diabetes research and clinical practice 80 (2008) available at journal homepage:

Transcription

1 available at journal homepage: Islet autoimmunity status in Asians with young-onset diabetes (12 40 years): Association with clinical characteristics, beta cell function and cardio-metabolic risk factors A.C. Thai a, *, V. Mohan b, B.A.K. Khalid c, C.S. Cockram d, C.Y. Pan e, P. Zimmet f, J.P. Yeo g the ASDIAB Study Group 1 a Department of Medicine, Yong Loo Lin School of Medicine, National University of Singapore, 5 Lower Kent Ridge Road, Singapore, Singapore b Madras Diabetes Research Foundation, Chennai, India c Faculty of Medicine, Universiti Kebangsaan Malaysia, Kebangsaan, Malaysia d Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong, SAR China e Chinese PLA General Hospital, Beijing, China f International Diabetes Institute, Melbourne, Victoria, Australia g Novo Nordisk International Operations Clinical Development Centre, Singapore article info Article history: Received 21 November 2007 Accepted 4 December 2007 Published on line 18 January 2008 Keywords: GADA IA-2A Beta cell function Metabolic syndrome Young Asians abstract In this paper, the islet autoimmunity status and relation to clinical characteristics, beta cell function and cardio-metabolic risk factors in young-onset Asian diabetic patients are evaluated at baseline. The study population consisted of 912 patients (from China, India, Malaysia and Singapore) with age years and diabetes duration <12 months. Autoantibodies to glutamic acid decarboxylase (GADA) and tyrosine phosphatase (IA-2A), beta cell function and cardio-metabolic risk parameters were assessed. Among our young patient cohort, 105 (11.5%) patients were GADA and/or IA-2A positives (Ab +ve). Ab +ve patients were younger, leaner, had more severe hyperglycaemia and lower beta cell function. The frequency of metabolic syndrome was significantly lower in Ab +ve patients (27%) compared to Ab ve patients (54%). However, a substantial proportion of patients in both groups of patients had atherogenic dyslipidaemia, hypertension and albuminuria (micro or macro). In our study cohort, only one in 10 Asian youth with new-onset diabetes had evidence of islet autoimmunity. At least 60% of Ab +ve and 50% of Ab ve patients demonstrated classical features of type 1 and type 2 diabetes respectively. Regardless of autoimmunity status, the cardio-metabolic risk factors, in particular atherogenic dyslipidaemia, hypertension and albuminuria were common in our patients with young-onset diabetes. # 2007 Elsevier Ireland Ltd. All rights reserved. * Corresponding author. Tel.: ; fax: address: mdctac@nus.edu.sg (A.C. Thai). 1 A list of other persons in ASDIAB Study Group and their affiliated institutions appears in the Appendix. Abbreviations: Ab +ve, autoantibody-positive; Ab ve, autoantibody-negative; GADA, autoantibodies against glutamic acid decarboxylase; IA-2A, autoantibodies against tyrosine phosphatase /$ see front matter # 2007 Elsevier Ireland Ltd. All rights reserved. doi: /j.diabres

2 Introduction Type 1 diabetes in Asian populations is less common than in European populations [1,2]. In contrast, the prevalence of type 2 diabetes in developing countries in Asia is amongst the highest worldwide [3]. Another disturbing observation is the increasing trend of type 2 diabetes occurring in Asian children and adolescents, associated with increasing prevalence of obesity and a strong family history of type 2 diabetes [4 7]. Otani et al. [8] first reported the presence of young-onset type 2 diabetes patients among Japanese in Young-onset type 2 diabetes has since been reported to be present among other Asian youths in China, India, Korea, Malaysia and Singapore [7,9]. Categorizing patients with young-onset diabetes into different types of diabetes may often be more difficult than those with adult-onset diabetes [10,11]. Clinical phenotypes at onset frequently overlap [12 14] and age at diagnosis poorly differentiates between types. In addition, there are also differences in clinical, biochemical and immunological variables between Asian and Caucasian diabetic patients. For instance, in Europe and America, majority of children and adolescents with type 2 diabetes are obese while in the Indian subcontinent, children with type 2 diabetes patients are generally non-obese [7,12,15,16] Similarly, young-onset Japanese type 2 diabetes patients were on average not obese compared to the Western population [15 18]. In addition, the reported frequencies of islet-cell cytoplasm (ICA) or glutamic acid decarboxylase (GADA) positivity are higher in Caucasians compared to Asians [16,19 21]. There