Abstract
Type 2 diabetes mellitus (T2DM) patients are often accompanied with hypertension. However, the association of antihypertensive drugs with β‐cell function has not been well studied. To investigate this question, the authors performed a cross‐sectional study involving 882 hypertensive T2DM patients. To assess β‐cell function, patients were given 75g glucose orally and C‐peptide levels before and 1, 2, and 3 hours after glucose intake were measured. Homa‐β was computed by Homeostasis Model Assessment model to evaluate β‐cell function using fasting C‐peptide and glucose levels in the plasma. Multivariable‐adjusted analysis was performed to evaluate the association of antihypertensive drugs with C‐peptide levels, HbA1c, and Homa‐β. Among 882 hypertensive patients, 547 (62.0%) received antihypertensive treatment. Multivariate‐adjusted analysis demonstrated that use of calcium channel blockers (CCBs) was negatively associated with HbA1c levels (CCBs: 0.95 [95% CI: 0.92‐0.98], P = 0.002). Our data further illustrated that the C‐peptide levels before and 1, 2, and 3 hours of OGTT were 1.10‐, 1.18‐, 1.19‐, and 1.15‐fold increase in T2DM patients taking CCBs (P = 0.084 for fasting C‐peptide levels; P ≤ 0.024 for C‐peptide levels at 1, 2, and 3 hours after OGTT) in comparison with non‐CCB users. Nevertheless, usage of any other antihypertensive drugs did neither associated with HbA1c nor associated with C‐peptide levels (P ≥ 0.11). In conclusion, CCB treatment was negatively associated with HbA1c levels but positively associated with β‐cell function in hypertensive T2DM patients, implying that CCBs could be considered to treat hypertensive T2DM patients with reduced β‐cell function.
Keywords: calcium channel blocker, hypertension, inhibitor of renin‐angiotensin‐aldosterone system, type 2 diabetes mellitus, β‐cell function
1. INTRODUCTION
The prevalence of diabetes mellitus is increasing worldwide, and the number of diabetic patients has reached up to 415 million at present.1, 2 In China, a nationally representative cross‐sectional survey reported that the prevalence of diabetes was 10.9% and the number of diabetic patients has reached more than 100 million by 2013.3 Except the increased economic burden, hyperglycemia poses high risks of cardiovascular and microvascular disease in diabetic patients.4 As the most common risk factor, the incidence of hypertension in type 2 diabetic patients is approximately 70%.5 Aside from improved glucose homeostasis, antihypertensive treatment reduces the incident diabetes in the subjects with impaired glucose tolerance,6, 7 and cardiovascular mortality and the progression of nephropathy in patients with type 2 diabetes mellitus (T2DM).8, 9
During the pathological progression in diabetes, patients experience from β‐cell hyperplasia to overcome hyperglycemia and then β‐cell death due to exhausted production of insulin. Therefore, how to protect β‐cell number and function is as crucial as how to restrain vascular complications. Renin‐angiotensin‐aldosterone system (RAAS) is a hormone system to control blood pressure and expressed in the kidney, adrenal glands, vasculature, nervous system, and pancreatic β cells.10, 11 Blockade of RAAS signaling pathways attenuates insulin resistance in general population.12 However, whether inhibition of RAAS system could preserve β‐cell function in human subjects remains controversial.13, 14
Apart from RAAS system, calcium channels are expressed on β‐cell membrane. They regulate cell maturation during development and insulin secretion in response to extracellular glucose concentrations.16 Moreover, mitochondrial Ca2+ transiently modulates oxidative metabolism within β cells.17 Nevertheless, excessive Ca2+ influx upon stimulators such as hyperglycemia and oxidative stress triggers endoplasmic reticulum stress and β‐cell apoptosis and dysfunction in murine islets.18, 19 The addition of calcium channel blocker, nifedipine, promotes cell survival in insulin‐secreting INS cells, rat islets and human islets in vitro.21 Nevertheless, the association of calcium channel blockers and β‐cell function in T2DM patients is not well defined. To address the questions above, we performed this cross‐sectional study to testify the relationship of antihypertensive treatment, in particular, blockade of RAAS or calcium channels and β‐cell function in T2DM patients.
2. METHODS
2.1. Study population
Type 2 diabetes mellitus patients (n = 1329) who were admitted to the Endocrinology Center at Luhe hospital in Beijing from July 2015 till January 2017 were enrolled in the cross‐sectional study. The criteria of T2DM included plasma glucose of at least 7.0 mmol/L while fasting and/or of 11.1 mmol/L or more 2 hours after an orally administered glucose load of 75 g. The exclusion criteria were type 1 diabetes mellitus, gestational diabetes mellitus, and T2DM with cancer. The study complied with the Helsinki Declaration for investigation of human subjects and was approved by the Institutional Review Boards of Beijing Luhe hospital and Capital Medical University. All the patients provided written informed consent.
We excluded 79 patients from analysis, because of missing fasting C‐peptide levels (n = 16) or repeated recruitment (n = 53) or inapplicability to calculate β‐cell function and insulin resistance using Homa formula (n = 10). Thus, the number of diabetic patients statistically analyzed totaled 1250. The flowchart is shown in Figure 1.
Figure 1.
