Background
Methods
Study design
Data source
Study population
Variables
Hemoglobin
Definitions of outcomes
Covariates
Definition of diabetes, hypertension, cardiovascular disease, and etiology of kidney disease
Statistical analysis
HB | Q1 (< 10.4) | Q2(10.4–12.0) | Q3(12.0–13.6) | Q4(≥ 13.6) | P-value |
---|---|---|---|---|---|
Participants | 231 | 232 | 247 | 252 | |
Age(years) | 71.54 ± 12.13 | 69.02 ± 13.27 | 67.03 ± 13.00 | 62.30 ± 14.03 | < 0.001 |
SBP(mmHg) | 140.81 ± 21.86 | 140.14 ± 22.19 | 136.89 ± 20.74 | 139.83 ± 22.53 | 0.206 |
BMI(kg/m2) | 23.15 ± 4.46 | 23.78 ± 4.05 | 23.54 ± 3.91 | 24.50 ± 3.78 | 0.002 |
HB(g/dL) | 9.19 ± 0.94 | 11.14 ± 0.45 | 12.76 ± 0.46 | 14.84 ± 0.99 | < 0.001 |
ALB(g/dL) | 3.53 ± 0.61 | 3.77 ± 0.54 | 4.01 ± 0.52 | 4.15 ± 0.56 | < 0.001 |
Scr(mg/dL) | 2.65 (1.91–3.99) | 2.08 (1.50–2.92) | 1.43 (1.10–1.96) | 1.25 (1.06–1.55) | < 0.001 |
eGFR (ml/min per 1.73 m2) | 19.83 ± 12.94 | 25.93 ± 13.25 | 37.53 ± 16.00 | 47.28 ± 15.64 | < 0.001 |
UPCR (g/gCr) | 1.75 (0.53–4.50) | 1.22 (0.21–3.15) | 0.28 (0.07–1.57) | 0.27 (0.05–1.07) | < 0.001 |
Gender | < 0.001 | ||||
Male | 126 (54.55%) | 148 (63.79%) | 171 (69.23%) | 225 (89.29%) | |
Female | 105 (45.45%) | 84 (36.21%) | 76 (30.77%) | 27 (10.71%) | |
Etiology of CKD | < 0.001 | ||||
Diabetic nephropathy, n(%) | 93 (40.26%) | 74 (31.90%) | 45 (18.22%) | 32 (12.70%) | |
Nephrosclerosis, n (%) | 79 (34.20%) | 84 (36.21%) | 114 (46.15%) | 108 (42.86%) | |
Glomerulonephritis, n (%) | 30 (12.99%) | 40 (17.24%) | 43 (17.41%) | 65 (25.79%) | |
Other, n (%) | 29 (12.55%) | 34 (14.66%) | 45 (18.22%) | 47 (18.65%) | |
Urinary occult blood, n(%) | 71 (30.74%) | 101 (43.53%) | 70 (28.34%) | 68 (26.98%) | < 0.001 |
Hypertension, n (%) | 220 (95.24%) | 217 (93.53%) | 220 (89.07%) | 209 (82.94%) | < 0.001 |
History of CVD, n (%) | 83 (35.93%) | 70 (30.17%) | 56 (22.67%) | 49 (19.44%) | < 0.001 |
Diabetes, n (%) | 112 (48.48%) | 97 (41.81%) | 90 (36.44%) | 65 (25.79%) | < 0.001 |
Use of RAAS inhibitor, n(%) | 156 (67.53%) | 175 (75.43%) | 160 (64.78%) | 135 (53.57%) | < 0.001 |
Use of calcium channel blocker, n (%) | 127 (54.98%) | 121 (52.16%) | 117 (47.37%) | 94 (37.30%) | < 0.001 |
Use of diuretics, n (%) | 123 (53.25%) | 81 (34.91%) | 60 (24.29%) | 48 (19.05%) | < 0.001 |
Results
Baseline characteristics of participants
The incidence rate of the renal composite endpoint
HB | Participants(n) | Renal composite endpoint events(n) | Incidence rate(95% CI)(%) | Cumulative incidence( Per 100 person-year) |
---|---|---|---|---|
Total | 962 | 252 | 26.20(23.41–28.98) | 0.99 |
Q1 | 231 | 112 | 48.48(41.99–54.98) | 2.33 |
Q2 | 232 | 81 | 34.91(28.73–41.09) | 1.37 |
Q3 | 247 | 42 | 17.00(12.29–21.72) | 0.58 |
