FormalPara Key Summary Points

Why carry out this study?

Diabetic kidney disease has become the leading cause of end-stage kidney disease in most countries

The prevalence of normoalbuminuric diabetic kidney disease has substantially increased over the past few decades, but the clinical characteristics remains under-characterized

What was learned from the study?

The prevalence of DKD in hospitalized patients with DM was 39.9%, normoalbuminuric and reduced eGFR, DKD with albuminuria but without reduced eGFR, and DKD with albuminuria and reduced eGFR groups accounted for 3.3%, 23.7%, and 12.9%, respectively

The clinical signs and symptoms of normoalbuminuria is different from other phenotypes of diabetic kidney disease

lncRNA XIST has potential as a biomarker to predict the progression of diabetic kidney disease

Introduction

Diabetic kidney disease (DKD) is one of the main complications of diabetes mellitus (DM). In recent years, the incidence of end-stage renal disease (ESRD) caused by diabetes has increased dramatically [1]. Traditionally, DKD begins with a progressive increase in albuminuria followed by a steady decrease in estimated glomerular filtration rate (eGFR) [2]. Recent research, however, suggested that a portion of patients with DM with normoalbuminuria have progressive renal insufficiency, referred to as normoalbuminuric diabetic kidney disease (NADKD) [3, 4]. The prevalence of DKD with this feature has substantially increased over the past few decades, according to several sizable, serial cross-sectional studies [5, 6]. Based on these, NADKD differs from albuminuric renal dysfunction in terms of its clinical features. The majority of patients with NADKD were women, and their conditions were frequently characterized by obesity, lipidemia, higher plasma glucose, and a reduced frequency of polyneuropathy [2, 7]. However, there is an apparent dearth of research on Asians, specifically Chinese individuals, regarding the clinical characteristics of normoalbuminuric DKD. Therefore, it is imperative to investigate the clinical traits associated with normoalbuminuric DKD, discover new molecular markers linked to the progression of DKD, and elucidate the role of these markers in the development of DKD.

Numerous studies have confirmed that long non-coding RNAs (lncRNAs) could regulate cell functions through a variety of pathways [8]. Among them, lncRNA XIST, a major regulator of X chromosome inactivation, has also been extensively found to play an important role in cell differentiation, proliferation, and metabolism [9,10,11]. In recent years, the study of lncRNA XIST in the kidney has revealed that it can regulate renal interstitial fibrosis in DKD and play a role as a biomarker in AKI treatment strategy [12, 13]. It has also been shown that lncRNA XIST is associated with the macrophage phenotype and thus may be involved in the DKD process [14,15,16], but its significance in patients with DKD remains to be clarified.

Therefore, in this study, 1184 hospitalized patients with DM were retrospectively analyzed to determine the prevalence and clinical characteristics of DKD, especially NADKD. In addition, we explored the relationship between lncRNA XIST expression of PBMCs in patients with DKD and the progression of DKD to provide novel biomarkers with the potential to indicate the progression of DKD.

Methods

Study Participants

This was a retrospective cross-sectional study enrolling the patients who were diagnosed with T1DM or T2DM between January 2021 and December 2021 in Nanfang Hospital, Guangdong Province, China. We have developed reasonable exclusion criteria (Table S1). In the end, data from 1184 participants were included in the study. The algorithm for the selected participants is shown in Fig. S1. The cases were further divided into four groups based on eGFR and UACR as follows: nDKD: normal control group (UACR < 3 mg/mmol and eGFR ≥ 60 ml/min/1.73 m2), NA-DKD: DKD with normoalbuminuric and reduced eGFR (UACR < 3 mg/mmol and eGFR < 60 ml/min/1.73 m2), A-DKD: DKD with albuminuria but without reduced eGFR (UACR ≥ 3 mg/mmol and eGFR ≥ 60 ml/min/1.73 m2), and Mixed: DKD with albuminuria and reduced eGFR (UACR ≥ 3 mg/mmol and eGFR < 60 ml/min/1.73 m2). Peripheral blood was collected from outpatients who met the inclusion criteria.

Approval of the research protocol: The protocol for this research project was approved by a suitably constituted Ethics Committee of the institution and conforms to the provisions of the Declaration of Helsinki and the Committee of Medical Ethics of Nanfang Hospital, approval no. NFEC-2021–049. Informed consent was obtained from all subjects to further follow-up.