Flowchart
2.2. Clinical measurement
The observers measured each participant's anthropometric characteristics and collected information on medical history, smoking and drinking habits, and intake of medications. Office systolic blood pressure and diastolic blood pressure were measured using a standard mercury sphygmomanometer as described before.22 Briefly, patients were rested for 5 minutes in the sitting position before blood pressure measurement. Five consecutive auscultatory readings were obtained from each patient, and each reading was performed at 30‐second interval. The average of systolic and diastolic blood pressure was calculated from five measurements. Mean arterial pressure was one third of systolic pressure plus two thirds of diastolic pressure. Hypertension was defined as systolic blood pressure ≥ 140 mm Hg or diastolic blood pressure ≥ 90 mm Hg or use of antihypertensive drugs. Body mass index was weight in kilograms divided by the square of height in meters. Glomerular filtration rate (eGFR) was derived from serum creatinine by the Chronic Kidney Disease Epidemiology Collaboration (CKD‐EPI) equation.23
2.3. Biochemical measurement
Fasting, venous blood samples were drawn to measure white blood cell count, serum total cholesterol, high‐density lipoprotein (HDL) cholesterol (HDL‐c), triglycerides, and creatinine, plasma glucose, HbA1c, γ‐glutamyltransferase, and serum levels of C‐peptide. Urinary samples were collected to measure creatinine and albumin. Low‐density lipoprotein (LDL) cholesterol (LDL‐c) was computed from serum total cholesterol, HDL‐c, and serum triglycerides by the Friedewald equation.24
2.4. Assessment of pancreatic β‐cell function and insulin resistance
Pancreatic β‐cell function was assessed by the measurement of fasting C‐peptide levels and C‐peptide levels during oral glucose tolerance test (OGTT). During OGTT, patients received oral administration of 75 g glucose. Blood glucose and C‐peptide values were measured before and 1, 2, and 3 hours after glucose administration.
In addition, Homa‐β and Homa‐IR were computed by Homeostasis Model Assessment (http://www.dtu.ox.ac.uk/homacalculator/) to evaluate β‐cell function and insulin resistance, respectively, using fasting C‐peptide and glucose levels in the plasma.
2.5. Enzyme‐linked immunosorbent assay
Serum levels of galectin‐3, high‐sensitivity C‐reactive protein (CRP), transforming growth factor β1 (TGF‐β1), tumor necrosis factor α (TNF‐α), and vascular endothelial growth factor (VEGF) were measured according to the manual instruction (MLBio, Shanghai, China). Intra‐assay coefficients of variation for galectin‐3, CRP, TGF‐β1, TNF‐α, and VEGF were 6.5%, 5.8%, 5.4%, 7.3%, and 7.1%, respectively. Inter‐assay coefficients of variation for galectin‐3, CRP, TGF‐β1, TNF‐α, and VEGF were 8.4%, 6.5%, 7.3%, 15.0%, and 9.4%, respectively.
2.6. Statistical analysis
For database management and statistical analysis, we used the SAS system, version 9.4 (SAS Institute Inc, Cary, NC). The continuous variables that followed Gaussian distribution were expressed as mean ± SD, and the continuous variables that did not follow Gaussian distribution were expressed as median with interquartile range (IQR). To compare the difference between two groups, unpaired t test was used for the continuous variables following Gaussian distribution and nonparametric t test was applied for continuous variables with a non‐Gaussian distribution. To compare proportions, Pearson chi‐square test or Fisher's exact test was used in the study.
In multivariable‐adjusted linear regression analysis, −log10 probability plots were constructed. The covariables considered included sex, age, the status of smoking and alcohol intake, history of cardiovascular disease, body mass index (BMI), mean arterial blood pressure, HbA1c, total cholesterol/HDL‐c ratio, serum triglyceride, γ‐glutamyltransferase, urinary albumin‐to‐creatinine ratio, eGFR, white blood cell count, diabetic duration, and antidiabetic treatment in drug class and statins. For continuous covariables that did not follow Gaussian distribution, they were normalized by a logarithmic transformation before linear regression analysis.
To exclude the effects of different classes of antihypertensive drugs on HbA1c or C‐peptide levels, we first tested the interactions of different antihypertensive drugs in the model where all covariables were present. If no significant interactions were detected, the main effects of calcium channel blockers (CCBs), renin‐angiotensin system (RAAS) inhibitors, β‐blockers, or diuretics were finally assessed in the model. Alternatively, we conducted a sensitivity analysis using the same multivariate‐adjusted linear regression model in which patients who received only one class of antihypertensive drug were included. In parallel, patients who did not take antihypertensive drug were served as controls. Furthermore, path analysis was performed to explore the direct association of CCB treatment with HbA1c and the indirect association with HbA1c due to VEGF levels or Homa‐β values. Significance was a two‐tailed P level of 0.05 or less.
3. RESULTS
3.1. General characterization of the study subjects
In 1250 T2DM patients analyzed (52.6% men), age averaged 57.6 (14.1) years and 235 (18.8%) had history of cardiovascular disease. The median of fasting blood glucose, HbA1c, fasting C‐peptide, Homa‐β, and Homa‐IR was 8.22 mmol/L, 10.0%, 1.07 ng/mL, 31.7%, and 0.96%, respectively. All patients received antidiabetic treatment, among which 715 (57.2%) took metformin and 671 (53.7%) took insulin injection. The general characteristics of T2DM patients are shown in Table S1.
Of 1250 T2DM patients, 882 were diagnosed with hypertension. Among 882 hypertensive patients, 547 (62.0%) received antihypertensive treatment: 349 (39.6%) taking inhibitors of RAAS system, 257 (29.1%) taking calcium channel blockers, 159 (18.0%) taking β‐blockers, 47 (5.3%) taking diuretics, and 213 (24.1%) taking more than one antihypertensive drugs. Compared with non‐hypertensive patients, hypertensive patients were older and more obese and had lower HbA1c levels, higher fasting C‐peptide levels, higher Homa‐β index, and higher Homa‐IR index. In the following analysis, we focused on hypertensive T2DM patients.
3.2. Blood glucose levels in hypertensive T2DM patients
Table 1 lists the general characteristics of hypertensive patients based on the usage of CCBs, RAAS inhibitors, β‐blockers, or diuretics. Compared with non‐CCB users, CCB users had reduced fasting blood glucose levels (7.73 mmol/L vs 8.28 mmol/L, P = 0.009) and HbA1c levels (9.6% vs 10.0%, P = 0.018). By contrast, fasting blood glucose levels and HbA1c did not differ in patients taking RAAS inhibitors or diuretics compared with their correspondent controls (P ≥ 0.30; Table 1 & Figure 2). Compared with patients not taking β‐blockers, β‐blocker users had comparable fasting blood glucose levels and lower HbA1c levels (fasting blood glucose levels: 8.16 mmol/L vs 7.92 mmol/L, P = 0.66; HbA1c: 9.9% vs 9.4%, P = 0.011; Table 1).
