The relationship between bile acids levels and the prognosis of patients with diabetes on maintenance hemodialysis: a retrospective study

Abstract

Objective

There is a paucity of research on the association between bile acids (BAs) levels and all-cause death in patients with diabetes mellitus (DM) on maintenance hemodialysis (MHD). This study aimed to investigate the clinical characteristics of patients with DM on MHD according to different BAs levels and their impact on prognosis.

Methods

A retrospective cohort of 1,081 patients on hemodialysis at Xindu People’s Hospital and the First Affiliated Hospital of Chengdu Medical College were enrolled. Demographic and clinical characteristics were collected. The relationship between BAs and all-cause death risk was fitted using restricted cubic splines (RCS), and the BAs cutoff value was calculated. Patients were divided into low and high BAs groups based on the cutoff value. The primary endpoint was all-cause death and the secondary outcomes were deaths from cardiovascular events.

Results

Finally, 387 patients with DM on MHD were included. The median BAs level of all patients was 4.0 μmol/L. The RCS-based BAs cutoff value was 3.5 μmol/L. The BAs levels correlated negatively with total cholesterol, low-density lipoprotein, and blood calcium levels. During the follow-up, 21.7% of the patients died. The multivariate Cox regression analysis demonstrated that patients with DM on MHD with higher BAs were associated independently with a decreased risk of all-cause death (HR =0.55; 95% CI, 0.35–0.81, p = 0.01) compared to those with lower BAs levels.

Conclusions

Higher BAs levels were associated with lower lipid levels in patients with DM on MHD. BAs is an independent risk factor for all-cause death in patients with DM on MHD.

Keywords:

• All-cause death
• bile acid
• prognosis
• diabetes mellitus
• glycolipid metabolism
• maintenance hemodialysis
• risk factors

1. Introduction

Maintenance hemodialysis (MHD) and diabetes mellitus (DM) are chronic diseases that have a huge impact on patients’ health and quality of life. DM is a global health problem that has seen a dramatic rise in prevalence over the past few decades, with type 2 diabetes mellitus (T2DM) accounting for the majority of the cases [1]. According to the International Diabetes Federation (IDF), there were 537 million adults with diabetes worldwide in 2021, and there will be an estimated 783 million patients with DM worldwide by 2045 [2]. Similarly, the number of patients with end-stage renal disease (ESRD) is increasing yearly, with 2.6 million people worldwide receiving renal replacement therapy (RRT) in 2010. In fact, it is projected that the number of people receiving RRT will double globally by 2030, possibly reaching 5.4 million people, with the fastest growth in Asia [3]. According to the annual report of the China Kidney Disease Network in 2015, 90.96% of patients on RRT were on MHD [4].

DM and CKD are independent risk factors for cardiovascular diseases, with patients with DM on MHD having a higher incidence of and mortality from cardiovascular events [5,6]. Numerous studies have confirmed that DM is an important risk factor for the poor prognosis of patients on MHD [7]. Similar results have also been obtained in the Chinese MHD population [8]. Therefore, considering the prognosis of such patients is important; however, current treatment options are limited to dietary and risk factor interventions and the control of complications. Moreover, although many programs are being made available, patient mortality remains high. Therefore, we need to explore new risk factors, drugs, or therapeutic targets that affect such patients urgently.

The abnormal metabolism of glucose and lipids is crucial in the occurrence and development of cardiovascular disease (CVD). As the main component of bile (50% of organic bile), bile acids (BAs) is synthesized mainly in vivo and studies have shown that it regulates glucose and lipid metabolism [9,10]. Controlling blood glucose and lipid levels can reduce the risk of cardiovascular diseases and death. In view of its role in the regulation of glucose and lipid metabolism, we can speculate that BAs levels may affect the degree of vascular disease directly, in patients with DM on MHD, thereby affecting the incidence of cardiovascular events and the survival of these patients. In addition, current basic experimental studies have discovered that BAs derivatives (INT-777) can act on high glucose cultured cardiomyocytes, promoting Takeda G protein-coupled receptor 5 (TGR5) expression, increasing Nrf2 expression, reducing ROS content, and inhibiting apoptosis. When the short hairpin ribonucleic acid was used to reduce TGR5 expression, Nrf2 expression decreased significantly, ROS content increased, and apoptosis occurred. This suggested that BAs may act on the cardiac TGR5 receptor, improving the apoptosis of myocardial cells [11].