Diagnostic Criteria

DKD was defined as previously described [17], as the presence of albuminuria and/or reduced eGFR in the absence of signs or symptoms of other primary causes of kidney damage; eGFR using the CKD-EPI-creatinine formulae. Reduced eGFR was defined as eGFR < 60 ml/min/1.73 m2. Albuminuria was defined as urinary albumin to creatinine ratio (UACR) ≥ 3 mg/mmol. Coronary heart disease (CHD) was defined as myocardial infarction, ischemic heart disease, or angina pectoris. Cardiovascular diseases (CVD) were defined as previously described [18], including myocardial infarction, atrial fibrillation, heart failure, left ventricular hypertrophy, and stroke. Metabolic syndrome (MetS) was defined as the presence of three or more of the following: (1) BMI > 30 kg/m2; (2) blood pressure > 130/85 mmHg and/or diagnosed hypertension; (3) fasting blood glucose ≥ 6.1 mmol/l or 2 h oral glucose tolerance test (OGTT) ≥ 7.8 mmol/l and/or have been diagnosed with diabetes and treated; (4) triglyceride > 1.7 mmol/l; (5) high-density lipoprotein cholesterol < 1.04 mmol/l.

Data Collection

General information and clinical manifestation were obtained through case history inquiry. Biochemical and physical examinations were conducted in related auxiliary departments of Nanfang Hospital. All the clinical characteristics and biochemical indexes were electronically collected through the electronic medical record system.

Isolation and Purification of Peripheral Blood Mononuclear Cells (PBMCs)

Anticoagulated 3–5-ml whole-blood samples from participants were centrifuged at 3500 rpm for 10 min, and then PBMCs were isolated using a commercially available Lymphoprep™ kit purchased from Haoyang Biological Manufacture Co. (Tianjin, China).

Quantitative Real-Time PCR

The total RNA of PBMCs was extracted using Trizol reagent (TAKARA), which was then reverse transcribed according to the PrimeScript RT Reagent Kit (TAKARA) manual. Quantitative real-time PCR (qRT-PCR) was performed in a Roche LightCycler 480II system (Roche, Basle, Switzerland) using a SYBR Green qPCR Kit (TAKARA). The 2−ΔΔCt method with β-actin as an internal control was used to carry out relative quantification. The levels of XIST mRNA were normalized by the method of 2−ΔCT; the normalized RNA-Seq data were subjected to correlation analysis. The primers used are listed in Table S2.

Statistical Analysis

Continuous variables are expressed as mean ± SD or median (interquartile range), and categorical variables are reported for frequency (%). For continuous variables, ANOVA and least significant difference test (LSD) were performed for normally distributed variables and the Kruskal-Wallis test for non-normally distributed variables, while the chi-square test was used for categorical variables and the rank sum test for grade data.

Results

Baseline Characteristics of Patients

This study included a total of 1184 cases, the vast majority being diagnosed with T2DM, n = 1101 (93%). The average age was 54.4 years, and the number of male patients was 736 (62.2%). Of the 1184 patients, 473 (39.9%) were confirmed as having DKD, 434 (36.6%) as having albuminuria, and 192 (16.2%) as having renal impairment. NA-DKD prevalence was 39 (3.3%), A-DKD prevalence was 281 (23.7%), and Mixed prevalence was 153 (12.9%).

Patients with DKD had longer duration of diabetes than those in the nDKD group, and their ages as well as ages at DM diagnosis tended to be older, notably in the NA-DKD and Mixed groups. Although the course of hypertension did not differ between the NA-DKD and nDKD groups, there was a tendency for a longer duration of hypertension in the Mixed and NA-DKD groups compared to the nDKD group. More baseline characteristics of all participants in the study are provided in Table 1.

Table 1 General condition data of the study groups
Full size table

Sodium-dependent glucose transporters 2 inhibitors (SGLT2i) and glucagon-like peptide 1 receptor agonist (GLP-1RA) were recommended as kidney beneficial glucose-lowering medications, so the participants’ use of these two drugs was further collected. The results show that 325 patients (27.3%) used GLP-1RA while 595 patients (50.3%) utilized SGLT2i. Of these, 186 patients (15.7%) took both types of medication while 450 patients (38.0%) took none. Notably, the Mixed group had the lowest percentage of SGLT2i (p < 0.001) (Table S3). Then, we looked into the connection between eGFR and SGLT2i use to further explore this. With declining eGFR level, there were noticeable drops in the utilization of SGLT2i or GLP-1RA (Fig. S2).

Comparison of Biochemical Indicators and Complications

An overview of the biochemical parameters observed in the study population is shown in Table 2. Those with DKD reported higher levels of white blood cell count (WBC), urea, cystatin C (CysC), homocysteine, and blood hypercoagulability but lower levels of hemoglobin (HGB) and free triiodothyronine (FT3) than control participants. Those with NA-DKD also had higher levels of WBC, and uric acid (UA), homocysteine but lower levels of high-density lipoprotein cholesterol (HDL-C) than those with the other DKD phenotypes. Compared to the other groups, the Mixed group’s urine had much higher levels of 2-microglobulin, indicating serious tubular damage (Table S4).

