Association between Hemoglobin A1c and Renal Arteriolar Sclerosis in Subjects Presenting without any Apparent Kidney Dysfunction

Abstract

Aims: Diabetic kidney disease is a major vascular complication in patients with diabetes mellitus (DM). However, the association between the hemoglobin (Hb)A1c levels, notably the prediabetic levels, and renal pathological changes remains unclear. We investigated the association between the HbA1c levels and renal arteriolar lesions in subjects without any apparent kidney dysfunction using a living kidney donor cohort.

Methods: Between January 2006 and May 2016, 393 living kidney donors underwent a “zero-time” biopsy at Kyushu University Hospital. The patients were divided into four groups (HbA1c levels ＜5.6%, 5.6%–5.7%, 5.8%–6.4%, and ≥ 6.5%, or diagnosed with DM [DM group]). Renal arteriolar hyalinization and wall thickening were assessed using semi-quantitative grading. We then investigated the association between the HbA1c levels and renal pathological changes.

Results: 158 (40.2%) patients had arteriolar hyalinization and 148 (37.6%) showed wall thickening. A significant correlation was observed between the HbA1c levels and wall thickening (p for trend ＜0.001). An elevated HbA1c level was significantly associated with wall thickening according to a multivariable logistic analysis in subjects with HbA1c levels of 5.6%–5.7% and 5.8%–6.4%, and the DM group, compared with those with HbA1c levels of ＜5.6% (odds ratio [OR], 1.91; 95% confidence interval [CI]: [1.03–3.54] for 5.6%–5.7%, OR, 1.96; 95% CI: [1.09–3.53] for 5.8%–6.4%, and OR, 2.86; 95% CI: [0.91–9.01] for the DM group), whereas arteriolar hyalinization did not increase within the nondiabetic HbA1c levels.

Conclusions: Elevated high-normal HbA1c levels are considered to be independent risk factors for arteriolar wall thickening. Subclinical renal arteriolar sclerosis may develop in patients with prediabetic HbA1c levels.

Introduction/Aim

Increasing evidence suggests that prediabetes is a significant risk factor for the future incidence of cardiovascular disease^{1, 2)}. Furthermore, high levels of hemoglobin A1c (HbA1c), even within the normal range, are associated with macrovascular atherosclerotic changes such as coronary artery disease and intima-media thickness of the carotid artery in nondiabetic subjects^{3, 4)}. Moreover, several studies have demonstrated an association between high-normal HbA1c levels and microvascular complications, such as retinopathy⁵⁾ and microalbuminuria⁶⁾. These studies indicate that high-normal HbA1c levels are associated with the progression of diabetic vascular complications in several organs. Based on these findings, there may also be an association between high and normal HbA1c levels and early pathological renal vascular changes. However, the direct pathological findings of the association between high-normal HbA1c levels and renal arteriolar changes remain unclear. As a renal biopsy is performed in patients with proteinuria or renal dysfunction, it is difficult to evaluate the early pathological changes in diabetic kidney disease (DKD). Further studies in healthy subjects are needed to clarify these early pathological changes.

In both living- and deceased-donor kidney transplantation (KT), many transplant centers perform wedge biopsy (‘zero-time’ biopsy) from the donated kidney during transplant surgery to investigate the baseline pathological changes. In living-donor KT, subjects with apparent kidney dysfunction, chronic kidney disease (CKD), or DKD are excluded as donor candidates. Therefore, a zero-time biopsy provides a unique and valuable opportunity to observe the kidney morphology in patients without any apparent kidney dysfunction. We hypothesized that findings from a zero-time biopsy would reveal the impact of HbA1c levels within the normal range on pathological renal changes. We previously reported an association between the serum uric acid levels and vascular changes in a zero-time biopsy-based cohort⁷⁾. In this study, although the presence of diabetes was not statistically associated with arteriolar hyalinization or arteriolar wall thickening, these results suggest that the presence of diabetes or prediabetes might be associated with vascular complications.

We herein examined the association between the HbA1c levels and renal pathological changes, notably microvascular changes, in healthy subjects from our zero-time biopsy-based cohort, without any apparent kidney dysfunction.

Methods

Study Design and Population

This was a cross-sectional study of subjects who donated a kidney for living-donor KT at Kyushu University Hospital between January 2006 and May 2016. During the study period, 623 living-donor KTs procedures were performed. A zero-time biopsy was not performed in 106 donors because of various factors such as a bleeding tendency, the use of antiplatelet therapy, or capsular injury of the donated kidney. Sixty-five donors were excluded due to inadequate biopsy samples (＜10 glomeruli). Another 59 donors were excluded because of missing relevant laboratory findings. The remaining 393 donors were included in the study. This study adhered to the guidelines of the Declaration of Helsinki and the study protocol was approved by the Human Ethics Committee of Kyushu University Hospital (protocol 21180-00). The ethics committee of all participating institutions granted approval to waive the requirement for written informed consent because of the retrospective nature of the present study. The database was accessed for analytical purposes only, and researchers did not access any personal information.

