Abstract
Objective
The primary objective was to assess beta-cell function of recently-diagnosed young-onset type 2 diabetes mellitus (T2DM) individuals using basal and stimulated C-peptide levels. The secondary objective was to examine the association between C-peptide with metabolic factors and diabetes complications.
Methodology
A cross-sectional study was conducted for young-onset T2DM individuals aged 18-35 years with a disease duration of not more than 5 years. Plasma C-peptide was measured before and after intravenous glucagon injection. Demographic data, medical history and complications were obtained from medical records and clinical assessment. Continuous data were expressed as median and interquartile range (IQR). Categorical variables were described as frequency or percentage. Multivariable linear regression analysis was used to determine factors associated with C-peptide levels.
Results
113 participants with young-onset T2DM with a median (IQR) age of 29.0 (9.5) years and 24 (36) months were included in this study. The median (IQR) basal and stimulated C-peptide was 619 (655) pmol/L and 1231 (1024) pmol/L. Adequate beta-cell function was present in 78-86% of the participants based on the basal and stimulated C-peptide levels. We found hypertension, obesity and diabetic kidney disease (DKD) to be independently associated with higher C–peptide levels. In contrast, females, smokers, those on insulin therapy and with longer duration of disease had lower C–peptide levels.
Conclusion
Most recently diagnosed young-onset T2DM have adequate beta-cell function. Elevated C-peptide levels associated with obesity, hypertension and diabetic kidney disease suggest insulin resistance as the key driving factor for complications.
Keywords: type 2 diabetes mellitus, young-onset, beta-cell function, C-peptide, C-peptide
INTRODUCTION
Young-onset type 2 diabetes mellitus (T2DM) is defined as onset of T2DM before the age of 40 years in the absence of secondary causes.1 Studies have shown an alarming increase in the prevalence of young-onset T2DM globally, more so in Asia in the last few decades.2 The International Diabetes Federation (IDF) estimated an increase from 23 to 63 million young adults worldwide to have T2DM from 2000 to 2013 with the biggest increase being in Africa, Southeast Asia, and the Western Pacific region.2 In an Asian study, approximately 20% of patients with type 2 diabetes were young-onset T2DM.3 This subset of individuals has been associated with accelerated disease progression and premature complications.2 There are also higher rates of insulin commencement and intensification of treatment regimen early in the disease compared to the usual onset.1
The pathophysiology of T2DM in the young has been thought to be similar to the usual onset with an interplay of beta-cell dysfunction, insulin resistance and obesity-related mechanisms.2,4 There is limited data with regards to beta-cell function in the early stage of the disease in this population. Objective assessment of beta-cell function is important to assist us in understanding the disease mechanism and rate of progression, as well as to guide therapeutic options. This is a pilot study performed in Malaysia to assess beta-cell function of recently-diagnosed young-onset T2DM individuals using fasting and glucagon stimulated C-peptide levels. C-peptide is produced in equal amounts to insulin and is an objective marker of beta-cell function.5 Glucagon stimulation test (GST) is an easily performed test to assess stimulated C-peptide with good sensitivity and reproducibility in clinical practice.6 The secondary objective of this study was to examine the association among C-peptide levels with metabolic parameters and diabetes-related complications.
METHODOLOGY
Study design and participants
This is a cross-sectional study involving young-onset T2DM individuals seen in the diabetes clinics of two urban tertiary hospitals in Malaysia between September 2019 to December 2020. Individuals aged 18 to 35 years diagnosed with T2DM for not more than five years were recruited using the universal sampling method. This was a descriptive study, therefore no formal sample size was required. We excluded individuals with chronic kidney disease stage 2 and above (estimated glomerular filtration rate (eGFR) <60 ml/min/1.73 m2), concurrent infection or inflammatory disease, recent diabetic ketoacidosis or hyperglycaemic hyperosmolar state in the last three months, positive diabetes autoantibodies, prior clinical diagnosis of other forms of diabetes (monogenic diabetes, type 1 diabetes, latent autoimmune diabetes of adult onset and secondary diabetes), fasting capillary blood glucose less than 4.0 mmol/L or more than 13.9 mmol/L on the day of testing, and those who were pregnant.7 Written informed consent was obtained from all study participants prior to the commencement of the study. This study was approved by the Medical Research and Ethics Committee (MREC) of the Ministry of Health of Malaysia.
Study recruitment and procedures
Study participants were recruited during their routine clinical visits to the diabetes clinic. Individuals with unknown diabetes mellitus autoantibodies status were screened prior to enrolment. Those with positive diabetes autoantibodies were excluded from the study. This study involved only a single visit. Study participants were required to fast overnight for at least eight hours and to omit all insulin and oral antidiabetic agents on the morning of testing. On the day of testing, anthropometric measurements (weight, height, waist circumference) were taken using calibrated tools. Vital signs and capillary blood glucose pre-procedure were checked. GST was performed if fasting capillary glucose levels were between 4.0 to 13.9 mmol/L. Basal C-peptide and blood glucose levels were sampled prior to administration of 1 mg of intravenous glucagon. After 6 minutes, stimulated blood glucose and C-peptide samples were collected. Study participants were monitored for adverse effects for 15 minutes post testing. All tests were conducted by a single operator. Information regarding demography, disease history, co-morbidities, complications and treatment were gathered from medical records and clinical assessment.
