Table 2.
Comparison of diagnostic methods for the detection of Mtb in hospitalized versus ambulatory and in smear-positive versus smear-negative patients
| A. | All patients | Hospitalized patients | Ambulatory patients | |||
| TB+ (n = 48) | Diagnostic yield (95 % CI) | TB+ (n = 35) | Diagnostic yield (95 % CI) | TB+ (n = 13) | Diagnostic yield (95 % CI) | |
| MODS | 46 | 95.8 % (85.7-99.5) | 34 | 97.1 % (85.1-99.9) | 12 | 92.3 % (64.0-99.8) |
| LJ | 38 | 79.2 % (65.0-89.5) | 31 | 88.6 % (73.3-96.8) | 7 | 53.8 % (25.1-80.8) |
| ZN | 30 | 62.5 % (47.3-76.0) | 22 | 62.9 % (44.9-78.5) | 8 | 61.5 % (31.6-86.1) |
| B. | All patients | Smear-positive | Smear-negative | |||
| TB+ (n = 48) | Diagnostic yield (95 % CI) | TB+ (n = 30) | Diagnostic yield (95 % CI) | TB+ (n = 18) | Diagnostic yield (95 % CI) | |
| MODS | 46 | 95.8 % (85.7-99.5) | 29 | 96.6 % (82.8-99.9) | 17 | 94.4 % (72.7-99.9) |
| LJ | 38 | 79.2 % (65.0-89.5) | 27 | 90.0 % (73.4-97.9) | 11 | 61.0 % (35.7-82.7) |
TB+ was defined as any patient with sputum or gastric sample that yielded a positive culture result by LJ or MODS. As the reference standard is a composite from all tests that are being evaluated, we can only comment on diagnostic yield rather than sensitivity and specificity. Panel A: Comparison of diagnostic yield of MODS versus LJ and ZN in all patients, hospitalized patients, and ambulatory patients who were TB positive. Panel B: Subgroup comparison of the diagnostic yield of MODS versus LJ in culture positive patients, sub-grouped into all patients, smear positive patients, and smear negative patients
n number of patients in each group; CI confidence interval; TB + samples tested positive for Mtb by each respective method (MODS, LJ, or ZN)