Introduction

Diabetic kidney disease (DKD), a major microvascular complication of type 2 diabetes (T2D), affects approximately 40% of T2D patients and remains the leading cause of end-stage renal disease (ESRD) worldwide.1,2 Given the irreversibility of advanced renal failure, identifying modifiable risk factors during the early disease stages is crucial for delaying DKD progression and improving clinical outcomes. Persistent microalbuminuria, defined as a urinary albumin-to-creatinine ratio (UACR) of 30–300 mg/g, serves as an early marker of glomerular injury and represents a critical window for intervention.3

While chronic hyperglycemia, typically assessed by glycated hemoglobin (HbA1c), an establisheddriver of the development and progression of DKD, a considerable proportion of T2D patients still experience DKD progression despite achieving target HbA1c levels. Accumulating evidence indicates that glycemic variability (GV) is a key driver of diabetic vascular complications, independent of chronic hyperglycemia.4–7 Specifically, acute glucose fluctuations can induce oxidative stress, endothelial dysfunction, and inflammatory activation, all central to the pathophysiology of DKD.8–10 Although experimental evidence supports these pathophysiological pathways, population-based studies on short-term GV and early renal injury have produced conflicting results. GV encompasses both long-term and short-term components.11,12 Long-term GV, characterized by visit-to-visit variability in HbA1c, has been consistently shown to predict albuminuria in patients with T2D.2,13–15 In contrast, the impact of short-term GV, defined as daily glucose fluctuations, on early renal injury remains controversial.16,17 Earlier studies relying on self-monitoring of blood glucose (SMBG) failed to detect significant associations between short-term GV and renal injury, likely due to an inability to capture nocturnal or postprandial glycemic excursions.4,6,18,19 Although continuous glucose monitoring (CGM) has overcome these limitations by providing more detailed and comprehensive glucose data, findings from CGM-based studies on short-term GV in the T2D populations remain inconsistent.20–22 A common yet often overlooked limitation in prior studies exploring the association between GV and albuminuria is inadequate consideration of estimated glomerular filtration rate (eGFR), often resulting in the inclusion of patients with significantly reduced eGFR. Since renal dysfunction itself can promote the occurrence of hypoglycemia and glycemic fluctuations, this may confound the interpretation of the true association. Furthermore, it remains unclear whether other risk factors modify the association between GV and early renal injury.

Therefore, we hypothesized that higher short-term GV, measured by CGM, is independently associated with early-stage DKD in T2D patients with preserved renal function, even after full adjustment for HbA1c and other conventional risk factors. To test this, the present study aimed to investigate the independent association between CGM-derived short-term GV and early-stage DKD, defined as persistent microalbuminuria (UACR 30–300 mg/g) alongside an eGFR≥60 mL/min/1.73 m2. According to the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines, this stage corresponds to KDIGO G1–G2/A2,23–25 wherein renal filtration function is largely maintained and microalbuminuria serves as a reliable clinical surrogate for underlying pathological changes. By focusing exclusively on this early-stage population, we sought to isolate the effect of short-term GV from the confounding influence of declining eGFR, thereby clarifying the role of short-term GV in the early pathogenesis of DKD.

Methods Study Design and Participants

This cross-sectional study retrospectively analyzed data from a cohort of 315 patients with T2D admitted to the Department of Endocrinology, The Second Affiliated Hospital of Anhui Medical University. Inclusion criteria were as follows: (1) age ≥18 years; (2) confirmed diagnosis of T2D in accordance with the World Health Organization (WHO) 2019 criteria; (3) availability of complete 72-hour CGM data; (4) valid UACR measurements (confirmed by at least two consecutive early-morning urine samples, as detailed in the DKD Staging section); and (5) eGFR ≥60 mL/min/1.73m2, calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.7 Key exclusion criteria included: (1) type 1, gestational, or other specific types of diabetes; (2) acute metabolic complications (eg, diabetic ketoacidosis, hyperosmolar hyperglycemic state); (3) recent major vascular events (eg, myocardial infarction, stroke) within 3 months prior to admission; (4) a known diagnosis of non-diabetic kidney disease (eg, glomerulonephritis, hypertensive nephropathy, polycystic kidney disease); (5) pregnancy or lactation; (6) current use of glucocorticoids, nonsteroidal anti-inflammatory drugs (NSAIDs), or other nephrotoxic agents (eg, aminoglycoside antibiotics) within 1 month of admission; and (7) other potential secondary causes of albuminuria (eg, uncontrolled hypertension [systolic blood pressure ≥ 160 mmHg or diastolic blood pressure ≥ 100 mmHg], active urinary tract infection confirmed by urine culture). This study was approved by the Ethics Review Committee of the Second Affiliated Hospital of Anhui Medical University and all patients were required to give full informed consent before the study (Approval No. YX2025-104).

