Association between dietary selenium intake and mortality
Table 2 presented the mortality hazard ratios for patients with diabetic kidney disease (DKD) based on quartiles of selenium intake levels. In model 1, the unadjusted model, the multivariate-adjusted hazard ratios (95% confidence intervals) for mortality ranged from < 60.4, 87.6–121.0, 121.0−715.8, to > 715.8 mcg/day, with values of 1 (reference), 0.817 (0.669, 0.998), 0.741 (0.611, 0.900), and 0.610 (0.496, 0.749) (P trend < 0.00001). In model 2, after adjusted for age(≤ 40, 40–60, > 60years), sex(male or female), family income-to-povery ratio(≤ 1.0, 1.0–3.0, > 3.0), smoking states(never smoker, fomer smoker, current smoker), hemoglobin(continuous), albumin(continuous), uric acid(continuous), dietary intake of phosphorus (continuous), Dietary Inflammatory Index (continuous), Systemic Inflammatory Index (continuous), serum iron (continuous), stroke (yes or no), from the lowest to the highest groups, the multivariate adjusted HRs (95% CIs) were 1 (reference), 0.810(0.670,0.979), 0.786(0.628,0.985), 0.714(0.560,0.909) for mortality (P for trend = 0.0150). Model 3 incorporated adjustments for all variables, including HbA1c (≤ 7% or > 7%), hypertension (presence or absence), hypoglycemic treatment (presence or absence), and lipid-lowering therapy (presence or absence), in addition to the adjustments made in Model 2, the multivariate-adjusted hazard ratios (HRs) with 95% confidence intervals (CIs) for mortality ranged from 1 (reference) to 0.813 (0.669, 0.987), 0.789 (0.629, 0.989), and 0.712 (0.558, 0.908) across the lowest to highest intake groups, demonstrating a significant trend (P trend = 0.021). Furthermore, a smooth curve fitting graph illustrated a significant decrease in mortality risk with increasing selenium intake among patients with diabetic kidney disease (DKD) (overall P = 0.036; nonlinear P = 0.3944) (Fig. 2).
Table 2
All-cause mortality according to quartiles of dietary selenium intake among patients with DKD in NHANES 2001–2014
Exposure | Quartile 1 | Quartile 2 | Quartile 3 | Quartile 4 | P for trend |
---|
Range | < 60.4mcg/d | 87.6–121.0 mcg/d | 121.0-715.8mcg/d | > 715.8mcg/d | |
No. death/total | 298/548 | 284/543 | 255/546 | 226/546 | |
Model 1 | 1 (reference) | 0.817(0.669,0.998) | 0.741(0.611,0.900) | 0.610(0.496,0.749) | < 0.00001 |
Model 2 | 1 (reference) | 0.810(0.670,0.979) | 0.786(0.628,0.985) | 0.714(0.560,0.909) | 0.015 |
Model 3 | 1 (reference) | 0.813(0.669,0.987) | 0.789(0.629,0.989) | 0.712(0.558,0.908) | 0.021 |
Date are presented as HR(95% CI) unless indicated otherwise.
Model 1: unadjusted model.
Model 2: adjusted for age( < = 40, 40–60, > 60years), sex(male or female), family income-to-povery ratio( < = 1.0, 1.0–3.0, or > 3.0), smoking states(never smoker, fomer smoker, current smoker), hemoglobin (continuous), albumin (continuous), uric acid(continuous), dietary intake of phosphorus (continuous), Dietary Inflammatory Index (continuous), Systemic Inflammatory Index (continuous), Serum iron (continuous), stroke (yes or no).
Model 3: Model 2 + HbA1c ( < = 7%, > 7%), hypertension(yes or no), hypoglycemic treatment(yes or no), lipid-lowering therapy(yes or no).
Kaplan-Meier survival curves with selenium intake categories showed that patients with the highest level of selenium intake had the highest likelihood of surviving(Fig. 3 ).
A consistent association was observed between total selenium intake and mortality when the analyses were stratified by age (> 60 years), sex, non-Hispanic Black race, estimated glomerular filtration rate (eGFR > 60 ml/min/1.73 m²), and glycated hemoglobin (HbA1c > 7%) (Table 3). No significant interactions were detected after adjustments for multiple testing were applied.
Table 3
Stratified analyses of the association between dietary selenium intake and all-cause mortality among patients with DKD in NHANES 2001–2014
| Dietary selenium intake | |
---|
Characteristic | Quartile 1 | Quartile 2 | Quartile 3 | Quartile 4 | P trend | P interaction |
Age, years | | | | | | 0.250 |
<= 40 (n = 114 ) | ref | 0.349(0.113,1.078) | 0.455(0.148,1.399) | 0.459(0.126,1.666) | 0.609 | |
40–60 (n = 520 ) | ref | 0.529(0.338,0.826) | 0.668(0.393,1.133) | 0.404(0.192,0.853) | 0.059 | |
> 60 (n = 1549 ) | ref | 0.918(0.709,1.189) | 0.751(0.570,0.989) | 0.688(0.487,0.973) | 0.021 | |
Sex | | | | | | 0.532 |
Male (n = 1030 ) | ref | 0.901(0.700,1.160) | 0.710(0.503,1.004) | 0.591(0.393,0.889) | 0.009 | |
Female (n = 1153) | ref | 0.716(0.543,0.945) | 0.742(0.543,1.014) | 0.619(0.437,0.876) | 0.014 | |
Race | | | | | | 0.898 |
Non-Hispanic White (n = 940 ) | ref | 0.847(0.625,1.147) | 0.799(0.563,1.134) | 0.666(0.408,1.086) | 0.109 | |
Non-Hispanic Black (n = 550) | ref | 0.686(0.481,0.977) | 0.649(0.431,0.977) | 0.444(0.276,0.715) | 0.002 | |
Mexican American (n = 398) | ref | 0.929(0.622,1.387) | 0.682(0.346,1.344) | 0.737(0.389,1.396) | 0.269 | |
Other (n = 295 ) | ref | 0.769(0.390,1.514) | 0.758(0.313,1.833) | 0.905(0.377,2.174) | 0.733 | |
HbA1c,% | | | | | | 0.415 |
<= 7.0 (n = 1254 ) | ref | 0.797(0.611,1.040) | 0.829(0.645,1.066) | 0.588(0.418,0.827) | 0.004 | |
> 7.0 (n = 929 ) | ref | 0.790(0.553,1.129) | 0.640(0.430,0.952) | 0.635(0.403,1.000) | 0.037 | |
eGFR-EPI, ml/min·1.73m2 | | | | | | 0.302 |
<= 60 (n = 1095) | ref | 0.763(0.505,1.151) | 0.930(0.623,1.387) | 0.731(0.451,1.184) | 0.419 | |
> 60 (n = 1088 ) | ref | 0.926(0.756,1.135) | 0.661(0.503,0.869) | 0.629(0.443,0.893) | 0.001 | |
Data are presented as HR (95% CI). Adjusted for age( < = 40, 40–60, > 60years), sex(male or female), family income-to-povery ratio( < = 1.0, 1.0–3.0, or > 3.0),smoking states(never smoker, fomer smoker,current smoker), hemoglobin (continuous), albumin (continuous), uric acid(continuous), dietary intake of phosphorus (continuous), Dietary Inflammatory Index (continuous), Systemic Inflammatory Index (continuous), Serum iron (continuous), stroke (yes or no), HbA1c ( < = 7%, > 7%), hypertension(yes or no), hypoglycemic treatment(yes or no), lipid-lowering therapy(yes or no). The strata variable was not included in the model when stratifying by itself.