Table 2 Biochemical indicators and complications of the study groups
Full size table

Since cardiovascular disease (CVD) is another important complication of diabetes [19], we further explored the differences in CVD-related indicators among the four groups. As shown in Table S3, patients with DKD have worse blood pressure than patients with nDKD, while the amount of hypertension in patients with NA-DKD is equivalent to that in patients with A-DKD but lower than that in Mixed group. Mixed and A-DKD groups had higher systolic blood pressure (SBP) compared to nDKD, and the heart restructure was more common. Interestingly, NA-DKD groups had lower SBP even though there was no statistically significant difference between nDKD and NA-DKD (Table S5).

Gender Differences

The results of the study show that the mean age and the age of DM diagnosis of men were significantly younger than that of women in DM population (52.46 ± 13.23 vs. 57.57 ± 13.26 years, 45.08 ± 11.78 vs. 47.53 ± 12.54 years) (P < 0.001). Before the age of 60 years, significantly more men than women were hospitalized with DM, while the number of male and female patients was about equal after the age of 60 years (Fig. 1a). Next, focusing on the age at DM diagnosis, more men than women were diagnosed with DM at either age. Similarly, there were some differences in the peak age of DM diagnosis in men and women, occurring at 40–49 and 50–59 years of age, respectively (Fig. 1b).

Fig. 1
figure 1

Gender differences among participants

Full size image

As demonstrated in Fig. 1c, the prevalence of DKD increased more rapidly with age in men than in women, with a prevalence of only 31.2% in men and 38.1% in women among those < 60 years of age, whereas in those > 60 years of age, the prevalence of DKD increased by 25% in men and only 9.5% in women. The prevalence of DKD increased with age at DM diagnosis in men but did not differ significantly in women (Fig. 1d). Furthermore, we focused on the duration of DM, and, as we widely perceived, the prevalence of DKD gradually increased with the duration of DM (Fig. 1e). Interestingly, however, the shorter the duration of DM, the greater the proportion of male patients among hospitalized patients (Fig. 1f).

Correlation of lncRNA XIST Expression Levels in PBMCs with Clinical Indicators of DKD

Based on the widely recognized important role of lncRNA XIST in sex dimorphism and previous related studies in our team [9, 20], we sought to further explore the relationship between lncRNA XIST and the gender differences in DKD obtained from the above analysis. We collected 50 female PBMC samples (patients with nDKD n = 25, patients with DKD n = 25), since lncRNA XIST expression in men was usually very low. As shown in Fig. 2, the expression of lncRNA XIST in patients with DKD was lower than in patients with nDKD. To further study the correlation between the expression of lncRNA XIST and clinical indicators in patients with DKD, we collected the clinical data and PBMC samples from 31 female patients with DKD (Table S6). Scatter plots shown in Fig. 3 demonstrate significant positive correlations between the eGFR level and the expression of lncRNA XIST (R = 0.390, P = 0.036) and negative correlations between the HbA1c and the expression of lncRNA XIST (R = − 0.425, P = 0.027) in female patients with DKD. Meanwhile, no correlation was suggested between age, BMI, and lncRNA XIST expression.

Fig. 2
figure 2

Comparison of lncRNA XIST expression in PBMCs between DKD and nDKD in female patients

Full size image
Fig. 3
figure 3

Correlation analysis of lncRNA XIST expression levels with clinical indicators of DKD progression

Full size image

Discussion

The clinical manifestations of DKD among patients with DM changed over time. We found that the prevalence of DKD among hospitalized patients with DM reached 39.9%, of which 16.2% exhibited decreased eGFR, and 36.6% exhibited albuminuria, and the prevalence of three DKD phenotypes, NADKD, ADKD, and Mixed, was 3.3%, 23.7%, and 12.9%, respectively. In the NHANES study, slightly different from our results, the prevalence of DKD in adults with DM was 26.2% in 2009–2014, while the prevalence of albuminuria and reduced eGFR was 15.9% and 14.1%, respectively [5]. We found that the prevalence of reduced eGFR is less than half of the prevalence of albuminuria, and the prevalence of reduced eGFR (20%) is also less than that of albuminuria (34%) in the people from Hong Kong [21], which is more than the prevalence of albuminuria in people from US and Japan [5, 22]. This shows the geographical variation of characteristics of DKD. We considered the variation and change in the prevalence of DKD mainly correlated with the enrollment population, detection rates, lifestyles, diet, and genes.