DKD Staging

DKD staging was performed in accordance with the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines.23,24 Early-stage DKD was defined as the presence of persistent microalbuminuria (UACR 30–300 mg/g) confirmed by at least two consecutive early-morning urine samples (collected at intervals of ≥2 weeks to exclude transient albuminuria), combined with preserved kidney function (eGFR ≥ 60 mL/min/1.73m2, KDIGO G1–G2 stages).

Data Collection

Demographic, clinical, and laboratory data were collected for all participants. Demographic and clinical data included age, gender, duration of T2D, relevant medical history and medication use. Medication use was categorized as follows:

Angiotensin Receptor Blockers/Angiotensin-Converting Enzyme Inhibitors (ARB/ACEI), Sodium-Glucose Cotransporter 2 (SGLT-2), Glucagon-like Peptide-1 (GLP-1) and insulin. Current use of each drug class was recorded at the time of the study visit (yes/no). Standard anthropometric measurements, including height, weight, and blood pressure, were recorded. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters (kg/m2). Blood pressure (BP) was measured using a calibrated electronic sphygmomanometer after participants rested for ≥5 minutes in a seated position. Hypertension was defined as systolic BP≥140 mmHg, diastolic BP≥90 mmHg, a self-reported or medically documented history of hypertension, or current use of antihypertensive medications.

After an overnight fast of 8–12 hours, venous blood samples were collected and centrifuged. Serum samples were analyzed using standard automated biochemical assays for the following parameters: glycated hemoglobin A1c (HbA1c), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), serum creatinine (Scr), blood urea nitrogen (BUN), uric acid (UA), fasting plasma glucose (FPG), 2-hour postprandial plasma glucose (2hPG), fasting C-peptide (FCP), and 2-hour postprandial C-peptide (2hPCP).

Early-morning midstream urine samples were collected, and urinary albumin and creatinine concentrations were measured using immunoturbidimetry and enzymatic methods, respectively. The UACR was calculated as the ratio of urinary albumin to creatinine (expressed in mg/g). eGFR was calculated using the CKD-EPI creatinine equation.26

CGM Parameters

All participants underwent 72-hour CGM system (SIBIONICS, Shenzhen, China) with glucose measurements recorded at 5-minute intervals. During this period, 288 glucose values were generated daily. Key GV and glucose control parameters were derived from the CGM data, including: mean amplitude of glycemic excursions (MAGE), standard deviation of blood glucose (SDBG), mean blood glucose (MBG), coefficient of variation (CV), and time in range (TIR). Definitions of each parameter were as follows: TIR was defined as the percentage of time spent within the target glucose range of 3.9–10.0 mmol/L (70–180 mg/dL) over a 24-hour period; MBG referred to the average value of all valid CGM glucose measurements during the 72-hour monitoring period; SDBG was a measure of the dispersion of glucose values around the MBG, reflecting overall glycemic fluctuations; CV was calculated as (SDBG/MBG) × 100%, representing relative glycemic variability independent of the absolute glucose level; MAGE was calculated as the average amplitude of glycemic excursions exceeding one standard deviation (SD) of the MBG, specifically involving identifying glucose peaks and nadirs, including only excursions with amplitude >1 SD of MBG, and defining the mean of these valid excursion amplitudes as MAGE, which reflects the magnitude of major glycemic fluctuations. Based on a multicenter study from the Chinese Guidelines for Glucose Monitoring, the following clinical thresholds were established: MAGE ≥3.9 mmol/L was defined as high glucose variability, MAGE <3.9 mmol/L as low variability, and SDBG ≥1.4 mmol/L as high variability.

Statistical Analysis

Continuous variables are presented as mean ± standard deviation or median (interquartile range), based on distribution normality. Categorical variables are expressed as numbers (percentages). Group comparisons (Early DKD vs Non-DKD) were performed using Student’s t-test for normally distributed data, the Mann–Whitney U-test for non-normally distributed data, and the Chi-square test for categorical variables. The association of MAGE and SDBG with early DKD was evaluated using multivariable logistic regression, with MAGE and SDBG analyzed both as a continuous variable and categorized into tertiles. Three regression models were constructed: Model 1 was adjusted for age, sex and BMI; Model 2 was adjusted for age, sex, BMI, disease duration, smoking, drinking and hypertension; Model 3 further adjusted for HbA1c, insulin use, ARB/ACEI use, SGLT-2 inhibitors, and GLP-1 receptor agonists. Subgroup analyses were performed based on clinically relevant strata, including age (<60 vs ≥60 years), sex (male vs female), diabetes duration (<5 vs ≥5 years), hypertension status (yes vs no), BMI categorized according to the Chinese national standard (overweight defined as ≥24.0 kg/m2),27 and HbA1c level (<7.0% vs ≥7.0%). Generalized additive models (GAMs) was employed to visualize the potential non-linear exposure-response relationship between MAGE, SDBG and the probability of early-stage DKD. All statistical analyses were conducted using SPSS version 26.0 and R software, with a two-tailed p-value < 0.05 considered statistically significant.