is a scarcity of data on young Asian patients with newly diagnosed diabetes from Chinese, Indian and Malay ethnic background. These are the three major ethnic groups within Asia, with documented high prevalence rates and are expected to contribute significantly to the predicted rise in the number of persons with diabetes in Asia [3]. Given the ethnic differences and heterogeneity in diabetes patients, data from previous studies may not be applied directly. Thus, the Asian Young Diabetes (ASDIAB) study was conducted to obtain detailed baseline as well as 5-year follow-up data on the clinical, immunological, metabolic characteristics and natural history of young patients with newly diagnosed diabetes from the three major ethnic groups within Asia. This paper aims to present the islet autoimmunity status and its relation to clinical characteristics, beta cell function and cardio-metabolic risk factors in our young-onset Asian diabetic patient cohort at baseline. 2. Materials and methods Details of the ASDIAB study population, methodology and baseline characteristics have been previously published [22] Patients The study recruited 925 eligible young-onset diabetic patients of Chinese, Indian or Malay ethnicity aged years within 12 months of the onset of disease. Autoantibodies data were available in 912 of these patients for analyses. Six centres from China (Beijing, Shanghai and Hong Kong), India (Chennai), Malaysia and Singapore were selected. These centres are involved mainly in hospital-based secondary care of diabetic patients in the respective cities. Patients were recruited from the pool of recently diagnosed patients with diabetes who either presented to these centres as hospital admissions or were referred from other clinics within their catchment area. All patients presented during the recruitment period, regardless of their mode of presentation and treatment, were recruited consecutively. Compliance of patients was monitored by a research nurse in the participating centre and patient retention was by means of education and maintenance of doctor patient and nurse patient relationships. Patients were recalled by phone, letter or home visit in one of the centres (Chennai) in case of defaulted follow-up. All patients gave written and signed informed consent and the protocol was approved by ethics committee in each centre. Those receiving oral hypoglycaemic agents prior to enrolment were given a wash-out period of 4 weeks in order to obtain baseline data. However, in cases where the fasting blood glucose exceeded 15 mmol/l after 1 week of stopping the oral agents, their oral treatment was reinstituted and they were then enrolled into the study at that point. This was not applicable to insulin-treated patients as they continued with their treatment. Exclusion criteria included: (1) participation in any interventional drug trial, (2) drug-induced diabetes, (3) gestational diabetes mellitus, (4) conditions which make it unlikely that the patients can be followed for the subsequent 5 years Laboratory and clinical examinations Autoantibodies to GADA and tyrosine phosphatase (IA-2A) were assessed by a central laboratory (International Diabetes Institute, Caulfield, Victoria, Australia). The 65-kDa isoform of GADA was measured by a liquid-phase radioimmunoprecipitation assay developed by Chen et al. [23] and Law et al. [24], with interassay coefficient of variations (CVs) of 6.9% for low positive control serum and 6.1% for high positive control serum. IA-2A was determined by a commercial kit (RSR Ltd., Cardiff, UK) with sensitivity and specificity of 57% and 99% respectively [25]. Fasting and glucagon-stimulated plasma C-peptide (GSCpeptide) concentrations were measured by radioimmunoassay (human C-peptide Linco RIA kit, cat #HCP-20K) with an intra-assay CV ranging from % for five duplicate determinations of five different samples. In this study, a fasting C-peptide concentration of 0.3 nm or GSC-peptide taken 6 min after intravenous one mg glucagon of 0.6 nm, was considered low and inferred as indicating poor beta cell functional reserve. Blood samples were also taken for measurement of biochemical parameters including fasting plasma glucose (FPG), lipids, HbA 1c and insulin. Insulin concentrations were measured using Linco radioimmunoassay (RIA) kits (human insulin specific RIA kit, cat #HI-14K). Insulin resistance was calculated as a fasting plasma insulin (mu/l) and glucose (mmol/l) product divided by This is numerically the same as that derived from the homeostasis model assessment (HOMA) equation: insulin resistance = fasting serum insulin/[22.5 e In(fasting plasma glucose) ] [26].