Pathological Interpretation of Zero-Time Biopsies

A wedge biopsy sample was obtained from the outer cortex of donated kidneys during bench surgery. The samples were fixed in formalin and embedded in paraffin. Next, samples were cut at 2-µm thickness and stained with hematoxylin-eosin, periodic acid-Schiff (PAS), methenamine silver, and Masson trichrome. Atherosclerotic changes in renal biopsies, namely % global glomerulosclerosis (%GGS), arteriolar hyalinization, and wall thickening were evaluated. Light microscopy was used to determine the%GGS. The vascular pathological findings were evaluated according to previously reported methods⁸⁾. Namely, Arteriolar hyalinization was scored as follows: absent was classified as grade 0; one or more partial arteriolar hyalinosis was classified as grade 1 (G1); approximately 50% hyalinosis was classified as grade 2 (G2); and more than 50% hyalinosis or penetrating hyalinosis, grade 3 (G3). Based on these findings, the arteriolar hyalinization index was calculated as the mean grade of renal arteriolar hyalinization in each patient, according to the following formula: arteriolar hyalinization index=(n0×0＋n1×1＋n2×2＋n3×3)÷N. Here, n0, n1, n2, and n3 indicate the number of arterioles with hyalinization scores of G0–G3, and N indicates the total number of arterioles. Arteriolar wall thickening was scored as follows: absent, grade 0 (G0); intimal thickening less than the thickness of the media, grade 1 (G1); and intimal thickening greater than the thickness of the media, grade 2 (G2). Based on these findings, arteriolar wall thickening index was calculated as the mean grade of renal arteriolar wall thickening in each patient according to the following formula: arteriolar wall thickening index=(n0×0＋n1 ×1＋n2×2)/N. Here, n0, n1, and n2 indicate the number of arterioles with wall thickening scores of G0–G2, respectively, and N indicates the total number of arterioles. %GGS was evaluated by three observers (YM, AT, and KM), and the vascular findings were evaluated by one observer (YM) who were all blinded to the donor’s clinical data.

Covariates

Information on medical history, current medications, and smoking habits was obtained from each donor prior to kidney donation. Because none of the donors were current smokers before surgery, they were divided into past smokers and nonsmokers. The baseline clinical data were collected from the medical records. Hypertension was defined as a blood pressure ≥ 140/90 mmHg or the current use of antihypertensive agents. Diabetes mellitus (DM) was defined as fasting blood glucose concentration ＞126 mg/dL, HbA1c ＞6.5%, past history of DM, and/or current use of oral glucose-lowering agents. In subjects whose HbA1c levels were measured according to Japanese Diabetes Society/Japanese Society of Clinical Chemistry guidelines, the values were standardized by adding 0.4% to the estimate to achieve the National Glycohemoglobin Standardization Program equivalent value⁹⁾. In subjects with DM, the criterion for microalbuminuria was ＜30 mg/g creatinine (Cr) and that for proteinuria was ＜150 mg/g Cr. Dyslipidemia was defined as a total cholesterol level of ≥ 220 mg/dL or the use of lipid-modifying agents. Hyperuricemia was defined as a serum uric acid concentration ＞7.0 mg/dL in men or ＞6.0 mg/dL in women. Obesity was defined as a body mass index ≥ 25 kg/m². Serum uric acid and Cr levels were measured using enzymatic assays. Because all donors were ≥ 18 years old, the estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation with a Japanese coefficient of 0.813 ¹⁰⁾, where SCr=serum creatinine; κ=0.7 for females and 0.9 for males, α=−0.329, and −0.411 for females, females, and males, respectively; min (SCr/κ, 1) indicates the minimum SCr/κ or 1; and max (SCr/κ, 1) indicates the maximum SCr/κ or 1.

eGFR (mL/min/1.73 m²)=0.813×141×min (SCr/κ, 1)α×max (SCr/κ, 1)−1.209×0.993Age (if female, ×1.018).

Statistical Analyses

The subjects were divided into four groups according to tertiles of the HbA1c level and the presence of DM as follows: HbA1c levels ＜5.6, 5.6–5.7, 5.8–6.4, and ≥ 6.5 % or diagnosed with DM (DM group). Data are presented as the mean and standard deviation, median and interquartile range, or percentage for categorical measures, as appropriate. The Jonckheere–Terpstra and Cochran–Armitage tests were used to evaluate trends in continuous and categorical values, respectively, across the four groups. Furthermore, we evaluated the associations between the HbA1c levels and the arteriolar hyalinization index and wall thickening index using Spearman’s rank correlation. We performed a logistic regression analysis using age- and sex-adjusted and multivariable-adjusted models. In the multivariable analysis, the potential confounding factors were age, sex, obesity, hypertension, dyslipidemia, uric acid, smoking habits, and eGFR. Heterogeneity in the correlations between subgroups was evaluated by adding a multiplicative interaction term to the relevant logistic model. All statistical analyses were performed using the JMP version 11 software program for Windows (SAS Institute, Cary, NC, USA). A two-tailed p value ＜0.05 was considered to be statistically significant in all analyses.

Results

Baseline Features of Subjects according to the HbA1c Levels

The baseline clinical features of the subjects according to the HbA1c levels are presented in Table 1. The median age of the 393 donors (131 men and 262 women) was 55 years (range: 20–79 years). The mean age, systolic blood pressure, and total cholesterol levels increased with higher HbA1c levels, while eGFR decreased with higher HbA1c levels. Participants with higher HbA1c levels had a lower proportion of a smoking habit and a higher proportion of dyslipidemia.