Results Baseline Characteristics of Participants

As detailed in Table 1, significant differences in baseline characteristics were observed between the 124 patients with early-stage DKD and the 191 patients in the Non-DKD group. Compared with the Non-DKD group, the early-stage DKD group had longer diabetes duration, higher HbA1C level, higher 2h C-peptide level, more insulin use, lower TIR, higher MBG, SDBG and MAGE (all P < 0.05).

Table 1 Baseline Characteristics of Participants

Association Between MAGE and SDBG with Early‑Stage DKD: Multivariable Logistic Regression

Given the significant inter-group differences observed in short-term GV indicators (Table 1), multivariable logistic regression analyses were performed to evaluate the independent associations of MAGE and SDBG with early-stage DKD. Both metrics were analyzed as continuous variables and as categorical variables (stratified by tertiles).

As shown in Table 2, when MAGE was treated as a continuous variable, it was significantly and positively associated with the risk of early-stage DKD. This association remained robust across all adjustment models. Specifically, in the Model 1 adjusting for age, gender and BMI, the odds ratio (OR) was 1.51 (95% confidence interval [CI]: 1.28–1.77). This positive correlation persisted after adjusting for age, gender and BMI, diabetes duration, smoking, alcohol consumption and hypertension in Model 2 (OR = 1.6, 95% CI: 1.35–1.91), and remained significant in the fully adjusted model (Model 3), which was further adjusted for HbA1c, insulin use, ARB/ACEI use, SGLT-2 inhibitors use and GLP-1 receptor agonists use, the association persisted (OR = 1.58, 95% CI: 1.31–1.89). Furthermore, when stratified into tertiles, participants in the highest MAGE tertile (Tertile-3) exhibited a significantly higher association of early-stage DKD compared to those in the lowest tertile. This elevated risk was consistently observed in Model1 (OR = 5.08, 95% CI: 2.66–9.71), Model 2 (OR = 6.06, 95% CI: 3.04–12.09), and Model 3 (OR = 5.78, 95% CI: 2.83–11.81).

Table 2 Multivariate Logistic Regression Analysis Between MAGE and Early-Stage DKD

Parallel findings were observed for SDBG (Table 3). As a continuous variable, SDBG was independently associated with an increased association of early-stage DKD across all three models (Model 3: OR = 2.14, 95% CI: 1.39–3.30). Analysis by tertiles further confirmed this trend, with individuals in the highest SDBG tertile demonstrating a significantly greater association of early-stage DKD than those in the reference tertile, even after comprehensive adjustment for confounding factors in Model 3 (OR = 3.56, 95% CI: 1.77–7.18).

Table 3 Multivariate Logistic Regression Analysis Between SDBG and Early-Stage DKD

Subgroup Analysis

To evaluate the robustness and potential effect modification of the observed associations, subgroup analyses were prespecified according to age, sex, diabetes duration, hypertension status, BMI, and HbA1c level. For MAGE (Figure 1), a significant interaction with BMI was observed (P for interaction = 0.014). Stratified analysis showed that the positive correlation between elevated MAGE and the risk of early-stage DKD was robust in participants with BMI≥24 kg/m2 (OR =1.90, 95% CI:1.50–2.41), whereas no significant association was found in participants with normal BMI. No other significant interactions were detected for age, sex, diabetes duration, hypertension, or HbA1c (all P interaction >0.05), indicating that the effect of MAGE on early DKD was generally stable across most clinical subgroups, with the exception of its modification by BMI. In contrast (Figure 2), the association between SDBG and early-stage DKD was consistent across all predefined subgroups. Interaction tests for all covariates were non-significant (all P for interaction >0.05).

Figure 1 Subgroup analysis of the association between MAGE and early-stage DKD. Above model adjusted for age, gender, duration, BMI, smoking status, alcohol use, hypertension, HbA1C, insulin use except the stratification variable itself. BMI, body mass index.

Figure 2 Subgroup analysis of the association between SDBG and early-stage DKD. Above model adjusted for age, gender, duration, BMI, smoking status, alcohol use, hypertension, HbA1C, insulin use except the stratification variable itself. BMI body mass index.