Q4 | 252 | 17 | 6.75(3.63–9.86) | 0.23 |
P for trend | < 0.0001 |
The results of univariate analyses using Cox proportional-hazards regression model
Variable | Statistics | Effect size HR(95%CI) | P value |
---|---|---|---|
Age(years) | 67.352 ± 13.558 | 0.992 (0.983, 1.001) | 0.07672 |
Gender | |||
Male | 670 (69.647%) | 1.0 | |
Female | 292 (30.353%) | 1.064 (0.816, 1.385) | 0.64813 |
Etiology of CKD | |||
Diabetic nephropathy, | 244 (25.364%) | 1.0 | |
Nephrosclerosis, | 385 (40.021%) | 0.185 (0.136, 0.254) | < 0.00001 |
Glomerulonephritis | 178 (18.503%) | 0.295 (0.208, 0.419) | < 0.00001 |
Other | 155 (16.112%) | 0.174 (0.110, 0.277) | < 0.00001 |
BMI(kg/m2) | 23.756 ± 4.072 | 1.027 (0.995, 1.059) | 0.09929 |
SBP(mmHg) | 139.388 ± 21.853 | 1.017 (1.012, 1.022) | < 0.00001 |
HB(g/dL) | 12.055 ± 2.213 | 0.691 (0.651, 0.733) | < 0.00001 |
Scr(mg/dL) | 2.144 ± 1.469 | 1.449 (1.396, 1.505) | < 0.00001 |
eGFR (ml/min per 1.73 m2) | 33.037 ± 18.007 | 0.917 (0.906, 0.929) | < 0.00001 |
ALB(g/dL) | 3.874 ± 0.605 | 0.349 (0.299, 0.407) | < 0.00001 |
Urinary occult blood | |||
No | 652 (67.775%) | 1.0 | |
Yes | 310 (32.225%) | 1.658 (1.290, 2.130) | 0.00008 |
UPCR(g/gCr) | 2.037 ± 3.189 | 1.208 (1.183, 1.234) | < 0.00001 |
Hypertension | |||
No | 96 (9.979%) | 1.0 | |
Yes | 866 (90.021%) | 4.070 (1.920, 8.627) | 0.00025 |
History of CVD | |||
No | 704 (73.181%) | 1.0 | |
Yes | 258 (26.819%) | 1.262 (0.962, 1.655) | 0.09277 |
Diabetes | |||
No | 598 (62.162%) | 1.0 | |
Yes | 364 (37.838%) | 2.699 (2.101, 3.466) | < 0.00001 |
Use of RAAS inhibitor | |||
No | 336 (34.927%) | 1.0 | |
Yes | 626 (65.073%) | 1.727 (1.293, 2.308) | 0.00022 |
Use of calcium channel blocker | |||
No | 503 (52.287%) | 1.0 | |
Yes | 459 (47.713%) | 1.724 (1.338, 2.221) | 0.00002 |
Use of diuretics | |||
No | 650 (67.568%) | 1.0 | |
Yes | 312 (32.432%) | 2.197 (1.715, 2.815) | < 0.00001 |
The results of multivariate analyses using Cox proportional-hazards regression model
Exposure | Crude model (HR,95%CI, P) | Model I(HR,95%CI, P) | Model II (HR,95%CI, P) | Model III (HR,95%CI, P) |
---|---|---|---|---|
HB | 0.691 (0.651, 0.733) < 0.00001 | 0.681 (0.640, 0.725) < 0.00001 | 0.836 (0.770, 0.907) 0.00002 | 0.863 (0.790, 0.944) 0.00123 |
HB Quartile | ||||
Q1 | Ref | Ref | Ref | Ref |
Q2 | 0.569 (0.427, 0.758) 0.00012 | 0.572 (0.426, 0.769) 0.00022 | 0.713 (0.518, 0.981) 0.03764 | 0.978 (0.697, 1.373) 0.89912 |
Q3 | 0.237 (0.166, 0.338) < 0.00001 | 0.241 (0.167, 0.349) < 0.00001 | 0.522 (0.348, 0.783) 0.00168 | 0.743 (0.483, 1.142) 0.17528 |
Q4 | 0.093 (0.056, 0.154) < 0.00001 | 0.082 (0.048, 0.139) < 0.00001 | 0.332 (0.181, 0.610) 0.00038 | 0.382 (0.203, 0.719) 0.00286 |
