ARIC A1c

In Review

Bash LD, et al. "Poor Glycemic Control in Diabetes and the Risk of Incident Chronic Kidney Disease Even in the Absence of Albuminuria and Retinopathy". Archives of Internal Medicine. 2008. 168(22):2440-2447.
PubMed • articleid=773455 Full text • PDF

Clinical Question

In patients with diabetes mellitus (DM), does elevated glycated hemoglobin (HbA_1c) increase the risk of chronic kidney disease (CKD) in the absence of albuminuria and retinopathy?

Bottom Line

In patients with DM, elevated HbA_1c is associated with an increase in the risk of CKD even in the absence of albuminuria and retinopathy.

Major Points

Diabetes mellitus is a leading cause of end-stage renal disease (ESRD). Prior to the study, intensive glycemic control has been shown to decrease the progression of microvascular complications and microalbuminuria in patients with Type 1 DM. This study aimed to determine if elevated HbA_1c is associated with an increase in the risk of CKD in diabetic patients without albuminuria and retinopathy.

The study data was obtained from the Atherosclerosis Risk in Communities (ARIC) Study. From 1990-1992, all ARIC study participants who had DM (n=1871) were recruited into the study. The study did not distinguish between different types of diabetes. Participants were stratified according to HbA_1c into <6%, 6-7%, 7-8% and >8%. The primary outcome was incidence of CKD.

After follow-up for 6 years, as compared to HbA_1c <6%, HbA_1c of 6-7%, 7-8% and >8% were associated with a higher risk of CKD with adjusted relative hazard ratios (HR) of 1.37; 95% CI 0.97-1.91, HR 2.49; 95% CI 1.7-3.66, and HR 3.67; 95% CI 2.76-4.9, respectively. Increase in HbA_1c by 1% was associated with a 31% increase in the risk of developing CKD. An increase in the risk of CKD was observed even in patients without albuminuria and retinopathy.

Guidelines

ADA Standards of Medical Care in Diabetes (2016, adapted)

• Assess urinary albumin (eg, spot urinary albumin-to-creatinine ratio) and eGFR at least once a year in: (Level of evidence: B)
  • patients with type 1 DM ≥5 years
  • patients with type 2 DM
  • all patients with comorbid hypertension

Design

• Prospective observational cohort study
• N=1,871; stratified according to HbA_1c concentration:
  • <6% (n=770)
  • 6-7% (n=407)
  • 7-8% (n=193)
  • >9% (n=501)

• Setting: 4 communities in the US 4: Forsyth County, North Carolina; Jackson, Mississippi; suburban Minneapolis, Minnesota; and Washington County, Maryland
• Enrollment: 1990-1992
• Mean follow-up: 6 years
• Analysis:
• Primary outcome: CKD, defined as eGFR < 60 mL/min/1.73 m² at 1996-1998, or kidney disease noted during a hospitalization or death

Population

Inclusion Criteria

=patients were recruited from the ARIC Study

• DM, all types, defined as:
  • fasting glucose ≥126 mg/dL or a nonfasting glucose ≥200 mg/dL, or
  • self-reported physician diagnosis of DM, or
  • use of oral hypoglycemic medication or insulin

Exclusion Criteria

• eGFR<60 mL/min/1.73 m² prior to recruitment
• HbA_1c data not available at baseline or followup
• race other than Caucasian or African American

Baseline Characteristics

From the HbA_1c <6% group

• Demographics: Age 58±5.7 years, 52% male, 29.5% African American
• BMI: 30.2±5.5 kg/m²
• DM:
  • HbA_1c: 5.28±0.43%; fasting glucose: 134±24 mg/dL
  • On antidiabetes agent: 14.7%
• Serum creatinine: 0.89±0.18 mg/dL; eGFR 89±18 mL/min/1.73 m²
• BP: systolic 128±20 mm Hg, diastolic 74±11 mm Hg
• LDL-C: 134±38 mg/dL; HDL-C: 44±14 mg/dL; triglyceride: 164±110 mg/dL
• Other medical history: CHD 8.1%, active smoker 20.3%, hypertension 52.6%

There are significant differences in the characteristics between the 4 groups of patients. Higher HbA_1c was associated with female sex, African American race, higher eGFR, fasting glucose, prevalent CHD, hypertension, use of antihypertensive or antidiabetes agents, BMI, triglyceride level and lower HDL-C.

Interventions

• Patients were recruited from the ARIC study during 1990-1992 after measuring the HbA_1c.
• Serum creatinine was measured using the modified kinetic Jaffe method during 1990-1992 and during 1996-1998.
• Urinary albumin excretion was measured from a spot urine sample during 1996-1998.
• Retinography was conducted during 1993-1995

Outcomes

Comparisons are made between increasing HbA_1c categories against HbA_1c<6%

All hazard ratios adjusted for age, sex, race, study center, baseline eGFR, BMI, comorbid hypertension, antihypertensive treatment, prevalent CHD, smoking status, LDL-C, HDL-C and triglyceride concentration.

Primary outcomes

Incidence of CKD in all patients

6-7% vs. <6%:

14.15 vs. 9.87 incidence per 1000 person-years (HR 1.37; 95% CI 0.97-1.91; P_trend<0.001)

7-8% vs. <6%:

21.87 vs. 9.87 incidence per 1000 person-yrs (HR 2.49; 95% CI 1.7-3.66; P_trend<0.001)

>8% vs. <6%:

30.29 vs. 9.87 incidence per 1000 person-yrs (HR 3.67; 95% CI 2.76-4.9; P_trend<0.001)

A 1% increase in HbA_1c was associated with a 31% increase in the risk of developing CKD

Additional comparisons

Incidence of CKD in patients who did not develop albuminuria or retinopathy during the study

6-7% vs. <6%:

9.95 vs. 6.1 incidence per 1000 person-yrs (HR 1.46; 95% CI 0.80-2.65; P_trend=0.004)

7-8% vs. <6%:

8.17 vs. 6.1 incidence per 1000 person-yrs (HR 1.17; 95% CI 0.43-3.19; P_trend=0.004)

>8% vs. <6%:

13.42 vs. 6.1 incidence per 1000 person-yrs (HR 3.51; 95% CI 1.67-7.39; P_trend=0.004)

A 1% increase in HbA_1c was associated with higher risk of developing CKD (HR 1.27, 95% CI 1.08-1.51; P_trend=0.004)

Incidence of CKD in patients who developed albuminuria and retinopathy during the study

6-7% vs. <6%:

44.06 vs. 30.77 incidence per 1000 person-yrs (HR 0.74; 95% CI 0.16-3.39; P_trend=0.1)

7-8% vs. <6%:

47.41 vs. 30.77 incidence per 1000 person-yrs (HR 2.09; 95% CI 0.43-10.05; P_trend=0.1)

>8% vs. <6%:

59.16 vs. 30.77 incidence per 1000 person-yrs (HR 2.13; 95% CI 0.63-7.28; P_trend=0.1)

Incidence of CKD in patients who did not develop albuminuria during the study

6-7% vs. <6%:

10.47 vs. 7.55 incidence per 1000 person-yrs (HR 1.34; 95% CI 0.83-2.15; P_trend=0.002)

7-8% vs. <6%:

12.66 vs. 7.55 incidence per 1000 person-yrs (HR 1.75; 95% CI 0.94-3.27; P_trend=0.002)

>8% vs. <6%:

13.46 vs. 7.55 incidence per 1000 person-yrs (HR 2.21; 95% CI 1.32-3.7; P_trend=0.002)

Incidence of CKD in patients who developed albuminuria during the study

6-7% vs. <6%:

30.32 vs. 20.81 incidence per 1000 person-yrs (HR 0.74; 95% CI 0.16-3.39; P_trend<0.001)

7-8% vs. <6%:

40.49 vs. 20.81 incidence per 1000 person-yrs (HR 2.09; 95% CI 0.43-10.05; P_trend<0.001)

>8% vs. <6%:

48.73 vs. 20.81 incidence per 1000 person-yrs (HR 2.13; 95% CI 0.63-7.28; P_trend<0.001)

Incidence of CKD in patients who did not develop retinopathy during the study

6-7% vs. <6%:

11.08 vs. 8.55 incidence per 1000 person-yrs (HR 1.12; 95% CI 0.71-1.77; P_trend=0.001)

7-8% vs. <6%:

12.21 vs. 8.55 incidence per 1000 person-yrs (HR 1.49; 95% CI 0.75-2.96; P_trend=0.001)

>8% vs. <6%:

17.38 vs.8.55 incidence per 1000 person-yrs (HR 2.44; 95% CI 1.48-4.01; P_trend=0.001)

Incidence of CKD in patients who developed retinopathy during the study

6-7% vs. <6%:

17.67 vs. 15.11 incidence per 1000 person-yrs (HR 1.01; 95% CI 0.52-1.97; P_trend<0.001)

7-8% vs. <6%:

29.78 vs. 15.11 incidence per 1000 person-yrs (HR 2.23; 95% CI 1.18-4.2; P_trend<0.001)

>8% vs. <6%:

35.51 vs. 15.11 incidence per 1000 person-yrs (HR 2.73; 95% CI 1.65-4.5; P_trend<0.001)

Subgroup analysis

No significant interaction between HbA_1c and any of the demographic characteristics

Criticisms

• The type of DM was not defined as the ARIC study did not collect the relevant information.
• Kidney function was not directly measured.
• Twenty percent of participants defined as having DM according to study definition did not meet the American Diabetes Association criteria for DM used at the time of the study.
• Only 40% of the study sample were prescribed antidiabetes medications.
• HbA_1c and urinary albumin excretion was measured only once during the study, as is the same for retinography. The use of 1 single measurement may have caused some cases to be missed.

Funding

• National Institutes of Health