 Cytopathology/histopathology and ADA test performance with respect to each extrapulmonary TB form and measurands

From: Bayesian latent class analysis produced diagnostic accuracy estimates that were more interpretable than composite reference standards for extrapulmonary tuberculosis tests

 

Cytopathology/histopathology with respect to EPTB form

ADA test with respect to EPTB form

Cytopathology/histopathology with respect to measurand

ADA test with respect to measurand

TB lymphadenitis

Sensitivity (%) (95% CrI)

98.7 (96.1, 99.7)

NA

99.8 (98.2, 100)

NA

Specificity (%) (95% CrI)

83.5 (74.6, 93.2)

NA

91.7 (79.6, 99.5)

NA

TB meningitis

Sensitivity (%) (95% CrI)

NA

83.1 (64.8, 94.5)

NA

91.9 (73.3, 99.6)

Specificity (%) (95% CrI)

NA

90.7 (83.8, 98.1)

NA

95.0 (86.5, 99.8)

TB pleuritis

Sensitivity (%) (95% CrI)

NA

94.6 (88.8, 98.1)

NA

97.5 (91.9, 98.1)

Specificity (%) (95% CrI)

NA

74.7 (65.1, 90.3)

NA

86.8 (70.1, 99.4)

  1. The performance of each test with respect to type of EPTB, otherwise referred to as the target condition, and with respect to the measurand is provided. Regarding non-specific tests ADA and cytopathology/histopathology, we have discerned between their performance at measuring their measurands versus the target condition, EPTB. This parametrization more accurately captures the nuance of the testing scenario, as ‘target condition’ and ‘measurand’ are not interchangeable entities [25]. Reassuringly, they both performed better at measuring their measurand (ADA level or change in cell morphology) than measuring EPTB
  2. ADA adenosine deaminase, CrI credible interval, EPTB extrapulmonary TB, NA not applicable, TB tuberculosis