Abstract
Aims/Introduction
We aimed to investigate whether there are differences in the risk factors or markers for the progression of diabetic retinopathy (DR) and diabetic nephropathy (DN) in type 2 diabetes mellitus.
Materials and Methods
We carried out a 3‐year retrospective cohort study of 604 patients with type 2 diabetes mellitus. The outcomes were the progression of DR (worsening of the DR stage) and DN (an estimated glomerular filtration rate decline >12%) at the 3‐year follow up. The mean hemoglobin A1c (HbA1c) level and HbA1c variability (HbA1c‐VAR) were calculated.
Results
The mean HbA1c and HbA1c‐VAR levels were higher in the DR progressors (n = 67) than in the DR non‐progressors (n = 537). The mean HbA1c was a significant predictor for DR progression independent of the duration of diabetes and HbA1c‐VAR levels. The urine albumin‐to‐creatinine ratio at baseline and HbA1c‐VAR levels were higher in the DN progressors (n = 34) than in the DN non‐progressors (n = 570). The triglyceride to high‐density lipoprotein cholesterol ratio at baseline tended to be higher in the DN progressors than in the DN non‐progressors. HbA1c‐VAR levels and the triglyceride‐to‐high‐density lipoprotein cholesterol ratio were significant predictors for DN progression independent of estimated glomerular filtration rate and the urine albumin‐to‐creatinine ratio.
Conclusions
Average glycemia was significantly associated with progression of DR, whereas glycemic variability and dyslipidemia were significantly associated with progression of DN in type 2 diabetes mellitus.
Keywords: Diabetic complication, Diabetic nephropathy, Diabetic retinopathy
Introduction
Diabetic retinopathy (DR) is a microvascular complication of diabetes and is the most frequent cause of blindness. Diabetic nephropathy (DN), or chronic kidney disease associated with diabetes, is characterized by albuminuria and progressive loss of renal function1. It is well established that long‐term exposure to hyperglycemia is the major risk factor for both DR and DN2, 3. Each complication has a strong impact on the initiation or progression of the other4.
However, the progression of DR and DN can be discordant in diabetes patients. During the Diabetes Control and Complications Trial in type 1 diabetes mellitus patients, 12.9% of patients did not have DR progression, but had DN development; 10.7% had DR progression, but not DN development, and 7.3% had both DR progression and DN development4. In the Renal Insufficiency and Cardiovascular Events (RIACE) study, 41.4% of patients with type 2 diabetes mellitus and advanced DR showed no evidence of DN5. That cross‐sectional study also showed that different risk factors or markers were associated with DN or DR. Subsequently, the RIACE study showed that glycemic variability over a long‐term period could predict the presence of DN, but not of DR6. Recently, we reported that DN was present in ~60% of patients with type 2 diabetes mellitus who had advanced DR, and that glycemic variability and dyslipidemia were associated with the initiation and progression of DN in these patients7. Another study reported that systolic blood pressure (BP) variability predicted the initiation and progression of DN, but not DR, in patients with type 2 diabetes mellitus8.
All of these studies suggest that different risk factors might be involved in the pathogenesis of DR and DN. However, few longitudinal studies have addressed this issue. Therefore, we carried out the present study to investigate whether there are differences in the risk factors or markers for the progression of DR and DN in patients with type 2 diabetes mellitus.
Methods
Study Design
Patients with type 2 diabetes mellitus who visited Yeouido St. Mary's Hospital, Seoul, Korea, from July 2013 to December 2013 were enrolled in the present retrospective, observational cohort study. Patients who had been diagnosed with type 2 diabetes mellitus for at least 1 year and were being followed up regularly at both the Department of Internal Medicine and Department of Ophthalmology were included. Patients with primary renal disease, advanced liver disease, cancer not in remission, secondary diabetes, an estimated glomerular filtration rate (eGFR; mL/min/1.73 m2) <30 or proliferative DR were excluded. The outcome was the progression of DR or DN after 3 years. This study was approved by the institutional review board of Yeouido St. Mary's Hospital.
Data Collection
The clinical and laboratory data of the study participants were collected from electronic medical records. Hypertension was defined as a systolic BP ≥140 mmHg or diastolic BP ≥90 mmHg or any use of BP‐lowering medications. Cardiovascular disease included coronary artery disease or cerebrovascular disease. Dyslipidemia was defined as the use of a statin or fibrate. During follow up, glycated hemoglobin (HbA1c) and serum creatinine levels were measured every 3–6 months in each individual.
After an overnight fast, blood samples were obtained for analysis of serum concentrations of creatinine and lipid profiles. The HbA1c level was determined by high‐performance liquid chromatography. Albuminuria was quantified by calculating the urine albumin‐to‐creatinine ratio (ACR; mg/g) in the urine, and eGFR was calculated using the Modification of Diet in Renal Disease study equation9.
The presence of DR was examined by two DR specialists at the Department of Ophthalmology. The stage of DR was graded as reported10: absent DR; mild, moderate or severe non‐proliferative DR; and proliferative DR. The progression of DR was defined as worsening of the stage of DR.
