Correlation Between Estimated Glomerular Filtration Rate and Asymptomatic Left Ventricular Diastolic Dysfunction in Patients With Type 2 Diabetes Mellitus

Abstract

Aim

To explore the correlation between estimated glomerular filtration rate (eGFR) and asymptomatic left ventricular diastolic dysfunction (ALVDD) in patients with type 2 diabetes mellitus (T2DM).

Methods

We performed a retrospective study involving 47 ALVDD and 30 non-ALVDD T2DM patients between October 2023 and March 2024. Patient characteristics, laboratory tests, and echocardiographic measurements were recorded. The relationship between eGFR and ALVDD was examined by Spearman correlation and logistic regression analyses. Receiver operating characteristic curve analysis was conducted to test the diagnostic value of eGFR for ALVDD.

Results

Patients with ALVDD had lower eGFR but higher echocardiographic measurements than patients without ALVDD (P < 0.05). The eGFR was negatively correlated with echocardiographic measurements (P < 0.05) and was an independent risk factor for ALVDD. Receiver operating characteristic curve analysis demonstrated that eGFR had an area under the ROC curve of 0.766, with a sensitivity of 63.8%, a specificity of 96.7%, and a threshold of 94.02 mL/min/1.73 m²

Conclusions

The eGFR was closely correlated with ALVDD and could be applied to estimate the risk of ALVDD in T2DM patients. We recommend the eGFR be considered when evaluating cardiac diastolic dysfunction in T2DM patients at an early stage and thereby guide prompt clinical intervention.

Highlights

• •Type 2 diabetes mellitus (T2DM) patients with asymptomatic left ventricular diastolic dysfunction (ALVDD) exhibited a significantly lower estimated glomerular filtration rate (eGFR) compared to those with normal diastolic function
• •eGFR was negatively correlated with key echocardiographic parameters of cardiac structure (left atrial end-systolic diameter, left ventricular diastolic diameter, left ventricular posterior wall thickness, interventricular septum thickness, ascending aortic diameter, and left ventricular mass index) and identified as an independent risk factor for ALVDD
• •Receiver operating characteristic analysis demonstrated that eGFR (<94.02 mL/min/1.73 m²) has high specificity (96.7%) for identifying ALVDD in T2DM patients
• •This suggests eGFR, a simple and widely available test, could serve as a valuable initial screening tool for early detection of diabetic cardiomyopathy

Clinical Relevance

Early identification of asymptomatic left ventricular diastolic dysfunction (ALVDD) is crucial to prevent heart failure in type 2 diabetes mellitus. This study proposes that a readily available estimated glomerular filtration rate value (<94.02 mL/min/1.73 m²) can effectively screen for ALVDD, enabling timely intervention and improved patient management in primary care settings.

Introduction

Type 2 diabetes mellitus (T2DM) is a global public health concern, and its incidence has increased significantly with lifestyle changes. Prolonged elevation in blood glucose levels can lead to various complications, including macrovascular and microvascular diseases, which can greatly impact quality of life and prognosis in affected patients. Diabetic cardiomyopathy (DCM) is a specific pathophysiological condition that develops in diabetic patients. In the early stage, DCM presents as left ventricular diastolic dysfunction and can then progress to systolic dysfunction and heart failure in its later stage.¹ Early detection of DCM is challenging and easily missed in the outpatient clinic. Clinical symptoms can be atypical in early DCM and are often manifested as asymptomatic left ventricular diastolic dysfunction (ALVDD).² Therefore, it is imperative to apply appropriate early and accurate screening examinations to identify T2DM patients with ALVDD for prompt management. The most common test adopted to diagnose ALVDD currently is the cardiac ultrasound examination (echocardiogram);³ however, echocardiography is relatively expensive and is not routinely performed in asymptomatic patients. There is no readily available and cost-effective screening test for the early detection of ALVDD in T2DM patients. However, the estimated glomerular filtration rate (eGFR) can be easily calculated from the CKD-EPI_scr formula which is based on serum creatinine from Epidemiology Collaboration Formula for chronic kidney disease and is widely available to patient.⁴

We herein performed a prospective study in which we analyzed the correlation between eGFR and ALVDD in T2DM patients. Our objective was to evaluate the diagnostic value of eGFR for ALVDD and to provide a reference for the early and accurate clinical detection of ALVDD for personalized evidence-based management of affected patients.