Measures
The primary outcome of this study, beta-cell function, was measured using C-peptide levels (fasting and stimulated). Adequate beta-cell function was defined as either a basal C-peptide level of more than 250 pmol/L or a stimulated C-peptide level of more than 600 pmol/L, or both.5
Independent variables examined included current age, gender, ethnicity, smoking status, family history of T2DM, age of disease onset, disease duration, fasting glucose, HbA1c, insulin therapy, waist circumference, obesity, hypertension, dyslipidaemia, macrovascular complications, microvascular complications, retinopathy, nephropathy, and neuropathy. Waist circumference was measured midway between the iliac crest and the lowermost margin of the ribs at the end of a normal respiratory expiration.8 Obesity was defined based on a body mass index (BMI) cut-off of ≥27.5 kg/m2.9 Abdominal obesity was defined as a waist circumference of ≥90 cm in males or ≥80 cm in females.8 Glycated haemoglobin (HbA1c) performed in the last three months was used as a measure of glycaemic control. Hypertension was defined as the persistent elevation of systolic blood pressure of 140 mmHg or greater and/or a diastolic blood pressure of 90 mmHg or greater.10 Dyslipidemia in type 2 diabetes was diagnosed based on low-density lipoprotein (LDL) cholesterol levels of 2.60 mmol/l or greater, high-density lipoprotein (HDL) cholesterol levels of 1.02 mmol/l or less and triglyceride levels of 1.7 mmol/l or greater.11
Macrovascular complication was defined as an established history of ischemic heart disease, stroke or peripheral vascular disease based on medical records. Microvascular complication was defined as the presence of one or more of the following complications: retinopathy, nephropathy or peripheral neuropathy. Retinopathy was assessed based on slit lamp examination records performed in the last 12 months by credentialed personnel at the ophthalmology clinic of respective tertiary hospitals. Participants with no recent assessment were referred for evaluation as per routine protocol. Both ophthalmology clinics used the Early Treatment for Diabetic Retinopathy Study (ETDRS) classification to diagnose and classify diabetic retinopathy. Diabetic kidney disease (DKD) was assessed based on two or more urinary spot quantification of proteinuria or urine albumin creatinine ratio (ACR) performed in the last 12 months. Microalbuminuria was defined as a urine ACR of 3-30 mg/mmol whereas overt nephropathy was defined as a urine ACR of >30 mg/mmol.12 Subjects with chronic kidney disease stage 2 and above (estimated glomerular filtration rate (eGFR) <60 ml/min/1.73 m2) were excluded from this study, as majority of C-peptide is metabolised by the kidney and may result in inaccurate results.5 Peripheral neuropathy was diagnosed based on diminished sensation to standardised monofilament examination or reduced vibration on graduated tuning fork testing. Successful weaning of initial insulin therapy was defined as the ability to discontinue insulin in those initiated on insulin within 6 months of diagnosis.
C-peptide analysis was performed in a centralised laboratory using IMMULITE-2000 C-peptide (Siemens), a solid phase, two-site chemiluminescence immunometric assay. The co-efficient of variation (CV) of this test is less than 5% based on the designated laboratory internal quality control (IQC) performance analysis. This is in keeping with target of desirable biological variation.13 Study participants were screened for diabetes autoantibodies including anti-islet cell antibodies (ICA), anti-glutamic acid (GAD) antibodies and anti-islet tyrosine phosphatase (IA2) antibodies using commercially available ELISA kits (Medipan GmbH, Germany). Fasting blood glucose level was determined by hexokinase method. Urinary albumin and creatinine were measured by immunoturbidimetric and enzymatic methods respectively. Lastly, HbA1c was analysed using high performance liquid chromatography (HLPC).
Statistical analysis
All statistical analysis was performed with Statistical Package for Social Science (SPSS) Version 22.0.14 Most continuous data was found to be not normally distributed, therefore expressed as median and interquartile range (IQR). Categorical variables were described as frequency or percentage. Bivariate analysis using Mann Whitney U and Kruskal Wallis test were used to examine association between categorical data. Spearman’s rank correlation coefficient was used to assess relationship between continuous data. We proceeded with linear regression analysis using variables with p<0.25 from bivariate analysis and clinically important outcome variables based on biological plausibility. Stepwise linear regression with forward selection was performed. The final model was tested for autocorrelation (Durbin-Watson test), multicollinearity and homoscedasticity. The significance level for all steps of analyses was p<0.05.
RESULTS
A total of 151 patients were screened to be recruited in this study. Twenty-eight patients were excluded due to positive autoantibodies (n = 20), CKD stage 3 and above (n = 2), pregnancy (n = 1), recent diabetic ketoacidosis (n = 2), and hyperglycemia on day of testing (n = 3). Six patients declined participation and the remainder (n = 4) were not contactable. A final sample size of 113 participants participated in this study.
Patient characteristics (Table 1)
Table 1.