Dose-Response Relationships of MAGE and SDBG with Early-Stage DKD: GAM Analysis

GAMs were employed to examine the dose-response relationships of MAGE and SDBG with the risk of early-stage DKD, with adjustment for all covariates included in Model 3. As shown in Figure 3A, GAM analysis revealed a statistically significant linear association between increasing MAGE levels and the risk of early-stage DKD (P linearity < 0.001). In contrast, the association between SDBG levels and the risk of early-stage DKD exhibited a non-linear, U-shaped curve (Figure 3B).

Figure 3 Exposure-response association curves between MAGE (A), SDBG (B) and early-stage DKD. Solid line represents the predicted value and the shaded area represents the 95% confidence interval for that prediction. Adjusted for “age”, “BMI”, “gender”, “diabetes duration”, “smoking status”, “alcohol consumption”, “hypertension”, “insulin use”, “HbA1C”.

Discussion

In this study, we observed that two CGM-derived short-term GV indices, MAGE and SDBG, were independently associated with early-stage DKD in patients with T2D and preserved renal function (KDIGO G1–G2; eGFR ≥60 mL/min/1.73 m2). Furthermore, we identified a significant interaction between MAGE and BMI and demonstrated distinct association patterns of MAGE and SDBG with early DKD using GAMs.

Epidemiological evidence linking GV to diabetic microvascular complications remains inconsistent overall. Long-term GV indices, such as visit-to-visit HbA1c or fasting glucose variability, have shown consistent associations with DKD risk, whereas findings for short-term GV have been conflicting. Earlier studies using 7-point low-frequency SMBG found no significant association between short-term GV and DKD progression in T1D,4,6,18,19 but intermittent SMBG sampling misses nocturnal and postprandial fluctuations and may underestimates actual GV exposure. Although CGM provides more accurate GV assessment,28 evidence on CGM-derived short-term GV in relation to albuminuria in T2D remains limited and conflicting. Zhou et al identified MAGE as an independent risk factor for microalbuminuria in T2D patients with well-controlled glycemia,21 and another study reported significant correlations between CGM-derived intraday and interday GV indices and albuminuria severity, independent of HbA1c.22 Conversely, two studies from the same research group found no significant association between CGM-derived short-term GV and albuminuria prevalence in T2D.20,29 A post-hoc analysis of the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial similarly found no independent association between 1,5-anhydroglucitol (1,5-AG), a surrogate marker of short-term GV, and microalbuminuria.10 These discrepancies are largely attributable to methodological and population differences. First, most earlier studies relied on intermittent SMBG sampling, which misses nocturnal and postprandial glucose fluctuations and systematically underestimates true GV exposure. Second, study populations differ substantially in renal disease stage, as many prior studies included patients across all DKD stages, whereas the present study focused exclusively on early-stage DKD with preserved eGFR. In early DKD, renal microvascular endothelial damage driven by acute glucose excursions and oxidative stress is more prominent, whereas in advanced disease, irreversible structural renal damage dominates in advanced disease, irreversible structural damage may attenuate the independent effect of GV. Third, differences in ethnicity, glycemic control level, and baseline nephroprotective medication use may also contribute to divergent findings. The present study specifically enrolled a Chinese T2D cohort with relatively preserved renal function, which may explain the detectable association observed in our study.

In our fully adjusted models, both MAGE and SDBG remained independently associated with early-stage DKD, suggesting that glucose excursions may have an independent association with early renal injury beyond average glycemic control alone. GAM analysis further revealed divergent association patterns, in which MAGE exhibited a linear positive relationship while SDBG showed a U-shaped association, indicating possible distinct pathophysiological pathways underlying the two indices. As a metric specifically designed to capture the magnitude of large, clinically meaningful glucose excursions,30 MAGE displayed a linear dose–response relationship with early DKD. MAGE’s linear association aligns with established evidence that acute glucose fluctuations trigger oxidative stress and endothelial dysfunction to drive renal microvascular damage.30–33 For SDBG, which reflects overall glycemic dispersion including both upward and downward glycemic fluctuations,34,35 its U-shaped relationship suggests that both excessively low and high SDBG may be linked to elevated renal risk. Elevated SDBG reflects impaired glycemic stability, whereas extremely low values may theoretically signal recurrent hypoglycemia or overly intensive glucose control, both of which carry heightened hypoglycemia risk that exacerbates renal microvascular injury.36 However, this interpretation remains strictly hypothetical. As no direct hypoglycemia metrics (eg, time below range, confirmed hypoglycemic episodes) were assessed in our cohort, the postulated link requires validation in dedicated prospective studies.