Table 1.
General characterization of hypertensive T2DM patients based on the antihypertensive treatments
| Characteristics | Calcium channel blockers | RAAS inhibitors | β‐Blockers | Diuretics | ||||
|---|---|---|---|---|---|---|---|---|
| No | Yes | No | Yes | No | Yes | No | Yes | |
| Number | 625 | 257 | 533 | 349 | 723 | 159 | 835 | 47 |
| Sex (male) | 329 (52.6) | 114 (44.4)* | 284 (53.3) | 159 (45.6)* | 369 (51.0) | 74 (46.5) | 425 (50.9) | 18 (38.3) |
| Smoking (1, 0) | 190 (30.4) | 61 (23.7)* | 162 (30.4) | 89 (25.5) | 205 (28.4) | 46 (28.9) | 242 (29.0) | 9 (19.2) |
| Alcohol intake (1, 0) | 171 (27.4) | 55 (21.4)* | 154 (28.9) | 72 (20.6)† | 181 (25.0) | 45 (28.3) | 218 (26.1) | 8 (17.0) |
| Cardiovascular disease (1, 0) | 137 (21.9) | 79 (30.7)† | 133 (25.0) | 83 (23.8) | 125 (17.3) | 91 (57.2)‡ | 196 (23.5) | 20 (42.6)† |
| Chronic kidney disease (1, 0) | 39 (6.2) | 34 (13.2)† | 46 (8.6) | 27 (7.7) | 54 (7.5) | 19 (20.0) | 62 (7.4) | 11 (23.4)‡ |
| Statins (1, 0) | 318 (50.9) | 151 (58.8)* | 272 (51.0) | 197 (56.5) | 354 (49.0) | 115 (72.3)‡ | 434 (52.0) | 35 (74.5) † |
| Mean ± SD | ||||||||
| Age (y) | 59.9 ± 13.4 | 63.3 ± 11.0‡ | 60.1 ± 13.8 | 62.2 ± 11.0* | 60.2 ± 13.3 | 64.2 ± 9.7‡ | 60.7 ± 12.9 | 64.9 ± 9.7* |
| Body mass index (kg/m2) | 26.4 ± 4.0 | 26.7 ± 3.7 | 26.4 ± 4.0 | 26.6 ± 3.8 | 26.4 ± 3.9 | 26.8 ± 3.8 | 26.4 ± 3.9 | 28.6 ± 3.0‡ |
| Systolic blood pressure (mm Hg) | 137.1 ± 16.5 | 140.8 ± 19.3† | 136.7 ± 16.1 | 140.3 ± 19.2† | 138.9 ± 16.7 | 134.5 ± 20.1† | 138.1 ± 17.3 | 138.3 ± 20.1 |
| Diastolic blood pressure (mm Hg) | 81.3 ± 10.0 | 80.0 ± 12.1 | 80.4 ± 10.3 | 81.6 ± 11.2 | 81.7 ± 10.4 | 77.4 ± 11.3‡ | 81.1 ± 10.6 | 77.0 ± 10.2* |
| MAP (mm Hg) | 99.9 ± 10.5 | 100.3 ± 12.7 | 99.2 ± 10.5 | 101.2 ± 12.1* | 100.8 ± 10.7 | 96.4 ± 12.5‡ | 100.1 ± 11.1 | 97.4 ± 12.1 |
| Heart rate (beat/min) | 80.0 ± 29.7 | 76.6 ± 13.3 | 78.4 ± 13.1 | 79.9 ± 38.2 | 79.3 ± 28.0 | 77.4 ± 14.0 | 79.1 ± 26.6 | 76.3 ± 12.6 |
| Median (IQR) | ||||||||
| Diabetic duration (y) | 10 (3‐15) | 10 (7‐17)‡ | 10 (3‐15) | 10 (5‐17)* | 10 (3‐15) | 11 (9‐20)‡ | 10 (4‐16) | 10 (8‐19)* |
| Fasting C‐peptide (ng/mL) | 0.99 (0.64, 1.61) | 1.14 (0.82, 1.78)* | 1.03 (0.69, 1.63) | 1.03 (0.70, 1.70) | 0.97 (0.66, 1.65) | 1.16 (0.80, 1.71)* | 0.98 (0.68, 1.64) | 1.40 (0.99, 2.23)† |
| C‐peptide at 1 h (ng/mL) | 1.67 (0.99, 2.99) | 2.05 (1.28, 3.13)† | 1.76 (1.05, 2.99) | 1.77 (1.07, 3.11) | 1.76 (1.07, 3.05) | 1.82 (1.03, 3.13) | 1.74 (1.04, 3.01) | 2.36 (1.52, 3.78)* |
| C‐peptide at 2 h (ng/mL) | 2.45 (1.45, 4.55) | 2.95 (1.66, 4.86)† | 2.56 (1.46, 4.69) | 2.61 (1.55, 4.54) | 2.58 (1.51, 4.59) | 2.57 (1.46, 4.81) | 2.55 (1.47, 4.59) | 3.01 (2.04, 4.87) |
| C‐peptide at 3 h (ng/mL) | 2.52 (1.50, 4.57) | 3.07 (1.82, 5.36)† | 2.60 (1.52, 4.70) | 2.76 (1.66, 4.82) | 2.64 (1.59, 4.69) | 2.76 (1.54, 5.37) | 2.63 (1.52, 4.72) | 3.37 (2.32, 5.37) |
| Fasting blood glucose (mmol/L) | 8.28 (6.28‐11.38) | 7.73 (5.81‐10.06)† | 8.19 (6.27‐11.10) | 7.94 (6.06‐11.06) | 8.16 (6.19‐11.28) | 7.92 (6.07‐10.40) | 8.17 (6.20‐11.13) | 7.65 (5.51‐9.87) |
| HbA1c (%) | 10.0 (8.5‐11.5) | 9.6 (8.2‐10.7)* | 9.9 (8.3‐11.3) | 9.8 (8.6‐11.3) | 9.9 (8.4‐11.5) | 9.4 (8.3‐10.7)* | 9.8 (8.4‐11.3) | 9.5 (8.2‐10.9) |
| Homa‐β (%) | 31.2 (16.0‐53.7) | 39.3 (20.6‐68.6)† | 31.9 (16.5‐57.5) | 35.9 (17.5‐56.8) | 32.8 (16.5‐55.0) | 36.2 (18.8‐67.7) | 32.7 (16.7‐55.6) | 54.9 (19.5‐77.9) |