Patient demography, metabolic parameters, disease history and complications
| Variables | Median (IQR)a | n (%) | ||
|---|---|---|---|---|
| Demography | Age (years) | 29.0 (9.5) | ||
| Gender | Male | 43 (38.1) | ||
| Female | 70 (61.9) | |||
| Ethnicity | Malay | 80 (70.8) | ||
| Chinese | 15 (13.3) | |||
| Indian | 18 (15.9) | |||
| Metabolic parameters | Obesity | BMI <27.5 kg/m2 | 29 (25.7) | |
| BMI ≥27.5 kg/m2 | 84 (74.3) | |||
| Abdominal Obesity | Male (WC ≥90 cm) | 38 (88.4) | ||
| Female (WC ≥80 cm) | 64 (91.4) | |||
| Hypertension | 34 (30.1) | |||
| Dyslipidaemia | 104 (92.0) | |||
| Medical history | Diabetes duration (months) | 24 (36) | ||
| HbA1c (%) | 8.5 (4.1) | |||
| Family history of diabetes | 94 (83.2) | |||
| Smoking | 16 (14.2) | |||
| Diabetes treatment history | Current insulin therapy | 66 (58.4) | ||
| Time to insulin initiation | ≤6 months from diagnosis | 55 (75.3) | ||
| >6 months from diagnosis | 18 (24.7) | |||
| Successful weaning of initial insulinb | 8 (15.7) | |||
| Complications | Microvascular | 43 (38.1) | ||
| Retinopathy | 8 (7.1) | |||
| Non-proliferativec | 4 (50.0) | |||
| Proiferativec | 4 (50.0) | |||
| Diabetic kidney disease | 39 (34.5) | |||
| Microalbuminuriad | 29 (74.4) | |||
| Overt nephropathyd | 10 (25.6) | |||
| Peripheral neuropathy | 9 (8.0) | |||
IQR, interquartile range; BMI, body mass index; WC, waist circumference.
not normally distributed data expressed as median (IQR)
within insulin initiation at ≤6 months from diagnosis subgroup
within retinopathy subgroup
within diabetic kidney disease subgroup
The median (IQR) age and disease duration of study population was 29 (9.5) years and 24 (36) months respectively. There was a female preponderance (n = 70, 61.9%) with the predominant ethnicity being Malay (n = 80, 70.8 %). More than two-thirds (n = 84, 74.3%) of the study population was obese. Approximately 90% of the subjects had abdominal obesity: 88.4% of the males (n = 38) and 91% of the females (n = 64). Almost all had concomitant dyslipidaemia (n = 104, 92.0%) while less than one-third had hypertension (n = 34, 30.1%). Majority had a family history of T2DM involving at least one first-degree relative (n = 94, 83.2%). There were 14.2% (n = 16) active smokers in the population studied. The average glycaemic control of the population studied was poor, with a median (IQR) HbA1c of 8.5% (4.1). More than half of the study population (n = 66, 58.4%) was currently on insulin therapy. It was found that 75.3% (n = 55) of all patients with history of insulin use were initiated on insulin therapy early (within six months from diagnosis). However, only 15.7% (n = 8) of those started on early insulinization were successfully weaned off. There were no documented macrovascular complications; however, microvascular complications were present in 38.1% (n = 43) of the cohort. DKD was the most common microvascular complication (34.5%, n = 39), mostly in the form of microalbuminuria. Retinopathy and peripheral neuropathy were observed in less than 10% of the cohort.
Glucagon stimulation test results (Table 2)
Table 2.
Glucagon stimulation test results
| Variables | Median (IQR) / n (%) | |
|---|---|---|
| Glucose (mmol/L), median (IQR) | Fasting glucose | 7.5 (3.3) |
| Stimulated glucose | 8.3 (3.7) | |
| C-peptide (pmol/L), median (IQR) | Basal C-peptide | 619.0 (655.0) |
| Glucagon stimulated C-peptide | 1231.0 (1024.0) | |
| Beta cell function, n (%) | Adequate basal beta cell functiona | 97.0 (85.8) |
| Adequate stimulated beta cell functionb | 88.0 (77.9) | |
| For participants on insulin as current therapy: |
51.0 (77.3) |
|
|
46.0 (69.7) |
Results expressed in median (IQR) for not normally distributed variables and percentage for categorical variables.
defined as C-peptide >250 pmol/L,
defined as C-peptide >600 pmol/L.9
The C-peptide levels had a median (IQR) basal value of 619 (655) pmol/L and 1231 (1024) pmol/L with stimulation. There was a strong positive correlation between fasting and stimulated C-peptide levels, rs = 0.92, p<0.001 (data not shown in Table 2). Majority of the study participants had adequate beta-cell function based on the basal and stimulated C-peptide levels at 86% and 78% respectively. Within the subgroup of participants on insulin as current therapy, there was still a high proportion registering adequate beta-cell function at 77% and 70% respectively based on the basal and stimulated C-peptide levels. Transient and self-limiting nausea was reported in 8.8% (n = 10) participants immediately after the procedure.
Factors associated with basal and stimulated C-peptide levels
Table 3 shows the bivariate analysis between basal and stimulated C-peptide levels with baseline clinical characteristics, metabolic parameters and diabetes-related complications. Both basal and stimulated C-peptide were significantly associated with gender, disease duration, age of disease onset, HbA1c, insulin therapy, obesity, waist circumference and hypertension (p<0.05). In addition, stimulated C-peptide was also significantly associated with current age, smoking and dyslipidaemia (p<0.05). There was no significant association between basal and stimulated C-peptide with diabetes related complications.
Table 3.
