TABLE 2.
Univariable and multivariable results for the 3 analyses: inflammation, clusters, and C4d/DSA
| Univariate |
Multivariable |
|
|---|---|---|
| Comparison | HR [95% CI]; P-value; adjusted P* | HR [95% CI]; P-value; adjusted P* |
| iatr score (0–3)a (N = 336) | ||
| iatr = 3 vs iatr = 2 | 3.2 [1.9, 5.6]; P < .001; P*< .001 | 1.5 [0.7, 3.4]; P = .300; P* = .728 |
| iatr = 3 vs iatr = 1 | 4.4 [2.5, 7.6]; P < .001; P*< .001 | 2.3 [1.0, 5.3]; P = .043; P* = .178 |
| iatr = 3 vs iatr = 0 | 7.7 [4.2, 14.2]; P < .001; P*< .001 | 4.0 [1.6, 10.1]; P = .003; P* = .018 |
| iatr = 2 vs iatr = 1 | 1.3 [0.9, 2.0]; P = .138; P* = .449 | 1.5 [0.9, 2.5]; P = .091; P* = .328 |
| iatr = 2 vs iatr = 0 | 2.4 [1.5, 3.8]; P < .001; P* = .001 | 2.6 [1.4, 5.0]; P = .004; P* = .019 |
| iatr = 1 vs iatr = 0 | 1.8 [1.1, 2.8]; P = .015; P* = .073 | 1.7 [1.0, 2.9]; P = .044; P* = .183 |
| iatr and i scoresb (N = 336) | ||
| Some i vs i = 0, iatr>0 | 1.2 [0.9, 1.8]; P = .249; P* = .481 | 1.0 [0.6, 1.5]; P = .908; P* = .993 |
| Some i vs no inflammation | 2.4 [1.6, 3.8]; P < .001; P* < .001 | 1.8 [1.0, 3.0]; P = .030; P* = .076 |
| i = 0, iatr > 0 vs no inflammation | 2.0 [1.2, 3.1]; P = .004; P* = .012 | 1.8 [1.1, 3.0]; P = .022; P* = .057 |
| Cluster analysis (1–6)c (N = 237) | ||
| Clusters 6 vs 5 | 2.3 [1.0, 5.4]; P = .050; P* = .368 | 1.9 [0.8, 4.6]; P = .159; P* = .723 |
| Clusters 6 vs 4 | 1.4 [0.6, 3.3]; P = .497; P* = .984 | 1.2 [0.5, 3.0]; P = .706; P* = .999 |
| Clusters 6 vs 3 | 2.3 [1.0, 5.2]; P = .040; P* = .313 | 2.0 [0.8, 4.7]; P = .116; P* = .618 |
| Clusters 6 vs 2 | 3.5 [1.5, 8.0]; P = .004; P* = .042 | 3.4 [1.4, 8.0]; P = .006; P* = .065 |
| Clusters 6 vs 1 | 6.9 [3.1, 15.5]; P < .001; P* < .001 | 5.8 [2.5, 13.7]; P < .001; P*< .001 |
| Clusters 5 vs 4 | 0.6 [0.3, 1.2]; P = .157; P* = .717 | 0.6 [0.3, 1.4]; P = .251; P* = .861 |
| Clusters 5 vs 3 | 1.0 [0.5, 1.9]; P = .997; P* > .999 | 1.0 [0.5, 2.1]; P = .890; P* > .999 |
| Clusters 5 vs 2 | 1.5 [0.8, 2.9]; P = .245; P* = .855 | 1.8 [0.8, 4.0]; P = .165; P* = .734 |
| Clusters 5 vs 1 | 3.0 [1.5, 5.7]; P = .001; P* = .013 | 3.1 [1.5, 6.1]; P = .001; P* = .016 |
| Clusters 4 vs 3 | 1.7 [0.8, 3.5]; P = .140; P* = .680 | 1.7 [0.8, 3.4]; P = .165; P* = .735 |
| Clusters 4 vs 2 | 2.6 [1.2, 5.3]; P = .011; P* = .110 | 2.8 [1.3, 6.1]; P = .009; P*= .091 |
| Clusters 4 vs 1 | 5.1 [2.5, 10.3]; P < .001; P* < .001 | 4.9 [2.4, 9.9]; P < .001; P* < .001 |
| Clusters 3 vs 2 | 1.5 [0.8, 2.8]; P = .221; P* = .825 | 1.7 [0.8, 3.4]; P = .149; P* = .700 |
| Clusters 3 vs 1 | 3.0 [1.6, 5.4]; P < .001; P* = .004 | 2.9 [1.6, 5.4]; P < .001; P* = .007 |
| Clusters 2 vs 1 | 2.0 [1.1, 3.7]; P = .029; P* = .247 | 1.7 [0.8, 3.5]; P = .130; P* = .654 |
| C4d and DSAd (N = 172) | ||
| C4d+ / DSA+ vs C4d+ / DSA- | 1.6 [0.8, 3.2]; P = .186; P* = .549 | 0.8 [0.3, 1.9]; P = .598; P* = .953 |
| C4d+ / DSA+ vs C4d- / DSA+ | 1.5 [0.8, 2.9]; P = .220; P* = .610 | 0.8 [0.4, 2.0]; P = .708; P* = .982 |
| C4d+ / DSA+ vs C4d- / DSA- | 4.8 [2.5, 9.3]; P < .001; P*< .001 | 3.6 [1.5, 8.3]; P = .003; P* = .016 |
| C4d+ / DSA- vs C4d- / DSA+ | 0.9 [0.4, 2.1]; P = .871; P* = .998 | 1.1 [0.4, 2.8]; P = .876; P* = .999 |
| C4d+ / DSA- vs C4d- / DSA- | 3.0 [1.3, 6.8]; P = .007; P*=.035 | 4.5 [1.7, 11.7]; P = .002; P* = .010 |
| C4d- / DSA+ vs C4d- / DSA- | 3.2 [1.5, 6.7]; P = .002; P* = .009 | 4.2 [1.7, 10.2]; P = .001; P* = .008 |
Models are stratified by clinical center. Unadjusted and adjusted (Tukey-Kramer, P*) P values are provided.
CI, confidence interval; DSA, donor specific antibody; HR, hazard ratio; iatr, inflammation in areas of atrophy.
iatr score multivariable analyses were adjusted for creatinine at biopsy, treatment for acute rejection after biopsy (yes/no), and central i, t, ci, and ct scores.
iatr and i group multivariable analyses were adjusted for creatinine at biopsy, treatment for acute rejection after biopsy (yes/no), and central t, ci, and ct scores.
Cluster multivariable analyses were adjusted for creatinine at biopsy, time from transplant to biopsy and treatment for acute rejection after biopsy (yes/no).
C4d and DSA multivariable analyses were adjusted for age at biopsy, race (black or nonblack), prior transplant status, serum creatinine at biopsy, treatment for acute rejection after biopsy (yes/no), and central i, t, g, and ptc scores.