Table 4.

ESRD risks by albuminuria and SBP reduction at month 6 from baselinea

VariablesNEvent (Rate)Adjusted HR (95% CI)P
Reduction in ALB
    <−30%38495 (85.9)3.50 (2.49 to 4.94)<0.000
    −30 to 0%23462 (96.5)2.11 (1.49 to 3.01)<0.000
    0 to 30%32688 (96.7)1.64 (1.20 to 2.24)0.002
    ≥30%48478 (53.8)1.00
Reduction in SBP
    <−15 mmHg24565 (92.1)0.97 (0.70 to 1.35)0.850
    −15 to 0 mmHg34577 (78.1)0.92 (0.68 to 1.26)0.623
    0 to 15 mmHg46387 (64.6)0.85 (0.63 to 1.15)0.304
    ≥15 mmHg37594 (88.0)1.00
Adjusted factors
    serum albumin (mg/dl)0.37 (0.27 to 0.51)<0.0001
    hemoglobin (mg/dl)0.89 (0.83 to 0.95)0.0004
    serum creatinine (mg/dl)3.93 (3.14 to 4.91)<0.0001
    log (urine alb/creat ratio) (mg/g)2.60 (2.15 to 3.14)<0.0001
    rALB*rSBP0.707
  • a Interaction between rALB and rSBP (rALB*rSBP) obtained using a multivariate Cox model in two steps: (1) Variable selection among covariates, including age, gender, race, BMI, smoking, previous antihypertensive therapies, duration of hypertension, cardiovascular disease history, serum creatinine, serum albumin, hemoglobin, total cholesterol, triglycerides, glucose, SBP, DBP, and albuminuria at baseline (backward selection) with indicators for interaction between reductions in SBP and albuminuria forced into the model; and (2) final analysis with significant baseline covariates (including baseline serum creatinine, serum albumin, hemoglobin, and albuminuria) and category factors for reductions in SBP and albuminuria (as shown in Table 4). P values for interaction between reductions in SBP and albuminuria were obtained using Wald χ2 statistic when interaction terms were added back into the model. Analysis adjusted for treatment group as strata.