Bivariate analysis between c-peptide and clinical features, metabolic parameters and diabetes complications
| Variables | Basal C-peptide | Stimulated C-peptide | |||||
|---|---|---|---|---|---|---|---|
| Median (IQR)a | r s b | p | Median (IQR)a | r s b | p | ||
| Current age | 0.164 | 0.083 | 0.259 | 0.006 | |||
| Gender | Male | 857.0 (709.0) | 0.002 | 1705.0 (1268.0) | 0.001 | ||
| Female | 541.0 (500.0) | 1021.0 (980.3) | |||||
| Ethnicity | Malay | 635.5 (658.5) | 0.499 | 1247.5 (1008.3) | 0.354 | ||
| Chinese | 559.0 (469.0) | 751.0 (1264.0) | |||||
| Indian | 639.0 (536.5) | 1276.0 (902.0) | |||||
| Smoking | Yes | 852.0 (625.5) | 0.177 | 1673.5 (851.75) | 0.034 | ||
| No | 583.0 (637.0) | 1086.0 (1071.5) | |||||
| Family history | Yes | 625.5 (618.5) | 0.693 | 1213.0 (952.0) | 0.785 | ||
| No | 543.0 (964.0) | 1241.0 (1612.0) | |||||
| Age of disease onset | 0.219 | 0.020 | 0.301 | 0.001 | |||
| Disease duration | 0.238 | 0.011 | 0.227 | 0.015 | |||
| Fasting glucose | 0.069 | 0.469 | 0.074 | 0.434 | |||
| HbA1c | 0.286 | 0.002 | 0.319 | 0.001 | |||
| Insulin therapy | Yes | 460.0 (382.0) | <0.001 | 917.0 (884.8) | <0.001 | ||
| No | 910.0 (560.0) | 1629.0 (954.0) | |||||
| Obesityc | Yes | 766.5 (604.8) | <0.001 | 1493.0 (1108.8) | <0.001 | ||
| No | 348.0 (322.5) | 761.0 (553.0) | |||||
| Waist circumference | 0.417 | <0.001 | 0.404 | <0.001 | |||
| Dyslipidaemia | Yes | 635.5 (681.5) | 0.160 | 1259.5 (1036.0) | 0.034 | ||
| No | 450.0 (239.5) | 970.0 (715.0) | |||||
| Hypertension | Yes | 897.0 (870.3) | 0.001 | 1622.0 (1085.5) | 0.001 | ||
| No | 546.0 (485.0) | 1039.0 (1083.0) | |||||
| Microvascular | Yes | 672.0 (688.0) | 0.707 | 1347.0 (1434.0) | 0.896 | ||
| No | 594.5 (578.5) | 1095.0 (975.3) | |||||
| Retinopathy | Yes | 708.0 (1004.8) | 0.392 | 1251.0 (1056.8) | 0.771 | ||
| No | 606.0 (645.0) | 1231.0 (1038.0) | |||||
| Nephropathy | Yes | 722.0 (685.0) | 0.474 | 1413.0 (1373.0) | 0.411 | ||
| No | 581.0 (644.8) | 1089.0 (1015.3) | |||||
| Neuropathy | Yes | 424.0 (710.0) | 0.321 | 1089.0 (1332.0) | 0.284 | ||
| No | 625.5 (641.5) | 1236.0 (1004.8) | |||||
Mann-Whitney U test used for non-parametric binomial data and Kruskal Wallis for data with >2 categories
Spearman’s rank correlation coefficient (rs) used for non-parametric data
Obesity defined as BMI ≥27.5 kg
We proceeded with linear regression analysis to determine factors independently associated C-peptide levels (Tables 4 and 5). We included 11 variables with p<0.25 from the bivariable analysis and clinically important outcome variables studied into the preliminary main effect model (current age, gender, smoking, age of disease onset, disease duration, HbA1c, obesity, insulin therapy, dyslipidaemia, hypertension and DKD). We found hypertension, obesity and DKD to be independently associated with higher C– peptide levels. In contrast, females, smokers, those on insulin therapy and those with longer disease duration had lower C–peptide levels.
Table 4.
Factors independently associated with basal c-peptide levels from multivariable linear regression
| Variables | Coefficient | 95% Confidence Interval | p | |
|---|---|---|---|---|
| Lower limit | Upper limit | |||
| Insulin therapy | -392.8 | -550.1 | -235.4 | <0.001 |
| Hypertension | 192.0 | 22.6 | 361.5 | 0.027 |
| Obesity | 252.6 | 78.9 | 426.3 | 0.005 |
| Female | -266.8 | -434.9 | -99.6 | 0.002 |
| Nephropathy | 186.0 | 27.2 | 344.9 | 0.022 |
| Smoker | -264.7 | -501.5 | -28.0 | 0.029 |
Stepwise multiple linear regression was applied. The Durbin-Watson statistic, d=1.945 was between 1.5 and 2.5 and therefore data was not autocorrelated.
Multicollinearity and homoscedasticity were checked and not found. A significant regression equation was found [F (6,104) = 13.076, p<0.001], with R2 of 0.43.
Table 5.
Factors independently associated with stimulated c-peptide levels from multivariable linear regression
| Variables | Coefficient | 95% Confidence Interval | p | |
|---|---|---|---|---|
| Lower limit | Upper limit | |||
| Insulin therapy | -393.4 | -689.6 | -97.3 | 0.010 |
| Hypertension | 533.5 | 201.6 | 865.4 | 0.002 |
| Obesity | 382.4 | 49.2 | 715.8 | 0.025 |
| Female | -355.1 | -644.1 | -66.1 | 0.017 |
| Duration of Disease | -7.2 | -14.4 | -0.2 | 0.044 |
Stepwise multiple linear regression was applied. The Durbin-Watson statistic, d=1.907 was between 1.5 and 2.5 and therefore data was not autocorrelated.
Multicollinearity and homoscedasticity were checked and not found. A significant regression equation was found [F (5,105) = 11.879, p<0.001], with R2 of 0.3.