Table 1.Baseline clinical features of subjects according to tertiles of HbA1c level and the presence of diabetes mellitus

HbA1c (%)	All (n = 393)	T1 (n = 119)＜5.6	T2 (n = 108) 5.6–5.7	T3 (n = 149) 5.8–6.4	DM (n = 17)≥ 6.5	p for trend
Age (years)	57 (47–64)	52 (41–61)	56 (45.5–64)	60 (52–64.5)	63 (61–64)	＜0.001
Male (%)	33.3	36.1	34.3	28.2	52.9	0.61
Obesity (%)	20.9	14.3	25.0	22.1	29.4	0.08
Smoking habits (%)	28.5	35.3	33.3	20.8	17.6	0.004
Hypertension (%)	24.1	26.1	22.2	22.1	41.2	0.95
Diabetes mellitus (%)	4.3	0.0	0.0	0.0	100.0	＜0.001
Dyslipidemia (%)	45.0	37.0	37.0	59.1	29.4	0.005
BMI (kg/m2)	22.1 (20.3–24.3)	21.9 (20.3–23.9)	21.8 (19.9–25)	22.2 (20.2–24.3)	23.4 (22.1–25.3)	0.08
SBP (mmHg)	118 (108–128)	115 (105–128)	115 (105.5–127)	121 (111–130)	122 (111–131)	0.005
DBP (mmHg)	72 (65–80)	71 (65–78)	71.5 (64–80)	73 (66–81)	74 (68.5–79)	0.06
Uric acid (mg/dL)	4.7 (4–5.7)	4.6 (3.9–5.6)	4.9 (4–5.7)	4.7 (4–5.7)	5.3 (4.5–5.8)	0.71
Total cholesterol (mg/dL)	209 (190–234)	200 (180–229)	206 (186–230)	220 (198–247)	205 (184–214)	＜0.001
eGFR (mL/min/1.73 m2)	84.2 (78.6–91.3)	87.3 (79.6–96.8)	84.2 (78.5–92.5)	82.7 (78.1–87.5)	79.0 (75.4–87.4)	＜0.001
Use of antihypertensive agent (%)	15.8	18.5	16.7	11.4	29.4	0.44
Use of RAS blocker (%)	8.4	8.4	10.2	6.0	17.6	0.94
Use of oral hypoglycemic agent (%)	3.1	0.0	0.0	0.0	70.6	＜0.001
Use of lipid-lowering agent (%)	9.7	6.7	9.3	12.8	5.9	0.20
Use of diuretic (%)	0.3	0.0	0.9	0.0	0.0	0.86
Use of uric acid-lowering agent (%)	1.0	1.7	0.9	0	5.9	0.22
arteriolar wall thickening index	0.11±0.18	0.07±0.14	0.10±0.15	0.15±0.22	0.14±0.13	0.001
arteriolar hyalinization index	0.11±0.16	0.10±0.15	0.12±0.19	0.11±0.15	0.16±0.16	0.42

Abbreviations: RAS, Renin-Angiotensin-Aldosterone System; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; HbA1c, hemoglobin A1c; eGFR, estimated glomerular filtration rate.

Association between the HbA1c Levels and Pathological Changes

Fig.1 shows the association between the HbA1c levels and the pathological changes. The association between the HbA1c level and %GGS was statistically significant (p for trend=0.002). We found a significant correlation between the HbA1c levels and arteriolar wall thickening (p for trend＜0.001), whereas no association was observed between the HbA1c levels and arteriolar hyalinization (p=0.41). Furthermore, a correlation was observed between the HbA1c levels and the arteriolar wall thickening index (ρ=0.18, p＜0.001) using the Spearman Rank Correlation. The association between the HbA1c levels and the arteriolar hyalinization index was not statistically significant (p=0.042, p=0.41) (Fig.2). Therefore, we performed a logistic analysis to investigate the association between the HbA1c levels and microvascular changes in more detail.

Fig.1.

(A) % global glomerulosclerosis, (B) Maximum grades of arteriolar hyalinization, and (C) Maximum grades of arteriolar wall thickening according to tertiles of the HbA1c level and the presence of DM; HbA1c levels ＜5.6, 5.6–5.7, 5.8–6.4, and ≥ 6.5 % or diagnosed with diabetes mellitus. Trends were evaluated using either the Jonckheere–Terpstra test (% global glomerulosclerosis) or the Cochran–Armitage test (for the presence of arteriolar hyalinization and arteriolar wall thickening).

Fig.2.

(A) Arteriolar hyalinization index and (B) Arteriolar wall thickening index according to the HbA1c level (%). The association between the HbA1c levels and the arteriolar hyalinization index or arteriolar wall thickening index were evaluated by using Spearman Rank Correlation.

