Reduction in urinary albumin excretion with a moderate low-carbohydrate diet in patients with type 2 diabetes: a 12-month intervention

Figures & data

Table 1 Body mass index, glycemic control, serum lipid profiles, blood pressure, and drugs at baseline and 12 months in all patients (n = 124)

	Baseline	At 12 months	P *
Body mass index	24.3 ± 3.7	23.5 ± 3.7	<0.001
Fasting plasma glucose (mg/dL)	151 ± 48	130 ± 27	<0.001
Hemoglobin Alc (%)	7.9 ± 1.5	6.7 ± 0.6	<0.001
Fasting insulin (IU/mL)	7.6 ± 5.1	6.8 ± 4.9	0.003
HOMA-IR	2.9 ± 2.3	2.2 ± 1.8	<0.001
LDL-cholesterol (mg/dL)†	131 ± 31	125 ± 27	0.126
HDL-cholesterol (mg/dL)†	56 ± 15	61 ± 18	0.001
Triglycerides (mg/dL)†	125 ± 109	108 ± 72	0.323
Systolic blood pressure (mmHg)	138 ± 16	132 ± 14	<0.001
Diastolic blood pressure (mmHg)	82 ± 12	79 ± 9	0.002
eGFR (mL/min/1.73 m^2)	82 ± 11	81 ± 22	0.343
Lipid lowering drugs (n)	30	34
Antihypertensive drugs (n)	47	60
Antidiabetic drugs (n)	45	49
Glibenclamide	8% (3.1 mg)	2% (1.7 mg)
Gliclazide	2% (80 mg)	2% (25 mg)
Glimepiride	19% (2.6 mg)	18% (1.1 mg)
Nateglinide	2% (180 mg)	0%
Metformin	6% (568 mg)	20% (680 mg)
Pioglitazone	11% (25 mg)	3% (26 mg)
Voglibose	19% (0.8 mg)	6% (0.6 mg)
Sitagliptin	0%	1% (50 mg)
Insulin	l% (17 μ)	3% (15 μ)

Notes: Data is shown as mean ± SD.

* Wilcoxon test between baseline and 12 months;

† patients who received lipid-lowering drugs were excluded (n = 39). The percentages indicate the proportion of patients that were prescribed each drug. Values in parentheses are the mean daily doses per person for antidiabetic drugs.

Abbreviations: HOMA-IR, homeostasis model of assessment of insulin resistance; LDL, low-density lipoprotein; HDL, high-density lipoprotein; eGFR, estimated glomerular filtration rate; SD, standard deviation.

Table 2 Baseline characteristics of patients by level of UAE (n = 124)

	Normoalbuminuria (n = 68)	Microalbuminuria (n = 50)	P *	Macroalbuminuria (n = 6)
Age (years)	61 ± 9	63 ± 10	0.253	66 ± 9
Sex (male/female)	38/30	21/29		3/3
Duration of diabetes (months)	57 ± 83	79 ± 99	0.590	124 ± 49
Body mass index	24.3 ± 3.8	24.2 ± 4.0	0.957	23.6 ± 2.1
Fasting plasma glucose (mg/dL)	135 ± 36	168 ± 54	<0.001	182 ± 54
Hemoglobin Alc (%)	7.6 ± 1.4	8.3 ± 1.6	0.015	8.5 ± 2.3
Fasting serum insulin (IU/mL)	7.7 ± 4.9	7.8 ± 5.5	0.858	4.6 ± 1.9
HOMA-IR	2.6 ± 2.0	3.3 ± 2.7	0.252	2.0 ± 0.9
LDL-cholesterol (mg/dL)†	132 ± 29	128 ± 35	0.200	157 ± 57
HDL-cholesterol (mg/dL)†	55 ± 13	59 ± I8	0.300	54 ± 14
Triglycerides (mg/dL)†	113 ± 76	143 ± 146	0.504	200 ± 131
Systolic blood pressure (mmHg)	134 ± 15	141 ± 16	0.086	152 ± 21
Diastolic blood pressure (mmHg)	81 ± 12	81 ± 11	0.937	85 ± 3
UAE (mg/g Cr) (95% CI)‡	11 (3, 47)	67 (56, 80)	<0.001	945 (234, 3802)
eGFR(mL/min/1.73 m^2)	84 ± 20	81 ± 21	0.601	71 ± 37
Antihypertensive drugs (%)	47%	46%		83%
Olmesartan	43%	46%		83%
Indapamide	43%	46%		66%
Calcium channel blockers	15%	14%		33%
Guanabenz	7%	2%		17%

Notes: Data is shown as mean ± SD.

* P between normoalbuminuric and microalbuminuric patients;

† patients who received anti-lipid drugs at baseline were excluded (n = 30);

‡ geometric means. The percentages for antihypertensive drugs indicate the proportion of patients who were prescribed each drug.

Abbreviations: Cr, creatinine; UAE, urinary albumin excretion; HOMA-IR, homeostasis model of assessment of insulin resistance; LDL, low-density lipoprotein; HDL, high-density lipoprotein; CI, confidence interval; eGFR, estimated glomerular filtration rate; SD, standard deviation.

