Association of sleep-disordered breathing with diabetes-associated kidney disease

Figures & data

Figure 1. Study flow chart.

Table 1. Clinical characteristics of the 679 analysed subjects at the DIACORE baseline visit.

Characteristic	n = 679
Age [years]	65.6 ± 8.8
Sex (male), n (%)	468 (61.0)
BMI [kg/m^2]	31.2 ± 5.5
Waist-hip-ratio	0.96 ± 0.08
DM2 duration [years]	10.2 ± 8.0
HbA1c [%] (mmol/mol)	6.8 ± 1.1 (50.8 ± 11.5)
HOMA-Index*	6.2 ± 7.5
SBP [mmHg]	138 ± 18
DBP dia [mmHg]	75 ± 10
Smoking, n (%)	10.7
Coronary artery disease (%)	20.7
Stroke or TIA (%)	7.8
eGFR [ml/min/1.73 m^2]	78.1 ± 19.3
UACR [mg/g]	10.1 (4.7; 30.7)

Results are provided as mean ± standard deviation (UACR is provided as median and quartiles).

BMI: body-mass-index; DBP: diastolic blood pressure; DM2: diabetes mellitus type 2; HbA1c: hemoglobin A1c; HOMA: homeostasis model assessment (fasting, use of long acting insulin); SBP: systolic blood pressure; SDB: sleep-disordered breathing; TIA: transient ischemic attack.

* Number of patients =413.

Table 2. Clinical and SDB parameters of the 679 analysed subjects at the DIACORE baseline visit by SDB class.

	No or mild SDB(AHI <15)	Moderate SDB(AHI ≥15 to 29)	Severe SDB(AHI ≥30)	p ValueANOVA^a	Moderate vs. no or mild SDB^b	Severe vs.no or mild SDB^b
n (%)	451 (66%)	163 (24%)	65 (10%)
Age [years]	64 ± 9	66 ± 8	67 ± 7	.008	0.030	0.088
Sex (male), n (%)	243 (54%)	120 (74%)	49 (75%)	<.001	<0.001	0.002
BMI [kg/m^2]	30.4 ± 5.1	31.3 ± 5.2	34.0 ± 6.8	<.001	0.139	<0.001
Waist-Hip-Ratio	0.95 ± 0.08	0.98 ± 0.07	1.00 ± 0.08	<.001	<0.001	<0.001
DM2 duration [years]	10 ± 7	12 ± 10	9 ± 7	.001	0.001	1.000
HbA1c [%] (mmol/mol)	6.8 ± 1.1 (51 ± 7)	6.9 ± 1.2 (52 ± 8)	6.6 ± 0.7 (49 ± 5)	.111	0.867	0.340
HOMA-Index*	5.6 ± 5.9	6.7 ± 9.1	7.9 ± 10.9	.137	0.607	0.242
SBP [mmHg]	137 ± 18	140 ± 19	141 ± 17	.041	0.135	0.179
DBP [mmHg]	74 ± 10	75 ± 10	76 ± 10	.393	0.980	0.794
Smoking, n (%)	53 (12%)	12 (8%)	3 (5%)	.087	0.329	0.220
Coronary artery disease (%)	72 (16%)	38 (23%)	13 (20%)	.101	0.083	0.235
Stroke or TIA (%)	28 (6%)	17 (10%)	7 (11%)	.135	0.248	0.589
Medication
Oral antidiabetic drugs	350 (78%)	121 (74%)	48 (74%)	.601	1.000	1.000
Insulin	102 (23%)	61 (37%)	19 (29%)	.003	0.002	1.000
RAAS inhibitors	278 (62%)	116 (71%)	45 (69%)	.067	0.088	0.693
Diuretics	163 (36%)	73 (45%)	30 (46%)	.074	0.158	0.366
Antihypertensive agents	345 (77%)	144 (88%)	57 (88%)	.001	0.003	0.098
HMGCR inhibitors	201 (45%)	88 (54%)	28 (43%)	.098	0.117	1.000
Anti-platelets	182 (40%)	87 (53%)	28 (43%)	.016	0.012	1.000
Nocturnal respiration
ODI [/h]	6.9 ± 4.7	20.0 ± 5.0	40.7 ± 11.9	.001	<0.001	<0.001
SpO2 <90% [min]	72.9 ± 106.0	124.5 ± 122.2	119.8 ± 84.9	.001	<0.001	0.004
Mean SpO2 [%]	92.7 ± 1.9	91.5 ± 2.3	91.5 ± 2.0	.001	<0.001	<0.001
Min SpO2 [%]	79.9 ± 12.2	75.3 ± 16.0	73.7 ± 14.7	.001	0.001	0.002
Periodic breathing [%]	5.5 ± 22.9	17.2 ± 37.8	36.9 ± 48.6	.001	<0.001	<0.001
ESS	5.1 ± 3.3	5.4 ± 3.6	5.9 ± 3.1	.148	1.000	0.195

Results are provided as mean ± standard deviation.