| Homa‐IR | 0.96 (0.59‐1.42) | 1.03 (0.69‐1.51) | 0.98 (0.60‐1.44) | 0.98 (0.63‐1.47) | 0.98 (0.60‐1.44) | 1.01 (0.67‐1.48) | 0.98 (0.61‐1.44) | 1.2 (0.79‐1.70)* |
| Total cholesterol (mmol/L) | 4.56 (3.80‐5.36) | 4.39 (3.58‐5.17) | 4.5 (3.73‐5.30) | 4.55 (3.77‐5.36) | 4.59 (3.80‐5.39) | 4.22 (3.48‐5.01)† | 4.53 (3.77‐5.35) | 4.38 (3.51‐5.15) |
| HDL‐c (mmol/L) | 1.03 (0.86‐1.21) | 1.04 (0.89‐1.23) | 1.03 (0.87‐1.21) | 1.05 (0.87‐1.23) | 1.05 (0.89‐1.23) | 0.98 (0.82‐1.20)* | 1.04 (0.87‐1.21) | 0.99 (0.77‐1.41) |
| LDL‐c (mmol/L) | 2.87 (2.28‐3.52) | 2.75 (2.07‐3.39) | 2.83 (2.18‐3.48) | 2.85 (2.28‐3.47) | 2.9 (2.29‐3.55) | 2.57 (2.00‐3.20) † | 2.85 (2.24‐3.50) | 2.77 (1.84‐3.33) |
| Serum triglyceride (mmol/L) | 1.48 (1.08‐2.14) | 1.49 (1.11‐2.17) | 1.49 (1.11‐2.28) | 1.47 (1.06‐2.06) | 1.48 (1.07‐2.15) | 1.50 (1.15‐2.11) | 1.47 (1.08‐2.14) | 1.70 (1.17‐2.78) |
| γ‐Glutamyltransferase (units/L) | 26 (18‐39) | 24 (17‐38) | 26 (18‐40) | 24 (18‐37) | 25 (18‐39) | 26 (18‐40) | 25 (18‐39) | 24 (17‐32) |
| Serum creatinine (μmol/L) | 63 (53, 76) | 66 (54, 82) | 64 (52, 76) | 65 (53, 78) | 63 (53, 76) | 68 (54, 82)* | 64 (53, 76) | 73 (55, 90)† |
| UACR (mg/mmol) | 2.27 (1.14‐6.82) | 4.52 (2.27‐9.09)‡ | 2.27 (1.14‐6.82) | 3.41 (1.69‐8.47) | 2.27 (1.14‐8.47) | 3.41 (1.69‐8.47) | 3.02 (1.14‐8.47) | 5.66 (2.27‐17.05)* |
| eGFR (ml/min/1.73 m2) | 96.5 (86.0‐106.4) | 92.2 (74.8‐102.1)† | 96.5 (84.6‐107.2) | 94.6 (81.4‐103.0) | 96.5 (84.8‐106.5) | 91.7 (76.3‐100.7)† | 96.2 (84.2‐105.7) | 81.0 (60.3‐97.4)† |
| White blood cell count (×109/L) | 7.00 (5.85‐8.24) | 6.97 (5.97‐8.50) | 7.00 (5.80‐8.27) | 7.00 (6.02‐8.42) | 6.97 (5.85‐8.24) | 7.26 (6.10‐8.63) | 6.99 (5.80‐8.30) | 7.40 (6.50‐8.90) |
Among 882 hypertensive T2DM patients, 710 performed OGTT test to assess β‐cell function. In T2DM patients who received CCBs, 196 had complete C‐peptide before and after glucose administration during OGTT and 61 only had fasting C‐peptide levels and did not perform OGTT. In T2DM patients who received RAAS inhibitors, β‐blockers or diuretics, the corresponding number was 280/69, 118/41, or 37/10, respectively. Homa‐β and Homa‐IR were computed by Homeostasis Model Assessment (HOMA‐β; http://www.dtu.ox.ac.uk/homacalculator/) to evaluate β‐cell function and insulin resistance, respectively, using fasting C‐peptide and glucose levels in the plasma; HDL‐c, high‐density lipoprotein cholesterol; LDL‐c, low‐density lipoprotein cholesterol; UACR, urinary albumin‐to‐creatinine ratio; glomerular filtration rate (eGFR) was derived from serum creatinine by the Chronic Kidney Disease Epidemiology Collaboration (CKD‐EPI) equation.23 Chronic kidney disease was defined as eGFR less than 60 mL/min/1.73 m2.23 Significance between two groups:
P ≤ 0.05;
P ≤ 0.01;
P ≤ 0.001.
Figure 2.
The effect of antihypertensive treatment on β‐cell function and insulin resistance in type 2 diabetic patients with hypertension. A, HbA1c levels; B, Homa‐β index; and C, Homa‐IR index based on CCBs/ RAAS inhibitor treatment. Data were presented by median with interquartile range (IQR)
Patients who received diuretics had higher Homa‐IR index, while the other antihypertensive drugs including CCBs, RAAS inhibitors, and β‐blockers did not affect Homa‐IR in hypertensive patients (Table 1 & Figure 2).