DISCUSSION
We found our young-onset T2DM population to be predominantly female (62%). Subjects also had a strong family history of diabetes. Almost three-quarters of them were obese by BMI criteria. More than 90% had abdominal obesity associated with dyslipidaemia while almost one-third had hypertension. This high prevalence of metabolic syndrome with multiple cardiovascular risk factors despite a young age and short disease duration is similar to the clinical characteristics reported in literature.1,3,15 The median HbA1c of 8.5% (69 mmol/mol) in our study cohort indicated a relatively poorer glycaemic control compared to the median HbA1c of 7.9% (63 mmol/mol) for the general diabetic population in the country.16 Our findings concur with several studies done in the Asian region showing poorer glycaemic control in young-onset T2DM compared to the usual onset T2DM.3,17
The high percentage of insulin use within six months of diagnosis coupled with a low rate of subsequent weaning of insulin therapy in this study indicate a delayed diagnosis with high prevalence of severe hyperglycaemia at presentation requiring insulin, followed by difficulties in restoring euglycemia despite intensive use of insulin. This is consistent with the high percentage of undiagnosed diabetes among the young age group (18 to 39 years) which is at 77% according to the National Health and Morbidity Survey in 2019.18 In the Treatment Options for Type 2 Diabetes in Adolescents and Youth (TODAY) cohort, only 13% were initiated on insulin at diagnosis and nearly all were weaned off insulin during the run-in period of two to six months using metformin titration and diabetes education.17 In the Asian region, Pan et al., showed that only 18% of newly-diagnosed young-onset diabetes patients used insulin during the first year of disease, despite approximately 10% of the study population having classical type 1 diabetes.17
There was a high prevalence of microvascular complication at 38% in our study cohort despite only a median disease duration of two years. There was a striking predominance of DKD among the microvascular complications, with majority having microalbuminuria. Literature shows the prevalence of microalbuminuria is higher among individuals with young onset T2DM at diagnosis and tends to occur after a shorter duration of disease.2 The TODAY cohort showed a 6% prevalence of microalbuminuria at baseline whichit increased to 17% after four years.19 The Pima Indian youth with T2DM (aged <20 years) had a much higher prevalence of microalbuminuria at diagnosis at 22% and was projected to reach 60% before the age of 30.2 Kim et al., also reported a high prevalence of microalbuminuria at 34.3%, nearly 2-fold higher in those with newly-diagnosed young-onset T2DM compared to those diagnosed after the age of 40 years.15 This indicates an increased risk of early-onset DKD with rapid progression among the young-onset T2DM individuals compared to usual onset, with possible ethnic variation.
C-peptide level was chosen as the measure of beta-cell function because it is a reliable indicator of insulin secretion. It has a longer half-life and is less affected by first-pass hepatic metabolism when compared to insulin. It also can be used to assess endogenous insulin production including patients on insulin treatment.5,6 The median C-peptide in our cohort was 619 pmol/L (basal) and 1231 pmol/L (glucagon stimulated). Based on the recommended thresholds of 250 pmol/L (basal) and 600 pmol/L (stimulated) C-peptide levels, majority (78-86%) of our patients have adequate beta-cell function.5 These results are comparable to other cohorts reported in the region.15,17
Kim et al., reported a median basal C-peptide of 788 pmol/L and 2-hour postprandial C-peptide of 2023 pmol/L in their newly-diagnosed young-onset T2DM.15 Pan et al., reported a mean basal C-peptide of 700 pmol/l with an almost identical percentage (80-83%) of adequate beta-cell function which was assessed using basal and glucagon stimulation C-peptide testing in their newly diagnosed young-onset diabetes individuals consisting mainly of T2DM.17 This is in contrast to some other studies that reported significant beta cell dysfunction in newly-diagnosed young-onset T2DM.20,21 Even within the subgroup of insulin users, there was a high proportion of at least 70% having adequate beta-cell function. Pan et al., showed a lower percentage, with approximately 50% of their young newly-diagnosed DM on insulin with adequate basal and stimulated beta-cell function using similar C-peptide cut off levels.17 However, their study population included those with T1DM.17
We postulated that the high fasting and stimulated C-peptide level shown in our study may be a surrogate marker of insulin resistance in this population.6,22 Therefore, we proceeded with post-hoc analysis to calculate modified HOMA-IR levels to assess insulin resistance in this cohort using basal C-peptide levels.23 We found a mean ± standard deviation (SD) modified HOMA-IR of 3.48 ± 1.36. The modified HOMA-IR values showed a strong positive correlation with basal C-peptide levels, rs = 0.90, p<0.001. Based on previous studies done in the Asian region, a HOMA-IR value of more than 2.5 has been used to define insulin resistance.24,25 Using this cut off, 78% of our study population were found to be insulin resistant. This further supports our hypothesis that our recently-diagnosed young-onset T2DM population was predominantly insulin resistant as opposed to having inadequate beta-cell function.