3 226 Documentation of demographic data and medical information concerning cardiovascular risk factors was performed. Clinical assessments included measurement of body mass index (BMI), waist circumference and blood pressure. Metabolic syndrome was defined according to the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) guideline [27] and all patients had new-onset diabetes, thus assumed to have fulfilled the cut-off for fasting plasma glucose at diagnosis. Overweight was defined as BMI 25 to <30 kg/m 2 ; obese as BMI 30 kg/m 2 ; abdominal obesity was defined as waist circumference >102 cm for male or >88 cm for female; hypertension was defined as systolic BP 130 mmhg and/or diastolic BP 85 mmhg or on anti-hypertensive medication; high triglycerides was defined as 1.7 mmol/l; low HDLcholesterol was defined as <1.04 mmol/l for male or <1.30 mmol/l for female and high LDL-cholesterol was defined as 2.59 mmol/l according to the NCEP ATP III criteria [27]. Overweight was also defined as BMI 23 to <25 kg/m 2, obesity as BMI 25 kg/m 2 and abdominal obesity as waist circumference 90 cm for male or 80 cm for female according to the modified Asia-Pacific WHO guidelines [28]. Normoalbuminuria was defined as urine albumin/creatinine <30 mg/g; microalbuminuria as urine albumin/creatinine mg/g and macroalbuminuria as urine albumin/creatinine 300 mg/g according to the criteria prescribed by the American Diabetes Association (ADA) [29] Data analysis Data were entered into a Microsoft 1 Access 1 database and analyzed using Statistical Analysis System (SAS 1, version 8.2, SAS Institute Inc., USA). Data are shown as median (interquartiles) and frequency (percentage). For continuous variables, non-parametric Mann Whitney U-test was used to evaluate differences between Ab +ve and Ab ve patient groups. Chi-Square test was used to compare the categorical variables between the patient groups when none of the expected values were less than 1 and no more than 20% of the expected values were less than 5. Otherwise, Fisher s Exact test was carried out. Statistical significance was set at 5% level. 3. Results 3.1. Demographic and clinical characteristics Baseline autoantibody status was available in 912 patients. Overall 105 (11.5%) patients were autoantibody positive (Ab +ve) for GAD and/or IA-2, whilst 807 (88.5%) patients were autoantibody negative (Ab ve). Of those Ab +ve patients: 81 (77.1%) had GADA alone, 5 (4.8%) had IA-2A alone and 19 (18.1%) had both GADA and IA-2A. The Ab +ve patients compared to Ab ve patients (Table 1) were younger ( p < 0.001), with median diabetes onset age of 27 vs. 34 years respectively, with a higher proportion having onset age 30 years (61% vs. 30%). Both groups had preponderance of male patients and there were no significant gender and ethnic differences Beta cell function, glycaemia and diabetes treatment Median fasting C-peptide and GSC-peptide concentrations were respectively lower ( p < 0.001) in the Ab +ve patients compared to Ab ve patients (Table 2). Ab +ve patients were more likely to have low C-peptide concentration ( p < 0.001) and low GSC-peptide concentration ( p < 0.001) compared to Ab ve patients. Fifty-eight percent of Ab +ve patients had low fasting C-peptide concentration compared with 12% of Ab ve patients. Likewise, 66% of Ab +ve patients had low GSCpeptide concentration compared with 13% of Ab ve patients. Median fasting plasma glucose and HbA 1c values were significantly higher ( p < 0.001) in Ab +ve patients than Ab ve patients. Fifty-eight percent of Ab +ve patients received insulin treatment, whilst majority (86%) of Ab ve patients were not insulin treated, but received diet and/or oral agents Metabolic syndrome and cardio-metabolic risk factors The percentage of patients identified as having metabolic syndrome was significantly lower among Ab +ve patients (27%) compared to Ab ve patients (54%), p < (Table 3). After adjusting for age, the odds of those with Ab ve in Table 1 Demographics and clinical characteristics according to baseline autoantibody status Ab +ve Ab ve p-value n (% of 912 patients) 105 (11.5) 807 (88.5) Gender, n (%) Male 66 (62.9) 532 (65.9) Ethnicity, n (%) Chinese 67 (63.8) 451 (55.9) Indian 26 (24.8) 242 (30.0) Malay 12 (11.4) 114 (14.1) Age at diagnosis (years) Median (Q1 Q3) ( ) ( ) <0.001 * years, n (%) 39 (37.1) 67 (8.3) <0.001 * years, n (%) 25 (23.8) 176 (21.8) years, n (%) 41 (39.1) 564 (69.9) Note: The delineations for antibody (Ab) grouping are: Ab +ve = GADA and/or IA-2A positive; Ab ve = both GADA and IA-2A negative. * Statistical significance at 5% level.