3.3. β‐Cell function in hypertensive T2DM patients
With regard to β‐cell function index, plasma levels of C‐peptide before and after oral glucose administration during OGTT and Homa‐β index were all increased in CCB users compared with non‐CCB users (fasting C‐peptide: 1.14 ng/mL vs 0.99 ng/mL, n = 882, P = 0.015; C‐peptide at 1 hour: 2.05 ng/mL vs 1.67 ng/mL, n = 710, P = 0.002; C‐peptide at 2 hours: 2.95 ng/mL vs 2.45 ng/mL, n = 710, P = 0.007; C‐peptide at 3 hours: 3.07 ng/mL vs 2.52 ng/mL, n = 710, P = 0.004; Homa‐β: 39.3% vs 31.2%, n = 882, P = 0.002; Table 1 & Figure 3).
Figure 3.
The effect of antihypertensive treatment on C‐peptide levels in type 2 diabetic patients with hypertension. To assess β‐cell function, T2DM patients were administered 75 g of glucose orally. C‐peptide levels in the blood were measured before and 1, 2, 3 h after glucose administration. Data were presented by median with interquartile range (IQR). Among 882 hypertensive T2DM patients, 710 performed OGTT test to assess β‐cell function. In T2DM patients who received CCBs, 196 had complete C‐peptide before and after glucose administration during OGTT and 61 only had fasting C‐peptide levels and did not proceed OGTT. In T2DM patients who received RAAS inhibitors, β‐blockers or diuretics, the corresponding number was 280/69, 118/41, or 37/10, respectively. *P < 0.05 and †P < 0.01
Similar as CCBs, patients who received β‐blockers or diuretics had higher fasting C‐peptide values than non‐β‐blocker users when compared with their corresponding controls. By contrast, following oral administration of 75 g of glucose, C‐peptide levels did not differ between RAAS inhibitor users and non‐RAAS inhibitor users (Table 1 & Figure 3).
3.4. Multivariate‐adjusted analysis
Thereafter, we explored covariables that influenced HbA1c and fasting C‐peptide levels in hypertensive T2DM patients. Table S2 summarizes the covariables entered into the multivariate‐adjusted analyses.
Giving that some patients took more than one antihypertensive drugs, when we analyzed the association of one class of antihypertensive drug with HbA1c or β‐cell function, the other antihypertensive drugs (in class) were included as covariables. There was no significant interaction between different antihypertensive drugs for HbA1c and β‐cell function index. Thus, the main effects of antihypertensive drugs were reported.
Multivariate‐adjusted analysis revealed that use of CCBs or RAAS inhibitors was negatively associated with HbA1c levels (CCBs: 0.95 [95% CI: 0.92‐0.98], P = 0.002; RAAS inhibitors: 0.997 [95% CI: 0.97, 1.03], P = 0.844). After adjusting for common covariables and other classes of antihypertensive drugs, the C‐peptide levels before and 1, 2, and 3 hours of OGTT were 1.10‐, 1.18‐, 1.19‐, and 1.15‐fold increase in T2DM patients taking CCBs (P = 0.084 for fasting C‐peptide levels; P ≤ 0.024 for C‐peptide levels at 1, 2, and 3 hours after OGTT). Likewise, Homa‐β was positively related to CCB usage (P = 0.007). Except CCBs, none of other antihypertensive drugs was associated neither with C‐peptide levels nor with Homa‐β (P ≥ 0.19; Table 2).
Table 2.
Multivariate‐adjusted analysis of β‐cell function
| Antihypertensive medications | Fasting C‐peptide levelsa | C‐peptide levels at 1 hb | C‐peptide levels at 2 hb | C‐peptide levels at 3 hb | Homa‐β valuesa | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| Estimate (95% CI) | P | Estimate (95% CI) | P | Estimate (95% CI) | P | Estimate (95% CI) | P | Estimate (95% CI) | P | |
| CCBs (n = 257) | 1.10 (0.99, 1.22) | 0.084 | 1.18 (1.05, 1.32) | 0.005 | 1.19 (1.06, 1.33) | 0.004 | 1.15 (1.02, 1.29) | 0.024 | 1.22 (1.05, 1.40) | 0.007 |
| RAAS inhibitors (n = 349) | 1.01 (0.92, 1.11) | 0.82 | 1.02 (0.92, 1.13) | 0.75 | 1.04 (0.94, 1.15) | 0.45 | 1.06 (0.96, 1.18) | 0.25 | 1.00 (0.88, 1.14) | 0.97 |
| β‐Blockers (n = 159) | 1.00 (0.88, 1.15) | 0.95 | 0.94 (0.81, 1.08) | 0.38 | 0.91 (0.78, 1.05) | 0.20 | 0.95 (0.81, 1.10) | 0.49 | 1.01 (0.84, 1.21) | 0.93 |
| Diuretics (n = 47) | 1.12 (0.91, 1.39) | 0.29 | 1.15 (0.91, 1.45) | 0.24 | 1.00 (0.80, 1.27) | 0.97 | 1.06 (0.83, 1.35) | 0.63 | 1.22 (0.91, 1.63) | 0.19 |
Fasting C‐peptide levels and Homa‐β values were obtained from 882 patients;
C‐peptide levels at 1/2/3 h during OGTT were obtained from 710 patients.
3.5. Sensitivity analysis
To further dissect the relationship studied above, we repeated the same analysis in the multivariate‐adjusted linear regression model in which patients who took only one class of antihypertensive drug were included. Patients who did not take any antihypertensive medication were used as controls. HbA1c levels were negatively associated with the usage of CCBs, β‐blockers, or diuretics but not with RAAS inhibitors (CCBs: 0.92 [95% CI: 0.87, 0.96], P = 0.001; β‐blockers: 0.89 [95% CI, 0.84, 0.95], P < 0.001; diuretics: 0.82 [95% CI, 0.71, 0.95], P = 0.009; RAAS inhibitors: 0.98 [95% CI, 0.94, 1.02], P = 0.31). To be noted, only CCB treatment remained positively associated with C‐peptide levels in the entire OGTT test (Table 3).