AHI: apnea-hypopnea-index; BMI: body-mass-index; DBP: diastolic blood pressure; DM2: diabetes mellitus type 2; ESS: Epworth Sleepiness Scale; HbA1c: hemoglobin A1c; HOMA: homeostasis model assessment (fasting; use of long acting insulin); HMGCR: 3-hydroxy-3-methyl-glutaryl-(HMG) coenzyme A reductase; ODI: Oxygen-Desaturation-Index; RAAS: renin-angiotensin-aldosterone system; SBP: systolic blood pressure; SDB: sleep-disordered breathing; SpO₂: peripheral oxygen saturation; TIA: transient ischemic attack.

* Number of patients =413.

^a Difference between three groups.

^b Post-hoc analysis (Bonferroni).

Figure 2. Kidney function measured as eGFR (A and C) and albuminuria measured as UACR (B and D, analysed on ln-scale) across SDB categories in 679 subjects at the DIACORE baseline visit. Bars indicate mean values by group and error bars indicate 95% confidence intervals. ANOVA p value was used for overall association and Bonferroni post-hoc test for pairwise comparisons (A and B). A generalized linear model (Analysis of covariance, ANCOVA) was used to compute adjusted mean eGFR and UACR for the categories of SDB (C and D, adjusted for age, BMI, systolic blood pressure, duration of DM2, HbA1c and AHI). eGFR: estimated glomerular filtration rate, UACR: urinary albumin-creatinine-ratio, SDB: sleep-disordered breathing.

Table 3. Relationship between SDB and DKD in the total sample and by sex at the DIACORE baseline visit.

	No or mild SDB(AHI <15)	Moderate SDB(AHI ≥15 to 29)	Severe SDB(AHI ≥30)	Moderate vs. no or mild SDB		Severe vs. no or mild SDB
Total (n = 679)	n = 451	n = 163	n = 65	OR (95% CI)	p Value	OR (95% CI)	p Value
DKD, n (%)	140 (31%)	65 (40%)	37 (57%)	1.47 (1.02;2.14)	.041	2.94 (1.73;4.99)	<.001
eGFR <60 ml/min, n (%)	68 (15%)	32 (20%)	20 (31%)	1.38 (0.87;2.19)	.178	2.50 (1.39;4.50)	.002
UACR >30 mg/g	95 (21%)	43 (26%)	27 (42%)	1.34 (0.89;2.03)	.164	2.66 (1.55;4.58)	<.001
Microalbuminuria, n (%)	85 (19%)	31 (19%)	21 (32%)	1.01 (0.64;1.60)	.962	2.06 (1.16;3.64)	.013
Macroalbuminuria, n (%)	10 (2%)	12 (7%)	6 (9%)	3.50 (1.48;8.28)	.004	4.49(1.57;12.79)	.005
Men (n = 412)	n = 243	n = 120	n = 49
DKD, n (%)	81 (33%)	52 (43%)	30 (61%)	1.53 (0.98;2.40)	.064	3.12 (1.68;5.95)	<.001
eGFR <60 ml/min, n (%)	35 (14%)	24 (20%)	15 (31%)	1.49 (0.84;2.64)	.176	2.62 (1.30;5.31)	.007
UACR >30 mg/g	59 (24%)	36 (30%)	22 (45%)	1.34 (0.82;2.18)	.244	2.54 (1.35;4.79)	.004
Microalbuminuria, n (%)	53 (22%)	25 (21%)	17 (35%)	0.94 (0.55;1.61)	.831	1.90 (0.98;3.69)	.057
Macroalbuminuria, n (%)	6 (2%)	11 (9%)	5 (10%)	3.99 (1.44;11.06)	.008	4.49(1.31;15.35)	.017
Women (n = 267)	n = 208	n = 43	n = 16
DKD, n (%)	59 (28%)	13 (30%)	7 (44%)	1.09 (0.53;2.24)	.805	1.96 (0.70;5.52)	.200
eGFR <60 ml/min, n (%)	33 (16%)	8 (19%)	5 (31%)	1.21 (0.52;2.85)	.659	2.41 (0.79;7.39)	.124
UACR >30 mg/g	36 (17%)	7 (16%)	5 (31%)	0.93 (0.38;2.25)	.871	2.17 (0.71;6.63)	.173
Microalbuminuria, n (%)	32 (15%)	6 (14%)	4 (25%)	0.89 (0.35;2.29)	.812	1.83 (0.56;6.04)	.319
Macroalbuminuria, n (%)	4 (2%)	1 (2%)	1 (6%)	1.21 (0.13;11.14)	.864	3.40 (0.36;32.36)	.287

Shown are unadjusted odds ratios (OR) comparing moderate vs. no or mild SDB and severe vs. no or mild SDB by logistic regression, 95% confidence intervals (95% CI) and p values.