However, no study has been conducted to confirm whether BAs is associated with all-cause death in patients with DM on MHD. Therefore, in this retrospective cohort study, we aimed to explore whether BAs can predict the all-cause death of patients with DM on MHD.

2. Materials and methods

2.1. Study design and patients

This was a retrospective cohort study including patients diagnosed with DM on MHD in Xindu People’s Hospital and the First Affiliated Hospital of Chengdu Medical College, from April 2009 to December 2021. The DM diagnosis and the classification were based on the guidelines of the American Diabetes Association (ADA). The definition of MHD was MHD ≥3 months, 2–3 times weekly for 4 h each time.

The inclusion criteria were patients aged ≥18 years, being diagnosed with T2DM, and being on MHD. The exclusion criteria were the presence of malignant tumors, concurrent glomerular diseases, hepatobiliary diseases (active hepatitis, liver cirrhosis, and cholecystolithiasis), acute renal injury or MHD < 3 months and, having incomplete data (Figure 1).

Figure 1. Flowchart of the patients included in this study.

The Ethics Committee of Xindu People’s Hospital and the First Affiliated Hospital of Chengdu Medical College approved this study (NO: 2022CYFYIRB-BA-Dec05). The research protocol complied with the ethical standards set forth in the 1964 Helsinki Declaration and its subsequent amendments. All the patients provided written informed consent at the time of hospitalization.

2.2. Clinical characteristics

The demographic and laboratory baseline indicators were obtained from the electronic medical record system for the patients starting hemodialysis and the frequency and duration of dialysis was determined as the weekly frequency and duration of dialysis after the first dialysis for three months. According to the patient’s condition, hemodialysis was performed 2–4 times weekly, hemodiafiltration was performed once every half month, and the laboratory tests and examinations related to renal complications were performed once every three months. Hypertension was defined as the diagnosis of hypertension, use of antihypertensive drugs, and an average systolic blood pressure (SBP) >140 mmHg or average diastolic blood pressure (DBP) >90 mmHg. The primary endpoint was defined as all-cause death. The secondary endpoint was defined as cardiovascular deaths. A cardiovascular death was defined as death caused by heart failure, myocardial infarction, arrhythmia, cerebral hemorrhage, infarction and other cardiovascular events. BAs levels were determined using the enzyme circulation method. The association between bile acids and all-cause mortality in all MHD patients was assessed, then the association between bile acids and all-cause mortality in MHD patients with and without DM was investigated separately. Ultimately, we focused our study on the relationship between BAs and all-cause mortality in patients on MHD with DM.

2.3. Statistical analyses

The continuous variables with a normal distribution were presented using means ± standard deviations or the medians and interquartile ranges. The classified data were presented using counts and percentages. A dynamic risk ratio (HR) curve was used to evaluate the influence of BAs on prognosis. The relationship between BAs and death risk was fitted using restricted cubic splines, and the BAs cutoff value was calculated. When comparing the two groups, we used the Student’s T, Mann–Whitney U, and Chi-square tests as appropriate. The correlation between the BAs levels and the clinical indexes was evaluated using correlation analyses. The Pearson’s correlation was used for the numerical variables with a normal distribution, and the Spearman’s correlation was used for the other variables. The Kaplan–Meier method was used to evaluate the survival curves of all-cause death and cardiovascular and cerebrovascular events, and a log-rank test was used for the comparison. A Cox proportional hazard model was used to analyze the effect of the BAs levels on the prognosis of patients with DM on MHD. SPSS (version 26.0) and R (version 4.1.3) were used for all the statistical tests. Statistical significance was set at a P value < 0.05.

3. Results

We first fitted the relationship between BAs and all-cause mortality in all MHD patients based on restricted cubic spline plots, and the results suggested that there was no significant trend of change in the risk of all-cause mortality with changes in BAs (Supplementary Figure 1); the survival analysis by subjecting BAs to dichotomous and trichotomous grouping showed no significant difference in the probability of survival among patients of different groups (Supplementary Figure 2 and 3). In the same way, we assessed the association between BAs and all-cause mortality in MHD without diabetes and showed that in this population, there was also no significant trend in the change in the risk of all-cause mortality with changes in BAs (Supplementary Figure 4); grouping BAs by dichotomous and trichotomous methods, survival analysis showed no significant difference in the probability of survival among patients of different groups either (Supplementary Figures 5 and 6). Subsequently, in our study, we ultimately focused on the relationship between BAs and all-cause mortality in patients on MHD with DM.