A Multivariable Analysis to Determine the Association between the HbA1c Levels and Pathological Changes

Overall, 158 (40.2%) subjects had arteriolar hyalinization and 148 (37.6%) had wall thickening. The association between the HbA1c levels and arteriolar changes was examined using continuous variables and categorical analyses using a logistic model. In the age- and sex-adjusted analysis for the HbA1c levels as a continuous variable, the odds ratio (OR) for arteriolar wall thickening increased linearly with increasing HbA1c levels (OR 2.01, 95% confidence interval [CI] 1.05–3.83, p=0.03). Furthermore, the age- and sex-adjusted ORs for arteriolar wall thickening were significantly higher in the high-normal HbA1c group than in the lowest HbA1c group (＜5.6%) (OR, 1.76; 95% CI: [0.97–3.18] for 5.6%–5.7%, OR, 1.84; 95% CI: [1.05–3.20] for 5.8%–6.4%, OR, 2.34; 95% CI: [0.78–7.05] for the DM group), and the trend test was statistically significant (p=0.03). This association was statistically significant, even after adjusting for potential confounding factors. Since the DM group was small (n=17), statistical significance may not have been achieved. In contrast, there was no statistically significant difference between the HbA1c levels and arteriolar hyalinization (Table 2). We also examined the association between the HbA1c levels and %GGS using a multivariable regression analysis, although statistical significance was not obtained (diabetic group; standardized β coefficient −0.02, p=0.75, high-normal HbA1c group; standardized β coefficient −0.05, p=0.39 for the lowest HbA1c group).

Table 2. Age- and sex-adjusted and multivariable-adjusted odds ratios for the presence of arteriolar hyalinization and arteriolar wall thickening according to tertiles of HbA1c level and the presence of diabetes mellitus

	N of Presence/ All	Age- and sex-adjusted			Multivariable-adjusted＊
		OR (95% CI)	p	p for trend	OR (95% CI)	p	p for trend
Arteriolar hyalinization
HbA1c (per 1-% increase)		0.89 (0.48–1.65)	0.71		1.00 (0.52–1.92)	0.99
T1 (＜5.6)	47/119	Ref.			Ref.
T2 (5.6–5.7)	41/108	0.84 (0.48–1.47)	0.54	0.91	0.85 (0.47–1.53)	0.59	0.75
T3 (5.8–6.4)	60/149	0.88 (0.52–1.49)	0.63		0.97 (0.56–1.70)	0.92
DM (≥ 6.5)	10/17	1.30 (0.43–3.89)	0.64		1.50 (0.48–4.65)	0.48
Arteriolar wall thickening
HbA1c (per 1-% increase)		2.01 (1.05–3.83)	0.03		2.21 (1.12–4.36)	0.02
T1 (＜5.6)	30/119	Ref.			Ref.
T2 (5.6–5.7)	43/108	1.76 (0.97–3.18)	0.06	0.03	1.91 (1.03–3.54)	0.04	0.02
T3 (5.8–6.4)	65/149	1.84 (1.05–3.20)	0.03		1.96 (1.09–3.53)	0.02
DM (≥ 6.5)	10/17	2.34 (0.78–7.05)	0.13		2.86 (0.91–9.01)	0.07

^＊Multivariable adjusted model; adjusted for age, sex, obesity, hypertension, dyslipidemia, smoking habits, eGFR and serum uric acid. Abbreviations: OR, odds ratio; CI, confidence interval; eGFR, estimated glomerular filtration rate.

Stratified Analysis

To assess the consistency of the association between the HbA1c levels and arteriolar wall thickening, we also conducted a stratified analysis according to potential confounders (Table 3). No significant interactions were detected between the HbA1c levels and the baseline features: age, sex, hypertension, reduced eGFR (＜90 mL/min/1.73 m²), and smoking habits (all P for interaction ≥ 0.1).

Table 3. Multivariable-adjusted odds ratios for the presence of arteriolar wall thickening per 1-% increment in HbA1c level among subgroups stratified by clinical parameters

		Presence of arteriolar wall thickening
		N of Presence/All	OR (95% CI)	p	p for interaction
Age	Low (＜60 years)	59/224	3.87 (1.28–11.74)	0.02	0.20
	High (≥ 60 years)	89/169	1.36 (0.55–3.38)	0.50
Sex	Male	58/131	1.46 (0.44–4.79)	0.53	0.66
	Female	90/262	2.52 (1.09–5.84)	0.03
eGFR	High (≥ 90 mL/min/1.73 m2)	125/284	5.72 (1.02–32.2)	0.04	0.34
	Low (＜90 mL/min/1.73 m2)	23/109	1.93 (0.90–4.13)	0.08
Smoking	Presence	46/112	2.11 (0.44–10.0)	0.35	0.59
	Absence	102/281	2.10 (0.96–4.58)	0.06
Hypertension	Presence	55/95	1.50 (0.42–5.39)	0.53	0.40
	Absence	93/298	2.50 (1.10–5.68)	0.03

^＊Multivariable adjusted model; adjusted for age, sex, obesity, hypertension, dyslipidemia, smoking habits, eGFR and serum uric acid. Abbreviations: OR, odds ratio; CI, confidence interval; eGFR, estimated glomerular filtration rate.

Discussion

Using our zero-time biopsy-based cohort, the present study clearly demonstrated that elevated HbA1c, even within nondiabetic levels, is an independent risk factor for the presence of arteriolar wall thickening in the healthy general population without any apparent kidney dysfunction or CKD. Our study suggests that subclinical renal arteriolar sclerosis may occur with prediabetic HbA1c levels.