Linear regression analysis in this study revealed females to have lower C-peptide levels than males. This may indicate a possible difference in beta-cell function between genders among the young-onset T2DM. Despite the known preponderance of females in the young-onset T2DM, there has been no conclusive evidence to suggest difference in beta-cell function between genders to date.2
Obesity and hypertension were independently associated with higher basal and stimulated C-peptide levels. This is presumably because an elevated C-peptide is known to be a surrogate marker of insulin resistance in individuals with the metabolic syndrome phenotype.6
As expected, patients on insulin therapy were found to have significantly lower basal and stimulated C-peptide levels. There has been concern in literature about suppression of endogenous insulin and C-peptide secretion in those on exogenous insulin therapy when beta-cell function is assessed. However, it has been proven that acute normalization of glucose concentration or hypoglycaemia is responsible for suppression of C-peptide levels during testing instead of the direct effect of the exogenous insulin.6,26
Gjessing et al., has shown that contrary to the hypothesis of beta-cell function attenuation due to glucotoxicity, fasting hyperglycaemia in fact potentiates stimulated C-peptide levels during mixed-meal tolerance testing or glucagon stimulation testing in T2DM.27 In this study, we ensured that the blood glucose of study participants was between 4.0 to 13.9 mmol/L at the time of testing to minimize the possible effect of extreme glucose concentrations on C-peptide values.7
We also found active smokers to have lower basal C-peptide levels than non-smokers. Although there are epidemiological studies confirming smoking as a risk factor for the development of T2DM, the effect of smoking on beta-cell function and insulin resistance is not conclusive.28–31
DKD was found to be independently associated with elevated basal C-peptide levels in this study. In our study cohort, there was a high prevalence of DKD early in the disease despite majority of them having adequate beta-cell function. Older studies done in the usual-onset T2DM with longer disease duration showed an opposite relationship with lower C-peptide levels associated with DKD.7,22
A recent large data-driven cluster analysis by Ahlqvist et al.,in Scandinavia identified five novel subgroups in their T2DM population. These five subgroups were classified as severe autoimmune diabetes (SAID), severe insulin-deficient diabetes (SIDD), severe insulin-resistant diabetes (SIRD), mild obesity-related diabetes (MOD) and mild age-related diabetes (MARD).32 The SIRD subgroup characterized by obesity and insulin resistance had a higher risk of DKD independent of their glycaemic control compared to the other subgroups.32 Zou et al., confirmed these findings in the Chinese and United States populations.33 However, Anjana et al., identified an additional subgroup, Combined Insulin Resistant and Deficient Diabetes (CIRDD), unique to the Asian Indian population. This subgroup was younger in onset, had combined insulin deficiency and resistance with poor glycaemic control and increased hazards of kidney disease and retinopathy.34 A similar analysis in our young-onset T2DM population will be beneficial to understand the subgroups within and their characteristics, including risk of complications.
Limitations
This study had a few limitations. Although this was a descriptive study not requiring a formal sample size, the modest sample size may have underscored the power in detecting the association between C-peptide with metabolic parameters and diabetes-related complications. Therefore, less emphasis can be placed on significance testing of results in this study. The cross-sectional study design was not able to establish causal relationship between measured variables and outcomes.
Our study population was comprised of patients attending diabetes clinics in two urban tertiary hospitals and was not fully representative of the general population of young-onset T2DM in the country. Despite attempts to exclude those with monogenic diabetes based on prior clinical diagnosis, the lack of genetic testing is a limiting factor.
Lastly, the slit lamp examination not being done by designated ophthalmologists may have lead to operator bias in the results of retinopathy screening. However, it was verified that the standard operating procedure of both of the ophthalmology clinics required their personnel to be credentialed and privileged to perform the test and used the Early Treatment for Diabetic Retinopathy Study (ETDRS) classification to diagnose and classify diabetic retinopathy.
This is the first study assessing beta-cell function in recently diagnosed, exclusively young-onset T2DM in the country and one of the few in Asia. Various measures to minimize the effects of confounding factors on C-peptide measurement were undertaken including stringent eligibility criteria, exclusion of extreme blood glucose levels and omission of exogenous insulin on the day of testing to improve the accuracy of C-peptide testing.
CONCLUSION
Majority of the participants who have young-onset diabetes have adequate basal and stimulated pancreatic beta-cell function. There was a high rate of metabolic syndrome in this study cohort associated with early-onset DKD. The glycaemic control was poor despite high rates of insulin use. Elevated C-peptide levels associated with obesity, hypertension and DKD suggest that insulin resistance rather than beta-cell dysfunction is the key driving factor of complications. Our study paves the direction for future prospective studies involving a larger cohort of individuals with direct measurement of insulin sensitivity to confirm our findings. Treatment strategies in this population should be tailored, focusing on early diagnosis before development of severe hyperglycaemia with a combination of intensive lifestyle modification and pharmacotherapy to address obesity, insulin resistance and avoiding overtreatment with insulin. These individuals should be screened for microvascular complications especially DKD upon diagnosis and monitored closely thereafter to prevent premature morbidity and mortality.
Statement of Authorship
The authors certified fulfillment of ICMJE authorship criteria.
CRediT Author Statement
SN: Conceptualization, Methodology, Validation, Formal Analysis, Investigation, Resources, Data Curation, Writing - original draft preparation, Visualization, Project administration, Funding acquisition; SR: Conceptualization, Methodology, Validation, Investigation, Resources, Data Curation, Writing – review and editing, Visualization; MBLB: Conceptualization, Methodology, Validation, Investigation, Resources, Data Curation, Writing – review and editing, Visualization; NSCR: Conceptualization, Methodology, Validation, Investigation, Resources, Visualization; ST: Conceptualization, Methodology, Validation, Investigation, Resources, Visualization; NA: Conceptualization, Methodology, Validation, Investigation, Resources, Visualization; YMC: Conceptualization, Methodology, Validation, Investigation, Resources, Visualization; SHF: Conceptualization, Methodology, Validation, Investigation, Resources, Data Curation, Writing – review and editing, Supervision, Project administration, Funding acquisition.