Table 3.
Sensitivity analysis
| Antihypertensive medications | Fasting C‐peptide levelsa | C‐peptide levels at 1 hb | C‐peptide levels at 2 hb | C‐peptide levels at 3 hb | Homa‐β valuesa | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| Estimate (95% CI) | P | Estimate (95% CI) | P | Estimate (95% CI) | P | Estimate (95% CI) | P | Estimate (95% CI) | P | |
| CCBs (n = 87) | 1.16 (0.98, 1.38) | 0.083 | 1.26 (1.04, 1.53) | 0.020 | 1.24 (1.03, 1.51) | 0.027 | 1.21 (1.00, 1.48) | 0.056 | 1.32 (1.05, 1.65) | 0.016 |
| RAAS inhibitors (n = 180) | 1.06 (0.93, 1.21) | 0.41 | 1.07 (0.93, 1.23) | 0.37 | 1.07 (0.93, 1.23) | 0.35 | 1.10 (0.95, 1.27) | 0.21 | 1.06 (0.89, 1.27) | 0.49 |
| β‐Blockers (n = 57) | 1.14 (0.92, 1.42) | 0.24 | 1.08 (0.84, 1.39) | 0.54 | 1.01 (0.78, 1.29) | 0.96 | 1.06 (0.82, 1.37) | 0.68 | 1.30 (0.98, 1.74) | 0.072 |
| Diuretics (n = 8) | 1.44 (0.87, 2.39) | 0.15 | 1.32 (0.71, 2.46) | 0.38 | 1.10 (0.59, 2.06) | 0.76 | 1.08 (0.57, 2.06) | 0.81 | 1.67 (0.86, 3.27) | 0.13 |
Fasting C‐peptide levels and Homa‐β values were obtained from 667 patients;
C‐peptide levels at 1/2/3 h during OGTT were obtained from 543 patients.
3.6. Path analysis
Galectin‐3, CRP, TGF‐β1, and TNF‐α participate in inflammation and substantially contribute to β‐cell apoptosis,25, 26 whereas VEGF promotes angiogenesis to improve regeneration and function of β cells.29 To dissect whether these factors participated in the association of CCBs with β‐cell function or HbA1c, their plasma levels were determined by ELISA. After adjusting for confounding factors, the direct association of HbA1c was −0.050 (P = 0.002) with CCBs and −0.032 (P = 0.031) with Homa‐β (Figure 4). The indirect associations of HbA1c mediated via VEGF and Homa‐β amounted to −0.001 (P = 0.197) and −0.018 (P = 0.004) with CCBs, respectively (Figure 4).
Figure 4.
Path analysis. Path analysis differentiating direct associations of CCB treatment with HbA1c levels in hypertensive T2DM patients. Full and dotted lines indicate direct and indirect associations, respectively. Values are multivariable‐adjusted regression coefficients (β) with corresponding P values. Indirect associations are obtained by multiplying the two regression coefficients in the path
4. DISCUSSION
Hypertension is often accompanied with T2DM patients with the approximate prevalence of 70%. Antihypertensive treatments could reduce the incidence of new‐onset diabetes and diabetic vascular complications.6, 7 It is well known that β‐cell dysfunction and insulin resistance are the common pathological features of diabetes. However, limited studies have assessed whether antihypertensive drugs could protect β‐cell function. Therefore, we investigated this issue in this cross‐sectional study and the key finding is that (a) CCB treatment was negatively associated with HbA1c but positively associated with C‐peptide levels during OGTT; (b) the intimate association of CCB usage with HbA1c or C‐peptide levels was independently present in T2DM patients.
From the view of precision medicine, it would be beneficial to choose a suitable antihypertensive drug for T2DM patients that could control blood pressure in a desirable and stable manner and, in the meantime, improve β‐cell number and function. Up to date, a growing body of evidence has shown that blockade of RAAS system or calcium channels protected pancreatic β cell from apoptosis in human and mice islets and insulinoma cell lines.19 When we searched PubMed for relevant publications (panel) without limitations of publication date or language using as terms “diabetic patients” AND “anti‐hypertensive” AND “beta cells,” we could not find any relevant papers. Furthermore, whether RAAS inhibitors could modify beta‐cell function in subjects with impaired glucose tolerance remains controversial.13, 14
Calcium channel is an ion channel selective for calcium influx into the cells. It is comprised of voltage‐gated channels and ligand‐gated channels, which are universally expressed in different cells. Excessive Ca2+ influx is critically involved in endoplasmic reticulum stress‐induced β‐cell apoptosis.19, 30, 31 However, the relationship of calcium channel blockers and β‐cell function in hypertensive T2DM patients has been rarely assessed. Therefore, we initiated this study comprising 1250 T2DM patients. Providing that some patients took combined antihypertensive drugs, we further adjusted the usage of antihypertensive drugs by class on top of the common covariables. To our knowledge, our study is the first one to report that the use of calcium channel blockers was positively associated with C‐peptide levels before and after OGTT and negatively associated with HbA1c, indicating that this class of drugs could improve β‐cell function in T2DM patients.
In RAAS system, angiotensinogen is catabolized by renin to generate angiotensin I, which is further converted to angiotensin II by angiotensin‐converting enzyme (ACE). Acting through angiotensin II type 1 receptor (AT1R), it stimulates smooth muscle cell constriction and increases blood pressure. Except smooth muscle cells, AT1R is also expressed in endothelial cells, neurons, fibroblasts, T lymphocytes, and macrophages, and AT1R deficiency in these cells accelerates vascular hypertrophy and kidney injury in mice models.32
Recent studies identified AT1R expression in pancreatic β cells. Activation of AT1R signaling pathway promotes cholesterol accumulation via downregulation of ABCA1 expression and reduces proinsulin biosynthesis in cultivated islets and insulinoma cell lines in vitro.33, 34 Blockade of RAAS activation by ACE inhibitors and AT1R blockers decreased the incident diabetes in the subjects with impaired glucose tolerance and in general populations.6, 7 Moreover, the increase in aldosterone was associated with insulin resistance in nondiabetic participants.12 Nevertheless, in human subjects with impaired glucose tolerance (n = 16), short‐term administration of AT1R antagonist valsartan for 6 weeks did not increase glucose‐stimulated insulin secretion following glucose infusion.13 After 3 months of candesartan medication to the subjects with impaired glucose tolerance (n = 6), serum insulin levels in the blood were increased at 30 minutes after oral administration of 75 g glucose.14 In our study, inhibition of RAAS did not relate to β‐cell function, that is, fasting C‐peptide levels and C‐peptide levels during OGTT. Nor use of RAAS inhibitors was associated with Homa‐IR despite its negative relationship with HbA1c in T2DM patients.