Recent clinical studies in nondiabetic populations revealed positive associations between high-normal HbA1c levels and diabetic microvascular complications such as neuropathy, retinopathy, and nephropathy^{6, 11)}. A population-based study from Australia showed that the prevalence of retinopathy and microalbuminuria increased from a high-normal range compared to a low level of HbA1c, and the threshold for the increasing prevalence of retinopathy was 6.1% for HbA1c⁶⁾. Another study showed that elevated HbA1c, but not fasting plasma glucose, is a significant risk factor for the future incidence of CKD in non-diabetic individuals, and with a threshold of 5.7%¹²⁾. In this study, prediabetes was a significant risk factor for CKD, even though subjects who progressed to diabetes during the observational period were excluded from the analysis. Overall, these studies indicate that prediabetes has the potential to promote renal injury. The results of our study may provide a valuable explanation for the early pathological microvascular changes that occur in the kidneys of patients with prediabetes.

A previous autopsy study assessed the association between glucose intolerance and pathological renal changes¹³⁾. This study showed that glucose intolerance was associated with glomerulosclerosis, arteriolar hyalinization, and arteriolar wall thickening in an age-adjusted analysis. However, these associations were not statistically significant in multivariate analyses. In our study, arteriolar wall thickening gradually increased with a high normal HbA1c level. Although the explanation for the discrepancy between our study and the previous one is unclear, the results may have been affected by differences in the study design or target population. For example, the mean age was different (previous study: 73±12 years in men, 78±11 years in women, 53±14 years in men, and 56±11 years in women). In our subgroup analysis, the association between the HbA1c level and arteriolar wall thickening was apparent among relatively healthy subgroups, namely younger subjects, those with high eGFR, females, and those with an absence of hypertension. The risk of prediabetes may be higher in such populations. The autopsy study included a higher number of unhealthy subjects than kidney donors in the present study. Therefore, the study design may have affected the results. Cardiovascular disease is a clinically important complication, but the risk factors differ between various populations^{14, 15)}; therefore, future pathological studies in various populations are required to clarify the association in more detail.

Historically, during the progression of diabetic nephropathy, an initial increase in the albumin excretion rate is believed to be linked to a subsequent decline in GFR. Recently, it has been reported that not all diabetic patients with renal dysfunction have microalbuminuria. In the United Kingdom Prospective Diabetes Study (n=5,102), 1,132 (28%) patients developed renal impairment, which was defined as Cockcroft-Gault estimated Cr clearance ＜60 mL/min or doubling of plasma Cr. Among them, 575 (51%) did not have any preceding albuminuria, a typical marker of glomerular injury. This study highlighted that not all diabetic renal impairments begin with glomerular injury and thus they can result from other causes, including vascular damage¹⁶⁾. In these patients, vascular lesions, such as nephrosclerosis, are considered the leading pathology, and the results of our study are consistent with these findings to some extent.

Renal arteriolar hyalinization and arteriolar wall thickening are typical findings in nephrosclerosis¹⁷⁾ and major vascular complications in patients with diabetes¹⁸⁾. Several studies have evaluated the association between vascular lesions and the renal prognosis in diabetic patients. Although some studies have reported that the association between vascular lesions and the renal prognosis was not significant¹⁹⁾ or significant only in a univariate analysis^{8, 20)}, others have reported that vascular changes have significant prognostic value, even in a multivariable analysis²¹⁾. As these studies evaluated patients with proteinuria or renal dysfunction who underwent a renal biopsy, glomerular damage may be a more apparent marker than vascular lesions. Furthermore, previous studies have demonstrated that arteriolar wall thickening or chronic histological changes in a zero-hour biopsy were independent risk factors for the relative loss of kidney function in either recipients²²⁾ or donors²³⁾. In our study, arteriolar wall thickening gradually increased with a high normal HbA1c level. In contrast, arteriolar hyalinization was also a more frequent lesion in the DM group, although an association with high-normal HbA1c levels was not observed. These data suggest that the thresholds between the HbA1c levels and these lesions may differ. Arteriolar wall thickening is also observed during normal aging and is a relatively mild and chronic pathological lesion. Furthermore, a recent study highlighted the association between high serum uric acid levels and arteriolar wall thickening²⁴⁾. Similar to other microvascular lesions, arteriolar wall thickening may increase when HbA1c is within non-diabetic levels^{5, 6, 11, 12)}. In contrast, arteriolar hyalinization is believed to be associated with endothelial dysfunction and high endothelial cell permeability¹⁷⁾. These pathological mechanisms may have contributed to these results. Several recent studies have elucidated the risk factors for atherosclerotic lesions, but the findings are still not considered to be sufficient^{25, 26)}. Further studies are necessary to clarify the effect of vascular lesions on renal decline in patients with diabetes.