Author Disclosure
The authors declared no conflict of interest.
Funding Source
This study was funded by a grant from the Malaysian Endocrine and Metabolic Society (MEMS), Malaysia.
References
- 1.Lascar N, Brown J, Pattison H, Barnett AH, Bailey CJ, Bellary S. Type 2 diabetes in adolescents and young adults. Lancet Diabetes Endocrinol. 2018;6(1):69–80. PMID: . 10.1016/S2213-8587(17)30186-9. [DOI] [PubMed] [Google Scholar]
- 2.Magliano DJ, Sacre JW, Harding JL, Gregg EW, Zimmet PZ, Shaw JE. Young-onset type 2 diabetes mellitus — Implications for morbidity and mortality. Nat Rev Endocrinol. 2020;16(6):321–31. PMID: . 10.1038/s41574-020-0334-z. [DOI] [PubMed] [Google Scholar]
- 3.Yeung RO, Zhang Y, Luk A, et al. Metabolic profiles and treatment gaps in young-onset type 2 diabetes in Asia (the JADE programme): A cross-sectional study of a prospective cohort. Lancet Diabetes Endocrinol. 2014;2(12):935–43. PMID: . 10.1016/S2213-8587(14)70137-8. [DOI] [PubMed] [Google Scholar]
- 4.Kahn SE. The relative contributions of insulin resistance and beta-cell dysfunction to the pathophysiology of Type 2 diabetes. Diabetologia. 2003;46(1):3–19. PMID: . https://doi.org10.1007/s00125-002-1009-0. [DOI] [PubMed] [Google Scholar]
- 5.Jones AG, Hattersley AT. The clinical utility of C-peptide measurement in the care of patients with diabetes. Diabet Med. 2013;30(7):803–17. PMID: . PMCID: . 10.1111/dme.12159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Leighton E, Sainsbury CA, Jones GC. A Practical Review of C-Peptide Testing in Diabetes. Diabetes Ther. 2017;8(3):475–87. PMID: . PMCID: . 10.1007/s13300-017-0265-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Yoon HJ, Cho YZ, Kim JY, et al. Correlations between glucagon stimulated C-peptide levels and microvascular complications in type 2 diabetes patients. Diabetes Metab J. 2012;36(5):379–87. PMID: . PMCID: . 10.4093/dmj.2012.36.5.379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Misra A, Vikram NK, Gupta R, Pandey RM, Wasir JS, Gupta VP. Waist circumference cutoff points and action levels for Asian Indians for identification of abdominal obesity. Int J Obes. 2006;30(1):106–11. PMID: . 10.1038/sj.ijo.0803111. [DOI] [PubMed] [Google Scholar]
- 9.WHO Expert Consultation . Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet. 2004;363(9403):157–63. PMID: . 10.1016/S0140-6736(03)15268-3. Erratum in: Lancet.2004;363(9412):902. [DOI] [PubMed] [Google Scholar]
- 10.Passarella P, Kiseleva TA, Valeeva FV, Gosmanov AR. Hypertension management in diabetes: 2018 update. Diabetes Spectr. 2018;31(3): 218–24. PMID: . PMCID: . 10.2337/ds17-0085. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Haffner SM; American Diabetes Association . Dyslipidemia management in adults with diabetes. Diabetes Care. 2004;1;27(Suppl 1): S68–71.Diabetes Care. PMID: . 10.2337/diacare.27.2007.s68. [DOI] [PubMed] [Google Scholar]
- 12.Forbes A, Gallagher H. Chronic kidney disease in adults: Assessment and management. Clin Med (Lond). 2020;20(2):128–32. PMID: . PMCID: . 10.7861/clinmed.cg.20.2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ricos C, Alvarez V, Cava F, et al. Current databases on biologic variation: Pros, cons and progress. Scand J Clin Lab Invest.1999;59(7): 491–500. PMID: . 10.1080/00365519950185229. [DOI] [PubMed] [Google Scholar]
- 14.IBM Corp. Released 2013 . IBM SPSS Statistics for Windows, version 22.0. Armonk, NY: IBM Corp. [Google Scholar]
- 15.Kim K-S, Oh H-J, Kim J-W, et al. The clinical characteristics of the newly diagnosed early onset (<40 years old) diabetes in outpatients’ clinic. Korean Diabetes J. 2010;34(2):119. PMID: . PMCID: . 10.4093/kdj.2010.34.2.119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ministry of Health Malaysia . National Diabetes Registry Report 2013-2019; 2020. https://www.moh.gov.my/moh/resources/Penerbitan/Rujukan/NCD/Diabetes/National_Diabetes_Registry_Report_2013-2019_26082021.pdf.
- 17.Pan CY, So WY, Khalid BAK, et al. Metabolic, immunological and clinical characteristics in newly diagnosed Asian diabetes patients aged 12 – 40 years. Diabet Med. 2004;21(9):1007–13. PMID: . 10.1111/j.1464-5491.2004.01287.x. [DOI] [PubMed] [Google Scholar]
- 18.Institute for Public Health . National Health and Morbidity Survey (NHMS) 2019: Non-communicable diseases, healthcare demand, and health literacy—Key Findings; 2020. https://iptk.moh.gov.my/images/technical_report/2020/4_Infographic_Booklet_NHMS_2019_-_English.pdf.