Microalbuminuria: 30 mg/g < UACR ≤300 mg/g. Macroalbuminuria: UACR >300 mg/g.

AHI: apnea-hypopnea-index; eGFR: estimated glomerular filtration rate; DKD: diabetic kidney disease (defined as eGFR <60 ml/min/1.73 m² or UACR >30 mg/g); OR: odds ratio; SDB: sleep-disordered breathing.

Table 4. Modulators for eGFR in total sample, in 412 men and in 267 women.

	Univariable analysis		Multivariable analysis
Variable	B (95% CI)	p Value	B (95% CI)	p Value
Total (n = 679)
Sex	0.0 (−2.8; 2.8)	.991	−0.6 (−3.2; 2.1)	.682
Age [years]	−1.2 (−1.3; −1.0)	<.001	−1.1 (−1.3; −1.0)	<.001
BMI [kg/m^2]	0.1 (−0.1; 0.3)	.426	−0.2 (−0.5; 0.0)	.094
Systolic BP [mmHg]	−0.1 (−0.2; 0.0)	.002	0.0 (−0.0; 0.1)	.252
Duration DM2 [years]	−0.6 (−0.7; −0.4)	<.001	−0.2 (−0.4; −0.1)	.009
HbA1c [%]	0.6 (−0.6; 1.8)	.347	0.8 (−0.4; 1.9)	.204
AHI [/h]	−0.2 (−0.4; −0.1)	<.001	−0.2 (−0.3; -0.1)	.004
Men (n = 412)
Age [years]	−1.2 (−1.4; −1.0)	<.001	−1.2 (−1.4; −0.9)	<.001
BMI [kg/m^2]	0.3 (−0.1; 0.6)	.115	−0.1 (−0.4; 0.3)	.718
Systolic BP [mmHg]	−0.1 (−0.2; −0.0)	.045	0.0 (−0.0; 0.1)	.272
Duration DM2 [years]	−0.4 (−0.6; −0.2)	<.001	−0.1 (−0.3; 0.1)	.172
HbA1c [%]	2.3 (0.8; 3.7)	.002	1.6 (0.3; 3.0)	.020
AHI [/h]	−0.2 (−0.4; −0.1)	.001	−0.1 (−0.3; −0.0)	.017
Women (n = 267)
Age [years]	−1.1 (−1.3; −0.9)	<.001	−1.0 (−1.3; −0.8)	<.001
BMI [kg/m^2]	−0.1 (−0.5; 0.3)	.661	−0.3 (−0.7; 0.1)	.114
Systolic BP [mmHg]	−0.2 (−0.3; −0.0)	.016	0.1 (−0.1; 0.2)	.426
Duration DM2 [years]	−0.8 (−1.1; −0.5)	<.001	−0.4 (−0.7; −0.1)	.023
HbA1c [%]	−2.6 (−4.8; −0.4)	.019	−0.9 (−3.1; 1.4)	.433
AHI [/h]	−0.4 (−0.06; −0.2)	.001	−0.2 (−0.4; 0.0)	.050

Shown are the unstandardized regression coefficient (B) by regression analysis, 95% confidence intervals (95% CI) and p values (ANOVA). The variables used in the regression model were age, BMI, systolic blood pressure, duration of DM2, HbA1c and AHI.

AHI: apnea-hypopnea-index; body-mass-index; BP: blood pressure; DM2: Diabetes mellitus type 2; HbA1c: hemoglobin A1c.

Table 5. Modulators for ln UACR in total sample, in 412 men and in 267 women.