Although the mechanisms underlying the positive association between the HbA1c levels and arteriolar changes remain unclear, there are some possible explanations. HbA1c is an early glycation end product and precursor of advanced glycation end products (AGEs). AGEs are senescent proteins that accumulate in various tissues during different stages of hyperglycemia. In vitro, AGEs have been shown to induce a significant increase in the number of viable human microvascular endothelial cells via the secretory forms of vascular endothelial growth factor²⁷⁾, and this phenomenon is accelerated by reactive oxygen species²⁸⁾. Furthermore, more apparent kidney injury was observed in AGE receptor (RAGE)-overexpressing mice than in diabetic littermates lacking the RAGE transgene²⁹⁾. In another study, decreased inflammation and oxidative stress were observed in RAGE^−/− mice and renal vascular lesions and glomerular sclerosis also decreased³⁰⁾. These studies demonstrate that AGEs may play an important role in diabetic microvascular changes, including renal arteriolar changes.

This study is associated with several limitations. First, it was a cross-sectional study. Therefore, a causal relationship could not be evaluated. However, a renal biopsy cannot be performed for research purposes because it is an invasive procedure. It is difficult to evaluate longitudinally via a renal biopsy in a healthy population. Second, this study was conducted on donors who were healthier than the general population. The prevalence of underlying classical cardiovascular risk factors differed from that observed in community-based populations. Therefore, it remains unclear whether our results are applicable to the general population. Further large-scale studies are required to clarify this issue. Third, the primary outcome of the present study was pathological changes. Our results could not conclusively demonstrate associations between the HbA1c levels and specific outcomes such as mortality rate, cardiovascular events, or renal survival. Future long-term observational studies may clarify the causal relationships. Finally, the interobserver agreement for this method in our institution was relatively low. We initially evaluated another semi-qualitative method³¹⁾, but the poor inter-observer agreement at our institution (Fleiss’ κ: arteriolar hyalinization 0.09 and wall thickening 0.33) suggested that the findings of this method may be biased in our study. Therefore, we used the recently established method⁸⁾. Vascular findings were reevaluated by one observer (YM). The inter-observer agreement in this method in our institution was fair to moderate (Fleiss’ κ: arteriolar hyalinization, 0.19; wall thickening 0.42, respectively). On the other hand, our inter-observer agreement of vascular findings was not as low as that in several previous studies^{32, 33)}. Further studies are needed to identify a better pathological evaluation method.

Conclusion

This study showed that a high-normal HbA1c level is an independent predictor of arteriolar sclerosis in subjects without any apparent kidney dysfunction and may be an important component of DKD. Further studies should be performed to clarify the mechanism underlying the association between high and normal HbA1c levels and the progression of renal arteriolar sclerosis.

Acknowledgements and Notice of Grant Support

We thank Richard Robins, Ph.D., from Edanz (https://jp.edanz.com/ac) for editing the draft of this manuscript.

The authors received no financial support for the research, authorship, or publication of this paper. This study was not supported by any funding sources.

Author Contributions

Study design: YM, AT, KM, and ST; data acquisition: YM, AT, and NH; pathological evaluation: YM, AT, KU, and KM; statistical analysis: YM and ST; supervision: KM, YO, MN, TK, and TN. Each author contributed important intellectual content during manuscript drafting or revision and accepted accountability for the overall work by ensuring that questions pertaining to the accuracy or integrity of any portion of the work were appropriately investigated and resolved. YM and TN accept responsibility that this study has been reported honestly, accurately, and transparently, that no important aspects of the study have been omitted, and that any discrepancies from the study as planned have been explained.

Conflict of Interest

The authors declare no conflicts of interest.