Materials and Methods

Study Design and Participant Selection

We conducted a prospective study in patients with T2DM who were admitted to the Department of Endocrinology at the Wuhu Second People's Hospital in China between October 2023 and March 2024. Our study protocol was approved by the hospital’s Ethics Committee (Department of Ethics Committee, Wuhu Hospital Affiliated to East China Normal University, 2024-KY-199), and written informed consent was obtained from every study participant.

Patient inclusion criteria were as follows: 1) age ≥18 years; 2) diagnosis of T2DM based on the T2DM diagnostic criteria from the World Health Organization;⁵ 3) glycated hemoglobin A1c (HbA1c) ≥7.5%; and 4) agreement for echocardiographic examination with normal ventricular ejection fraction (EF) results based on the recommendations of the Evaluation of Left Ventricular Diastolic Function by Echocardiography jointly issued by the American Society of Echocardiography and the European Association of Cardiovascular Imaging.³ Exclusion criteria included evidence of the following: 1) a history of hypertension and an average blood pressure of ≥120/80 mmHg for the 3 times after admission; 2) a history of hyperlipidemia and low-density lipoprotein ≥1.7 mmol/L after admission; 3) cardiovascular diseases such as heart failure (echocardiography-confirmed EF <50%), acute myocardial infarction, unstable angina, arrhythmia, or valvular heart disease; 4) serious liver and kidney diseases (transaminase is 3 times or more than the standard value, eGFR <60 mL/min/1.73 m²); 5) concurrent tumor; 6) concurrent acute or chronic infection; 7) diabetic ketoacidosis and hyperosmolar hyperglycemic state; 8) pregnancy or breastfeeding; 9) mental illness and inability to cooperate with the research study due to communicative disorders or cognitive impairment; 10) thyroid disease (hyperthyroidism or hypothyroidism); 11) perioperation; 12) concurrent participation in other clinical research.

Study Protocol

Clinical information, including demographics (age and sex), height, weight, and T2DM duration, was collected. Laboratory tests, such as serum creatinine (by automatic biochemical analyzer 7600, Hitachi) and HbA1c (by high-performance liquid chromatography) levels, were conducted on peripheral blood samples collected after greater than 8 hours of fasting.

All echocardiographic examinations were performed by the same technician. Measurements of the left atrial end-systolic diameter (LAD), left ventricular diastolic diameter (LVDd), EF, left ventricular posterior wall thickness (LVPWd), interventricular septum thickness (IVSd), and ascending aortic diameter (AAO) were recorded. The left ventricular mass index (LVMI) was calculated based on height, weight, and echocardiographic results:^6,7

$$\text{Left}\text{ventricular}\text{mass}(\text{LVM},\text{in}\mathrm{g})=0.8\times (1.04\times {\left[\text{LVDd}+\text{LVPWd}+\text{IVSd}\right]}^3–{\left[\text{LVDd}\right]}^3)+0.6\mathrm{g},\text{body}\text{surface}\text{area}(\text{BSA},\text{in}{\mathrm{m}}^2)=0.0061\times \text{height}+0.0124\times \text{weight}-0.0099\text{,}$$

$$\text{LVMI}(\text{in}\mathrm{g}/{\mathrm{m}}^2)=\text{LVM}/\text{BSA.}$$

The eGFR was calculated based on the CKD-EPI_scr formula:⁴

$$\text{eGFR}(\text{in}\text{mL}/\min /1.73{\mathrm{m}}^2)=141\times \min (\text{Scr}/\mathrm{\kappa },1)\mathrm{\alpha }\times \max (\text{Scr}/\mathrm{\kappa },1)-1.209\times 0.993\text{age}\times (1.018\text{for}\text{female},1\text{for}\text{male})\text{,}$$

$$\mathrm{\kappa }=0.7\left(\text{female}\right)\text{or}0.9\left(\text{male}\right)$$

$$\mathrm{\alpha }=-0.329\left(\text{female}\right)\text{or}-0.411\left(\text{male}\right)$$

where min indicates the smaller number between Scr/κ and 1, and max indicates the larger number between Scr/κ and 1.

Left ventricular diastolic dysfunction was determined based on recommendations from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.³ Patients were then assigned to an ALVDD group or a normal left ventricular diastolic function (non-ALVDD) group.