Table 3 Differences of dietary assessment, BMI, HbA_1c, and HOMA-IR in less strict and strict CARD patients

	Patients on less strict CARD*	Patients on strict CARD*	P †
N	101	23
Male/female	50/51	12/11
Normoalbuminuria/microalbuminuria (n)	61/35	7/15
Age (years)	62 ± 9	59 ± 9	0.084
Total energy intake (kcal/day)	1716 ± 406	1770 ± 445	0.60
Carbohydrate intake/day (g)	170 ± 51	143 ± 43	0.012
% carbohydrate	40.2 ± 10.0	32.8 ± 9.0	0.010
Fat intake/day (g)	70 ± 29	84 ± 32	0.056
% fat	36.2 ± 9.7	42.3 ± 8.8	0.006
Protein intake/day (g)	80 ± 22	90 ± 27	0.010
% protein	18.8 ± 3.9	20.3 ± 2.9	0.046
Baseline body mass index	24.2 ± 3.9	24.4 ± 3.4	0.69
Baseline fasting plasma glucose (mg/dL)	138 ± 35	208 ± 53	<0.001
Baseline hemoglobin Alc (%)	7.3 ± 0.7	10.6 ± 1.2	<0.001
Baseline HOMA-IR	2.5 ± 1.8	4.4 ± 3.5	0.011
Δ body mass index	−0.8 ± 1.3	−0.6 ± 1.2	0.54
Δ fasting plasma glucose (mg/dL)	−10 ± 35	−66 ± 50	<0.001
Δ hemoglobin Alc (%)	−0.7 ± 0.7	−3.6 ± 1.4	<0.001
Δ HOMA-IR	−0.4 ± 1.3	−1.6 ± 2.7	0.001

Notes: Data is shown as mean ± SD.

* Patients with HbA_1c levels < 9.0% were instructed to follow a less strict carbohydrate restriction, while those with HbA_1c levels ≥ 9.0% were instructed to follow a strict carbohydrate restriction;

† P between patients on less strict CARD and patients on strict CARD. The parameter change (Δ) for each biomarker was defined as the level after 12 months minus the level at baseline.

Abbreviations: BMI, body mass index; HOMA-IR, homeostasis model of assessment of insulin resistance; CARD, carbohydrate-reduced diet; SD, standard deviation.

Figure 1 Individual changes in UAE (mg/g Cr) in (A) normoalbuminuric patients (n = 68), (B) microalbuminuric patients (n = 50), and (C) macroalbuminuric patients (n = 6), over 12 months. Dot lines show borderlines between normoalbuminuria and microalbuminuria (30 mg/g Cr) in (A and B), and between microalbuminuria and macroalbuminuria (300 mg/g Cr) in (B and C).

Abbreviations: Cr, creatinine; UAE, urinary albumin secretion.

Table 4 Changes in UAE, BMI, glycemic control, fasting serum insulin, serum lipid profiles, blood pressure, and eGFR over 12 months

	Normoalbuminuria (n = 68)	P *	Microalbuminuria (n = 50)	P *	Macroalbuminuria (n = 6)	P *
Δ% UAE (%) (95% CI)†	−20 (−33, −4)	0.021‡	−53 (−62, −43)	<0.001‡	−41 (−57, −18)	0.025‡
Δ body mass index	−0.9 ± 1.3	<0.001	−0.6 ± 1.3	0.003	−0.6 ± 0.4	0.028
Δ fasting plasma glucose (mg/dL)	−13 ± 37	0.003	−29 ± 48	<0.001	−47 ± 63	0.116
Δ hemoglobin Alc (%)	−1.2 ± 1.2	<0.001	−1.4 ± 1.5	<0.001	−1.8 ± 2.5	0.074
Δ fasting serum insulin (IU/mL)	−1.0 ± 4.4	0.008	−0.9 ± 3.4	0.049	1.5 ± 2.3	0.115
Δ HOMA−IR	−0.52 ± 1.9	0.001	−0.9 ± 1.4	<0.001	−0.03 ± 0.7	0.753
Δ LDL-cholesterol (mg/dL)§	−6 ± 26	0.089	−3 ± 31	0.695	−11, −7
Δ HDL-cholesterol (mg/dL)§	5 ± 13	0.016	6 ± 11	0.008	0, −7
Δ triglycerides (mg/dL)§	−11 ± 58	0.297	−30 ± 136	0.432	35, 89
Δ systolic blood pressure (mmHg)	−4 ± 12	0.016	−7 ± 16	0.008	−11 ± 22	0.293
Δ diastolic blood pressure (mmHg)	−3 ± 9	0.016	−3 ± 10	0.084	−5 ± 14	0.916
Δ eGFR (mL/min/1.73 m^2)	−0.0 ± 10.7	0.656	−2.9 ± 11.9	0.218	−12 ± 16	0.075

Notes: Data is shown as mean ± SD.

* P Wilcoxon test between baseline and 12 months except for UAE;

† geometric means;

‡ paired t-test;

§ patients who received anti-lipid drugs were excluded (18 for normoalbuminuria, 17 for microalbuminuria, and four for macroalbuminuria). The parameter change (Δ) was defined as the level after 12 months minus the level at baseline.

Abbreviations: UAE, urinary albumin excretion; BMI, body mass index; eGFR, estimated glomerular filtration rate; Δ% UAE, percentage change in urinary albumin excretion; CI, confidence interval; HOMA-IR, homeostasis model of assessment of insulin resistance; LDL, low-density lipoprotein; HDL, high-density lipoprotein; SD, standard deviation.

Figure 2 Correlation of Δ% UAE with Δ HOMA-IR in the 50 microalbuminuric patients.

Notes: The figure shows the regression line and the 95% confidence interval, superimposed on the graph. The Δ% UAE was significantly and positively correlated with Δ HOMA-IR levels (r_s = 0.308; P = 0.031).
Abbreviations: Δ% UAE, percentage change in urinary albumin secretion; HOMA-IR, homeostasis model of assessment of insulin resistance.