Table 4.
Characteristic at VAD implantation |
||||||
---|---|---|---|---|---|---|
Adverse Event Type | Gender (male) | Age (>54) | Implantation Era (before 2001) | Intention to Treat (Bridge to Transplant) | VAD Type (LVAD) | Model Fit, χ2 (df=5) |
Any AE | 0.87 (0.61, 1.26) |
0.91 (0.66, 1.24) |
0.94 (0.67, 1.31) |
0.74 (0.48, 1.13) |
1.03 (0.75, 1.42) |
3.83 |
Arrhythmia | 1.65 (0.82, 3.29) |
0.95 (0.58, 1.54) |
0.70 (0.42, 1.16) |
2.14 (0.96, 4.76) |
2.36** (1.32, 4.21) |
16.45** |
Tamponade | 0.98 (0.50, 1.93) |
1.15 (0.66, 2.01) |
1.24 (0.70, 2.19) |
1.50 (0.62, 3.66) |
1.07 (0.60, 1.93) |
2.26 |
RV Failure | 1.60 (0.60, 4.34) |
1.48 (0.70, 3.14) |
1.59 (0.68, 3.76) |
0.35* (0.14, 0.89) |
---b | 6.36 |
Thromboembolism | 0.51 (0.14, 1.81) |
3.26* (1.09, 9.80) |
0.12* (0.02, 0.91) |
0.69 (0.23, 2.04) |
1.99 (0.61, 6.55) |
14.47* |
Respiratory | 0.60 (0.31, 1.15) |
1.01 (0.55, 1.86) |
0.71 (0.36, 1.37) |
0.95 (0.44, 2.07) |
0.60 (0.33, 1.09) |
9.19 |
Neurologic | 0.53 (0.27, 1.03) |
1.21 (0.66, 2.23) |
2.17* (1.20, 3.93) |
2.62 (0.76, 8.85) |
1.28 (0.69, 2.39) |
16.07** |
Renal | 1.83 (0.66, 5.09) |
1.03 (0.47, 2.25) |
0.36 (0.12, 1.09) |
0.43 (0.18, 1.04) |
0.40* (0.18, 0.89) |
15.34** |
Hepatic | 1.43 (0.42, 4.95) |
0.97 (0.35, 2.64) |
0.27 (0.06, 1.23) |
0.58 (0.19, 1.82) |
0.37 (0.13, 1.03) |
9.81 |
Bleeding | 0.92 (0.56, 1.50) |
0.84 (0.54, 1.30) |
0.87 (0.55, 1.38) |
0.78 (0.45, 1.38) |
0.56** (0.36, 0.85) |
11.79* |
Infection | 1.43 (0.81, 2.51) |
0.80 (0.50, 1.27) |
1.41 (0.87, 2.27) |
0.85 (0.44, 1.63) |
0.64 (0.41, 1.00) |
7.37 |
Reoperation | 0.64 (0.36, 1.15) |
0.95 (0.56, 1.61) |
0.92 (0.52, 1.60) |
0.76 (0.39, 1.49) |
1.29 (0.74, 2.24) |
4.07 |
Device Malfunction | 0.48 (0.16, 1.41) |
1.21 (0.47, 3.28) |
0.30 (0.09, 1.07) |
2.47 (0.54, 11.23) |
0.76 (0.29, 1.95) |
6.67 |
p<.05
p<.01
p<.001
Hazard ratios <1.00 indicate a reduced risk of an AE associated with a given characteristic. Hazard ratios >1 indicate increased risk. Time to each AE was regressed on the set of baseline characteristics.
VAD type could not be included as a predictor of RV failure because patients who received BiVADs had right ventricular support and thus could not develop new RV failure.