A significant interaction was observed between MAGE and BMI, with the positive association between MAGE and early DKD evident only in overweight individuals (BMI≥24 kg/m2). No such interaction was detected for SDBG, nor were significant interactions observed between MAGE and other covariates including age, sex, diabetes duration, hypertension and HbA1c. These findings suggest that only BMI significantly modified the association between MAGE and early DKD. However, the subgroup analysis regarding MAGE and BMI is purely exploratory. Given the limited sample size in certain subgroups, this interaction finding requires external independent cohorts for further verification. On the basis of this exploratory observation, we propose the following mechanistic hypotheses for future investigation. MAGE captures acute glucose excursions that impose direct oxidative stress and endothelial injury, a burden likely amplified in overweight individuals by pre-existing insulin resistance and chronic inflammation, thereby lowering the threshold for early renal damage. In contrast, SDBG may operate through more direct pathways, such as cumulative advanced glycation end-product formation or sustained podocyte injury, which appear less influenced by metabolic status.

Accumulating evidence links GV to oxidative stress, inflammation, endothelial dysfunction and epigenetic modifications,30,37–39 all central to the pathophysiology of DKD.40,41 Acute glucose excursions have been consistently shown to induce more pronounced oxidative stress than sustained hyperglycemia,30,37 as reflected by increased MDA and ROMs levels following hyperglycemic challenge.8 Accordingly, MAGE, a metric capturing such excursions, has been correlated with urinary 8-iso-prostaglandin F2α, a validated oxidative stress marker, whereas no such correlation was observed with HbA1c or fasting glucose.30 Similar associations have been reported for day-to-day GV indices, including MAGE and MODD, with circulating oxidative stress markers in patients with T2D.42 This enhanced oxidative stress response activates the PI3K/Akt, TGF-β1/p38-MAPK, and NF-κB pathways, driving endothelial cell apoptosis, inflammation, autophagy and fibrosis, ultimately leading to renal structural and functional injury.10 Fluctuating glucose conditions also provoke greater monocyte activation, pro-inflammatory cytokine production,43,44 and mitochondrial dysfunction, thereby impairing endothelial and renal tubular function. MAGE has further been reported to be associated with immune dysregulation, including abnormally high T helper (Th) 1 cell/Th2 cell ratio and reduced regulatory T-cell (Treg) proportions.45,46 Additionally, transient hyperglycemia may induce sustained epigenetic changes in oxidative and inflammatory genes, exerting long-term renal effects beyond immediate glycemic exposure.38,47,48 For SDBG, which reflects overall intraday glycemic dispersion, renal effects may arise from cumulative fluctuation exposure that sustains oxidative and inflammatory pathway activation. Collectively, these established biological mechanisms support the plausibility of the associations observed in our study. However, as no mechanistic biomarkers were assessed in this cohort, these interpretations remain speculative.

Current clinical guidelines recommend routine UACR screening and HbA1c monitoring for patients with T2D but do not yet incorporate GV assessment.23 Our cross-sectional findings suggest that CGM-derived short-term GV metrics (MAGE and SDBG) may have potential value as auxiliary indicators for early DKD risk stratification in patients with preserved renal function. The observed linear association between MAGE and early DKD raises the hypothesis that mitigating acute glucose fluctuations may confer renal protective benefits, while the MAGE–BMI interaction indicates that overweight patients may hypothetically derive greater renal benefit from interventions targeting glycemic excursions. Nevertheless, these findings are strictly hypothesis-generating, and their clinical utility as well as implications for clinical practice require validation in future prospective and interventional studies.

Several limitations should be acknowledged. First, the cross-sectional design cannot establish causal relationships between GV indices and early DKD. Second, only the use status of nephroprotective medications was adjusted in multivariable models. Detailed data on drug dose, treatment duration and treatment adherence were not consistently available owing to the retrospective design, and residual confounding from medication exposure therefore cannot be excluded. Third, the subgroup analysis of MAGE and BMI had insufficient statistical power, so this interaction result is exploratory. Fourth, no hypoglycemia metrics or mechanistic biomarkers were measured, so related pathophysiological interpretations remain speculative. Prospective cohort studies or interventional trials are urgently needed to determine whether reducing MAGE and SDBG can mitigate early DKD risk. Further large-sample, multicenter prospective studies incorporating comprehensive GV metrics and detailed medication exposure data are warranted to elucidate the causal relationship between GV and early DKD.

Conclusions

In this study, CGM-derived short-term GV indices, MAGE and SDBG, were independently associated with early-stage DKD in patients with T2D. The association patterns differed: MAGE exhibited a linear positive relationship, whereas SDBG showed a U-shaped association. Moreover, the MAGE–DKD association was modified by BMI and was only observed in overweight/obese individuals. Given the cross‑sectional design and limited sample size of subgroup analyses, these findings require external validation in larger independent cohorts. Further prospective and interventional studies are warranted to clarify whether targeting short-term GV confers incremental renal protective benefits.