Statistical Analysis

Due to the sample size limit (n = 77), subgroup analysis was not conducted. All statistical analyses were performed using SPSS version 24.0 (IBM). The normally distributed continuous data are depicted as mean ± SD, and intergroup differences were compared by t test. Categorical data are presented as numbers with percentages (%). Spearman correlation analysis was performed to evaluate the relationship between eGFR and echocardiographic measurements. Binary logistic regression analysis was exploited to explore the variables associated with ALVDD in T2DM patients. Combined receiver operating characteristic (ROC) curve analysis was applied to assess the diagnostic value of eGFR on ALVDD in T2DM patients. Differences were considered to be statistically significant when P was <0.05.

Results

Baseline Characteristic Comparisons Between the 2 Groups

A total of 77 patients were included, with 47 and 30 in the ALVDD and non-ALVDD groups, respectively. There were no statistically significant intergroup differences in age, weight, T2DM duration, EF, or serum creatinine and HbA1c levels between the 2 groups (all P > 0.05). Patients in the ALVDD group exhibited a lower eGFR but higher LAD, LVDd, LVPWd, IVSd, AAO, and LVMI than the patients in the non-ALVDD group (all P < 0.05) (Table 1).

Table 1.

Baseline Characteristic Comparisons Between ALVDD and Non-ALVDD Groups

Characteristics	ALVDD group (n = 47)	Non-ALVDD group (n = 30)	t	P
Age, y, mean ± SD	54.4 ± 1.3	50.0 ± 2.0	1.896	0.062
Sex, male/female, n (%)	34 (72.3%)/13 (27.7%)	17 (56.7%)/13 (43.3)
Weight, kg, mean ± SD	64.2 ± 1.6	64.4 ± 2.0	0.069	0.945
T2DM duration, y, mean ± SD	4.4 ± 0.3	4.8 ± 0.4	0.743	0.459
Scr, μmol/L, mean ± SD	64.1 ± 2.1	70.9 ± 3.2	1.842	0.697
HbA1c, %, mean ± SD	7.8 ± 0.1	7.8 ± 0.2	0.191	0.849
eGFR, mL/min/1.73 m2, mean ± SD	87.5 ± 1.4	111.2 ± 0.7	12.410	<0.001a
LAD, mm, mean ± SD	33.8 ± 0.4	29.9 ± 0.8	4.761	<0.001a
LVDd, mm, mean ± SD	45.4 ± 0.4	41.8 ± 0.6	5.018	<0.001a
EF, %, mean ± SD	63.0 ± 0.4	64.2 ± 0.4	1.951	0.054
LVPWd, mm, mean ± SD	8.3 ± 0.1	7.4 ± 0.1	5.951	<0.001a
IVSd, mm, mean ± SD	8.6 ± 0.1	7.7 ± 0.1	5.293	<0.001a
AAO, mm, mean ± SD	30.3 ± 0.4	26.6 ± 0.4	6.122	<0.001a
LVMI, g/m2, mean ± SD	52.7 ± 1.9	49.2 ± 1.4	7.779	<0.001a

Abbreviations: AAO = ascending aortic diameter; ALVDD = asymptomatic left ventricular diastolic dysfunction; EF = ejection fraction; eGFR = estimated glomerular filtration rate; HbA1c = glycosylated hemoglobin; IVSd = interventricular septum thickness; LAD = left atrial end-systolic diameter; LVDd = left ventricular diastolic diameter; LVMI = left ventricular mass index; LVPWd = left ventricular posterior wall thickness; Scr = serum creatinine; T2DM = type 2 diabetes mellitus.

^aP < 0.05.

Correlation Between eGFR and Echocardiographic Measurements

Spearman correlation analysis showed that eGFR was negatively correlated with echocardiographic measurements, including LAD, LVDd, LVPWd, IVSd, AAO, and LVMI (all P < 0.05) (Table 2).

Table 2.

Spearman Correlation Analysis Between eGFR and Echocardiographic Measurement

Measurement	r	P
LAD	−0.340	0.002a
LVDd	−0.248	0.029a
LVPWd	−0.377	0.001
IVSd	−0.319	0.005a
AAO	−0.283	0.013
LVMI	−0.425	<0.001

Abbreviations: AAO = ascending aortic diameter; eGFR = estimated glomerular filtration rate; IVSd = interventricular septum thickness; LAD = left atrial end-systolic diameter; LVDd = left ventricular diastolic diameter; LVMI = left ventricular mass index; LVPWd = left ventricular posterior wall thickness.

^aP < 0.05.