3.1. Baseline characteristics for patients on MHD with DM

Our cohort included 387 patients with DM on MHD (Figure 1). Table 1 presents the clinical characteristics of these patients. The median BAs level of all the patients was 4.0 (2.3 − 6.2) μmol/L. The median age was 61 (50 − 72) years, and 58.7% (227/387) of the patients were men. Hypertension was noted in 95.3% (369/387) of the patients. The median eGFR at baseline was 4.78 (3.74 − 6.62) mL/min/1.73 m². The mean glycosylated hemoglobin (HbA1c) of all patients was 7.89 ± 1.99, and the median random blood glucose was 6.45 (5.38 − 8.75) mmol/L. The relationship between the BAs levels and the risk of all-cause death was fitted using a restricted cubic spline, and the cutoff value for the BAs was calculated (Figure 2). The HR curve indicated that an increase in the BAs levels reduced the risk of all-cause death in patients (Figure 3). Based on the cutoff values, the patients were divided into two groups: the low BAs (≤3.5 μmol/L) group and the high BAs (> 3.5 μmol/L) groups. Compared with patients with low BAs levels, patients with high BAs levels had higher albumin levels and lower cholesterol, low-density lipoprotein cholesterol (LDL-c), and blood calcium levels (Table 1).

Figure 2. The optimal cut-point value of bile acids.

Figure 3. Dynamic HR changes with bile acids levels. HR, dynamic risk ratio.

Table 1. Baseline clinical features of enrolled 387 patients on MHD with DM.

Variables	All (n = 387)	Lower BA(n = 172) (≤3.5μmol/L)	Higher BA (n = 215) (>3.5μmol/L)	P-value
BAs (μmol/L)	4.00 (2.30 − 6.20)	1.40 (2.20 − 2.70)	4.50 (5.80 − 8.20)	<0.001
Age (years)	61.00 (50.00 − 72.00)	60.00 (49.00 − 71.75.00)	62.00 (50.00 − 73.00)	0.354
Gender (male, %)	227 (58.7)	98 (57.0)	129(60.0)	0.548
BMI (kg/m^2)	22.34 (16.96 − 24.62)	22.80 (19.76 − 25.38)	21.97 (19.76 − 25.38)	0.094
Hypertension [n (%)]	369 (95.3)	165 (95.9)	204 (94.9)	0.809
e-GFR (ml/min/1.73m^2)	4.78 (3.74 − 6.62)	4.74 (3.78 − 6.70)	4.79 (3.74 − 6.49)	0.756
Serum albumin (g/L)	34.90 (30.40 − 39.10)	34.40 (29.70 − 38.20)	35.50 (31.28 − 39.50)	0.018
Hemoglobin (g/L)	82.00 (68.00 − 98.00)	82.00 (68.00 − 98.00)	82.00 (69.00 − 97.25)	0.943
HbA1c (%)	7.89 ± 1.99	7.76 ± 2.25	8.02 ± 1.67	0.411
Blood glucose (mmol/L)	6.45 (5.38 − 8.75)	6.33 (5.38 − 8.76)	6.59 (5.38 − 8.75)	0.513
Triglyceride (mmol/L)	1.48 (1.08 − 2.21)	1.49 (1.06 − 2.14)	1.46 (1.08 − 2.25)	0.725
Total cholesterol (mmol/L)	3.79 (3.18 − 4.56)	4.01 (3.28 − 4.71)	3.68 (3.00 − 4.50)	0.029
LDL-c (mmol/L)	2.15 (1.55 − 2.80)	2.31 (1.63 − 2.88)	1.98 (1.51 − 2.64)	0.012
HDL-c (mmol/L)	1.16 (0.88 − 1.40)	1.14 (0.89 − 1.40)	1.16 (0.88 − 1.68)	0.900
Serum calcium (mmol/L)	2.04 (1.83 − 2.18)	2.06 (1.90 − 2.18)	2.01 (1.81 − 2.18)	0.04
Serum phosphorus (mmol/L)	1.89 (1.55 − 2.31)	1.85 (1.48 − 2.30)	1.91 (1.59 − 2.92)	0.344
Parathyroid hormone (ng/ml)	249.00 (147.00 − 421.00)	230.00 (1140.25 − 438.50	253.00 (154.00 − 415.25)	0.709
C-reactive protein (mmol/L)	4.00 (1.40 − 16.00)	4.65 (2.00 − 17.80)	3.55 (1.00 − 15.00)	0.259
Death (n, %)				0.118
All-causes of death	84 (21.7%)	49 (28.5)	35 (16.28)	0.130
Cardiovascular death	49 (12.7)	29 (16.7)	20 (9.3)	0.261