Beta receptors are expressed in cardiomyocytes and smooth muscles in airways, arteries, kidneys, and sympathetic nervous system. Stimulated by epinephrine, beta receptors participate in vasoconstriction and stress response. β‐Blockers competitively bind to the beta receptor of epinephrine to weaken the effects of stress hormones. Except RAAS inhibitors and calcium channel blockers, we also analyzed the relationship of β‐blockers or diuretics and β‐cell function. Neither of these treatments was associated with Homa‐β.
Our findings have to be interpreted within the context of some potential limitations. First, this is a cross‐sectional study. Whether long‐term use of calcium channels blockers could improve β‐cell function in T2DM patients still waits for further confirmation from longitudinal studies after years of follow‐up. Second, as calcium channels are widely expressed, we assumed that calcium channel blockers could enhance blood flow in pancreas as well as improve β‐cell function in T2DM patients. Third, we did not specify the generations of calcium channel blockers. And fourth, the dosages of antihypertensive drugs were not considered in the analysis. And fifth, the number of T2DM patients was relatively small, especially for the patients using diuretics, all of which reduced the power of the study. Therefore, our results wait for further investigation in T2DM patient population with bigger sample size and in prospective studies.
In conclusion, the present cross‐sectional study showed that CCBs were negatively associated with HbA1c levels in hypertensive T2DM patients. On top of that, administration of CCBs rather than other antihypertensive drugs was positively associated with β‐cell function, that is, C‐peptide levels at fasting stage and upon OGTT challenge in these patients. Thus, whether CCB treatment could improve β‐cell function in T2DM patients is to be determined for precision medicine.
CONFLICT OF INTEREST
None.
AUTHOR CONTRIBUTIONS
Dong Zhao, Cai‐Guo Yu, Sha‐Sha Yuan, Ning Zhang, Yuan‐Yuan Zhang collected the data; Yu Cao performed data analysis and participated in manuscript preparation; Ying‐Mei Feng designed the study, analyzed the data and drafted the manuscript.
Supporting information
ACKNOWLEDGMENTS
Hereby, we thank Jia‐Nan Lang and Li‐Jie Zhang for technical assistance.
Zhao D, Cao Y, Yu C‐G, et al. The association of calcium channel blockers with β‐cell function in type 2 diabetic patients: A cross‐sectional study. J Clin Hypertens. 2019;21:638–647. 10.1111/jch.13517
Zhao and Cao are equally contributed to the study.
Funding information
The study received support from the National Natural Science Funding in China (81470566 & 81670765) and the Science & Technology Commission of Tongzhou District (KJ2018CX007) for Ying‐Mei Feng.
REFERENCES
- 1. Giovannucci E, Harlan DM, Archer MC, et al. Diabetes and cancer: a consensus report. CA Cancer J Clin. 2010;60:207‐221. [DOI] [PubMed] [Google Scholar]
- 2. Hu C, Jia W. Diabetes in China: epidemiology and genetic risk factors and their clinical utility in personalized medication. Diabetes. 2018;67:3‐11. [DOI] [PubMed] [Google Scholar]
- 3. Wang L, Gao P, Zhang M, et al. Prevalence and ethnic pattern of diabetes and prediabetes in China in 2013. JAMA. 2017;317:2515‐2523. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Zoppini G, Cacciatori V, Raimondo D, et al. Prevalence of cardiovascular autonomic neuropathy in a cohort of patients with newly diagnosed type 2 diabetes: the Verona Newly Diagnosed type 2 Diabetes Study (VNDS). Diabetes Care. 2015;38:1487‐1493. [DOI] [PubMed] [Google Scholar]
- 5. Ferrannini E, Cushman WC. Diabetes and hypertension: the bad companions. Lancet. 2012;380:601‐610. [DOI] [PubMed] [Google Scholar]
- 6. Bosch J, Yusuf S, Gerstein HC, et al. Effect of ramipril on the incidence of diabetes. N Engl J Med. 2006;355:1551‐1562. [DOI] [PubMed] [Google Scholar]
- 7. McMurray JJ, Holman RR, Haffner SM, et al. Effect of valsartan on the incidence of diabetes and cardiovascular events. N Engl J Med. 2010;362:1477‐1490. [DOI] [PubMed] [Google Scholar]
- 8. Barnett AH, Bain SC, Bouter P, et al. Angiotensin‐receptor blockade versus converting‐enzyme inhibition in type 2 diabetes and nephropathy. N Engl J Med. 2004;351:1952‐1961. [DOI] [PubMed] [Google Scholar]
- 9. Remonti LR, Dias S, Leitao CB, et al. Classes of antihypertensive agents and mortality in hypertensive patients with type 2 diabetes‐Network meta‐analysis of randomized trials. J Diabetes Complications. 2016;30:1192‐1200. [DOI] [PubMed] [Google Scholar]
- 10. Nguyen G. Renin, (pro)renin and receptor: an update. Clin Sci. 2011;120:169‐178. [DOI] [PubMed] [Google Scholar]
- 11. Luther JM, Brown NJ. The renin‐angiotensin‐aldosterone system and glucose homeostasis. Trends Pharmacol Sci. 2011;32:734‐739. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Joseph JJ, Echouffo‐Tcheugui JB, Kalyani RR, et al. Aldosterone, renin, and diabetes mellitus in African Americans: the Jackson Heart Study. J Clin Endocrinol. Metab. 2016;101:1770‐1778. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Bokhari S, Israelian Z, Schmidt J, Brinton E, Meyer C. Effects of angiotensin II type 1 receptor blockade on beta‐cell function in humans. Diabetes Care. 2007;30:181. [DOI] [PubMed] [Google Scholar]