	Univariable analysis		Multivariable analysis
Variable	B (95% CI)	p Value	B (95% CI)	p Value
Total (n = 679)
Sex	−0.33 (−0.55; -0.11)	.003	−0.05 (−0.28; 0.17)	.645
Age [years]	0.02 (0.01; 0.03)	<.001	0.01 (−0.01; 0.02)	.277
BMI [kg/m^2]	0.02 (0.00; 0.04)	.034	0.00 (−0.02; 0.02)	.948
Systolic BP [mmHg]	0.02 (0.02; 0.03)	<.001	0.02 (0.01; 0.02)	<.001
Duration DM2 [years]	0.04 (0.03; 0.06)	<.001	0.02 (0.01; 0.04)	.001
HbA1c [%]	0.19 (0.10; 0.29)	<.001	0.16 (0.06; 0.26)	.002
AHI [/h]	0.02 (0.01; 0.02)	<.001	0.01 (0.00; 0.02)	.009
Men (n = 412)
Age [years]	0.03 (0.01; 0.04)	.006	0.01 (−0.00; 0.03)	.164
BMI [kg/m^2]	0.05 (0.02; 0.07)	.002	0.02 (−0.02; 0.05)	.335
Systolic BP [mmHg]	0.03 (0.02; 0.03)	<.001	0.02 (0.01; 0.03)	<.001
Duration DM2 [years]	0.05 (0.03; 0.07)	<.001	0.03 (0.01; 0.05)	.001
HbA1c [%]	0.14 (0.01; 0.27)	.032	0.15 (0.02; 0.27)	.026
AHI [/h]	0.02 (0.00; 0.03)	.002	0.01 (0.00; 0.02)	.042
Women (n = 267)
Age [years]	0.02 (0.00; 0.03)	.016	0.00 (−0.02; 0.02)	.804
BMI [kg/m^2]	0.00 (−0.02; 0.02)	.960	−0.02 (−0.05; 0.01)	.172
Systolic BP [mmHg]	0.02 (0.01; 0.03)	<.001	0.01 (001, 0.02)	.003
Duration DM2 [years]	0.03 (0.00; 0.05)	.004	0.00 (−0.02; 0.03)	.712
HbA1c [%]	0.28 (0.15; 0.41)	<.001	0.26 (0.10; 0.42)	.001
AHI [/h]	0.01 (−0.00; 0.02)	.146	0.01 (−0.01; 0.02)	.243

Shown are the unstandardized regression coefficients (B) by regression analysis, 95% confidence intervals (95% CI) and p values (ANOVA). The variables used in the regression model were age, BMI, systolic blood pressure, duration of DM2, HbA1c and AHI.

AHI: apnea-hypopnea-index; BMI: body-mass-index; BP: blood pressure; DM2: Diabetes mellitus type 2; HbA1c: hemoglobin A1c.

Table 6. Cross-sectional studies looking for association between DKD and SDB.

Study, Year (Ref.)	Setting	Patients	Source of patients	SDB Diagnosis	Women, n (%)	Modulators for DKD
Tahrani et al., 2013 [14]	Multicenter	224	Diabetes clinic	Ambulatory SDB-monitoring device	94 (42%)	OSA, age, diabetesduration, HbA1c, triglycerides
Leong et al., 2014 [24]	Single center	90	Weight management service	Ambulatory SDB-monitoring device	51 (57%)	AHI, duration of nocturnal hypoxemia
Furukawa et al., 2013 [15]	Multicenter	513	Diabetes clinics	Pulse oximeter	221 (43%)	ODI, female gender
Buyukaydin et al., 2012 [39]	Single center	52	Diabetes clinic	Polysomnography	38 (73%)	No significant modulator
Schober et al., 2011 [26]	Multicenter	556	Primary care centers	Ambulatory SDB-monitoring device	264 (48%)	SDB
This study	Multicenter	721	Outpatient DM2	Ambulatory SDB-monitoring device	267 (39%)	AHI, age, diabetes duration, Hba1c, systolic blood pressure

Tahrani AA, Ali A, Raymond NT, et al. Obstructive sleep apnea and diabetic nephropathy: a cohort study. Diabetes Care. 2013;36:3718–3725.

 PubMed Web of Science ®Google Scholar

Leong WB, Nolen M, Thomas GN, et al. The impact of hypoxemia on nephropathy in extremely obese patients with type 2 diabetes mellitus. J Clin Sleep Med. 2014;10:773–778.

 PubMed Web of Science ®Google Scholar

Furukawa S, Saito I, Yamamoto S, et al. Nocturnal intermittent hypoxia as an associated risk factor for microalbuminuria in Japanese patients with type 2 diabetes mellitus. Eur J Endocrinol. 2013;169:239–246.

 PubMed Web of Science ®Google Scholar

Buyukaydin B, Akkoyunlu ME, Kazancioglu R, et al. The effect of sleep apnea syndrome on the development of diabetic nephropathy in patients with type 2 diabetes. Diabetes Res Clin Pract. 2012;98:140–143.

 PubMed Web of Science ®Google Scholar

Schober A-K, Neurath MF, Harsch IA. Prevalence of sleep apnoea in diabetic patients. Clin Respir J. 2011;5:165–172.

 PubMed Web of Science ®Google Scholar