BAs: bile acids; e-GFR: estimated glomerular filtration rate; LDL-c: low-density lipoprotein cholesterol; HDL-c: high-density lipoprotein cholesterol; MHD: maintenance hemodialysis; DM: diabetes mellitus.

3.2. BAs-related clinical characteristics for patients on MHD with DM

The BAs levels correlated negatively with the total cholesterol (r = −0.102, p = 0.046), LDL-c (r = −0.110, p = 0.030), and serum calcium levels (r = −0.107, p = 0.036) (Table 2).

Table 2. Correlations of BAs levels with different clinicopathological data.

Variables	Correlation coefficient	P-value
Age (years)	0.068	0.182
Gender (male, %)	−0.069	0.178
BMI (kg/m^2)	−0.067	0.191
Hypertension [n (%)]	0.032	0.531
e-GFR (ml/min/1.73m^2)	−0.048	0.344
Serum albumin (g/L)	0.054	0.288
Hemoglobin (g/L)	−0.057	0.260
HbA1c (%)	−0.416	0.083
Blood glucose (mmol/L)	0.080	0.118
Triglyceride (mmol/L)	0.013	0.795
Total cholesterol (mmol/L)	−0.102	0.046
LDL-c (mmol/L)	−0.110	0.030
HDL-c (mmol/L)	−0.045	0.377
Serum calcium (mmol/L)	−0.107	0.036
Serum phosphorus (mmol/L)	0.039	0.443
Parathyroid hormone (ng/ml)	−0.001	0.983
C-reactive protein (mmol/L)	−0.002	0.963

BAs: bile acids; e-GFR: estimated glomerular filtration rate; LDL-c: low-density lipoprotein cholesterol; HDL-c: high-density lipoprotein cholesterol.

3.3. Risk of endpoint events for patients on MHD with DM

During a median follow-up of 39.0 months (19.0 − 56.0 months), 84 of the 387 patients (21.7%) died of any cause. Patients with higher BAs levels had lower all-cause mortality rates than patients with low BAs levels (Figure 4). The Kaplan–Meier survival analysis revealed that the risk of all-cause death increased significantly in patients with low BAs levels at baseline (Figure 4). In addition, the Kaplan–Meier survival analysis of all the patients with DM on MHD revealed that the occurrence of cardiovascular events between the two groups was significant (Figure 5). The Cox regression analysis was used to assess the correlation between the baseline clinicopathological variables and all-cause death. A single-factor analysis revealed that the BAs, age, random blood glucose, and parathyroid hormone were risk factors for all-cause death (p < 0.05) (Supplementary Table 1). After adjusting for factors such as sex, age, body mass index, hypertension, eGFR, hemoglobin, serum albumin, triglyceride, total cholesterol, LDL-c, high-density lipoprotein cholesterol, blood calcium, blood phosphorus, parathyroid hormone, C-reactive protein, and HbA1c, low BAs levels remained associated independently with the risk of all-cause death. Compared with the lower BAs group, the HR of the higher BAs group was 0.55 (95% [confidence interval]: 0.35, 0.87, p = 0.01) (Table 3). In addition, age, albumin, and HbA1c were also independent risk factors for all-cause death in patients on MHD. The dynamic Area Under Curve (AUC) values for time-varying bile acids predicting all-cause mortality ranged from 0.55–0.75, and the dynamic C-index values ranged from 0.55–0.70 (Supplementary Figures 7 and 8).

Figure 4. The relationship between varying levels of BAs and all-cause death in MHD with DM patients.

Figure 5. The relationship between varying levels of BAs and cardiovascular death in MHD with DM patients.

Table 3. Associations between BAs level and all-cause death in MDH with DM.

