Cohort Characteristics According to Quartiles of Serum Anion Gap and Total CO2 (n=1057).

Quartiles of Serum Anion Gap / p-value
Quartile 1
(n=242) / Quartile 2
(n=280) / Quartile 3
(n=242) / Quartile 4
(n=293)
Age, yr (mean, SD) / 54.3 (16.2) / 52.8 (15.7) / 55.1 (16.2) / 54.1 (16.1) / 0.41
Female (%) / 59 / 60 / 56 / 58 / 0.71
Race (%) / 0.10
White / 51 / 57 / 50 / 54
Black / 6 / 5 / 3 / 2
Hispanic / 1 / 3 / 0 / 2
Other / 2 / 2 / 5 / 4
Missing / 39 / 34 / 42 / 38
BMI (mean, SD) / 26.7 (5.3) / 27.5 (6.3) / 27.9 (5.8) / 27.0 (5.6) / 0.10
Creatinine, mg/dL (mean, SD) / 0.96 (0.20) / 0.94 (0.22) / 0.99 (0.35) / 1.02 (0.83) / 0.33
Impaired fasting glucose (%) / 6 / 5 / 5 / 7 / 0.54
Systolic BP, mmHg (mean, SD) / 125.9 (19.3) / 128.3 (19.3) / 130.1 (16.9) / 128.5 (19.0) / 0.11
Diastolic BP, mmHg (mean, SD) / 77.0 (12.1) / 79.0 (12.2) / 80.2 (11.0) / 78.7 (11.9) / 0.02
MAP, mmHg (mean, SD) / 93.3 (13.2) / 95.4 (13.3) / 96.8 (11.6) / 95.3 (12.8) / 0.02

P-values were generated using one-way ANOVA (continuous variables) or chi-square tests of trend (categorical variables)

Crude Associations between Cohort Characteristics and Blood Pressure.

Systolic BP / Diastolic BP
b (95% CI) / p-value / b (95% CI) / p-value
Age, yr (10 years) / 4.0 (3.4, 4.7) / <0.0001 / 0.001 (-0.5, 0.5) / 1.00
Female / -3.3 (-5.6, -1.0) / 0.004 / -3.3 (-4.7, -1.8) / <0.0001
Race
White / 0.0 (Ref) / 0.0 (Ref)
Black / 2.2 (-3.7, 8.1) / 0.46 / 3.5 (-0.2, 7.3) / 0.06
Hispanic / -6.3 (-15.5, 3.0) / 0.18 / 1.2 (-4.7, 7.1) / 0.69
Other / -8.6 (-15.0, -2.1) / 0.009 / -3.1 (-7.2, 1.0) / 0.14
Missing / 1.6 (-0.8, 3.9) / 0.20 / 1.3 (-0.2, 2.8) / 0.08
BMI (kg/m2) / 0.54 (0.35, 0.73) / <0.0001 / 0.52 (0.40, 0.64) / <0.0001
Creatinine (mg/dL) / -0.1 (-2.6, 2.4) / 0.95 / -0.71 (-2.3, 0.9) / 0.38
Estimated GFR
<30 / 4.9 (-4.2, 14.0) / 0.29 / -3.3 (-9.1, 2.5) / 0.26
30-60 / 5.9 (2.8, 9.0) / 0.0002 / -4.3 (-6.3, -2.3) / <0.0001
>60 / 1.0 (Ref) / 0.0 (Ref)
Missing / -6.3 (-9.8, -2.8) / 0.0004 / -3.2 (-5.4, -0.9) / 0.005
Impaired fasting glucose / 10.2 (5.4, 15.0) / <0.0001 / 1.7 (-1.4, 4.8) / 0.28

b, beta-coefficient from linear regression