The progression of DN was defined as an eGFR decline >12% at the 3‐year follow up (4% per year)11, 12, 13. The eGFR decline was calculated as follows: eGFR decline (%/year) = (eGFR at 3 year – eGFR at baseline) / 3 × 100 (%). An eGFR slope per year was also calculated after creating a linear regression model for time versus eGFR using the least‐squares method.
Glycemic variability during follow up was measured. HbA1c variability (HbA1c‐VAR) was calculated as the standard deviation (SD) of multiple HbA1c levels during follow up. To correct for differences in the number of HbA1c measurements, an “adjusted HbA1c‐VAR” was calculated as HbA1c‐VAR divided by square root of [n/(n − 1)], where n was the number of HbA1c measurements14. To correct for large SDs due to high levels of mean HbA1c, the coefficient of variation of HbA1c (HbA1c‐CV) was also calculated as the HbA1c‐VAR divided by mean HbA1c level15.
Statistical Analysis
All statistical analyses were carried out with the use of SAS software (SAS Institute Inc., Cary, NC, USA). Data are expressed as mean ± SD or medians (interquartile range [IQR]) for continuous variables, and as numbers (percentage) for categorical variables. Because triglyceride and urine ACR values are not normally distributed, these were analyzed after logarithmic transformation. The t‐test was used to compare continuous variables. The numbers of categorical variables were compared with the use of the χ2‐test or Fisher's exact test. Multivariate logistic regression analysis was carried out to assess the independent predictive effects of the variables on the risk for progression of DR or DN. P‐values of <0.05 were considered to show statistical significance.
Results
A total of 895 patients met the inclusion criteria. From these patients, we selected and analyzed the data for 604 patients who were followed up for 3 years. There was no difference in baseline clinical characteristics between 604 patients with follow up and 895 patients enrolled initially (Table S1).
A total of 68 patients (11%) showed progression of DR, including proliferative DR in six patients. Progressors of DR had a longer duration of diabetes, higher body mass index, more frequent history of hypertension and more frequent use of insulin compared with DR non‐progressors (Table 1). The HbA1c level at baseline and the mean HbA1c level were higher in the DR progressors than in the DR non‐progressors (7.70 ± 1.10 vs 7.27 ± 1.03%, P = 0.001, and 7.72 ± 1.05 vs 7.27 ± 0.96%, P = 0.001, respectively). Among three indices of HbA1c variability, HbA1c‐VAR and adjusted HbA1c‐VAR were higher in the progressors than in the non‐progressors (0.63 ± 0.41 vs 0.52 ± 0.33%, P = 0.036 for HbA1c‐VAR; 0.59 ± 0.39 vs 0.49 ± 0.31%, P = 0.033 for adjusted HbA1c‐VAR). However, another index of HbA1c variability, HbA1c‐CV, was not different between the progressors and non‐progressors. Multiple logistic regression analysis that included the mean HbA1c level, HbA1c‐VAR and duration of diabetes as independent variables showed that mean HbA1c level was a significant and independent predictor of the progression of DR (Table 2). This result was similar after adjusting for eGFR, triglyceride‐to‐high‐density lipoprotein (HDL) cholesterol ratio, the presence of DR, hypertension and use of an angiotensin‐converting enzyme inhibitor or angiotensin‐converting enzyme receptor blocker (Table S2).
Table 1.
Clinical characteristics of study participants according to the progression of diabetic retinopathy
| Non‐progressors (n = 536) | Progressors (n = 68) | P‐value | |
|---|---|---|---|
| Baseline | |||
| Age (years) | 60.7 ± 10.7 | 60.8 ± 11.4 | 0.967 |