Abbreviations

GV, glycemic variability; DKD, diabetic kidney disease; CGM, continuous glucose monitoring; T2D, type 2 diabetes; UACR, urine albumin-to-creatinine ratio; eGFR, estimated glomerular filtration rate; TIR, time in range; SDBG, glucose standard deviation; MBG, mean glucose; MAGE, mean amplitude of glycemic excursions; GAM, generalized additive model; ESRD, end-stage kidney disease; HbA1c, glycated hemoglobin A1c; SMBG, self-monitoring of blood glucose; CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration; NSAIDs, anti-inflammatory drugs; KDIGO, Kidney Disease: Improving Global Outcomes; BMI, Body mass index; BP, Blood pressure; TG, triglycerides; TC, total cholesterol; HDL-C, high-density lipoprotein cholesterol, LDL-C, low-density lipoprotein cholesterol; Scr, serum creatinine; BUN, blood urea nitrogen; UA, uric acid; FPG, fasting plasma glucose; 2hPPG, 2-hour postprandial plasma glucose; FCP, fasting C-peptide; 2hPCP, 2-hour postprandial C-peptide; DCCT, Diabetes Control and Complications Trial; ACCORD, Action to Control Cardiovascular Risk in Diabetes; CKD, chronic kidney disease.

Data Sharing Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Ethics Approval

This study was conducted and approved by the Ethics Committee of The Second Affiliated Hospital of Anhui Medical University (approval No. YX2025-104) in accordance with the principles of the Declaration of Helsinki. This study does not disclose any personal privacy of the participants and does not violate data protection laws.

Acknowledgments

We would like to thank all the participants and the research team involved in the present project.

Author Contributions

Nan Chen: Formal analysis, Funding acquisition, Visualization, Writing-original-draft, Writing-review and editing. Nuojin Wang: Data curation, Investigation, Writing-original-draft. Hao Duan: Data curation, Investigation, Writing-original-draft. Tianrong Pan: Project administration,Supervision,Conceptualization,Writing-original-draft. Limin Zhai: Conceptualization, Investigation, Data curation, Writing original draft, Writing-review and editing. All authors gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This study was supported by the Clinical and Translational Research Project of Anhui Province (202427b10020026).

Disclosure

The authors declare there no conflict of interest.

References

1. Afkarian M, Zelnick LR, Hall YN, et al. Clinical manifestations of kidney disease among US adults with diabetes, 1988-2014. JAMA. 2016;316:602–12. doi:10.1001/jama.2016.10924

2. Habte-Asres HH, Murrells T, Nitsch D, Wheeler DC, Forbes A. Glycaemic variability and progression of chronic kidney disease in people with diabetes and comorbid kidney disease: retrospective cohort study. Diabet Res Clin Pract. 2022;193:110117. doi:10.1016/j.diabres.2022.110117

3. Association AD. Standards of medical care in diabetes—2011. Diabetes Care. 2011;34:S11–S61. doi:10.2337/dc11-S011

4. Kilpatrick ES, Rigby AS, Atkin SL. Effect of glucose variability on the long-term risk of microvascular complications in type 1 diabetes. Diabetes Care. 2009;32:1901–1903. doi:10.2337/dc09-0109

5. Hirsch IB. Glycemic variability and diabetes complications: does it matter? Of course it does! Diabetes Care. 2015;38:1610–1614. doi:10.2337/dc14-2898

6. Kilpatrick ES, Rigby AS, Atkin SL. A1C variability and the risk of microvascular complications in type 1 diabetes: data from the diabetes control and complications trial. Diabetes Care. 2008;31:2198–2202. doi:10.2337/dc08-0864

7. Zhang T, Wang X, Zhang Y, et al. Establishment of a potent weighted risk model for determining the progression of diabetic kidney disease. J Transl Med. 2023;21:381. doi:10.1186/s12967-023-04245-w

8. Adeshara K, Di Marco E, Bordino M, et al; FinnDiane Study Group. Altered oxidant and antioxidant levels are associated with vascular stiffness and diabetic kidney disease in type 1 diabetes after exposure to acute and chronic hyperglycemia. Cardiovasc Diabetol. 2024;23:350. doi:10.1186/s12933-024-02427-4

9. Klimontov VV, Saik OV, Korbut AI. Glucose variability: how does it work? Int J Mol Sci. 2021;22:7783. doi:10.3390/ijms22157783

10. Ma X, Ma J, Leng T, et al. Advances in oxidative stress in pathogenesis of diabetic kidney disease and efficacy of TCM intervention. Ren Fail. 2023;45:2146512. doi:10.1080/0886022X.2022.2146512