We next summarize the clinical characteristics of each DKD phenotype in an attempt to provide a basis for clinical diagnosis and treatment. Among patients with diabetes, compared with patients with nDKD, patients with DKD were older, were more likely to be complicated with aortic atherosclerosis and carotid plaque, and had higher levels of urea, CysC, and homocysteine but lower levels of HGB. Patients with NADKD were more likely to have urolithiasis and higher UA but lower HDL-C and HbA1c attainment rates. Patients with ADKD had a longer duration of diabetes and were more likely to be complicated with diminished protective sensation, diabetic retinopathy (DR), CVD, and cardiac structural changes. At the same time, the levels of SBP and triglycerides (TG) were higher, while the levels of albumin (ALB) and HbA1c attainment rate were lower. Patients in the Mixed group had a longer duration of diabetes and hypertension, had an older age at DM diagnosis, and were more likely to be complicated with CVD, DR, diminished protective sensation, and urolithiasis and higher levels of SBP, UA, and TG.

The patients with DKD were recommended an SGLT2i and GLP1-RA based on ADA guidelines; however, we found that > 30% of patients with DKD did not use kidney beneficial glucose-lowering medication. Special attention needs to be paid that Mixed groups use the least kidney beneficial glucose-lowering medication among patients with DKD, and since they had the worst kidney function, this indicated that their eGFR had significantly declined and they were experiencing especially severe medication distress. This suggests that additional medications tailored to these individuals are required and that the therapeutic quality, safety, and efficacy of currently available medications also need to be enhanced to continue treating patients with seriously reduced eGFR.

Improved understanding of sex and gender-specific differences in the etiology, mechanisms, and epidemiology of kidney disease could help nephrologists better address the needs of their patients [23]. A number of studies in recent years have focused on the role of gender differences in DKD. A single-center cohort study in 2012 showed that lower HDL-C levels appear to be associated with the progression of diabetic nephropathy in men but not in women [24]. In our study, we found that the shorter the duration of diabetes, the more patients with diabetes, especially male patients. This is similar to some findings among Chinese populations where the prevalence of diabetes is increasing fastest in younger patients [25], which may be related to the rapid changes in the economy and the increase in household income, which has led to an increase in diet and changes to high-fat and high-sugar foods [26]. Our results also suggested that in the patients < 60 years old or diagnosed with diabetes, the prevalence of DKD in women was higher than in men; the opposite was true in the patients aged > 60 years old or diagnosed with diabetes after 60 years old. These results prompted us to see the sex differences in the clinical characteristics of the patients with DM.

LncRNA XIST is involved in diverse physiological and pathological processes; the expression of lncRNA XIST in PBMCs isolated from patients with T2DM has been proven higher than in that isolated from normoglycemic patients [25]. Studies have also shown that lncRNA XIST affects the immune function of B cells [26], and various immune pathways are involved in the occurrence and development of DM and its related complications [27], so lncRNA XIST may play a significant role in DM and DKD. In recent years, some studies have found that lncRNA XIST can play a role in DKD through the underlying ceRNA network. However, the correlation of lncRNA XIST with clinical indicators of DKD remains unknown [12, 28, 29]. As reported previously [30], the expression of lncRNA XIST in men was largely absent. In women, the expression of lncRNA XIST in patients with DKD was lower than in patients with nDKD. Otherwise, the expression of lncRNA XIST was positively correlated with eGFR level and negatively correlated with HbA1c. eGFR is known to be critical to the diagnosis and progression of DKD. Higher HbA1c was associated with an increased risk of CKD progression in patients with diabetes and remained an independent risk factor for adverse renal outcomes in patients with diabetes and CKD in patients with low carbamylation levels and no anemia [31]. According to a systematic review and meta-analysis of 20 cohorts, when HbA1c increased by 1%, the risk for DKD increased by 17% [32]. It is suggested that lncRNA XIST has great potential to participate in the development of DKD, but its mechanism needs to be further explored. Our investigation showed no correlations between age and lncRNA XIST expression, consistent with the results of a meta-analysis in cancer [33]. Thus, the variation in DKD prevalence with age, age of diabetes diagnosis, and duration of diabetes in women may not be attributed to differences in lncRNA XIST expression overages. Taking these into consideration, it is reasonable to assume that lncRNA XIST expression might be a slow and lasting protective factor for women, which is one of the reasons leading to a smaller increase in the prevalence of DKD with age, age at diabetes diagnosis, and duration of diabetes compared to men.

Limitations of our study cannot be shied away from, including that it is a small size, cross-sectional, single-center study and has possible enrollment bias. Meanwhile, further prospective and mechanistic studies are needed to confirm the causal relationship between the expression of lncRNA XIST and the changes in eGFR and HbA1c.

Overall, in this study we found that the prevalence of DKD in hospitalized patients with DM was 39.9% in economically advanced regions of southern China. LncRNA XIST expression in PBMCs of female patients with DKD is closely correlated with eGFR and HbA1c and therefore has potential as a biomarker of DKD progression.

Conclusions

Our study revealed that 39.9% of DM inpatients who were admitted to the hospital had DKD. Importantly, lncRNA XIST expression in PBMCs of female patients with DKD was significantly correlated with eGFR and HbA1c.