Table 2.

Independent clinical correlates of PTMIa

CharacteristicHR for PTMI (95% CI)P Value
Recipient characteristics
        age (yr)
            18–301.00 = reference
            31–441.97 (1.55–2.49)<0.0001
            45–603.20 (2.54–4.02)<0.0001
            60+3.96 (3.14–5.01)<0.0001
            female0.73 (0.66–0.82)<0.0001
                female × time (yr)b1.14 (1.05–1.24)0.002
            male1.00 = reference
            black0.79 (0.72–0.88)<0.0001
            other/unknown0.80 (0.67–0.96)0.02
            white1.00 = reference
            Hispanic0.71 (0.62–0.81)<0.0001
            non-Hispanic1.00 = reference
            working full or part time0.84 (0.76–0.92)0.0001
            unemployed1.00 = reference
        primary cause of ESRD
            diabetes1.38 (1.23–1.55)<0.0001
            hypertension1.06 (0.95–1.19)>0.2
            glomerulonephritis0.96 (0.85–1.09)>0.2
            other or unknown cause1.00 = reference
        diabetes1.13 (1.00–1.28)0.05
            diabetes × time (yr)b1.19 (1.09–1.29)0.0001
        pretransplant MI3.77 (3.34–4.25)<0.0001
            pretransplant MI × time (yr)b0.70 (0.62–0.78)<0.0001
        angina/coronary disease, without known MI1.66 (1.48–1.85)<0.0001
        peripheral vascular disease1.23 (1.13–1.34)<0.0001
        dyslipidemia1.22 (1.12–1.32)<0.0001
        cardiac arrhythmia1.41 (1.27–1.57)<0.0001
        cardiac arrhythmia × time (yr)b0.89 (0.82–0.97)0.007
Donor characteristics
        deceased1.17 (1.05–1.31)0.006
        living1.00 = reference
    donor age (yr)
        0–301.00 = reference
        31–441.10 (0.99–1.23)0.07
        45–591.16 (1.05–1.29)0.005
        60+1.28 (1.13–1.46)0.0002
        unknown1.11 (0.94–1.32)0.2
Transplantation complications
    delayed graft function1.15 (1.06–1.25)0.001
    graft failurec2.78 (2.41–3.19)<0.0001
    posttransplantation diabetesc1.60 (1.35–1.88)<0.0001
  • a HR, hazards ratio. Results of extended Cox hazards analysis, where HR >1.00 or <1.00 indicate an increased or reduced risk for PTMI, respectively. Each risk estimate was adjusted for all other characteristics in the model, including those shown along with the following characteristics with P ≥ 0.01 (recipient education, pretransplantation dialysis duration, obesity, tobacco use history, hypertensive heart disease history, and panel reactive antibody status; donor race and gender; transplant year, number of donor-recipient HLA mismatches, donor-recipient CMV seropairing, administration of induction therapy, and maintenance immunosuppression [cyclosporine microemulsions, cyclosporine suspensions, tacrolimus, azathioprine, mycophenolate mofetil, and rapamycin]; data not shown).

  • b Term indicating HR associated with the parameter estimate for the interaction of a given predictor with each 1-yr interval since transplantation. At a given time, the HR associated with a time-dependent predictor is calculated as e raised to the sum of [ln(HR at time zero) + ln(HR for interaction term) × (years from time zero)].

  • c Indicates a time-varying covariate, defined as a variable that may change in value during the observation interval on the basis of the diagnosis of a condition after transplantation.