Variables Associated With ALVDD in T2DM Patients

Binary logistic regression analysis was performed with ALVDD as a dependent variable and age, weight, T2DM duration, creatinine, HbA1c, and eGFR as independent variables. The results revealed that eGFR was an independent variable associated with ALVDD in T2DM patients (Table 3).

Table 3.

Logistic Regression Analysis of Variables Associated With ALVDD in T2DM Patients

Variable	β	SE	Wals	OR	P	95% CI
Age	0.705	0.027	4.185	0.980	0.540	0.125–1.019
Weight	0.005	0.022	0.057	1.005	0.812	0.964–1.048
Disease duration	0.039	0.100	0.150	0.962	0.698	0.792–1.170
Serum creatinine	−0.029	0.015	3.549	0.972	0.060	0.943–1.001
HbA1c	0.047	0.309	0.023	0.954	0.878	0.521–1.746
eGFR	0.077	0.021	11.914	1.074	0.001a	1.031–1.118

Abbreviations: ALVDD = asymptomatic left ventricular diastolic dysfunction; eGFR = estimated glomerular filtration rate; HbA1c = glycosylated hemoglobin; OR = odds ratio; T2DM = type 2 diabetes mellitus.

^aP < 0.05.

Diagnostic Value of eGFR for ALVDD in T2DM Patients

ROC curve analysis showed that the area under the ROC curve for eGFR was 0.766, with a sensitivity of 63.8%, a specificity of 96.7%, and a threshold of 94.02 mL/min/1.73 m² (Fig.).

Fig.

ROC curve analysis shows the diagnostic value of eGFR on ALVDD in T2DM patients. eGFR = estimated glomerular filtration rate; ALVDD = asymptomatic left ventricular diastolic dysfunction; T2DM = type 2 diabetes mellitus; ROC = receiver operating characteristic.

Discussion

DCM is a common complication in T2DM patients. In 1972, Rubler et al first described DCM after performing autopsies on 4 diabetic patients with congestive heart failure but without coronary atherosclerosis, valvular heart disease, hypertension, congenital heart disease, or alcoholic heart disease.⁸ The pathogenesis of DCM is complex, including an inflammatory response, oxidative stress,⁹ mitochondrial dysfunction,¹⁰ apoptosis and autophagy,¹¹ and myocardial fibrosis.¹² These mechanisms interact with each other, leading to abnormal myocardial structure and function. In the early stage of DCM, patients can manifest increased LVM and diastolic dysfunction with a normal EF and no clinical symptoms (ALVDD).¹³ As the disease progresses, LVM increases further, with ventricular wall thickening, worsening diastolic dysfunction, and systolic dysfunction leading to heart failure.¹⁴ The prevalence of DCM was 1.1%,¹⁵ whereas the prevalence of ALVDD was 50% to 70% in T2DM patients.^16,17 In spite of the high prevalence of ALVDD in T2DM patients, it can easily be missed by physicians and patients due to its atypical and often subtle clinical presentation. Additionally, current diagnostic tests possess inadequate sensitivity and cannot detect subtle changes in cardiac function in patients with ALVDD. These factors then lead to a relatively low rate of diagnosis of ALVDD, and the prognosis of DCM patients is poor, with a 31% prevalence of heart failure.¹⁵ Therefore, early diagnosis and intervention are critical for T2DM patients with ALVDD.

The eGFR value is a key measurement for evaluating renal function. There are many different methods used to calculate eGFR, including the modification of diet in renal disease (MDRD) formula,¹⁸ MDRD Chinese formula,¹⁹ CKD-EPIScr formula,⁴ CKD-EPICysC formula, and CKD-EPICr-Cysc formula.²⁰ When evaluating renal function in our study, we did not measure the CysC level. Compared with the MDRD formula, the CKD-EPIScr formula is more widely applied with less variability in results in the clinical setting. Therefore, we chose the CKD-EPIScr formula to calculate eGFR in our current study. A previous study showed that decreased eGFR was associated with an increased 10-year risk of atherosclerotic cardiovascular disease in patients with T2DM,²¹ and our study revealed that T2DM patients with ALVDD had a significantly lower eGFR than T2DM patients without ALVDD, indicating a potential association between decreased eGFR and ALVDD in T2DM patients. In addition, we also demonstrated that echocardiographic measurements—including LAD, LVDd, LVPWd, IVSd, and AAO—were significantly higher in T2DM patients with ALVDD relative to those individuals without ALVDD; this was also consistent with cardiac structural changes such as ventricular hypertrophy and myocardial fibrosis in patients with DCM.²² LVMI is calculated based on echocardiographic measurements and BSA, which is significantly correlated with left ventricular hypertrophy.²³ An LVMI >49.2 g/m² in males or >46.7 g/m² in females supports the diagnosis of left ventricular hypertrophy. Zhang et al confirmed that left ventricular hypertrophy induced ALVDD in T2DM patients,²⁴ which was congruent with our findings that demonstrated that LVMI was statistically significantly increased in the T2DM patients with ALVDD compared to those without ALVDD.

