Table 3.
Lipid/oxidative marker at enrollment Median (25th, 75th) |
Derivation cohort | Replication cohort | ||||
---|---|---|---|---|---|---|
Hypolipoprotein cluster (n = 58) | Normolipoprotein cluster (n = 110) | p Value+ | Hypolipoprotein cluster (n = 26) | Normolipoprotein cluster (n = 60) | p Value^ | |
Total cholesterol (mg/dL) | 67 (53, 84) | 108 (81, 122) | < 0.001 | 87 (70, 102) | 101 (82, 133) | 0.047 |
HDL-C (mg/dL) | 7 (5, 16) | 26 (16, 33) | < 0.001 | 10 (5, 30) | 34 (20, 46) | < 0.001 |
LDL-C (mg/dL) | 30 (19, 44) | 52 (36, 68) | 0.001 | 41 (29, 68) | 46 (35, 66) | 0.267 |
PON-1 (mg/dL) | 35.81 (22.4, 67.87) | 65.99 (40.03, 114.6) | 0.001 | 46.77 (23.59, 104.98) | 89.42 (43.49, 128.59) | 0.017 |
ApoA-I (ng/mL) | 754,631 (583,580, 952,226) | 1,400,000 (1,000,000, 1,625,000) | < 0.001 | 475785 (185360, 763480) | 793130 (658255, 1030550) | < 0.001 |
ICAM-1 (ng/mL) | 589.75 (455.82, 750.58) | 346.502 (271.251, 485.513) | < 0.001 | 649.15 (516.80, 929.30) | 351.92 (222.79, 384.97) | < 0.001 |
Cardiovascular SOFA | 1 (1, 3) | 1 (0, 1) | < 0.001 | 2 (0, 4) | 1 (0, 3) | 0.172 |
Neurologic SOFA | 2 (0, 3) | 0 (0, 2) | 0.003 | 0 (0, 2) | 2 (0, 3) | 0.020 |
Coagulation SOFA | 0 (0, 1) | 0 (0, 1) | 0.165 | 1 (0, 2) | 0 (0, 1) | 0.005 |
Hepatic SOFA | 0 (0, 1) | 0 (0, 0) | 0.019 | 2 (1, 2) | 0 (0, 0) | < 0.001 |
Renal SOFA | 2 (0, 3) | 1 (0, 2) | 0.018 | 1 (0, 1) | 1 (1, 3) | 0.190 |
Respiratory SOFA | 2 (0, 3) | 0 (0, 2) | 0.005 | 1 (0, 3) | 2 (1, 3) | 0.007 |
Systolic BP (mm Hg) | 97 (88.25, 114) | 113 (100.5, 126.75) | < 0.001 | 101 (92, 111) | 108 (97, 123) | 0.079 |
Temperature (˚F) | 98.6 (98.1, 99.8) | 99.6 (98.6, 101.3) | 0.003 | 99.0 (97.6, 100.2) | 99.4 (98.4, 100.5) | 0.296 |
Total SOFA score | 8 (6, 10) | 4 (2, 6) | < 0.001 | 7 (5, 12) | 7 (5, 10) | 0.741 |
Bold values are significant after Bonferroni correction for multiple comparisons
Comparisons made using Wilcoxon rank-sum test; +after Bonferroni adjustment for multiple tests, significant if < 0.001 (0.05/47 tests). ^For the replication cohort, after Bonferroni adjustment for multiple tests, significant if p-values < 0.003 (0.05/15 tests) were considered significant. Comparisons are bivariate between Hypolipoprotein and Normolipoprotein Clusters separately for derivation and validation cohorts