References

• 1)  Coutinho M, Gerstein HC, Wang Y, Yusuf S. The relationship between glucose and incident cardiovascular events. A metaregression analysis of published data from 20 studies of 95,783 individuals followed for 12.4 years. Diabetes Care, 1999; 22: 233-240
• 2)  Ikeda F, Doi Y, Ninomiya T, Hirakawa Y, Mukai N, Hata J, Shikata K, Yoshida D, Matsumoto T, Kitazono T, Kiyohara Y. Haemoglobin A1c even within non-diabetic level is a predictor of cardiovascular disease in a general Japanese population: the Hisayama Study. Cardiovasc Diabetol, 2013; 12: 164
• 3)  Scicali R, Giral P, Gallo A, Di Pino A, Rabuazzo AM, Purrello F, Cluzel P, Redheuil A, Bruckert E, Rosenbaum D. HbA1c increase is associated with higher coronary and peripheral atherosclerotic burden in non diabetic patients. Atherosclerosis, 2016; 255: 102-108
• 4)  Jørgensen L, Jenssen T, Joakimsen O, Heuch I, Ingebretsen OC, Jacobsen BK. Glycated hemoglobin level is strongly related to the prevalence of carotid artery plaques with high echogenicity in nondiabetic individuals: the Tromsø study. Circulation, 2004; 110: 466-470
• 5)  Miyazaki M, Kubo M, Kiyohara Y, Okubo K, Nakamura H, Fujisawa K, Hata Y, Tokunaga S, Iida M, Nose Y, Ishibashi T. Comparison of diagnostic methods for diabetes mellitus based on prevalence of retinopathy in a Japanese population: the Hisayama Study. Diabetologia, 2004; 47: 1411-1415
• 6)  Tapp RJ, Zimmet PZ, Harper CA, de Courten MP, McCarty DJ, Balkau B, Taylor HR, Welborn TA, Shaw JE. Diagnostic thresholds for diabetes: the association of retinopathy and albuminuria with glycaemia. Diabetes Res Clin Pract, 2006; 73: 315-321
• 7)  Matsukuma Y, Masutani K, Tanaka S, Tsuchimoto A, Haruyama N, Okabe Y, Nakamura M, Tsuruya K, Kitazono T. Association between serum uric acid level and renal arteriolar hyalinization in individuals without chronic kidney disease. Atherosclerosis, 2017; 266: 121-127
• 8)  Furuichi K, Yuzawa Y, Shimizu M, Hara A, Toyama T, Kitamura H, Suzuki Y, Sato H, Uesugi N, Ubara Y, Hisano S, Ueda Y, Nishi S, Yokoyama H, Nishino T, Kohagura K, Ogawa D, Mise K, Shibagaki Y, Kimura K, Haneda M, Makino H, Matsuo S, Wada T. Nationwide multicentre kidney biopsy study of Japanese patients with type 2 diabetes. Nephrol Dial Transplant, 2018; 33: 138-148
• 9)  Kashiwagi A, Kasuga M, Araki E, Oka Y, Hanafusa T, Ito H, Tominaga M, Oikawa S, Noda M, Kawamura T, Sanke T, Namba M, Hashiramoto M, Sasahara T, Nishio Y, Kuwa K, Ueki K, Takei I, Umemoto M, Murakami M, Yamakado M, Yatomi Y, Ohashi H. International clinical harmonization of glycated hemoglobin in Japan: From Japan Diabetes Society to National Glycohemoglobin Standardization Program values. J Diabetes Investig, 2012; 3: 39-40
• 10)  Horio M, Imai E, Yasuda Y, Watanabe T, Matsuo S. Modification of the CKD epidemiology collaboration (CKD-EPI) equation for Japanese: accuracy and use for population estimates. Am J Kidney Dis, 2010; 56: 32-38
• 11)  Sabanayagam C, Liew G, Tai ES, Shankar A, Lim SC, Subramaniam T, Wong TY. Relationship between glycated haemoglobin and microvascular complications: is there a natural cut-off point for the diagnosis of diabetes? Diabetologia, 2009; 52: 1279-1289
• 12)  Koshi T, Sagesaka H, Sato Y, Hirabayashi K, Koike H, Yamauchi K, Nishimura R, Noda M, Yamashita K, Aizawa T. Elevated haemoglobin A1c but not fasting plasma glucose conveys risk of chronic kidney disease in non-diabetic individuals. Diabetes Res Clin Pract, 2018; 146: 233-239
• 13)  Kubo M, Kiyohara Y, Kato I, Tanizaki Y, Katafuchi R, Hirakata H, Okuda S, Tsuneyoshi M, Sueishi K, Fujishima M, Iida M. Risk factors for renal glomerular and vascular changes in an autopsy-based population survey: the Hisayama study. Kidney Int, 2003; 63: 1508-1515
• 14)  Kitamura H, Tanaka S, Hiyamuta H, Shimamoto S, Tsuruya K, Nakano T, Kitazono T. Cardiovascular Risk Factor Burden and Treatment Control in Patients with Chronic Kidney Disease: A Cross-Sectional Study. J Atheroscler Thromb, 2023; 30: 1210-1288
• 15)  Matsumura T, Sankai T, Yamagishi K, Kubota Y, Hayama-Terada M, Muraki I, Umesawa M, Cui R, Imano H, Ohira T, Kitamura A, Okada T, Kiyama M, Iso H. Impact of Major Cardiovascular Risk Factors on the Incidence of Cardiovascular Disease among Overweight and Non-Overweight Individuals: The Circulatory Risk in Communities Study (CIRCS). J Atheroscler Thromb, 2022; 29: 422-437
• 16)  Retnakaran R, Cull CA, Thorne KI, Adler AI, Holman RR. Risk factors for renal dysfunction in type 2 diabetes: U.K. Prospective Diabetes Study 74. Diabetes, 2006; 55: 1832-1839
• 17)  Hill GS. Hypertensive nephrosclerosis. Curr Opin Nephrol Hypertens, 2008; 17: 266-270