P for trend | < 0.00001 | < 0.00001 | 0.00005 | 0.00669 |
The results of competing risks multivariate Cox’s regression
Exposure | Crude model (SHR,95%CI, P) | Model I(SHR,95%CI, P) | Model II (SHR,95%CI, P) |
---|---|---|---|
HB | 0.69 (0.65, 0.73) < 0.0001 | 0.68 (0.64, 0.72) < 0.0001 | 0.84 (0.77, 0.91) < 0.0001 |
HB Quartile | |||
Q1 | Ref | Ref | Ref |
Q2 | 0.57(0.43, 0.76) 0.0001 | 0.57 (0.43, 0.77) 0.0002 | 0.72 (0.53, 0.99) 0.0462 |
Q3 | 0.24 (0.17, 0.34) < 0.0001 | 0.24 (0.17, 0.35) < 0.0001 | 0.53 (0.35, 0.79) 0.0019 |
Q4 | 0.09 (0.06, 0.15) < 0.0001 | 0.08 (0.05, 0.14) < 0.00001 | 0.34(0.18, 0.62) 0.0004 |
P for trend | < 0.0001 | < 0.0001 | < 0.0001 |
The results of multivariate analyses using linear regression model
Exposure | Crude model (β,95%CI, P) | Model I (β,95%CI, P) | Model II (β,95%CI, P) | Model III (β,95%CI, P) |
---|---|---|---|---|
Hb | -0.298 (-0.551, -0.045) 0.02118 | -0.340 (-0.620, -0.061) 0.01731 | -0.436 (-0.778, -0.093) 0.01288 | -0.428 (-0.777, -0.079) 0.01638 |
Hb Quartile | ||||
Q1 | Ref | Ref | Ref | Ref |
Q2 | 0.767 (-0.847, 2.381) 0.35164 | 0.789 (-0.802, 2.381) 0.33126 | 0.745 (-0.876, 2.365) 0.36799 | 0.976 (-0.631, 2.583) 0.23427 |
Q3 | -0.186 (-1.775, 1.404) 0.81898 | 0.205 (-1.414, 1.825) 0.80377 | -0.163 (-1.949, 1.623) 0.85844 | 0.059 (-1.719, 1.837) 0.94813 |
Q4 | -1.322 (-2.903, 0.260) 0.10184 | -1.509 (-3.250, 0.232) 0.08964 | -2.118 (-4.201, -0.034) 0.04661 | -1.795 (-3.868, 0.278) 0.08996 |
P for trend | 0.05020 | 0.07445 | 0.04345 | 0.06906 |
Sensitivity analysis
Exposure | Model I(HR,95%CI, P) | Model II (β,95%CI, P) |
---|---|---|
eGFR ≥ 15(ml/min per 1.73 m2) | ||
Hb | 0.837 (0.757, 0.926) 0.00053 | -0.408 (-0.789, -0.027) 0.03608 |
Hb Quartile | ||
Q1 | Ref | Ref |
Q2 | 0.928 (0.586, 1.470) 0.75033 | 0.680 (-1.351, 2.711) 0.51203 |
Q3 | 0.600 (0.352, 1.021) 0.05944 | -0.009 (-2.116, 2.097) 0.99313 |
Q4 | 0.380 (0.193, 0.748) 0.00508 | -1.789 (-4.141, 0.563) 0.13641 |
P for trend | 0.00195 | 0.07496 |
UPCR < 0.5 g/gCr | ||
Hb | 0.622 (0.474, 0.815) 0.00058 | -0.201 (-0.625, 0.223) 0.35305 |
Hb Quartile | ||
Q1 | Ref | Ref |
Q2 | 0.380 (0.140, 1.029) 0.05683 | 0.956 (-1.518, 3.429) 0.44930 |
Q3 | 0.172 (0.044, 0.678) 0.01188 | 0.014 (-2.370, 2.398) 0.99059 |
Q4 | 0.145 (0.022, 0.946) 0.04357 | -0.396 (-3.061, 2.269) 0.77110 0.49540 |
P for trend | 0.0066 |
The non-linear relationship
Renal prognosis (HR,95%CI, P) | |
Fitting model by two-piecewise Cox regression | |
Inflection point of Hb | 8.6 g/dL |
≤ 8.6 g/dL | 1.257 (0.841, 1.878) 0.2650 |
> 8.6 g/dL | 0.789 (0.715, 0.870) < 0.0001 |
P for log-likelihood ratio test | 0.026 |
Renal function decline(β,95%CI, P) | |
Fitting model by two-piecewise linear regression | |