11. Ceriello A, Monnier L, Owens D. Glycaemic variability in diabetes: clinical and therapeutic implications. Lancet Diabetes Endocrinol. 2019;7:221–230. doi:10.1016/s2213-8587(18)30136-0

12. 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:10.2337/dc15-1188

13. Okuno T, Vansomphone A, Zhang E, et al. Association of both short-term and long-term glycemic variability with the development of microalbuminuria in the ACCORD trial. Diabetes. 2023;72:1864–1869. doi:10.2337/db23-0374

14. Yan Y, Kondo N, Oniki K, et al. Predictive ability of visit-to-visit variability of HbA1c measurements for the development of diabetic kidney disease: a retrospective longitudinal observational study. J Diabetes Res. 2022;2022:6934188. doi:10.1155/2022/6934188

15. Hsu CC, Chang HY, Huang MC, et al. HbA1c variability is associated with microalbuminuria development in type 2 diabetes: a 7-year prospective cohort study. Diabetologia. 2012;55:3163–3172. doi:10.1007/s00125-012-2700-4

16. Rodbard D. Continuous glucose monitoring: a review of recent studies demonstrating improved glycemic outcomes. Diabetes Technol Ther. 2017;19:S25–s37. doi:10.1089/dia.2017.0035

17. Subramanian S, Hirsch IB. Diabetic kidney disease: is there a role for glycemic variability? Curr Diab Rep. 2018;18:13. doi:10.1007/s11892-018-0979-3

18. Kilpatrick ES, Rigby AS, Atkin SL. The effect of glucose variability on the risk of microvascular complications in type 1 diabetes. Diabetes Care. 2006;29:1486–1490. doi:10.2337/dc06-0293

19. Lachin JM, Bebu I, Bergenstal RM, et al. Association of glycemic variability in type 1 diabetes with progression of microvascular outcomes in the diabetes control and complications trial. Diabetes Care. 2017;40:777–783. doi:10.2337/dc16-2426

20. Jin SM, Kim TH, Oh S, et al. Association between the extent of urinary albumin excretion and glycaemic variability indices measured by continuous glucose monitoring. Diabet Med. 2015;32:274–279. doi:10.1111/dme.12607

21. Zhou J, Jia WP, Ma XJ, et al. [Relationship between blood glucose variability and microalbuminuria in type 2 diabetic patients with well-controlled glycosylated hemoglobin]. Zhonghua Yi Xue Za Zhi. 2008;88:2977–2981. [Chinese].

22. Wakasugi S, Mita T, Katakami N, et al. Associations between continuous glucose monitoring-derived metrics and diabetic retinopathy and albuminuria in patients with type 2 diabetes. BMJ Open Diabetes Res Care. 2021:9. doi:10.1136/bmjdrc-2020-001923

23. de Boer IH, Khunti K, Sadusky T, et al. Diabetes management in chronic kidney disease: a consensus report by the American Diabetes Association (ADA) and Kidney Disease: Improving Global Outcomes (KDIGO). Diabetes Care. 2022;45:3075–3090. doi:10.2337/dci22-0027

24. Kidney Disease: Improving Global Outcomes CKDWG. KDIGO 2024 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int. 2024;105:S117–s314. doi:10.1016/j.kint.2023.10.018

25. Sheng Y, Cheng J, Zhang C, et al. Establishment and comparison of prediction models for early-stage diabetic kidney disease. Digit Health. 2025;11:20552076251355448. doi:10.1177/20552076251355448

26. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150:604–612. doi:10.7326/0003-4819-150-9-200905050-00006

27. Zhou BF. Predictive values of body mass index and waist circumference for risk factors of certain related diseases in Chinese adults--study on optimal cut-off points of body mass index and waist circumference in Chinese adults. Biomed Environ Sci. 2002;15:83–96.

28. Danne T, Nimri R, Battelino T, et al. International consensus on use of continuous glucose monitoring. Diabetes Care. 2017;40:1631–1640. doi:10.2337/dc17-1600

29. Yoo JH, Choi MS, Ahn J, et al. Association between continuous glucose monitoring-derived time in range, other core metrics, and albuminuria in type 2 diabetes. Diabetes Technol Ther. 2020;22:768–776. doi:10.1089/dia.2019.0499

30. Monnier L, Mas E, Ginet C, et al. Activation of oxidative stress by acute glucose fluctuations compared with sustained chronic hyperglycemia in patients with type 2 diabetes. JAMA. 2006;295:1681–1687. doi:10.1001/jama.295.14.1681

