Table 3.
Univariate and Multivariable-adjusted Hazard (models 1 and 2) of postoperative periprosthetic fracture following Primary Total Hip Replacement using Cox regression analysis
| Variable | Total (n=14,065) | Periprosthetic Fractures (n=305) | Univariate Hazard Ratio (95% CI) | Multivariablea model 1 Hazard Ratio (95% CI) | Multivariableb model 2 Hazard Ratio (95% CI) |
|---|---|---|---|---|---|
| Gender | p=0.002 | p=0.001 | p=0.001 | ||
| Male | 6,819 | 117 (2%) | 1.00 (REF) | 1.00 (REF) | 1.00 (REF) |
| Female | 7,246 | 188 (3%) | 1.44 (1.15, 1.82) | 1.48 (1.17, 1.88) | 1.49 (1.18, 1.89) |
| Age Category (years) | p=0.02 | p=0.06 | |||
| ≤60 | 4,455 | 104 (2%) | 1.00 (REF) | 1.00 (REF) | |
| 61–70 | 4,252 | 98 (2%) | 0.88 (0.67, 1.17) | 0.91 (0.68, 1.22) | |
| 71–80 | 4,119 | 71 (2%) | 0.77 (0.57, 1.04) | 0.78 (0.55, 1.10) | |
| >80 | 1,239 | 32 (3%) | 1.45 (0.97, 2.15) | 1.37 (0.88, 2.14) | |
| Body Mass Index, kg/m2 | p=0.25 | ||||
| Normal, < 25.0 | 3,429 | 88 (3%) | 1.00 (REF) | ||
| Overweight, 25–29.9 | 5,334 | 108 (2%) | 0.84 (0.63, 1.12) | ||
| Obese, 30–39.9 | 4,589 | 94 (2%) | 1.12 (0.62, 2.01) | ||
| Morbidly Obese, ≥40.0 | 641 | 13 (2%) | 0.78 (0.59, 1.03) | ||
| Deyo-Charlson Index | p<0.001 | p<0.001 | p<0.001 | ||
| 0 | 7,520 | 136 (2%) | 1.00 (REF) | 1.00 (REF) | 1.00 (REF) |
| 1 | 2,393 | 49 (2%) | 1.17 (0.84, 1.62) | 1.04 (0.74, 1.46) | 1.03 (0.73, 1.45) |
| 2 | 1,640 | 51 (3%) | 1.82 (1.32, 2.52) | 1.74 (1.25, 2.43) | 1.74 (1.24, 2.43) |
| 3+ | 2,512 | 69 (3%) | 1.85 (1.38, 2.47) | 1.71 (1.26, 2.32) | 1.69 (1.23, 2.31) |
| Operative Diagnosis | p<0.001 | p<0.001 | p<0.001 | ||
| Osteoarthritis | 12,252 | 235 (2%) | 1.00 (REF) | 1.00 (REF) | 1.00 (REF) |
| Rheumatoid arthritis | 371 | 15 (4%) | 1.54 (1.06, 2.23) | 1.36 (0.78, 2.36) | 1.39 (0.95, 2.03) |
| Avascular necrosis | 1,021 | 32 (3%) | 2.55 (1.66, 3.91) | 2.58 (1.67, 3.98) | 2.51 (1.61, 3.89) |
| Other | 421 | 23 (5%) | 1.76 (1.05, 2.97) | 1.42 (0.98, 2.01) | 1.37 (0.78, 2.39) |
| Cemented | p=0.01 | p=0.002 | p=0.005 | ||
| Uncemented | 8,796 | 190 (2%) | 1.00 (REF) | 1.00 (REF) | 1.00 (REF) |
| Cemented | 5,269 | 115 (2%) | 0.75 (0.59, 0.94) | 0.68 (0.54, 0.87) | 0.69 (0.54, 0.89) |
| ASA class | p=0.006 | p=0.03 | p=0.04 | ||
| 1 | 653 | 8 (1%) | 1.00 (REF) | 1.00 (REF) | 1.00 (REF) |
| 2 | 7,947 | 161 (2%) | 1.73 (0.85, 3.51) | 1.84 (0.9, 3.76) | 1.90 (0.92, 3.90) |
| 3 | 5,254 | 131 (2%) | 2.44 (1.19, 4.98) | 2.45 (1.18, 5.1) | 2.48 (1.18, 5.21) |
| 4 | 151 | 3 (2%) | 2.69 (0.71, 10.15) | 2.68 (0.7, 10.28) | 2.54 (0.65, 9.87) |
| Prior Cardiac event | p=0.08 | p=0.37 | |||
| No | 11,879 | 255 (2%) | 1.00 (REF) | 1.00 (REF) | |
| Yes | 2,186 | 50 (2%) | 1.31 (0.97, 1.78) | 1.16 (0.84, 1.62) | |
| Prior Thromboembolic Event | p=0.20 | ||||
| No | 13,574 | 293 (2%) | 1.00 (REF) | ||
| Yes | 491 | 12 (2%) | 1.46 (0.82, 2.59) |
p-values indicate the overall significance of the variable, for example, in model 1, ASA class was significantly associated with risk of fractures with p value of 0.03.
Multivariable model 1 was a Cox regression model that considered all significant variables from the univariate analyses with p-value <0.05 and retained only those significantly associated in the multivariable model by using a backward selection process; the c-statistic of the multivariable model was 0.64
Multivariable model 2 was a Cox regression model obtained by retaining all variables from the univariate analyses with p-value <0.20 and forcing all of them in the model; the c-statistic of the multivariable model was 0.64
ASA, American Society of Anesthesiologists; CI, confidence interval, REF, reference category; p, p-value