- 19.Kelsey MM, Geffner ME, Guandalini C, et al. Presentation and effectiveness of early treatment of type 2 diabetes in youth: Lessons from the TODAY study. Pediatr Diabetes. 2016;17(3):212–21. PMID: . PMCID: . 10.1111/pedi.12264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Elder DA, Herbers PM, Weis T, Standiford D, Woo JG, D’Alessio DA. β-Cell dysfunction in adolescents and adults with newly diagnosed type 2 diabetes mellitus. J Pediatr. 2012;160(6):904–10. PMID: . PMCID: . 10.1016/j.jpeds.2011.12.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Mohan V, Amutha A, Ranjani H, et al. Associations of β-cell function and insulin resistance with youth-onset type 2 diabetes and prediabetes among Asian Indians. Diabetes Technol Ther. 2013;15(4): 315–322. PMID: . 10.1089/dia.2012.0259. [DOI] [PubMed] [Google Scholar]
- 22.Kim BY, Jung CH, Mok JO, Kang SK, Kim CH. Association between serum C-peptide levels and chronic microvascular complications in Korean type 2 diabetic patients. Acta Diabetol. 2012; 49(1):9–15. PMID: . 10.1007/s00592-010-0249-6. [DOI] [PubMed] [Google Scholar]
- 23.Xia L, Zhi-guang Z, Hai-ying Q, Xiao-Yan C, Gan H. Replacement of insulin by fasting C-peptide in modified homeostasis model assessment to evaluate insulin resistance and islet beta cell function. Zhong Nan Da Xue Xue Bao Yi Xue Ban. 2004;29(4):419–23. PMID: . [PubMed] [Google Scholar]
- 24.Minh HV, Tien HA, Sinh CT, et al. Assessment of preferred methods to measure insulin resistance in Asian patients with hypertension. J Clin Hypertens (Greenwich). 2021;23(3):529–537. PMID: . 10.1111/jch.14155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Singh Y, Garg MK, Tandon N, Marwaha RK. A study of insulin resistance by HOMA-IR and its cut-off value to identify metabolic syndrome in urban Indian adolescents. JCRPE J Clin Res Pediatr Endocrinol. 2013;5(4):245–51. PMID: . PMCID: . 10.4274/Jcrpe.1127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Albareda M, Rigla M, Rodríguez-Espinosa J, et al. Influence of exogenous insulin on C-peptide levels in subjects with type 2 diabetes. Diabetes Res Clin Pract. 2005;68(3):202–6. PMID: . 10.1016/j.diabres.2004.10.005. [DOI] [PubMed] [Google Scholar]
- 27.Gjessing HJ, Reinholdt B, Pedersen O. The plasma C-peptide and insulin responses to stimulation with intravenous glucagon and a mixed meal in well-controlled Type 2 (non-insulin-dependent) diabetes mellitus: Dependency on acutely established hyperglycaemia. Diabetologia. 1989;32(12):858–63. PMID: . 10.1007/BF00297450. [DOI] [PubMed] [Google Scholar]
- 28.Sliwinska-Mosson M, Milnerowicz H. The impact of smoking on the development of diabetes and its complications. Diabetes Vasc Dis Res. 2017;14(4):265–76. PMID: . 10.1177/1479164117701876. [DOI] [PubMed] [Google Scholar]
- 29.Östgren CJ, Lindblad U, Ranstam J, Melander A, Råstam L; Skaraborg Hypertension and Diabetes Project . Associations between smoking and β-cell function in a non-hypertensive and non-diabetic population. Diabet Med. 2000;17(6):445–50. PMID: . 10.1046/j.1464-5491.2000.00294.x. [DOI] [PubMed] [Google Scholar]
- 30.Frati AC, Iniestra F, Ariza CR. Acute effect of cigarette smoking on glucose tolerance and other cardiovascular risk factors. Diabetes Care. 1996;19(2):112–8. PMID: . 10.2337/diacare.19.2.112. [DOI] [PubMed] [Google Scholar]
- 31.Eliasson B, Mero N, Taskinen MR, Smith U. The insulin resistance syndrome and postprandial lipid intolerance in smokers. Atherosclerosis. 1997;129(1):79–88. PMID: . 10.1016/s0021-9150(96)06028-5. [DOI] [PubMed] [Google Scholar]
- 32.Ahlqvist E, Storm P, Käräjämäki A, et al. Novel subgroups of adult-onset diabetes and their association with outcomes: a data-driven cluster analysis of six variables. Lancet Diabetes Endocrinol. 2018;6(5): 361–9. PMID: . 10.1016/S2213-8587(18)30051-2. [DOI] [PubMed] [Google Scholar]
- 33.Zou X, Zhou X, Zhu Z, Ji L. Novel subgroups of patients with adult-onset diabetes in Chinese and US populations. Lancet Diabetes Endocrinol. 2019;7(1):9–11. PMID: . 10.1016/S2213-8587(18)30316-4. [DOI] [PubMed] [Google Scholar]
- 34.Anjana RM, Baskar V, Nair ATN, et al. Novel subgroups of type 2 diabetes and their association with microvascular outcomes in an Asian Indian population: A data-driven cluster analysis: The INSPIRED study. BMJ Open Diabetes Res Care. 2020;8(1):e001506. PMID: . PMCID: . 10.1136/bmjdrc-2020-001506. [DOI] [PMC free article] [PubMed] [Google Scholar]