Inflection point of Hb | 15.5 g/dL |
≤ 15.5 g/dL | -0.551 (-0.920, -0.182) 0.0035 |
> 15.5 g/dL | 1.527 (-0.830, 3.883) 0.2045 |
P for log-likelihood ratio test | 0.096 |
The results of subgroup analyses
Characteristic | participants | HR (95%CI) P for interacion | β(95%CI) P for interacion |
---|---|---|---|
Age(years) | 0.6636 | 0.5920 | |
< 60(years) | 218 | 0.859 (0.747, 0.988) | -0.308 (-0.881, 0.264) |
≥ 60(years) | 744 | 0.829 (0.751, 0.914) | -0.466 (-0.831, -0.102) |
Gender | 0.8543 | 0.6664 | |
Male | 670 | 0.844 (0.766, 0.930) | -0.480 (-0.868, -0.092) |
Female | 292 | 0.831 (0.722, 0.956) | -0.304 (-1.026, 0.417) |
Urinary occult blood | 0.1975 | 0.5015 | |
No | 652 | 0.874 (0.786, 0.972) | -0.380 (-0.760, -0.001) |
Yes | 310 | 0.788 (0.697, 0.892) | -0.566 (-1.080, -0.052) |
BMI(kg/m2) | 0.5280 | 0.1642 | |
< 23.5 | 492 | 0.825 (0.737, 0.924) | -0.715 (-1.168, -0.262) |
≥ 23.5 | 470 | 0.866 (0.775, 0.968) | -0.304 (-0.752, 0.144) |
SBP(mmHg) | 0.0276 | 0.3823 | |
< 140 | 514 | 0.741 (0.651, 0.843) | -0.317 (-0.740, 0.107) |
≥ 140 | 448 | 0.871 (0.791, 0.959) | -0.536 (-0.961, -0.111) |
UPCR (g/gCr) | 0.0145 | 0.2201 | |
< 0.5 | 458 | 0.683 (0.548, 0.851) | -0.217 (-0.742, 0.308) |
≥ 0.5 | 504 | 0.903 (0.832, 0.981) | -0.637 (-1.075, -0.199) |
Etiology of CKD | 0.3714 | 0.9391 | |
Diabetic nephropathy | 44 | 0.781 (0.692, 0.881) | -0.498 (-1.049, 0.053) |
Nephrosclerosis | 85 | 0.943 (0.786, 1.130) | -0.391 (-0.846, 0.064) |
Glomerulonephritis | 178 | 0.799 (0.605, 1.055) | -0.566 (-1.227, 0.096) |
Other | 155 | 0.868 (0.613, 1.230) | -0.338 (-0.990, 0.314) |
Use of RAAS inhibitor | 0.9565 | 0.7207 | |
No | 336 | 0.830 (0.712, 0.967) | -0.496 (-0.974, -0.018) |
Yes | 626 | 0.834 (0.761, 0.913) | -0.404 (-0.789, -0.018) |
Hypertension | 0.0173 | 0.6052 | |
No | 96 | 0.435 (0.257, 0.734) | -0.716 (-1.657, 0.225) |
Yes | 866 | 0.842 (0.775, 0.915) | -0.457 (-0.823, -0.091) |
History of CVD | 0.0978 | 0.1901 | |
No | 704 | 0.862 (0.787, 0.944) | -0.323 (-0.706, 0.059) |
Yes | 258 | 0.755 (0.652, 0.874) | -0.694 (-1.212, -0.175) |
Diabetes | 0.7648 | 0.1948 | |
No | 598 | 0.819 (0.728, 0.922) | -0.592 (-1.005, -0.179) |
Yes | 364 | 0.837 (0.759, 0.923) | -0.208 (-0.697, 0.280) |
Use of calcium channel blocker | 0.2353 | 0.7644 | |
No | 503 | 0.791 (0.698, 0.897) | -0.474 (-0.901, -0.047) |
Yes | 459 | 0.869 (0.783, 0.965) | -0.398 (-0.821, 0.024) |
Use of diuretics | 0.0271 | 0.3428 | |
No | 650 | 0.903 (0.811, 1.005) | -0.615 (-1.010, -0.220) |
Yes | 312 | 0.761 (0.676, 0.856) | -0.329 (-0.854, 0.195) |
ALB(g/dL) | 0.8595 | 0.1032 | |
< 4 | 465 | 0.837 (0.768, 0.911) | -0.716 (-1.129, -0.303) |
≥ 4 | 497 | 0.852 (0.711, 1.021) | -0.280 (-0.718, 0.159) |