	Hazard ratio (95% Confidence interval)
	Lower BA(n = 172) (≤3.5μmol/L)	Higher BA (n = 215) (>3.5μmol/L)	P-value
Unadjusted	1	0.59 (0.38-0.91)	0.01
Model 1^a	1	0.53 (0.34-0.83)	0.005
Model 2^b	1	0.51 (0.33-0.81)	0.004
Model 3^c	1	0.55 (0.35, 0.87)	0.01

Model 1^a:adjusted for baseline age, gender, BMI; Model 2^b: adjusted for covariates in model 1 plus dialysis frequency and e-GFR; Model 3^c: adjusted for covariates in model 2 plus hemoglobin, serum albumin, parathyroid hormone, C-reactive protein. BAs: bile acids; e-GFR: estimated glomerular filtration rate; MHD: maintenance hemodialysis; DM: diabetes mellitus.

4. Discussion

Our cohort study was the first to explore the relationship between BAs levels and clinical prognosis in patients with DM on MHD. Further, we investigated the relationship between the BAs levels, clinicopathological features, and all-cause death in 387 patients with DM on MHD. The results revealed that BAs was an independent risk predictor of death in patients with DM on MHD, in addition to conventional factors such as age and serum albumin. We observed that serum BAs, as a noninvasive marker, was related closely to the risk of all-cause death. This also suggested that BAs analogs and their downstream signal pathways may be potential therapeutic drugs and targets for treating patients with DM on MHD.

BAs is synthesized from cholesterol and is the main organic component of bile. Factors that increase BAs secretion include the consumption of foods (especially those high in protein), excitation of the vagus nerve, gastric factors (e.g., gastrin, pancreatic enzymes, cholecystokinin, and bile salt), pathological factors such as hepatobiliary diseases, and some drugs and treatments such as metformin [12,13] and weight-loss surgery [14,15].

The all-cause risk of patients with DM on MHD decreases with increased BAs levels. We obtained the cutoff value of death by fitting BAs and death risk with a restricted cubic spline. If the BAs level of a patient with DKD is less than 3.5 μmol/L, we needed to be alert to the increased risk of all-cause death. The cutoff value serves as a reference in clinical practice for identifying patients with an increased risk of end-stage renal disease. However, we must rule out a pathological increase in BAs caused by hepatobiliary system diseases. Our findings showed that, in addition to the conventional death risk factors for patients on MHD, such as age, albumin, and HbA1c, the all-cause death of patients in the higher BAs group was 0.55 times that of patients in the lower BAs group, indicating that BAs levels are play an important role in the prognosis of patients with DM on MHD. In addition, in the analysis of secondary endpoint events, we discovered that patients with higher BAs levels had a higher risk of cardiovascular death.

For patients with MHD and DM, controlling blood glucose and lipid levels to achieve the target is important as these may affect the incidence of cardiovascular events, and ultimately affecting patient prognosis. Moreover, BAs has been shown to be an important molecule in glucose regulation and lipid metabolism. Recent studies have suggested that BAs play an important role in regulating postprandial glucose metabolism [16]. Another study discovered that the risk of abnormal blood glucose in individuals undergoing cholecystectomy in Chinese community residents was increased, indicating that BAs is important in maintaining blood glucose [17]. More studies have also revealed that the role of BAs in blood glucose regulation may be to activate BAs receptors involved in downstream physiological functions [18]. However, many studies have also reported that poor blood glucose control affects the survival of patients on MHD [19]. In a national prospective cohort study including 503 patients on hemodialysis, the impact of baseline glycosylated albumin on the prognosis of patients was evaluated. The results revealed that glycosylated albumin was a risk factor for the death and morbidity of patients on hemodialysis [20]. Kim et al. [21] suggested that the HbA1c levels of patients on hemodialysis was related linearly to the risk of all-cause death. Compared with other causes of death, the cardiovascular death risk increased earlier and faster with increasing HbA1c levels. These studies show that better blood glucose control is related closely to a better prognosis in patients with DM on MHD. In view of the importance of BAs in the blood glucose regulation of patients, regulating blood glucose may be one of the reasons for the better prognosis in patients with high BAs levels. However, we observed no correlation between BAs levels and blood glucose control in our study, possibly because most of the patients who were using hypoglycemic drugs and anemia were on dialysis, which may have led to inaccurate HbA1c measurements. Our study also found that advanced age was also a risk factor for mortality in MHD with DM patients, which was consistent with previous research findings [22]. Hemodialysis patients are prone to malnutrition, and hypoalbuminemia is one of the most common indicators of malnutrition [23]. Studies have shown that ESRD patients with hypoalbuminemia have a higher risk of mortality [24]. We came to the same conclusion. Elevated parathyroid hormone levels are closely related to complications such as bone disease and vascular calcification. Our research also showed that parathyroid hormone levels is a risk factor for MHD with DM patients, which was consistent with previous studies [25].

