Table 4.
Time points (hours) | |||||
---|---|---|---|---|---|
0 | 3 | 6 | 12 | 24 | |
H-FABP | |||||
| |||||
Sensitivity | 37% (27–49%) | 99% (92–100%) | 93% (85–97%) | 77% (66–85%) | 40% (29–52%) |
Specificity | 96% (78–99%) | 91% (71–98%) | 95% (77–99%) | 96% (78–99%) | 96% (78–99%) |
Positive predictive value | 96% (82–99%) | 97% (90–99%) | 99% (92–100%) | 98% (90–100%) | 97% (83–99%) |
Negative predictive value | 32% (22–44%) | 95% (76–99%) | 80% (61–91%) | 57% (41–71%) | 33% (23–46%) |
Positive likelihood ratio | 8.17 (1.2–56.8) | 10.35 (2.77–38.70) | 19.52 (2.88–132.33) | 16.97 (2.49–115.66) | 8.80 (1.27–61.01) |
Negative likelihood ratio | 0.66 (0.54–0.81) | 0.02 (0.00–0.11) | 0.07 (0.03–0.17) | 0.24 (0.15–0.37) | 0.63 (0.51–0.78) |
Risk ratio | 1.48 (1.25–1.75) | 19.44 (2.88–131.41) | 4.93 (2.25–10.79) | 2.32 (1.60–3.35) | 1.50 (1.26–1.79) |
| |||||
GPBB | |||||
| |||||
Sensitivity | 40% (29–52%) | 97% (90–99%) | 86% (76–92%) | 70% (59–80%) | 23% (15–34%)* |
Specificity | 96% (78–99%) | 81% (60–92%) | 86% (65–95%) | 95% (77–99%) | 96% (78–99%) |
Positive predictive value | 97% (83–99%) | 95% (87–98%) | 95% (87–98%) | 98% (90–100%) | 94% (73–99%) |
Negative predictive value | 33% (23–46%) | 90% (69–97%) | 64% (46–79%) | 49% (34–64%) | 28% (19–39%) |
Positive likelihood ratio | 8.80 (1.27–61.01) | 5.10 (2.11–12.33) | 6.01 (2.10–17.22) | 14.79 (2.17–100.74) | 5.03 (0.71–35.80) |
Negative likelihood ratio | 0.63 (0.51–0.78) | 0.04 (0.01–0.14) | 0.16 (0.09–0.03) | 0.31 (0.21–0.45) | 0.81 (0.69–0.95) |
Risk ratio | 1.50 (1.26–1.79) | 8.98 (2.42–33.35) | 2.67 (1.62–4.40) | 1.91 (1.42–2.59) | 1.39 (1.21–1.60) |
Level of statistical significance
P < 0.01 between sensitivity of H-FABP and GPBB at 24 hours after the admission as calculated with Z-test. Values in parentheses represent 95% confidence intervals; h, hours after admission.