- 14. Suzuki K, Nakagawa O, Aizawa Y. Improved early‐phase insulin response after candesartan treatment in hypertensive patients with impaired glucose tolerance. Clin Exp Hypertens. 2008;30:309‐314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Yang YS, Lim MH, Lee SO, et al. Fimasartan increases glucose‐stimulated insulin secretion in patients with type 2 diabetes and hypertension compared with amlodipine. Diabetes Obes Metab. 2018;20:1670‐1677. [DOI] [PubMed] [Google Scholar]
- 16. Garcia‐Delgado N, Velasco M, Sanchez‐Soto C, Diaz‐Garcia CM, Hiriart M. Calcium channels in postnatal development of rat pancreatic beta cells and their role in insulin secretion. Front Endocrinol. 2018;9:40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Nita II, Hershfinkel M, Kantor C, Rutter GA, Lewis EC, Sekler I. Pancreatic β‐cell Na+ channels control global Ca2+ signaling and oxidative metabolism by inducing Na+ and Ca2+ responses that are propagated into mitochondria. FASEB J. 2014;28:3301‐3312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Tong X, Kono T, Evans‐Molina C. Nitric oxide stress and activation of AMP‐activated protein kinase impair beta‐cell sarcoendoplasmic reticulum calcium ATPase 2b activity and protein stability. Cell Death Dis. 2015;6:e1790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Zhou Y, Sun P, Wang T, Chen K, Zhu W, Wang H. Inhibition of calcium influx reduces dysfunction and apoptosis in lipotoxic pancreatic beta‐cells via regulation of endoplasmic reticulum stress. PLoS ONE. 2015;10:e132411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Zhang HH, Ma XJ, Wu LN, et al. Sirtuin‐3 (SIRT3) protects pancreatic beta‐cells from endoplasmic reticulum (ER) stress‐induced apoptosis and dysfunction. Mol Cell Biochem. 2016;420:95‐106. [DOI] [PubMed] [Google Scholar]
- 21. Syeda K, Mohammed AM, Arora DK, Kowluru A. Glucotoxic conditions induce endoplasmic reticulum stress to cause caspase 3 mediated lamin B degradation in pancreatic beta‐cells: protection by nifedipine. Biochem Pharmacol. 2013;86:1338‐1346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. O'Brien E, Asmar R, Beilin L, et al. European Society of Hypertension recommendations for conventional, ambulatory and home blood pressure measurement. J Hypertens. 2003;21:821‐848. [DOI] [PubMed] [Google Scholar]
- 23. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150:604‐612. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Johnson R, McNutt P, MacMahon S, Robson R. Use of the Friedewald formula to estimate LDL‐cholesterol in patients with chronic renal failure on dialysis. Clin Chem. 1997;43:2183‐2184. [PubMed] [Google Scholar]
- 25. Saksida T, Nikolic I, Vujicic M, et al. Galectin‐3 deficiency protects pancreatic islet cells from cytokine‐triggered apoptosis in vitro. J Cell Physiol. 2013;228:1568‐1576. [DOI] [PubMed] [Google Scholar]
- 26. Chan C, Wang C, Chen L, Hsu N, Tian F, Hsieh PS. Importance of NADPH oxidase‐mediated redox signaling in the detrimental effect of crp on pancreatic insulin secretion. Free Radic Biol Med. 2017;112:200‐211. [DOI] [PubMed] [Google Scholar]
- 27. Yang J, Sun Y, Xu F, et al. Involvement of estrogen receptors in silibinin protection of pancreatic β‐cells from TNFα‐ or IL‐1β‐induced cytotoxicity. Biomed Pharmacother. 2018;102:344‐353. [DOI] [PubMed] [Google Scholar]
- 28. You W, Wang K, Yu C, Song L. Baicalin prevents tumor necrosis factor‐α‐induced apoptosis and dysfunction of pancreatic β‐cell line Min6 via upregulation of miR‐205. J Cell Biochem. 2018;119:8547‐8554. [DOI] [PubMed] [Google Scholar]
- 29. Lui KO. VEGF‐A: the inductive angiogenic factor for development, regeneration and function of pancreatic beta cells. Curr Stem Cell Res Ther. 2014;9:396‐400. [DOI] [PubMed] [Google Scholar]
- 30. Wang Y, Gao L, Li Y, Chen H, Sun Z. Nifedipine protects INS‐1 beta‐cell from high glucose‐induced ER stress and apoptosis. Int J Mol Sci. 2011;12:7569‐7580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Arora DK, Mohammed AM, Kowluru A. Nifedipine prevents etoposide‐induced caspase‐3 activation, prenyl transferase degradation and loss in cell viability in pancreatic beta‐cells. Apoptosis. 2013;18:638‐8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Eguchi S, Kawai T, Scalia R, Rizzo V. Understanding angiotensin II type 1 receptor signaling in vascular pathophysiology. Hypertension. 2018;71:804‐810. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Lau T, Carlsson PO, Leung PS. Evidence for a local angiotensin‐generating system and dose‐dependent inhibition of glucose‐stimulated insulin release by angiotensin II in isolated pancreatic islets. Diabetologia. 2004;47:240‐248. [DOI] [PubMed] [Google Scholar]
- 34. Lyu J, Imachi H, Fukunaga K, et al. Angiotensin II induces cholesterol accumulation and impairs insulin secretion by regulating ABCA1 in beta cells. J Lipid Res. 2018;59:1906‐1915. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials