ORIGINAL
Correlation between serum C-peptide-releasing effects and the risk of elevated uric acid in type 2 diabetes mellitus
2022 Volume 69 Issue 7 Pages 773-784
Details
2022 Volume 69 Issue 7 Pages 773-784
Our study aimed to investigate the C-peptide-releasing effect associated with the risk of elevated serum uric acid (SUA) levels in patients with type 2 diabetes mellitus (T2DM). In the cross-sectional study, 345 patients with T2DM hospitalized at the First Affiliated Hospital of Harbin Medical University were consecutively enrolled, and their baseline data were collected. The study design used two parameters for C-peptide releasing effects: the multiplication effect of 1 h postprandial C-peptide to fasting C-peptide ratio (1hCp/FCp) and 2hCp/FCp; the incremental effect of 1hCp minus FCp (1hΔCp) and 2hΔCp. The patients with T2DM in the upper quartiles of SUA had higher FCp, 1hCp, 1hΔCp, 2hCp, and 2hΔCp. Multiple linear regression analysis revealed that after adjusting all the confounding factors, the serum C-peptide including 1hCp (β = 5.14, p = 0.036), 1hΔCp (β = 7.80, p = 0.010), 2hCp (β = 4.27, p = 0.009) and 2hΔCp (β = 5.20, p = 0.005) were still positively correlated with SUA levels in patients with T2DM. In female patients, only the 2hCp (β = 4.78, p = 0.017) and 2hΔCp (β = 5.28, p = 0.019) were associated with SUA level; however, in male patients, no C-peptide parameter was associated with SUA levels in T2DM (all p > 0.05). Within a certain range, the elevated SUA levels might be associated with the better C-peptide incremental effect of islet β cell function in T2DM, especially in female patients.
SERUM URIC ACID (SUA) levels are positively associated with multiple metabolic abnormalities including obesity, hypertension, hyperglycemia, and hyperlipidemia [1]. Elevated SUA levels are independently correlated with the incidences of hypertension and metabolic syndrome (MetS) in some Asian countries including China, Japan, and Korea [2-5]. For patients with type 2 diabetes mellitus (T2DM), higher SUA levels are associated with the lower estimated glomerular filtration rate (eGFR) and higher prevalence of diabetic nephropathy, but not diabetic retinopathy [6]. Moreover, in general, the decline in SUA levels slows the progression of diabetic kidney disease [7]. SUA levels are positively associated with a higher cardiovascular mortality risk irrespective of the presence of MetS [8]. A meta-analysis of cohort studies involving 62,834 participants reported that SUA levels were positively associated with the incidence of impaired fasting glucose and T2DM [9]. In patients with T2DM, hyperuricemia was positively correlated with hyperinsulinemia [10], fasting C-peptide (FCp) level [11], and insulin-resistant syndrome [12].
C-peptide, the 31 amino-acid residues peptide formed during cleavage of insulin from proinsulin, is co-secreted from pancreatic β cells in equimolar amounts with insulin [13]. C-peptide may serve as a simple predictor of an insulin resistance condition [14, 15] and exhibits a simple correlation with components of MetS [16]. FCp was correlated with SUA level and might be an independent risk factor of elevated SUA level in T2DM [17]. The SUA level was shown to be significantly correlated with C-peptide and also regarded as an independent risk factor of islet β cell functions in female patients with type 2 diabetes [18], but other studies showed there were no significant correlations between SUA level and insulin secretory capacity in the females [19].
The detecting level of serum insulin was particularly interfered by exogenous insulin, internal insulin antibody and proinsulin levels. However, the level of C-peptide assay was not influenced by the above factors and could also reflect the function of islet β cells. Therefore, we usually used the oral glucose tolerance test (OGTT) combined with a C-peptide release assay to evaluate the insulin secretion capacity of pancreatic β cells in patients with diabetes. After OGTT combined with the C-peptide release test, the C-peptide level of the healthy control group increased to the peak at postprandial 0.5–1 h, which was 4–6 times as much as fasting C-peptide, and gradually returned to normal at postprandial 3 h. The C-peptide level of T2DM patients increased to the peak at postprandial 2–3 h, which was generally lower than 5 times of fasting C-peptide [20, 21].
Based on the above mentioned characteristics of C-peptide release in T2DM patients, we applied two C-peptide parameters, the incremental and multiplication effect, to determine which parameter can better predict the long-term risk of serum uric acid: 1) The incremental effect was postprandial C-peptide minus fasting C-peptide, which could quickly determine the extra insulin secretion capacity of islet β cell in postprandial state compared with fasting state. The incremental effect includes 1hCp minus FCp (1hΔCp) and 2hCp minus FCp (2hΔCp). 2) The multiplication effect was postprandial C-peptide to fasting C-peptide ratio, and its value could also reflect the insulin secretion capacity of islet β cell, which may be more intuitive than the incremental effect. The multiplication effect includes 1 h postprandial C-peptide to FCp ratio (1hCp/FCp) and 2 h postprandial C-peptide to FCp ratio (2hCp/FCp). The cross-sectional study aimed to investigate which C-peptide releasing effect is associated with the risk of elevated SUA levels in patients with T2DM.
This cross-sectional study, conducted from August 2019 to June 2020, involved 345 consecutive patients with T2DM who were hospitalized for treatment at the First Affiliated Hospital of Harbin Medical University. The 1999 World Health Organization criteria were used for T2DM diagnosis (fasting plasma glucose (FPG) ≥7.0 mmol/L or 2 h-postprandial plasma glucose ≥11.1 mmol/L). Hyperuricemia was defined as the SUA level of 7.0 mg/dL (420 umol/L) or higher regardless of sex. Subjects who were pregnant; had diabetic ketoacidosis, ketonuria or acute kidney injury, pancreatitis, or infection, had a malignant disease, or were undergoing dialysis were excluded. The study protocol was approved by the Ethics Committee of The First Affiliated Hospital of Harbin Medical University (Clinical Trials Registry number: ChiCTR2100047648).
Clinical dataAll patients underwent a standardized interview to collect the clinical data including diabetes duration; smoking history; alcohol intake; medication treatment history including only insulin, only oral hypoglycemic medicine, oral + insulin hypoglycemic medicine, calcium channel blocker (CCB), angiotensin-converting enzyme inhibitor/angiotensin II receptor blocker (ACEI/ARB), aspirin and statins and other related variables. Smoking was defined as using >1 cigarette/day for at least one year and still currently smoking; drinking was defined as the consumption of liquor, beer, or wine on two days/week for one year. Blood pressure, body weight, and height were assessed. Body mass index (BMI) was calculated as weight (kg) divided by the square of height (m2). After 10 minutes of rest, blood pressure was measured three times using a mercury sphygmomanometer, with the patient in a seated position. The values of the three measurements were then averaged. Hypertension was diagnosed as systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg or the use of antihypertensive drugs.
Biochemical measurementsVenous blood samples were collected in the morning after overnight fasting for at least 8–10 h, refrigerated immediately after phlebotomy, and centrifuged within 2 h of collection. SUA, triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-c), low-density lipoprotein cholesterol (LDL-C), creatinine, and FPG levels were measured using a Hitachi automated analyzer (NO.87128, Hitachi, 149 Ltd, Tokyo, Japan). Glycated haemoglobin (HbA1c) was measured by high-performance liquid chromatography (VARIANT II, BIO-RAD, Hercules, CA, USA). Serum C-peptide was assessed by chemiluminescence immunoassay (AIA-1800ST, TOSOH, kaisei-machi, Yamaguchi, Japan). The GFR was estimated based on the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.
Statistical analysisStatistical analyses were performed using IBM SPSS version 23.0 (IBM Corp., Armonk, NY, USA). P-value was 2-tailed and considered statistically significant at p < 0.05. Continuous data are summarized as mean ± standard deviation (SD) or medians (interquartile ranges) for skewed variables. Categorical variables are presented as percentages (%) and were assessed using chi-square test, as appropriate. One-way analysis of covariance (ANOVA) and an independent-samples t-test were applied to compare normally distributed continuous variables. The Mann-Whitney U test and Kruskal-Wallis H test were used for continuous variables not distributed normally.
Pearson’s correlation was adopted to identify the correlations among the normally distributed variables. The correlations between the C-peptide-related parameters with SUA were analyzed using Spearman’s rank correlation. Multiple linear regression analysis was applied to evaluate the association between the serum C-peptide releasing effects and SUA levels while adjusting for other confounding factors. Three models were constructed to avoid inference: Model 1 was unadjusted; Model 2 included adjustments for sex, age, smoking, drinking, BMI, TG, HDL-c, and eGFR. Model 3 included the variables of Model 2 and the duration of diabetes, HbA1c, oral medicine, oral + insulin, CCB, ACEI/ARB, and FPG.
The clinical characteristics of the T2DM patients with vs. without hyperuricemia are summarized in Table 1. Compared to patients without hyperuricemia, patients with hyperuricemia were more often male; were young; had a smoking history; took more oral and CCB medicines; and had high BMI, TG, creatinine, and FCp levels as well as lower levels of eGFR (all p < 0.05). However, no significant differences were found in diabetes duration, the prevalence of hypertension, drinking status, levels of HbA1c, TC, HDL-c, LDL-c, FPG, and intake of other medicines between both groups (all p > 0.05). Furthermore, the serum C-peptide-releasing effects including 1hCp, 1hΔCp, 2hCp and 2hΔCp concentrations in the hyperuricemia group were significantly higher than those in the non-hyperuricemia group (all p < 0.05), whereas 1hCp/FCp and 2hCp/FCp did not differ significantly between the groups. All the patients were divided into two groups by gender (Table 2). Compared to the male patients, the female patients tended to be older and have higher HDL-c level and 2hCp/FCp but the lower BMI and lower levels of creatinine, eGFR, SUA, and FCp; the prevalence of smoking and drinking was also lower (all p < 0.05).
| Non-Hyperuricemia | Hyperuricemia | All (n = 345) | p-value | |
|---|---|---|---|---|
| UA <420 (n = 289) | UA ≥420 (n = 56) | |||
| male (n [%]) | 140 (48.44) | 43 (76.79) | 183 (53.00) | <0.001 |
| Age (year) | 54.92 ± 11.21 | 50.91 ± 11.55 | 54.27 ± 11.34 | 0.015 |
| Duration (year) | 8.00 (3.00–14.00) | 5.00 (2.00–11.00) | 7.00 (3.00–14.00) | 0.129 |
| Hypertension (n [%]) | 113 (39.10) | 28 (50.00) | 141 (40.87) | 0.129 |
| Smoking (n [%]) | 84 (29.07) | 27 (48.21) | 111 (32.20) | 0.005 |
| drinking (n [%]) | 82 (28.37) | 23 (41.07) | 105 (30.4) | 0.059 |
| BMI (kg/m2) | 25.57 ± 3.38 | 26.95 ± 3.75 | 25.79 ± 3.47 | 0.019 |
| HbA1c (%) | 8.68 ± 1.79 | 8.59 ± 1.84 | 8.67 ± 1.80 | 0.737 |
| TC (mmol/L) | 4.94 ± 1.23 | 5.21 ± 1.07 | 4.99 ± 1.20 | 0.128 |
| TG (mmol/L) | 1.78 (1.18–2.61) | 2.49 (1.47–4.20) | 1.81 (1.29–2.89) | 0.002 |
| HDL-c (mmol/L) | 1.08 (0.92–1.25) | 1.02 (0.94–1.20) | 1.07 (0.92–1.25) | 0.454 |
| LDL-c (mmol/L) | 3.04 ± 0.80 | 3.23 ± 0.75 | 3.07 ± 0.79 | 0.101 |
| Creatinine (mmol/L) | 57.30 (47.58–67.53) | 75.40 (63.40–88.80) | 59.40 (48.70–70.90) | <0.001 |
| eGFR (mL·min–1·1.73(m2)–1) | 105.78 (98.26–115.74) | 103.85 (87.62–117.30) | 105.18 (97.10–116.15) | 0.034 |
| SUA (umol/L) | 303.83 ± 66.49 | 474.58 ± 53.35 | 331.54 ± 90.18 | <0.001 |
| FPG (mmol/L) | 6.80 (5.93–7.80) | 6.90 (6.10–7.80) | 6.90 (6.00–7.80) | 0.688 |
| FCp (ng/mL) | 1.50 (1.00–2.00) | 2.00 (1.50–3.10) | 1.60 (1.10–2.10) | <0.001 |
| 1hCp (ng/mL) | 3.85 (2.65–5.80) | 5.20 (3.60–7.42) | 4.00 (2.90–6.10) | <0.001 |
| 1hΔCp (ng/mL) | 2.30 (1.40–3.80) | 3.20 (1.79–4.50) | 2.46 (1.50–3.90) | 0.026 |
| 1hCp/FCp | 2.78 (2.12–3.52) | 2.57 (2.03–3.25) | 2.73 (2.09–3.45) | 0.193 |
| 2hCp (ng/mL) | 5.48 (3.70–8.30) | 7.80 (4.90–10.90) | 5.70 (3.89–8.60) | <0.001 |
| 2hΔCp (ng/mL) | 4.10 (2.52–6.33) | 5.48 (2.90–8.00) | 4.20 (2.60–6.60) | 0.011 |
| 2hCp/FCp | 3.94 (3.00–5.21) | 3.29 (2.71–4.50) | 3.88 (2.94–5.14) | 0.051 |
| Insulin (n [%]) | 32 (11.07) | 3 (5.36) | 35 (10.14) | 0.195 |
| Oral medicine (n [%]) | 103 (35.64) | 28 (50.00) | 131 (37.97) | 0.043 |
| Oral + Insulin (n [%]) | 156 (53.98) | 25 (44.64) | 181 (52.46) | 0.200 |
| CCB (n [%]) | 56 (19.38) | 19 (33.93) | 75 (21.74) | 0.016 |
| ACEI/ARB (n [%]) | 69 (23.88) | 19 (33.93) | 88 (25.51) | 0.114 |
| Aspirin (n [%]) | 87 (30.10) | 14 (25.00) | 101 (29.28) | 0.442 |
| statins (n [%]) | 183 (63.32) | 31 (55.36) | 214 (62.03) | 0.261 |
Baseline characteristics of participants according to the presence or not of hyperuricemia. Data were expressed as the mean ± SD, median (interquartile range) or n (%) according to the distribution of variables. Comparison between Non-Hyperuricemia and Hyperuricemia group was shown with p value. Abbreviations: SUA, serum uric acid; BMI, body mass index; HbA1c, glycosylated hemoglobin; TC, total cholesterol; TG, triglycerides; HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate; FPG, fasting plasma glucose; FCp, fasting C-peptide; 1hCp, 1 h postprandial C-peptide; 2hCp, 2 h postprandial C-peptide; CCB, calcium channel blocker; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; SD, standard deviation.
| Male (n = 183) | Female (n = 162) | p-value | |
|---|---|---|---|
| Age (year) | 52.80 ± 10.89 | 55.93 ± 11.65 | 0.010 |
| Duration (year) | 8.00 (2.00–15.00) | 7.00 (3.00–12.25) | 0.729 |
| Hypertension (n [%]) | 74 (40.44) | 67 (41.36) | 0.913 |
| Smoking (n [%]) | 93 (50.82) | 18 (11.11) | <0.001 |
| drinking (n [%]) | 86 (46.99) | 19 (11.73) | <0.001 |
| BMI (kg/m2) | 26.31 ± 3.22 | 25.27 ± 3.64 | 0.016 |
| HbA1c (%) | 8.64 ± 1.84 | 8.70 ± 1.75 | 0.775 |
| TC (mmol/L) | 4.87 ± 1.07 | 5.12 ± 1.33 | 0.052 |
| TG (mmol/L) | 1.88 (1.35–3.02) | 1.81 (1.15–2.56) | 0.090 |
| HDL-c (mmol/L) | 1.00 (0.87–1.15) | 1.13 (0.99–1.32) | <0.001 |
| LDL-c (mmol/L) | 3.02 ± 0.73 | 3.14 ± 0.86 | 0.170 |
| Creatinine (mmol/L) | 68.20 (60.15–77.40) | 48.85 (44.68–57.60) | <0.001 |
| eGFR (mL·min–1·1.73 (m2)–1) | 107.84 (98.22–119.02) | 103.54 (96.02–111.67) | 0.016 |
| SUA (umol/L) | 362.20 ± 85.77 | 296.91 ± 82.36 | <0.001 |
| FPG (mmol/L) | 6.90 (6.05–7.90) | 6.80 (5.90–7.70) | 0.273 |
| FCp (ng/mL) | 1.60 (1.20–2.20) | 1.50 (0.90–2.09) | 0.023 |
| 1hCp (ng/mL) | 4.00 (2.94–6.10) | 3.99 (2.62–5.94) | 0.388 |
| 1hΔCp (ng/mL) | 2.50 (1.40–3.90) | 2.45 (1.60–3.90) | 0.706 |
| 1hCp/FCp | 2.62 (1.88–3.37) | 2.83 (2.23–3.51) | 0.074 |
| 2hCp (ng/mL) | 5.70 (3.90–8.30) | 5.65 (3.70–9.33) | 0.851 |
| 2hΔCp (ng/mL) | 4.10 (2.45–6.10) | 4.58 (2.70–7.40) | 0.283 |
| 2hCp/FCp | 3.54 (2.74–4.68) | 4.12 (3.10–5.51) | 0.001 |
| Insulin (n [%]) | 19 (10.38) | 16 (9.88) | 0.877 |
| Oral medicine (n [%]) | 77 (42.08) | 54 (33.33) | 0.095 |
| Oral + Insulin (n [%]) | 89 (48.63) | 92 (56.79) | 0.130 |
| CCB (n [%]) | 42 (22.95) | 33 (20.37) | 0.562 |
| ACEI/ARB (n [%]) | 50 (27.32) | 38 (23.46) | 0.411 |
| Aspirin (n [%]) | 60 (32.79) | 41 (25.31) | 0.128 |
| statins (n [%]) | 117 (63.93) | 97 (59.88) | 0.438 |
Clinical characteristics of participants according to gender. Data were expressed as the mean ± SD, median (interquartile range) or n (%) according to the distribution of variables. Comparison between male group and female group was shown with p value. Abbreviations: SUA, serum uric acid; BMI, body mass index; HbA1c, glycosylated hemoglobin; TC, total cholesterol; TG, triglycerides; HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate; FPG, fasting plasma glucose; FCp, fasting C-peptide; 1hCp, 1 h postprandial C-peptide; 2hCp, 2 h postprandial C-peptide; CCB, calcium channel blocker; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; SD, standard deviation.
The characteristics of the patients based on SUA quartiles are shown in Table 3. The patients with T2DM in the upper quartiles of SUA are more likely to be males, of younger age, smokers, and drinkers; with a higher prevalence of oral medicine treatment, higher levels of BMI, TG, creatinine, and FCp; but lower levels of HbA1c and HDL-c (all p < 0.05). Additionally, the serum C-peptide including 1hCp (p < 0.001), 1hΔCp (p = 0.011), 2hCp (p < 0.001) and 2hΔCp (p = 0.013) were obviously elevated together with the increase in SUA quartiles, but not for 1hCp/FCp (p = 0.753) and 2hCp/FCp (p = 0.465). Moreover, patients turn to have higher SUA levels with elevating quartiles of 1hCp (Q1: 295.77 ± 73.88, Q2: 328.93 ± 89.51, Q3: 340.31 ± 93.62, Q4: 356.54 ± 92.85, umol/L, p = 0.003), 1hΔCp (Q1: 309.47 ± 87.20, Q2: 320.12 ± 81.70, Q3: 338.50 ± 94.79, Q4: 362.37 ± 90.59, umol/L, p < 0.001), 2hCp (Q1: 301.22 ± 83.09, Q2: 325.77 ± 79.81, Q3: 336.72 ± 95.43, Q4: 362.95 ± 92.20, umol/L, p < 0.001), and 2hΔCp (Q1: 311.39 ± 86.87, Q2: 327.13 ± 87.85, Q3: 331.23 ± 90.05, Q4: 357.95 ± 91.26, umol/L, p = 0.007), whereas this was not the case for 1hCp/FCp (p = 0.804) and 2hCp/FCp (p = 0.549) (Fig. 1).
| Q1 (n = 86) | Q2 (n = 87) | Q3 (n = 86) | Q4 (n = 86) | p-value | |
|---|---|---|---|---|---|
| male (n [%]) | 22 (25.58) | 44 (50.57) | 54 (62.79) | 63 (73.26) | <0.001 |
| Age (year) | 57.0 ± 11.6 | 55.9 ± 10.3 | 52.9 ± 10.9 | 51.3 ± 11.9 | 0.003 |
| Duration (year) | 6.0 (3.0–13.8) | 8.0 (4–14.3) | 8.0 (3–15) | 5.0 (1–12) | 0.127 |
| Hypertension (n [%]) | 33 (23.5) | 32 (22.7) | 38 (26.9) | 38 (26.9) | 0.658 |
| smoking (n [%]) | 15 (13.5) | 27 (24.3) | 31 (27.9) | 38 (34.3) | 0.002 |
| drinking (n [%]) | 12 (11.4) | 28 (26.7) | 32 (30.5) | 33 (31.4) | 0.001 |
| BMI (kg/m2) | 24.1 ± 3.1 | 25.9 ± 3.1 | 26.6 ± 3.6 | 26.6 ± 3.5 | <0.001 |
| HbA1c (%) | 9.2 ± 1.9 | 8.5 ± 1.6 | 8.5 ± 1.8 | 8.5 ± 1.8 | .025 |
| TC (mmol/L) | 4.97 ± 1.38 | 4.80 ± 1.01 | 5.14 ± 1.33 | 5.05 ± 1.04 | 0.298 |
| TG (mmol/L) | 1.44 (0.94–2.21) | 1.69 (1.18–2.45) | 1.92 (1.38–2.94) | 2.49 (1.48–4.09) | <0.001 |
| HDL-c (mmol/L) | 1.17 (1.03–1.42) | 1.05 (0.91–1.22) | 1.06 (0.90–1.19) | 1.00 (0.86–1.14) | <0.001 |
| LDL-c (mmol/L) | 2.98 ± 0.91 | 3.01 ± 0.70 | 3.20 ± 0.80 | 3.11 ± 0.76 | 0.242 |
| Creatinine (mmol/L) | 50.10 (44.15–59.88) | 56.80 (47.68–67.83) | 61.40 (50.00–70.88) | 71.70 (59.25–80.20) | <0.001 |
| eGFR (mL·min–1·1.73 (m2)–1) | 104.98 (96.72–114.98) | 107.38 (99.88–116.26) | 105.29 (99.88–116.20) | 103.85 (90.88–116.72) | 0.172 |
| SUA (umol/L) | 223.17 ± 33.24 | 296.11 ± 18.15 | 357.12 ± 20.23 | 450.18 ± 54.68 | <0.001 |
| FPG (mmol/L) | 6.70 (5.90–8.10) | 6.85 (5.80–7.93) | 7.00 (6.28–7.73) | 6.90 (6.10–7.70) | 0.935 |
| FCp (ng/mL) | 1.45 (0.80–2.00) | 1.40 (1.00–1.71) | 1.60 (1.06–2.20) | 1.90 (1.40–2.60) | <0.001 |
| 1hCp (ng/mL) | 3.65 (2.13–5.65) | 3.60 (2.80–5.15) | 4.10 (2.92–7.08) | 5.09 (3.50–6.55) | <0.001 |
| 1hΔCp (ng/mL) | 2.00 (1.18–3.73) | 2.21 (1.40–3.40) | 2.60 (1.62–4.58) | 3.10 (1.71–4.15) | 0.011 |
| 1hCp/FCp | 2.82 (2.09–3.44) | 2.63 (2.12–3.50) | 2.84 (2.07–3.74) | 2.63 (2.04–3.30) | 0.753 |
| 2hCp (ng/mL) | 5.00 (3.10–7.58) | 5.43 (3.93–7.90) | 5.80 (3.68–9.15) | 7.40 (4.70–10.25) | <0.001 |
| 2hΔCp (ng/mL) | 3.45 (2.15–5.63) | 4.20 (2.89–5.93) | 4.15 (2.70–6.95) | 5.20 (2.85–7.75) | 0.013 |
| 2hCp/FCp | 3.93 (2.89–5.23) | 3.99 (3.09–5.50) | 3.87 (2.88–5.14) | 3.60 (2.84–5.14) | 0.465 |
| Insulin (n [%]) | 13 (15.12) | 7 (8.05) | 8 (9.30) | 7 (8.14) | 0.362 |
| Oral medicine (n [%]) | 22 (25.58) | 29 (33.33) | 38 (44.19) | 42 (48.84) | 0.007 |
| Oral + Insulin (n [%]) | 51 (59.30) | 52 (59.77) | 41 (47.67) | 37 (43.02) | 0.062 |
| CCB (n [%]) | 13 (15.12) | 20 (22.99) | 18 (20.93) | 24 (27.91) | 0.235 |
| ACEI/ARB (n [%]) | 16 (18.60) | 21 (24.14) | 23 (26.74) | 28 (32.56) | 0.207 |
| Aspirin (n [%]) | 28 (32.56) | 23 (26.44) | 27 (31.40) | 23 (26.74) | 0.744 |
| statins (n [%]) | 55 (63.95) | 48 (55.17) | 61 (70.93) | 50 (58.14) | 0.150 |
Data were expressed as the mean ± SD, median (interquartile range), or n (%) according to the distribution of variables. Comparison among all the groups was shown with p value. Abbreviations: SUA, serum uric acid; BMI, body mass index; HbA1c, glycosylated hemoglobin; TC, total cholesterol; TG, triglycerides; HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate; FPG, fasting plasma glucose; FCp, fasting C-peptide; 1hCp, 1 h postprandial C-peptide; 2hCp, 2 h postprandial C-peptide; CCB, calcium channel blocker; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; SD, standard deviation.
Association of different C-peptide quartiles with SUA levels: (A) Serum uric acid with quartiles of 1hCp; (B) Serum uric acid with quartiles of 1hΔCp; (C) Serum uric acid with quartiles of 1hCp/FCp; (D) Serum uric acid with quartiles of 2hCp; (E) Serum uric acid with quartiles of 2hΔCp; (F) Serum uric acid with quartiles of 2hCp/FCp. Abbreviations: SUA, serum uric acid; FCp, fasting C-peptide; 1hCp, 1 h postprandial C-peptide; 2hCp, 2 h postprandial C-peptide.
The linear correlation analysis between the C-peptide levels and SUA levels is shown in Fig. 2. As expected, SUA levels was significantly correlated with age (r = –0.225), smoking (r = 0.208), drinking (r = 0.184), BMI (r = 0.270), TG (r = 0.332), HDL-c (r = –0.241), creatinine (r = 0.469), FCp (r = 0.282), oral medicine (r = 0.202), oral + insulin (r = –0.148), CCB (r = 0.117), and ACEI/ARB (r = 0.118). Furthermore, the serum C-peptide levels including 1hCp, 1hΔCp, 2hCp and 2hΔCp were all significantly correlated with SUA level (all p < 0.001). However, 1hCp/FCp and 2hCp/FCp were not significantly correlated with SUA level (both p > 0.05).
Correlation of serum uric acid with different C-peptide releasing effects: (A) Serum uric acid with 1hCp; (B) Serum uric acid with 1hΔCp; (C) Serum uric acid with 1hCp/FCp; (D) Serum uric acid with 2hCp; (E) Serum uric acid with 2hΔCp; (F) Serum uric acid with 2hCp/FCp. Abbreviations: SUA, serum uric acid; FCp, fasting C-peptide; 1hCp, 1 h postprandial C-peptide; 2hCp, 2 h postprandial C-peptide.
Multiple linear regression analysis revealed that in Model 1 (Table 4), the serum C-peptide-related effects including FCp, 1hCp, 1hΔCp, 2hCp and 2hΔCp were positively associated with SUA levels (p < 0.001 for trend ). After adjusting for age, BMI, TG, HDL-c, eGFR, sex, smoking, drinking, diabetes duration, HbA1c and FPG, oral medicine, oral + insulin, CCB, and ACEI/ARB in Model 3, the serum C-peptide involved 1hCp, 1hΔCp, 2hCp and 2hΔCp (all p < 0.05) were still significantly associated with SUA levels, but FCp (p = 0.977), 1hCp/FCp (p = 0.852), and 2hCp/FCp (p = 0.676) were not significantly associated (Table 4).
| Model 1 | Model 2 | Model 3 | ||||
|---|---|---|---|---|---|---|
| β (95%CI) | p | β (95%CI) | p | β (95%CI) | p | |
| FCp | 26.82 (17.60–36.04) | <0.001 | 5.61 (–4.26–15.47) | 0.264 | –0.18 (–12.02–11.67) | 0.977 |
| 1hCp | 9.89 (6.24–13.53) | <0.001 | 5.51 (1.64–9.37) | 0.005 | 5.14 (0.33–9.95) | 0.036 |
| 1hΔCp | 9.06 (4.34–13.78) | <0.001 | 7.07 (2.12–12.02) | 0.005 | 7.80 (1.90–13.70) | 0.010 |
| 1hCp/FCp | –3.36 (–8.17–1.45) | 0.17 | –0.39 (–5.06–4.29) | 0.871 | 0.47 (–4.50–5.44) | 0.852 |
| 2hCp | 6.86 (4.23–9.48) | <0.001 | 4.49 (1.92–7.05) | 0.001 | 4.27 (1.09–7.46) | 0.009 |
| 2hΔCp | 6.11 (2.99–9.23) | <0.001 | 5.27 (2.31–8.23) | <0.001 | 5.20 (1.63–8.78) | 0.005 |
| 2hCp/FCp | –2.93 (–6.35–0.49) | 0.093 | 0.31 (–3.04–3.67) | 0.86 | 0.77 (–2.86–4.40) | 0.676 |
In every regression analysis, only one C-peptide parameter was used to analyze the relationship with SUA, but not all the C-peptide parameters. Model 1 was unadjusted; Model 2 was adjusted for age, BMI, TG, HDL-c, eGFR, sex, smoking, drinking; Model 3 was adjusted for the Model 2 variables plus diabetic duration, HbA1c, oral medicine, oral + insulin, CCB, and ACEI/ARB, and FPG. Abbreviations: β, beta coefficient; FCp, fasting C-peptide; 1hCp, 1 h postprandial C-peptide; 2hCp, 2 h postprandial C-peptide; SUA, serum uric acid; BMI, body mass index; TG, triglycerides; HDL-c, high-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate; CCB, calcium channel blocker; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker.
According to linear correlation analysis, the levels of SUA and C-peptide parameters including FCp (male, r = 0.230; female, r = 0.298), 1hCp (male, r = 0.242; female, r = 0.287), 1hΔCp (male, r = 0.201; female, r = 0.227), 2hCp (male, r = 0.243; female, r = 0.313) and 2hΔCp (male, r = 0.220; female, r = 0.240) were positively correlated in both male and female groups (all p < 0.01). However, 1hCp/FCp and 2hCp/FCp were not significantly correlated with SUA levels (both p > 0.05).
Multiple linear regression analysis was used to determine the relationship between SUA and C-peptide levels in male and female groups (Table 5). In Model 1, SUA level was significantly associated with C-peptide level including FCp, 1hCp, 1hΔCp, 2hCp, and 2hΔCp in both male and female groups unadjusted for any factors (all p < 0.01). However, after adjusting for the confounding factors in Model 3, SUA level only remained positively associated with 2hCp (β = 4.78, p = 0.017) and 2hΔCp (β = 5.28, p = 0.019) in the female group. The FCP, 1hCp, 1hΔCp, and the multiplication effect involving 1hCp/FCp and 2hCp/FCp in the female group and any C-peptide parameters in the male group were not associated with SUA levels in T2DM (all p > 0.05).
| Male | Female | ||||
|---|---|---|---|---|---|
| β (95%CI) | p | β (95%CI) | p | ||
| Model 1 | FCp | 18.83 (7.12–30.54) | 0.002 | 29.49 (16.21–42.77) | <0.001 |
| 1hCp | 8.30 (3.71–12.88) | <0.001 | 10.66 (5.52–15.80) | <0.001 | |
| 1hΔCp | 8.51 (2.68–14.33) | 0.004 | 11.21 (4.17–18.25) | 0.002 | |
| 1hCp/FCp | 0.537 (–7.64–8.71) | 0.897 | –2.95 (–8.32–2.42) | 0.280 | |
| 2hCp | 8.00 (4.05–11.95) | <0.001 | 7.03 (3.98–10.07) | <0.001 | |
| 2hΔCp | 8.26 (3.50–13.03) | 0.001 | 6.56 (2.94–10.17) | <0.001 | |
| 2hCp/FCp | –0.21 (–6.03–5.61) | 0.943 | –1.71 (–5.58–2.16) | 0.384 | |
| Model 2 | FCp | 0.21 (–13.52–13.95) | 0.976 | 13.56 (0.12–27.01) | 0.048 |
| 1hCp | 2.73 (–2.95–8.41) | 0.343 | 7.24 (2.31–12.18) | 0.004 | |
| 1hΔCp | 4.06 (–2.90–11.02) | 0.250 | 9.76 (3.02–16.51) | 0.005 | |
| 1hCp/FCp | 0.66 (–12.16–13.48) | 0.919 | 0.44 (–4.12–4.99) | 0.850 | |
| 2hCp | 2.16 (–2.59–6.90) | 0.370 | 4.98 (2.21–7.75) | 0.001 | |
| 2hΔCp | 2.78 (–2.64–8.19) | 0.312 | 5.59 (2.36–8.83) | 0.001 | |
| 2hCp/FCp | –0.70 (–9.28–7.88) | 0.872 | 1.12 (–2.21–4.45) | 0.508 | |
| Model 3 | FCp | –2.30 (–18.72–14.13) | 0.782 | 3.15 (–14.86–21.17) | 0.729 |
| 1hCp | 3.92 (–3.04–10.89) | 0.267 | 5.57 (–1.28–12.42) | 0.110 | |
| 1hΔCp | 5.93 (–2.18–14.04) | 0.150 | 8.26 (–0.61–17.13) | 0.067 | |
| 1hCp/FCp | 3.25 (–11.10–17.60) | 0.654 | 1.69 (–3.18–6.56) | 0.494 | |
| 2hCp | 2.70 (–3.07–8.46) | 0.356 | 4.78 (0.86–8.69) | 0.017 | |
| 2hΔCp | 3.66 (–2.71–10.03) | 0.258 | 5.28 (0.88–9.68) | 0.019 | |
| 2hCp/FCp | 0.308 (–9.53–10.15) | 0.951 | 1.89 (–1.73–5.51) | 0.303 | |
In every regression analysis, only one C-peptide parameter was used to analyze the relationship with SUA, but not all the C-peptide parameters. Model 1 was unadjusted; Model 2 was adjusted for age, BMI, TG, HDL-c, eGFR, smoking, drinking; Model 3 was adjusted for the Model 2 variables plus diabetic duration, HbA1c, oral medicine, oral + insulin, CCB, and ACEI/ARB, and FPG. Abbreviations: β, beta coefficient; FCp, fasting C-peptide; 1hCp, 1 h postprandial C-peptide; 2hCp, 2 h postprandial C-peptide; SUA, serum uric acid; BMI, body mass index; TG, triglycerides; HDL-c, high-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate; CCB, calcium channel blocker; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker.
We assessed the association between the long-term risk of elevated SUA levels and serum C-peptide effect obtained clinically. We identified that SUA levels were positively correlated with 1hCp, 2hCp, and the incremental effect of serum C-peptide-releasing effect 1hΔCp and 2hΔCp in patients with T2DM, but were not significantly associated with FCp and the multiplication effect of C-peptide such as 1hCp/FCp and 2hCp/FCp after adjusting for the confounding variables of age, BMI, TG, HDL-c, eGFR, sex, smoking, drinking, diabetes duration, HbA1c, oral medicine, oral + insulin, CCB, ACEI/ARB, and FPG. After grouping by sex, only the 2hCp and 2hΔCp were associated with SUA levels in female patients; but no C-peptide parameters were associated with SUA levels in males with T2DM.
Hyperuricemic patients were older and had higher values of macroalbuminuria, BMI, TC, C-peptide, the prevalence of hypertension, and MetS, but lower values of HDL-c and HbA1c [11]. SUA was also strongly associated with full MetS as well as hypertriglyceridemia [22, 23]. Nevertheless, some prospective and retrospective studies have yielded contrary results. There were no significant association between SUA levels and cholesterol, HDL-c, LDL-C, TG, FPG, HbA1c, and urine microalbumin levels [24]. In subjects with MetS, the average SUA level was higher than that in the subjects without MetS, but this difference was not significant (6.62 vs. 6.28 mg/dL; p = 0.556) [25]. A study on 1,381 subjects who constituted a longitudinal health check-up cohort failed to show that increased SUA level was a causal risk factor for MetS or its components [26]. In T2DM, the subjects with higher SUA levels tended to have a higher prevalence of oral medicine treatment; higher BMI, TG, creatinine, and FCp levels; and lower levels of HbA1c and HDL-c (all p < 0.05). The serum C-peptide including FCp, 1hCp, 1hΔCp, 2hCp and 2hΔCp were obviously elevated together with the increase in SUA quartiles, but not for 1hCp/FCp (p = 0.753) and 2hCp/FCp (p = 0.465). When these C-peptide levels were decreased, the islet β cell function was poorer, thus leading to higher HbA1c (Q1, 9.2 ± 1.9%) at lower SUA levels. The HbA1c mean value of Q4 (8.5%) was significantly lower than Q1 (9.2%), which might be associated with the improved islet β cell function at the upper SUA level.
According to The US Third National Health and Nutrition Examination Survey (1988–1994) involving 14,664 participants, SUA levels increased linearly with FCp and serum insulin levels (multivariate p-values for trend, <0.001) [27]. A study on 164 patients with type 1 diabetes mellitus revealed that C-peptide secretors had higher SUA levels (259 vs. 238 μmol/L, p = 0.048) compared to subjects with undetectable C-peptide [28]. SUA was found to be positively correlated with FCp, 2hCp, BMI, waist-to-hip ratio, TG, creatinine, smoking, and drinking (p < 0.05), and FCp was identified as an independent risk factor for elevated SUA, but sex did not affect the relationship between FCp and SUA [17]. In males with type 2 diabetes, SUA level also showed significant positive correlations with insulin secretory capacity, including 2hCp, FCPI, PPCPI, ΔC-peptide, and HOMA2%B; however, no such significant correlations were observed among females [19]. The results of our present study are consistent with some of the observations described above involving the significant relationship between uric acid and ΔC-peptide levels. After adjusting for all the confounding factors, the 1hCp, 2hCp, 1hΔCp, and 2hΔCp of C-peptide-releasing parameters in patients with T2DM were positively associated with SUA levels. However, FCp was not significantly associated with SUA in patients with T2DM which is contrary to most previous findings. Insulin can enhance renal urate reabsorption via stimulation of the urate transporter in the human kidneys [29], thus insulin also decreased renal urate excretion [30-32]. Here, in order to reliably observe the insulin-secreting ability of islet β cell, we replaced insulin index with C-peptide. The finding may explain the positive correlation between the above C-peptide parameters and SUA levels. However, an elevated level of uric acid caused β-cell injury via the Nuclear factor-kappaB- Inducible nitric oxide synthase- nitric oxide (NF-κB-iNOS-NO) signalling axis that causes abnormal pancreatic β cells function [33]. Hyperuricemia induced oxidative stress in vivo and in vitro, thereby leading to insulin resistance and glucose metabolism dysfunction [34]. Hence, we considered that at a certain range, the elevated SUA levels might be associated with the better C-peptide levels of islet β cell function.
Moreover, the level of SUA was positively associated with 2hCp (β = 4.78, p = 0.017) and 2hΔCp (β = 5.28, p = 0.019) in female patients with T2DM. Compared with all negative results in the male patients, the positive results between SUA and C-peptide parameters in the females might be associated with better clinical indicators including the higher HDL-c, but the lower BMI, creatinine, SUA, FCp, and prevalence of smoking and drinking (all p < 0.05). Females had lesser risk factors and better clinical indicators. Therefore their insulin resistance and metabolic syndrome were milder than male patients with T2DM. The C-peptide release test for type 2 diabetes generally peaks at 2–3 h postprandial. Hence, the level of SUA was positively associated with 2hCp and 2hΔCp in female patients with T2DM. However, male patients got the above risk factors, including smoking, drinking, the higher BMI, SUA, FCp and creatinine, but lower mean age and HDL-c, which could exacerbate insulin resistance and metabolic syndrome, decrease renal function and worsen cardiovascular disease. Thus, the SUA and C-peptide parameters do not show a similar correlation as female patients. The result suggests that male patients should reduce the above risk factors.
The studies have shown that hyperuricemia may mediate insulin resistance, fatty liver, and dyslipidemia in both fructose-dependent and fructose-independent models of MetS [35]. SUA may augment the basal insulin secretion and insulin resistance indexes both in T2DM subjects in a population-based study [36]. After adjustment for age, alcohol, tobacco consumption, creatinine levels, BMI, and menopausal status, individuals with hyperuricemia had a higher prevalence of insulin resistance (OR = 1.84, 95%CI 1.25–2.73). And an increased prevalence of insulin resistance was associated with increasing levels of uric acid (p for trend <0.001) [37]. Though the mechanism between SUA and C-peptide-releasing effects remains unclear, we found a linear increase between SUA and islet β cells function in line with the existing data.
Nevertheless, the study has several limitations. First, the present study did not include healthy control subjects, and only inpatients with T2DM were included. Second, there might be a selection bias; the subjects were confined to a single-center or ethnic group; hence, the results could not be representative of other regions or peoples. Third, the causality of ΔC-peptide and SUA levels could not be established in a cross-sectional study, thus, further prospective research is needed to confirm these effects. Fourth, the retrospective study did not collect quantitative data on renal urate excretion that could be associated with SUA levels. Finally, the SUA levels were influenced by many factors including diet [38], genetics [39], race [40], and commonly used drugs such as statins [41] and aspirin [42]. However, we could not adjust for such confounders in the present study due to the data insufficiency.
In conclusion, our results point out that after adjusting for the confounding factors, the 1hCp, 2hCp, and the incremental effect of serum C-peptide- releasing parameters involving 1hΔCp and 2hΔCp were still positively correlated with SUA levels in patients with T2DM. However, FCp, 1hCp, 1hΔCp, and the multiplication effect of C-peptide involving 1hCp/FCp and 2hCp/FCp in female patients and all C-peptide parameters in male patients were not significantly associated with SUA levels in patients with T2DM. Within a certain range, the elevated SUA levels might be associated with the better C-peptide incremental effect of islet β cell function in T2DM, especially in female patients with T2DM.
SUA, serum uric acid; T2DM, type 2 diabetes mellitus; MetS, metabolic syndrome; FCp, fasting C-peptide; OGTT, oral glucose tolerance test; 1hCp, 1 h postprandial C-peptide; 2hCp, 2 h postprandial C-peptide; 1hCp/FCp, 1 h postprandial C-peptide to fasting C-peptide ratio; 2hCp/FCp, 2 h postprandial C-peptide to fasting C-peptide ratio; BMI, body mass index; HbA1c, glycosylated hemoglobin; TC, total cholesterol; TG, triglycerides; HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate; FPG, fasting plasma glucose; CCB, calcium channel blocker; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration; SD, standard deviation; ANOVA, One-way Analysis of covariance; β, beta coefficient
The authors thank all team members and participants in the study.
None of the authors have any potential conflicts of interest associated with this research.
Study design: YY-L, ZQ-Y, and HJ-Z; collection of data: X-Z, SR-W, and C-H; analysis and interpretation of data: YY-L and ZQ-Y; study supervision: X-Z and HJ-Z; writing of the manuscript: YY-L. All authors contributed to the manuscript and approved the submission.