In the current study, we also ascertained that eGFR declined with increased measurements in LAD, LVDd, LVPWd, IVSd, AAO, and LVMI. A drop in eGFR indicates impaired renal function, and an increase in LAD, LVDd, LVPWd, IVSd, AAO, and LVMI is associated with cardiac structural changes. Researchers previously explored the relationship between eGFR decline and cardiac structural changes and found that LVM increased by 1.54 g for every 10 mL/min/1.73 m² decrease in eGFR.²⁵ In our study, we included BSA and calculated LVMI. Compared with LVM, LVMI entails individual differences and achieves standardized assessment of LVM. Hyperglycemia can cause damage to multiple systems in the body, especially the renal and cardiovascular systems. The metabolic disorder caused by hyperglycemia not only affects kidney filtration function but also impacts the energy supply to the myocardium—promoting inflammation, fibrosis, and oxidative stress. This metabolic disorder is the common pathogenic pathway for both the eGFR diminution and left ventricular hypertrophy. In our study, correlation analysis suggested that the eGFR decrease may have caused increased LAD, LVDd, LVPWd, IVSd, AAO, and LVMI, contributing to the development of ALVDD in T2DM patients. Further binary logistic regression analysis showed that eGFR was associated with ALVDD in T2DM patients, suggesting that a decreased eGFR was a risk factor for ALVDD in T2DM patients. A decreased eGFR can reduce renal perfusion, minimizing blood pressure and blood flow in the renal afferent arterioles, and lead to an activation of the renin-angiotensin-aldosterone system. Previous studies have confirmed the close association between an activated renin-angiotensin-aldosterone system and myocardial hypertrophy and fibrosis in diabetic patients.^26,27 In addition, decreased eGFR can lead to fluid retention and left ventricular volume overload, causing impaired left ventricular diastolic function and abnormal myocardial energy metabolism—particularly in diabetic patients.²⁸ These data suggest that a decreased eGFR might be involved in ALVDD development in T2DM patients. Our ROC curve analysis showed that eGFR generated a satisfactory diagnostic value for ALVDD in T2DM patients, suggesting that eGFR can be applied to predict the occurrence and development of ALVDD in T2DM patients. Although American Diabetes Association guidelines recommend pro-BNP as a screening tool for asymptomatic heart failure,²⁹ our study demonstrated that eGFR <94.02 mL/min/1.73 m² exhibited a specificity of 96.7% in diagnosing ALVDD. This indicates that it could serve as a cost-effective and efficient primary screening indicator, suitable for use in primary care settings for the screening and stratification of high-risk populations. For patients with an eGFR <60 mL/min/1.73 m², supplementary pro-BNP testing was performed to improve diagnostic sensitivity. Although the ALVDD group demonstrated a higher mean age (54.4 ± 1.3 years vs 50.0 ± 2.0 years, P = 0.062), multivariate regression analysis indicated that age was not an independent predictor of ALVDD (odds ratio = 0.980, 95% CI: 0.125–1.019, P = 0.54). This implies that renal dysfunction (eGFR decline) may exert a stronger influence on cardiac function than age in T2DM. These findings align with the observations reported by Manjunath et al.³⁰ Nonetheless, age remains a crucial covariate influencing the decline in eGFR. Additional studies with larger sample sizes are necessary to confirm the interaction between age and eGFR.

The limitations of our study included the fact that it involved a single center with a small sample size. We also did not perform a subgroup analysis in patients stratified by sex. Future multicenter studies with a large sample size are still required to confirm our findings.

In conclusion, eGFR may play a role in the development of ALVDD in T2DM patients. eGFR is a simple and widely available test and can be utilized for the early screening of ALVDD, expediting prompt intervention for the prevention of heart failure in T2DM patients.

Disclosure

The authors have no conflicts of interest to disclose.