| Sex (female) | 249 (46.5) | 26 (38.2) | 0.2 |
| Duration of diabetes (years) | 13.7 ± 8.4 | 16.3 ± 7.7 | 0.019 |
| BMI (kg/m2) | 25.0 ± 3.5 | 26.1 ± 3.3 | 0.017 |
| Smoking | 83 (17.4) | 17 (27.4) | 0.055 |
| Hypertension | 333 (62.5) | 52 (76.5) | 0.024 |
| ACEi or ARB use | 281 (85.9) | 46 (88.5) | 0.623 |
| Dyslipidemia | 421 (78.5) | 53 (77.9) | 0.909 |
| Insulin use | 120 (22.4) | 24 (35.3) | 0.019 |
| CVD | 154 (28.7) | 18 (26.5) | 0.697 |
| Hemoglobin (g/dL) | 13.7 ± 1.6 | 13.8 ± 1.6 | 0.592 |
| Systolic BP (mmHg) | 126.4 ± 11.7 | 128.2 ± 10.0 | 0.229 |
| Diastolic BP (mmHg) | 73.8 ± 8.1 | 75.2 ± 7.5 | 0.189 |
| HbA1c (%) | 7.27 ± 1.03 | 7.70 ± 1.10 | 0.001 |
| Triglyceride (mmol/L) | 1.51 (1.11, 2.08) | 1.31 (0.90, 1.91) | 0.02 |
| HDL cholesterol (mmol/L) | 1.19 ± 0.27 | 1.23 ± 0.33 | 0.427 |
| LDL cholesterol (mmol/L) | 2.17 ± 0.57 | 2.04 ± 0.49 | 0.077 |
| Triglyceride‐to‐HDL cholesterol ratio | 3.5 ± 2.3 | 3.1 ± 1.9 | 0.153 |
| eGFR (mL/min/1.73 m2) | 77.4 ± 16.0 | 78.8 ± 19.1 | 0.555 |
| Urine ACR (mg/g) | 12.9 (7.1, 32.4) | 15.8 (8.1, 38.9) | 0.639 |
| Follow up | |||
| Mean HbA1c (%) | 7.27 ± 0.96 | 7.72 ± 1.05 | 0.001 |
| HbA1c‐VAR (%) | 0.52 ± 0.33 | 0.63 ± 0.41 | 0.036 |
| Adjusted HbA1c‐VAR (%) | 0.49 ± 0.31 | 0.59 ± 0.39 | 0.033 |
| HbA1c‐CV (%) | 0.07 ± 0.04 | 0.08 ± 0.05 | 0.104 |
| eGFR slope (mL/min/1.73 m2 per year) | 5.17 ± 6.50 | 5.12 ± 4.04 | 0.923 |
| eGFR decline (% per year) | 5.17 ± 6.50 | 5.32 ± 5.84 | 0.853 |
Data are mean ± standard deviation, n (%) or median (interquartile range). ACEi, angiotensin‐converting enzyme inhibitor; ACR, albumin‐to‐creatinine ratio; ARB, angiotensin‐converting enzyme receptor blocker; BMI, body mass index; BP, blood pressure; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; HbA1c, glycated hemoglobin; HbA1c‐CV, coefficient of variation of glycated hemoglobin; HbA1c‐VAR, glycated hemoglobin variability; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein.
Table 2.
Multiple logistic regression analysis of risk factors influencing progression of diabetic retinopathy
| Predictors | Odds ratio | 95% CI | P‐value |
|---|---|---|---|
| Mean HbA1c | 1.35 | 1.02–1.78 | 0.033 |
| HbA1c‐VAR | 1.57 | 0.71–3.45 | 0.264 |
| Duration of diabetes | 1.03 | 1.00–1.06 | 0.08 |
CI, confidence interval; HbA1c‐VAR, glycated hemoglobin variability.
A total of 34 patients (6%) showed the progression of DN, five of whom received hemodialysis. Progressors of DN were older, used insulin more frequently and smoked more frequently than DN non‐progressors (Table 3). The eGFR at the baseline was lower, and urine ACR at the baseline and HbA1c‐VAR level were higher in the DN progressors than in the DN non‐progressors (66.8 ± 22.7 vs 78.1 ± 15.7 mL/min/1.73 m2, P = 0.007; 50.1 [IQR 8.9–676.1] vs 12.9 [IQR 7.1–30.9] mg/g, P < 0.001, and 0.73 ± 0.42 vs 0.52 ± 0.33%, P = 0.006, respectively). The other indices of HbA1c variability, adjusted HbA1c‐VAR and HbA1c‐CV, were also higher in the DN non‐progressors. The triglyceride‐to‐HDL cholesterol ratio tended to be higher in the DN progressors than in the DN non‐progressors (4.9 ± 4.2 vs 3.4 ± 2.1, P = 0.051). Multiple logistic regression analysis including HbA1c‐VAR, triglyceride‐to‐HDL cholesterol ratio and eGFR or urine ACR at baseline as independent variables showed that HbA1c‐VAR and the triglyceride‐to‐HDL cholesterol ratio were significant and independent predictors of the progression of DN (Table 4). This result was similar after adjusting for the duration of diabetes, mean HbA1c, presence of DR, hypertension and use of an angiotensin‐converting enzyme inhibitor or angiotensin‐converting enzyme receptor blocker (Table S3).
Table 3.
Clinical characteristics of study participants according to the progression of diabetic nephropathy
| Non‐progressors (n = 570) | Progressors (n = 34) | P‐value | |
|---|---|---|---|
| Baseline | |||
| Age (years) | 60.5 ± 10.9 | 64.7 ± 8.1 | 0.006 |
| Sex (female) | 263 (46.1) | 12 (35.3) | 0.217 |
| Duration of diabetes (years) | 14.0 ± 8.3 | 14.6 ± 8.9 | 0.665 |
| BMI (kg/m2) | 25.1 ± 3.5 | 25.2 ± 3.9 | 0.831 |
| Smoking | 88 (17.3) | 12 (37.5) | 0.004 |
| Hypertension | 360 (63.5) | 25 (73.5) | 0.236 |
| ACEi or ARB use | 305 (86.2) | 22 (88.0) | 0.796 |
| Dyslipidemia | 445 (78.1) | 29 (85.3) | 0.319 |
| Insulin use | 125 (21.9) | 19 (55.9) | <0.0001 |
| CVD | 158 (27.7) | 14 (41.2) | 0.091 |
| Hemoglobin (g/dL) | 13.7 ± 1.6 | 13.2 ± 1.6 | 0.095 |
| Systolic BP (mmHg) | 126.5 ± 11.5 | 128.2 ± 11.8 | 0.403 |
| Diastolic BP (mmHg) | 74.0 ± 8.0 | 73.5 ± 7.4 | 0.731 |
| HbA1c (%) | 7.30 ± 1.05 | 7.52 ± 1.03 | 0.24 |
| Triglyceride (mmol/L) | 1.49 (1.08, 2.04) | 1.61 (1.11, 3.09) | 0.071 |
| HDL cholesterol (mmol/L) | 1.20 ± 0.23 | 1.10 ± 0.24 | 0.045 |
| LDL cholesterol (mmol/L) | 2.16 ± 0.54 | 2.24 ± 0.85 | 0.564 |
| Triglyceride to HDL cholesterol ratio | 3.4 ± 2.1 | 4.9 ± 4.2 | 0.051 |
| eGFR (mL/min/1.73 m2) | 78.1 ± 15.7 | 66.8 ± 22.7 | 0.007 |
| Urine ACR (mg/g) | 12.9 (7.1, 30.9) | 50.1 (8.9, 676.1) | 0.0003 |
| Follow up | |||
| Mean HbA1c (%) | 7.33 ± 0.98 | 7.25 ± 0.92 | 0.671 |
| HbA1c‐VAR (%) | 0.52 ± 0.33 | 0.73 ± 0.42 | 0.006 |
| Adjusted HbA1c‐VAR (%) | 0.49 ± 0.31 | 0.69 ± 0.39 | 0.006 |
| HbA1c‐CV (%) | 0.07 ± 0.04 | 0.10 ± 0.06 | 0.004 |
| eGFR slope (mL/min/1.73 m2 per year) | 5.71 ± 4.21 | −5.64 ± 4.50 | <0.0001 |
| eGFR decline (% per year) | 6.08 ± 5.16 | −9.88 ± 6.85 | <0.0001 |
Data are mean ± standard deviation, n (%) or median (interquartile range). BMI, body mass index; ACEi, angiotensin‐converting enzyme inhibitor; ACR, albumin‐to‐creatinine ratio; ARB, angiotensin‐converting enzyme receptor blocker; BP, blood pressure; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; HbA1c, glycated hemoglobin; HbA1c‐CV, coefficient of variation of glycated hemoglobin; HbA1c‐VAR, glycated hemoglobin variability; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein.
Table 4.
Multiple logistic regression analysis of risk factors influencing the progression of diabetic nephropathy
| Predictors | Odds ratio | 95% CI | P‐value |
|---|---|---|---|
| Model 1 | |||
| HbA1c‐VAR | 3.02 | 1.28–7.10 | 0.012 |
| Triglyceride‐to‐HDL cholesterol ratio | 1.15 | 1.02–1.29 | 0.026 |
| eGFR | 0.96 | 0.94–0.98 | 0.0003 |
| Model 2 | |||
| HbA1c‐VAR | 2.59 | 1.01–6.64 | 0.048 |
| Triglyceride‐to‐HDL cholesterol ratio | 1.12 | 0.97–1.28 | 0.112 |
| Urine ACR | 3.82 | 2.30–6.34 | <0.0001 |
ACR, albumin‐to‐creatinine ratio; CI, confidence interval; eGFR, estimated glomerular filtration rate; HbA1c‐VAR, glycated hemoglobin variability; HDL, high‐density lipoprotein.
There were 31 progressors of DN alone, 65 progressors of DR alone, and three progressors of both DN and DR. DN‐alone progressors were older and used insulin more frequently than DR‐alone progressors (Table 5). The urine ACR at baseline was higher, and eGFR at baseline was lower in the DN‐alone progressors than in the DR‐alone progressors: 53.7 (IQR 16.6–776.2) versus 15.8 (IQR 8.3–37.1) mg/g, P = 0.001, for urine ACR level, and 67.6 ± 21.9 versus 79.7 ± 18.1 mL/min/1.73 m2, P = 0.005, for eGFR. The triglyceride‐to‐HDL cholesterol ratio at baseline was higher in the DN‐alone progressors than in the DR‐alone progressors: 5.1 ± 4.3 versus 3.1 ± 1.9, P = 0.023. The mean HbA1c level during the follow‐up period was lower in the DN‐alone progressors than in the DR‐alone progressors: 7.23 ± 0.85 versus 7.73 ± 1.03%, P = 0.021. However, indices of HbA1c variability during the follow‐up period did not differ significantly between the two groups. Multiple logistic regression analysis that included the duration of diabetes, presence of DR, hypertension and use of an angiotensin‐converting enzyme inhibitor or angiotensin‐converting enzyme receptor blocker as independent variables showed that the mean HbA1c, triglyceride‐to‐HDL cholesterol ratio and urine ACR or eGFR at baseline were significant and independent predictors of the progression of DN alone (Table S4).
Table 5.
Clinical characteristics of study participants according to the progression of diabetic nephropathy alone and diabetic retinopathy alone
| DN alone (n = 31) | DR alone (n = 65) | P‐value | |
|---|---|---|---|
| Baseline | |||
| Age (years) | 64.7 ± 7.9 | 60.6 ± 11.4 | 0.043 |
| Sex (female) | 11 (35.5) | 25 (38.4) | 0.778 |
| Duration of diabetes (years) | 14.9 ± 9.3 | 16.5 ± 7.8 | 0.399 |
| BMI (kg/m2) | 24.8 ± 3.9 | 25.9 ± 3.3 | 0.164 |
| Smoking | 11 (37.9) | 16 (27.1) | 0.301 |
| Hypertension | 22 (71.0) | 49 (75.4) | 0.645 |
| ACEi or ARB use | 20 (90.9) | 44 (89.8) | 1 |
| Dyslipidemia | 26 (83.9) | 50 (76.9) | 0.433 |
| Insulin use | 18 (58.1) | 22 (35.4) | 0.036 |
| CVD | 12 (38.7) | 15 (24.6) | 0.155 |
| Hemoglobin (g/dL) | 13.2 ± 1.6 | 13.8 ± 1.5 | 0.097 |
| Systolic BP (mmHg) | 128.4 ± 11.3 | 128.3 ± 9.6 | 0.948 |
| Diastolic BP (mmHg) | 73.4 ± 7.5 | 75.2 ± 7.5 | 0.275 |
| HbA1c (%) | 7.5 ± 1.0 | 7.7 ± 1.1 | 0.39 |
| Triglyceride (mmol/L) | 1.61 (1.20, 3.19) | 1.32 (0.95, 1.91) | 0.016 |
| HDL cholesterol (mmol/L) | 1.10 ± 0.22 | 1.23 ± 0.33 | 0.024 |
| LDL cholesterol (mmol/L) | 2.24 ± 0.88 | 2.03 ± 0.49 | 0.232 |
| Triglyceride‐to‐HDL cholesterol ratio | 5.1 ± 4.3 | 3.1 ± 1.9 | 0.023 |
| eGFR (mL/min/1.73 m2) | 67.6 ± 21.9 | 79.7 ± 18.1 | 0.005 |
| Urine ACR (mg/g) | 53.7 (16.6, 776.2) | 15.8 (8.3, 37.1) | 0.001 |
| Follow up | |||
| Mean HbA1c (%) | 7.23 ± 0.85 | 7.73 ± 1.03 | 0.021 |
| HbA1c‐VAR (%) | 0.71 ± 0.41 | 0.61 ± 0.40 | 0.271 |
| Adjusted HbA1c‐VAR (%) | 0.67 ± 0.39 | 0.58 ± 0.38 | 0.269 |
| HbA1c‐CV (%) | 0.10 ± 0.06 | 0.08 ± 0.05 | 0.092 |
| eGFR slope (mL/min/1.73 m2 per year) | −5.70 ± 4.7 | 5.60 ± 3.47 | <0.0001 |
| eGFR decline (% per year) | −9.87 ± 7.03 | 6.02 ± 4.83 | <0.0001 |
Data are mean ± standard deviation, n (%) or median (interquartile range). ACEi, angiotensin‐converting enzyme inhibitor; ACR, albumin‐to‐creatinine ratio; ARB, angiotensin‐converting enzyme receptor blocker; BMI, body mass index; BP, blood pressure; CVD, cardiovascular disease; DN, diabetic nephropathy; DR, diabetic retinopathy; eGFR, estimated glomerular filtration rate; HbA1c, glycated hemoglobin; HbA1c‐CV, coefficient of variation of glycated hemoglobin; HbA1c‐VAR, glycated hemoglobin variability; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein.
Discussion
The present 3‐year retrospective, observational cohort study provides evidence that different factors were associated with the progression of DR and DN in patients with type 2 diabetes mellitus. The mean HbA1c level was a risk factor for the progression of DR independent of the duration of diabetes and HbA1c variability, whereas HbA1c variability and the triglyceride‐to‐HDL cholesterol ratio were risk factors for the progression of DN independent of eGFR and urine ACR.
Long‐term glycemic control, expressed as the HbA1c level, is crucial for preventing the initiation and progression of both DR and DN1. HbA1c variability is another risk factor that is related to chronic hyperglycemia, and is expressed as the SD of serially measured HbA1c levels14, 16. In patients with type 1 diabetes mellitus, HbA1c variability is an independent risk factor for DR and DN14, 15, 17. In patients with type 2 diabetes mellitus, HbA1c variability is associated with the initiation of DN, such as microalbuminuria and decreased GFR5, 18, 19, 20, 21. Although the reason for the harmful effects of HbA1c variability on the development of DR or DN are not clear, one possible mechanism involves “metabolic memory” from repeated exposure to glycemic instability22, which can lead to increased oxidative stress23.
Factors that increase the risk of or are associated with DR are the duration of diabetes, level of glycemic control, DN, hypertension and dyslipidemia24. In the present study, DR progressors had a higher HbA1c level at baseline, and a higher mean HbA1c level and longer duration of diabetes than DR non‐progressors. Two of the three indices of HbA1c variability were higher in the DR progressors than in the DR non‐progressors. However, the mean HbA1c level alone was a significant predictor of DR progression after adjusting for HbA1c‐VAR and duration of diabetes. Most studies have reported that HbA1c variability is an independent risk factor for DR in type 1 diabetes mellitus, but not type 2 diabetes mellitus14, 15, 17, 25.
DN is the leading cause of end‐stage renal disease. It is often accompanied by DR, as the pathogenesis of both complications is related to chronic hyperglycemia. However, some patients do not show any phenotype of DN, even in the presence of proliferative DR. Phenotypes of DN are renal dysfunction (eGFR <60 mL/min/1.73 m2), albuminuria (urine ACR >30 mg/g creatinine) or both26. In the present study, we defined the progression of DN as an eGFR decline >4% per year, in accordance with previous studies11, 12, 13, because there is no definite criterion for the progression of DN. At baseline, as expected, eGFR was lower and urine ACR was higher in the DN progressors than in the DN non‐progressors. All three indices of HbA1c variability were also higher in the DN progressors. HbA1c‐VAR was a significant predictor of DN progression, even after adjusting for eGFR, urine ACR and the triglyceride‐to‐HDL cholesterol ratio. These findings are consistent with those of previous studies5, 18, 19, 20, 21. Recently, we also reported that HbA1c variability was significantly associated with urine ACR and eGFR, and was an independent predictor of the presence of DN (urine ACR >30 mg/g and eGFR <60 mL/min/1.73 m2) in patients with advanced DR7. These results are consistent with those of the RIACE study, showing that HbA1c variability affected albuminuric DN more than average HbA1c level did6. Of note, HbA1c variability was not an independent predictor of DR in the RIACE study, as in the present study.
Dyslipidemia exists frequently in patients with type 2 diabetes mellitus, and plays a critical role in the pathogenesis of atherosclerotic cardiovascular disease. High triglyceride and low HDL cholesterol levels are hallmarks of diabetic dyslipidemia27. Studies have reported that a lower HDL cholesterol level is associated with DN in type 2 diabetes patients28, 29, 30. In particular, the Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation study31 showed that a lower HDL cholesterol level was a risk factor predicting the initiation and progression of DN, whereas it was not a risk factor for DR. The deleterious effect of low HDL cholesterol level on the progression of DN could be explained by the fact that HDL plays a protective role in renal damage by reducing oxidative stress and inflammation32, 33. High triglyceride level might influence the progression of DR and DN. The Fenofibrate Intervention and Event Lowering In Diabetes study34 and Action to Control Cardiovascular Risk in Diabetes trial35 showed that a fenofibrate‐induced decrease in triglyceride level has favorable effects on the progression of DR. In addition, the Fenofibrate Intervention and Event Lowering In Diabetes study and the Diabetes Atherosclerosis Intervention Study showed that fenofibrate reduces albuminuria and slows the decline in eGFR36, 37. Therefore, it is not unpredictable that low HDL cholesterol and high triglyceride levels are independent risk factors for the development of albuminuria and DN in patients with type 2 diabetes mellitus29, 38. Consistent with the preceding studies, in the present study, the triglyceride‐to‐HDL cholesterol ratio tended to be higher in the DN progressors than in the DN non‐progressors, and was a significant predictor of DN progression even after adjusting for eGFR and HbA1c‐VAR.
Finally, comparison of clinical characteristics between DN‐alone progressors and DR‐alone progressors confirmed that mean HbA1c level was more strongly associated with DR progression, whereas the triglyceride‐to‐HDL cholesterol ratio and urine ACR or eGFR levels were more strongly associated with DN progression.
The current study had a few limitations. First, it had a retrospective design and no causality could be ascertained. Second, outcomes were infrequent, as the follow‐up duration was relatively short and duration of diabetes was heterogeneous. Third, DN progression was defined as the eGFR decline only. The initiation and progression of albuminuria were not included in the definition of DN progression, because urine ACR was not measured in every patient during the follow‐up period.
In conclusion, the present study showed that average HbA1c level was a risk factor for the progression of DR independent of the duration of diabetes and HbA1c variability, whereas HbA1c variability and dyslipidemia were risk factors for the progression of DN independent of eGFR and urine ACR. However, long‐term prospective studies are required to confirm the discordance in risk factors for the progression of DR and DN in patients with type 2 diabetes mellitus in the future.
Disclosure
The authors declare no conflict of interest.
Supporting information
Table S1 ¦ Baseline clinical characteristics of study participants.
Table S2 ¦ Multiple logistic regression analysis of risk factors influencing the progression of diabetic retinopathy.
Table S3 ¦ Multiple logistic regression analysis of risk factors influencing the progression of diabetic nephropathy.
Table S4 ¦ Multiple logistic regression analysis of risk factors influencing the progression of diabetic nephropathy alone.
Acknowledgments
Statistical consultation was supported by the Department of Biostatistics of the Catholic Research Coordinating Center.
J Diabetes Investig 2019; 10: 745–752
References
- 1. American Diabetes Association . 10. Microvascular complications and foot care: standards of medical care in diabetes‐2018. Diabetes Care 2018; 41: S105–S118. [DOI] [PubMed] [Google Scholar]
- 2. Fong DS, Aiello L, Gardner TW, et al Retinopathy in diabetes. Diabetes Care 2004; 27(Suppl 1): S84–S87. [DOI] [PubMed] [Google Scholar]
- 3. Molitch ME, DeFronzo RA, Franz MJ, et al Diabetic nephropathy. Diabetes Care 2003; 26(Suppl 1): S94–S98. [DOI] [PubMed] [Google Scholar]
- 4. Kramer CK, Retnakaran R. Concordance of retinopathy and nephropathy over time in Type 1 diabetes: an analysis of data from the Diabetes Control and Complications Trial. Diabet Med 2013; 30: 1333–1341. [DOI] [PubMed] [Google Scholar]
- 5. Penno G, Solini A, Zoppini G, et al Rate and determinants of association between advanced retinopathy and chronic kidney disease in patients with T2DM: the Renal Insufficiency And Cardiovascular Events (RIACE) Italian multicenter study. Diabetes Care 2012; 35: 2317–2323. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Penno G, Solini A, Bonora E, et al HbA1c variability as an independent correlate of nephropathy, but not retinopathy, in patients with T2DM: the Renal Insufficiency and Cardiovascular Events (RIACE) Italian multicenter study. Diabetes Care 2013; 36: 2301–2310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Yun KJ, Kim HJ, Kim MK, et al Risk factors for the development and progression of diabetic kidney disease in patients with type 2 diabetes mellitus and advanced diabetic retinopathy. Diabetes Metab J 2016; 40: 473–481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Takao T, Matsuyama Y, Yanagisawa H, et al Visit‐to‐visit variability in systolic blood pressure predicts development and progression of diabetic nephropathy, but not retinopathy, in patients with T2DM. J Diabetes Complications 2014; 28: 185–190. [DOI] [PubMed] [Google Scholar]
- 9. Levey AS, Coresh J, Greene T, et al Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med 2006; 145: 247–254. [DOI] [PubMed] [Google Scholar]
- 10. Wilkinson CP, Ferris FL III, Klein RE, et al Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales. Ophthalmology 2003; 110: 1677–1682. [DOI] [PubMed] [Google Scholar]
- 11. Yokoyama H, Kanno S, Takahashi S, et al Determinants of decline in glomerular filtration rate in nonproteinuric subjects with or without diabetes and hypertension. Clin J Am Soc Nephrol 2009; 4: 1432–1440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Zoppini G, Targher G, Chonchol M, et al Predictors of estimated GFR decline in patients with T2DM and preserved kidney function. Clin J Am Soc Nephrol 2012; 7: 401–408. [DOI] [PubMed] [Google Scholar]
- 13. Goderis G, Van Pottelbergh G, Truyers C, et al Long‐term evolution of renal function in patients with T2DM mellitus: a registry‐based retrospective cohort study. BMJ Open 2013; 3: e004029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kilpatrick ES, Rigby AS, Atkin SL. A1C variability and the risk of microvascular complications in T1DM: data from the Diabetes Control and Complications Trial. Diabetes Care 2008; 31: 2198–2202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Waden J, Forsblom C, Thorn LM, et al A1C variability predicts incident cardiovascular events, microalbuminuria, and overt diabetic nephropathy in patients with T1DM. Diabetes 2009; 58: 2649–2655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Gorst C, Kwok CS, Aslam S, et al Long‐term glycemic variability and risk of adverse outcomes: a systematic review and meta‐analysis. Diabetes Care 2015; 38: 2354–2369. [DOI] [PubMed] [Google Scholar]
- 17. Marcovecchio ML, Dalton RN, Chiarelli F, et al A1C variability as an independent risk factor for microalbuminuria in young people with T1DM. Diabetes Care 2011; 34: 1011–1013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Hsu CC, Chang HY, Huang MC, et al HbA1c variability is associated with microalbuminuria development in T2DM: a 7‐year prospective cohort study. Diabetologia 2012; 55: 3163–3172. [DOI] [PubMed] [Google Scholar]
- 19. Sugawara A, Kawai K, Motohashi S, et al HbA(1c) variability and the development of microalbuminuria in T2DM: Tsukuba Kawai Diabetes Registry 2. Diabetologia 2012; 55: 2128–2131. [DOI] [PubMed] [Google Scholar]