• 18)  Tervaert TW, Mooyaart AL, Amann K, Cohen AH, Cook HT, Drachenberg CB, Ferrario F, Fogo AB, Haas M, de Heer E, Joh K, Noël LH, Radhakrishnan J, Seshan SV, Bajema IM, Bruijn JA. Pathologic classification of diabetic nephropathy. J Am Soc Nephrol, 2010; 21: 556-563
• 19)  An Y, Xu F, Le W, Ge Y, Zhou M, Chen H, Zeng C, Zhang H, Liu Z. Renal histologic changes and the outcome in patients with diabetic nephropathy. Nephrol Dial Transplant, 2015; 30: 257-266
• 20)  Stefan G, Stancu S, Zugravu A, Petre N, Mandache E, Mircescu G. Histologic predictors of renal outcome in diabetic nephropathy: Beyond renal pathology society classification. Medicine (Baltimore), 2019; 98: e16333
• 21)  Zhang Y, Jiang Q, Xie J, Qi C, Li S, Wang Y, Him YH, Chen Z, Zhang S, Li Q, Zhu Y, Li R, Liang X, Bai X, Wang W. Modified arteriosclerosis score predicts the outcomes of diabetic kidney disease. BMC Nephrol, 2021; 22: 281
• 22)  Sofue T, Inui M, Kiyomoto H, Moritoki M, Nishioka S, Nishijima Y, Moriwaki K, Hara T, Kushida Y, Haba R, Yoda T, Hirao T, Kakehi Y, Nishiyama A, Kohno M. Pre-existing arteriosclerotic intimal thickening in living-donor kidneys reflects allograft function. Am J Nephrol, 2012; 36: 127-135
• 23)  Ohashi Y, Thomas G, Nurko S, Stephany B, Fatica R, Chiesa A, Rule AD, Srinivas T, Schold JD, Navaneethan SD, Poggio ED. Association of metabolic syndrome with kidney function and histology in living kidney donors. Am J Transplant, 2013; 13: 2342-2351
• 24)  Maki K, Hata J, Sakata S, Oishi E, Furuta Y, Nakano T, Oda Y, Kitazono T, Ninomiya T. Serum Uric Acid Levels and Nephrosclerosis in a Population-Based Autopsy Study: The Hisayama Study. Am J Nephrol, 2022; 53: 69-77
• 25)  Ganbaatar N, Kadota A, Hisamatsu T, Araki SI, Kume S, Fujiyoshi A, Kadowaki S, Torii S, Kondo K, Segawa H, Salman E, Miyazawa I, Yamamoto T, Nakagawa Y, Maegawa H, Miura K, Ueshima H. Relationship between Kidney Function and Subclinical Atherosclerosis Progression Evaluated by Coronary Artery Calcification. J Atheroscler Thromb, 2022; 29: 1359-1371
• 26)  Higashioka M, Sakata S, Honda T, Hata J, Shibata M, Yoshida D, Goto K, Kitazono T, Osawa H, Ninomiya T. The Association of Small Dense Low-Density Lipoprotein Cholesterol and Coronary Heart Disease in Subjects at High Cardiovascular Risk. J Atheroscler Thromb, 2021; 28: 79-89
• 27)  Yamagishi S, Yonekura H, Yamamoto Y, Katsuno K, Sato F, Mita I, Ooka H, Satozawa N, Kawakami T, Nomura M, Yamamoto H. Advanced glycation end products-driven angiogenesis in vitro. Induction of the growth and tube formation of human microvascular endothelial cells through autocrine vascular endothelial growth factor. J Biol Chem, 1997; 272: 8723-8730
• 28)  Urata Y, Yamaguchi M, Higashiyama Y, Ihara Y, Goto S, Kuwano M, Horiuchi S, Sumikawa K, Kondo T. Reactive oxygen species accelerate production of vascular endothelial growth factor by advanced glycation end products in RAW264.7 mouse macrophages. Free Radic Biol Med, 2002; 32: 688-701
• 29)  Yamamoto Y, Kato I, Doi T, Yonekura H, Ohashi S, Takeuchi M, Watanabe T, Yamagishi S, Sakurai S, Takasawa S, Okamoto H, Yamamoto H. Development and prevention of advanced diabetic nephropathy in RAGE-overexpressing mice. J Clin Invest, 2001; 108: 261-268
• 30)  Teissier T, Quersin V, Gnemmi V, Daroux M, Howsam M, Delguste F, Lemoine C, Fradin C, Schmidt AM, Cauffiez C, Brousseau T, Glowacki F, Tessier FJ, Boulanger E, Frimat M. Knockout of receptor for advanced glycation end-products attenuates age-related renal lesions. Aging Cell, 2019; 18: e12850
• 31)  Kohagura K, Kochi M, Miyagi T, Kinjyo T, Maehara Y, Nagahama K, Sakima A, Iseki K, Ohya Y. An association between uric acid levels and renal arteriolopathy in chronic kidney disease: a biopsy-based study. Hypertens Res, 2013; 36: 43-49
• 32)  Working Group of the International IgA Nephropathy Network and the Renal Pathology Society; Roberts IS, Cook HT, Troyanov S, Alpers CE, Amore A, Barratt J, Berthoux F, Bonsib S, Bruijn JA, Cattran DC, Coppo R, D’Agati V, D’Amico G, Emancipator S, Emma F, Feehally J, Ferrario F, Fervenza FC, Florquin S, Fogo A, Geddes CC, Groene HJ, Haas M, Herzenberg AM, Hill PA, Hogg RJ, Hsu SI, Jennette JC, Joh K, Julian BA, Kawamura T, Lai FM, Li LS, Li PK, Liu ZH, Mackinnon B, Mezzano S, Schena FP, Tomino Y, Walker PD, Wang H, Weening JJ, Yoshikawa N, Zhang H. The Oxford classification of IgA nephropathy: pathology definitions, correlations, and reproducibility. Kidney Int, 2009 Sep; 76: 546-556
• 33)  Mansour SG, Hall IE, Reese PP, Jia Y, Thiessen-Philbrook H, Moeckel G, Weng FL, Revelo MP, Khalighi MA, Trivedi A, Doshi MD, Schröppel B, Parikh CR. Reliability of deceased-donor procurement kidney biopsy images uploaded in United Network for Organ Sharing. Clin Transplant, 2018 Dec; 32: e13441