BAs promote the digestion and absorption of lipid nutrients and play a crucial role in lipid metabolism [9,26,27]. However, hyperlipidemia has been recognized traditionally as a risk factor for cardiovascular disease. Nakano et al. [28] suggested that higher cholesterol levels can predict cardiovascular events and are related negatively to mortality in patients on hemodialysis. Further, although some studies have suggested that cholesterol levels are related negatively to the death risk of patients on dialysis, the author also explained that this paradox may have been due to the presence of systemic inflammation and/or malnutrition, which are related to lower cholesterol levels and higher mortality, rather than the protective effect of a high cholesterol concentration [29]. In view of the strong evidence that lipid-lowering therapy reduces the risks and benefits in the non-dialysis population, the emerging consensus emphasizes that patients with dyslipidemia should be treated actively to achieve an LDL-c target lower than 100 mg/dl [30]. In addition to its role in glycolipid metabolism and the hepatobiliary system, BAs receptor activation is related closely to the occurrence and development of other diseases. Xiao et al. [31] have confirmed that patients with higher BAs levels have a lower risk of progressing from diabetic nephropathy to ESRD and closely related to lipid levels. We also discovered that BAs correlates negatively with total cholesterol and LDL-c, suggesting that BAs can regulate lipid metabolism. Based on these findings, regulating lipids metabolism may be the other reason for the better prognosis of patients with high BAs levels.

In view of the fact that the main physiological function of BAs is to mediate glucolipid metabolism, which is more pronounced in patients with DM. Therefore, in our study, we found that BAs can not predict all-cause mortality in patients with MHD without DM, and BAs can not predict the risk of all-cause mortality in all patients with MHD because of the confounding of outcomes by patients without DM.

Additionally, BAs may have a direct cardiovascular benefit. Hu et al. [32] revealed that INT-777 attenuated the neuroinflammation mediated by NLRP3-ASC inflammatory bodies through the TGR5/cAMP/PKA signaling pathway after the occurrence of subarachnoid hemorrhage in rats. Activating the TGR5 signaling pathway can reduce inflammation and improve post-infarction cardiac function. Strategies to control BAs metabolism and TGR5 signaling to improve the inflammatory response may benefit patients with myocardial infarction. Moreover, Farnesoid X receptor (FXR) activation was also found to have a similar effect. The FXR signaling pathway is an attractive therapeutic target for treating atherosclerosis [33]. Mesenchymal stromal cells overexpressing FXR play a cardioprotective role in acute ischemic heart injury by binding endogenous BAs [34].

The findings of all the above studies have suggested that increased BAs may regulate glucolipid metabolism through downstream signaling pathways as well as present additional cardiovascular and cerebrovascular event benefits which may explain why patients with higher BAs levels have better prognoses.

This study had some limitations. First, it was a retrospective study and hence selection bias was inevitable. Second, the sample size was small and there may have been uncommon conditions that resulted in increased BAs levels that may not have been excluded. Finally, we did not control for all the treatment interventions (such as statins, antiplatelet aggregation drugs, and RASS inhibitors), which may have been confounding factors affecting the results.

In conclusion, our study provided a new marker for predicting all-cause death in patients with DM on MHD, and the prognoses of patients may improve if the serum BAs levels remain above 3.5 μmol/L. Our study also predicted that BAs analogs and their downstream signal pathways may be promising drugs for treating patients with DM on MHD.

Author contributions

Conception and design of the study: BL, CP, YW, RM, YF; Acquisition and analysis of data: RM, YF; Drafting the manuscript or figures: BL, CP, YW, RM, YF.

Acknowledgements

The authors thank all those who helped to enable the completion of this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability

All datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Additional information

Funding

This study was supported by the Health Commission of Sichuan Province Program (S21097), (CYFY2021YB05); Sichuan Provincial Administration of Traditional Chinese Medicine (2020JC0022), Chengdu Medical College Science and Technology Foundation (CYZZD20-03).