31. Al-Hamodi Z, Ismail IS, Saif-Ali R, Ahmed KA, Muniandy S. Association of plasminogen activator inhibitor-1 and tissue plasminogen activator with type 2 diabetes and metabolic syndrome in Malaysian subjects. Cardiovasc Diabetol. 2011;10:23. doi:10.1186/1475-2840-10-23

32. Lu J, Ma X, Zhang L, et al. Glycemic variability assessed by continuous glucose monitoring and the risk of diabetic retinopathy in latent autoimmune diabetes of the adult and type 2 diabetes. J Diabetes Investig. 2019;10:753–759. doi:10.1111/jdi.12957

33. Hu YM, Zhao LH, Zhang XL, et al. Association of glycaemic variability evaluated by continuous glucose monitoring with diabetic peripheral neuropathy in type 2 diabetic patients. Endocrine. 2018;60:292–300. doi:10.1007/s12020-018-1546-z

34. Bragd J, Adamson U, Bäcklund LB, Lins PE, Moberg E, Oskarsson P. Can glycaemic variability, as calculated from blood glucose self-monitoring, predict the development of complications in type 1 diabetes over a decade? Diabetes Metab. 2008;34:612–616. doi:10.1016/j.diabet.2008.04.005

35. Monnier L, Colette C, Owens DR. Glycemic variability: the third component of the dysglycemia in diabetes. Is it important? How to measure it? J Diabetes Sci Technol. 2008;2:1094–1100. doi:10.1177/193229680800200618

36. Sheng X, Li T, Hu Y, Xiong CS, Hu L. Correlation between blood glucose indexes generated by the flash glucose monitoring system and diabetic vascular complications. Diabet Metabol Syndrom Obes Targets Ther. 2023;16:2447–2456. doi:10.2147/DMSO.S418224

37. Ceriello A, Esposito K, Piconi L, et al. Oscillating glucose is more deleterious to endothelial function and oxidative stress than mean glucose in normal and type 2 diabetic patients. Diabetes. 2008;57:1349–1354. doi:10.2337/db08-0063

38. Costantino S, Paneni F, Battista R, et al. Impact of glycemic variability on chromatin remodeling, oxidative stress, and endothelial dysfunction in patients with type 2 diabetes and with target HbA(1c) levels. Diabetes. 2017;66:2472–2482. doi:10.2337/db17-0294

39. Mo Y, Lu J, Zhou J. Glycemic variability: measurement, target, impact on complications of diabetes and does it really matter? J Diabetes Investig. 2024;15:5–14. doi:10.1111/jdi.14112

40. Forbes JM, Coughlan MT, Cooper ME. Oxidative stress as a major culprit in kidney disease in diabetes. Diabetes. 2008;57:1446–1454. doi:10.2337/db08-0057

41. Sani S. Diabetic nephropathy: a short review. J Med Res. 2020;15(2):45–52.

42. Ohara M, Fukui T, Ouchi M, et al. Relationship between daily and day-to-day glycemic variability and increased oxidative stress in type 2 diabetes. Diabet Res Clin Pract. 2016;122:62–70. doi:10.1016/j.diabres.2016.09.025

43. Gu QW, Sun Q, Wang J, Gu WS, Wang W, Mao XM. Effects of glycemic variability on regulatory T cells in patients with type 2 diabetes and kidney disease. Diabetes Metab Syndr Obes. 2023;8:2365–2375. doi:10.2147/DMSO.S413407

44. Li-Bo Y, Wen-Bo Q, Xiao-Hong L, You-Lun F, Tie Z. Intermittent high glucose promotes expression of proinflammatory cytokines in monocytes. Inflamm Res. 2011;60:367–370. doi:10.1007/s00011-010-0279-0

45. Sun Q, Yang P, Gu QW, et al. Increased glycemic variability results in abnormal differentiation of T cell subpopulation in type 2 diabetes patients. J Diabetes Complications. 2024;38:108738. doi:10.1016/j.jdiacomp.2024.108738

46. Zhang C, Xiao C, Wang P, et al. The alteration of Th1/Th2/Th17/Treg paradigm in patients with type 2 diabetes mellitus: relationship with diabetic nephropathy. Hum Immunol. 2014;75:289–296. doi:10.1016/j.humimm.2014.02.007

47. El-Osta A, Brasacchio D, Yao D, et al. Transient high glucose causes persistent epigenetic changes and altered gene expression during subsequent normoglycemia. J Exp Med. 2008;205:2409–2417. doi:10.1084/jem.20081188

48. Yunlei D, Qiuling F, Xu W, et al. Transient high-glucose stimulation induces persistent inflammatory factor secretion from rat glomerular mesangial cells via an epigenetic mechanism. Cell Physiol Biochem. 2018;49:1747–1754. doi:10.1159/000493619