- 20. Luk AO, Ma RC, Lau ES, et al Risk association of HbA1c variability with chronic kidney disease and cardiovascular disease in T2DM: prospective analysis of the Hong Kong Diabetes Registry. Diabetes Metab Res Rev 2013; 29: 384–390. [DOI] [PubMed] [Google Scholar]
- 21. Rodriguez‐Segade S, Rodriguez J, Garcia Lopez JM, et al Intrapersonal HbA(1c) variability and the risk of progression of nephropathy in patients with Type 2 diabetes. Diabet Med 2012; 29: 1562–1566. [DOI] [PubMed] [Google Scholar]
- 22. Ihnat MA, Thorpe JE, Ceriello A. Hypothesis: the ‘metabolic memory’, the new challenge of diabetes. Diabet Med 2007; 24: 582–586. [DOI] [PubMed] [Google Scholar]
- 23. Monnier L, Mas E, Ginet C, et al Activation of oxidative stress by acute glucose fluctuations compared with sustained chronic hyperglycemia in patients with T2DM. JAMA 2006; 295: 1681–1687. [DOI] [PubMed] [Google Scholar]
- 24. Solomon SD, Chew E, Duh EJ, et al Diabetic retinopathy: a position statement by the American diabetes association. Diabetes Care 2017; 40: 412–418. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Foo V, Quah J, Cheung G, et al HbA1c, systolic blood pressure variability and diabetic retinopathy in Asian type 2 diabetics. J Diabetes 2017; 9: 200–207. [DOI] [PubMed] [Google Scholar]
- 26. Tuttle KR, Bakris GL, Bilous RW, et al Diabetic kidney disease: a report from an ADA Consensus Conference. Diabetes Care 2014; 37: 2864–2883. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Beckman JA, Creager MA, Libby P. Diabetes and atherosclerosis: epidemiology, pathophysiology, and management. JAMA 2002; 287: 2570–2581. [DOI] [PubMed] [Google Scholar]
- 28. Zoppini G, Targher G, Chonchol M, et al Higher HDL cholesterol levels are associated with a lower incidence of chronic kidney disease in patients with T2DM. Nutr Metab Cardiovasc Dis 2009; 19: 580–586. [DOI] [PubMed] [Google Scholar]
- 29. Lee IT, Wang CY, Huang CN, et al High triglyceride‐to‐HDL cholesterol ratio associated with albuminuria in type 2 diabetic subjects. J Diabetes Complications 2013; 27: 243–247. [DOI] [PubMed] [Google Scholar]
- 30. Sacks FM, Hermans MP, Fioretto P, et al Association between plasma triglycerides and high‐density lipoprotein cholesterol and microvascular kidney disease and retinopathy in T2DM mellitus: a global case‐control study in 13 countries. Circulation 2014; 129: 999–1008. [DOI] [PubMed] [Google Scholar]
- 31. Morton J, Zoungas S, Li Q, et al Low HDL cholesterol and the risk of diabetic nephropathy and retinopathy: results of the ADVANCE study. Diabetes Care 2012; 35: 2201–2206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Tabet F, Rye KA. High‐density lipoproteins, inflammation and oxidative stress. Clin Sci (Lond) 2009; 116: 87–98. [DOI] [PubMed] [Google Scholar]
- 33. Vaziri ND. Lipotoxicity and impaired high density lipoprotein‐mediated reverse cholesterol transport in chronic kidney disease. J Ren Nutr 2010; 20: S35–S43. [DOI] [PubMed] [Google Scholar]
- 34. Keech AC, Mitchell P, Summanen PA, et al Effect of fenofibrate on the need for laser treatment for diabetic retinopathy (FIELD study): a randomised controlled trial. Lancet 2007; 370: 1687–1697. [DOI] [PubMed] [Google Scholar]
- 35. Group AS , Group AES , Chew EY, et al Effects of medical therapies on retinopathy progression in T2DM. N Engl J Med 2010; 363: 233–244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Davis TM, Ting R, Best JD, et al Effects of fenofibrate on renal function in patients with T2DM mellitus: the Fenofibrate Intervention and Event Lowering in Diabetes (FIELD) Study. Diabetologia 2011; 54: 280–290. [DOI] [PubMed] [Google Scholar]
- 37. Ansquer JC, Foucher C, Rattier S, et al Fenofibrate reduces progression to microalbuminuria over 3 years in a placebo‐controlled study in T2DM: results from the Diabetes Atherosclerosis Intervention Study (DAIS). Am J Kidney Dis 2005; 45: 485–493. [DOI] [PubMed] [Google Scholar]
- 38. Russo GT, De Cosmo S, Viazzi F, et al Plasma triglycerides and HDL‐C levels predict the development of diabetic kidney disease in subjects with type 2 diabetes: the AMD annals initiative. Diabetes Care 2016; 39: 2278–2287. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1 ¦ Baseline clinical characteristics of study participants.
Table S2 ¦ Multiple logistic regression analysis of risk factors influencing the progression of diabetic retinopathy.
Table S3 ¦ Multiple logistic regression analysis of risk factors influencing the progression of diabetic nephropathy.
Table S4 ¦ Multiple logistic regression analysis of risk factors influencing the progression of diabetic nephropathy alone.