4 227 Table 2 Glycaemic indices, beta cell function and diabetes treatment by autoantibody status Ab +ve Ab ve p-value n (% of 912 patients) 105 (11.5) 807 (88.5) HbA 1c (%) ( ) ( ) <0.001 * Fasting plasma glucose (mmol/l) ( ) 8.60 ( ) <0.001 * Fasting C-Peptide (nmol/l) 0.23 ( ) 0.70 ( ) <0.001 * n (%) with 0.3 nmol/l 60 (57.7) 96 (11.9) <0.001 * Glucagon-stimulated C-peptide (nmol/l) 0.43 ( ) 1.34 ( ) <0.001 * n (%) with 0.6 nmol/l 69 (65.7) 107 (13.3) <0.001 * Diabetes drug treatment, n (%) Insulin 61 (58.1) 116 (14.4) <0.001 * Non-insulin 44 (41.9) 691 (85.6) Note: The delineations for antibody (Ab) grouping are: Ab +ve = GADA and/or IA-2A positive, Ab ve = both GADA and IA-2A negative. Results are median (interquartile range). * Statistical significance at 5% level. experiencing metabolic syndrome was 2.70 times higher (95% CI ). Patients who were Ab ve had significantly higher ( p < 0.001) median BMI, systolic and diastolic blood pressure, triglyceride level and lower HDL-cholesterol ( p < 0.001). However, a similar high proportion of patients in both groups had elevated LDL-cholesterol levels (Ab ve: 76% vs. Ab +ve: 70%). Insulin resistance by HOMA index was higher in Ab ve patients compared with Ab +ve patients ( p = 0.027). A higher proportion of Ab ve patients had hypertension (41% vs. 20%) and albuminuria (34% vs. 20%) compared to Ab +ve patients. 4. Discussion In a large cohort of 912 young Asians years old with incident diabetes, our study found a relatively low proportion (11.5%) of our patients with islet autoantibodies. This is in contrast to the high 67% antibody positivity rate reported in 313 young Swedish patients aged years with new-onset diabetes [16]. Among the Ab +ve patients, at least 60% had classical features of type 1 diabetes lean body weight [BMI <25 kg/m 2 (NCEP guideline): 86%, BMI <23 kg/m 2 (Asian guideline): 76%], low beta cell reserve (58 66%) or need for insulin treatment (58%). They were severely hyperglycaemic with high median HbA 1c and fasting plasma glucose at diagnosis. These characteristics were similar to those reported in the type 1 Ab +ve diabetic patients in the Swedish study [16]. In the SEARCH study on youth in United States [30], type 1 Ab +ve patients were also insulin requiring (100%) and had low mean fasting C-peptide level (0.3 nmol/l) at diagnosis. In Ab ve patients, at least 50% had the typical characteristics described in type 2 diabetes overweight and obese [BMI 25 kg/m 2 (NCEP guideline): 49%, BMI 23 kg/m 2 (Asian guideline) 70%], adequate beta cell function (87 88%) or not requiring insulin therapy (86%). Comparing the proportion of Ab ve patients defined as obese by NCEP ATP III criteria [27] (15%) and modified Asia-Pacific WHO guidelines [22] (50%), the former was noted to be much lower than those (73%) reported in young Swedish Ab ve type 2 diabetes patients [16]. This observation lends further support to the lower cut-off criteria recommended for defining obesity in Asians [28]. Of note, obesity was present in only half of our Ab ve patients (according to the Asian guideline). This is similar to that observed in young type 2 diabetes patients in India where half of the patients were not obese [7]. In addition, young Japanese type 2 diabetes patients were also on average not obese (mean BMI: kg/m 2 ) [17]. These findings are in contrast to those of Western populations where the majority of type 2 diabetes patients are obese [15,16]. In our study cohort, the frequency of metabolic syndrome was lower among Ab +ve patients (27% vs. 54%) and they were also less insulin resistant than Ab ve patients. However, atherogenic dyslipidaemia (elevated triglycerides 26% vs. 46%, low HDL-cholesterol 47% vs. 59% and high LDL-cholesterol 70% vs. 76%) and hypertension (20% vs. 41%) were common in both groups of patients. The SEARCH study [31] also reported a substantial proportion of young-onset Asian diabetes patients aged <20 years with atherogenic dyslipidaemia (elevated triglycerides 30% and low HDL-cholesterol 24%) and hypertension (39%). These observations are of particular concern, as they suggest increased cardiovascular disease risk among Asian youth with diabetes at an early age. A substantial proportion of our diabetes patients had albuminuria and this raised concern for increased risk of overt kidney disease, cardiovascular disease and death. Ng et al. [32] reported that in 65 Chinese, obese patients with early-onset type 2 diabetes, there was clustering of cardiovascular risk factors, and increased rates of retinopathy (14%) and albuminuria (62%). In addition, a Japanese study documented high incidence rate of diabetic nephropathy in early-onset type 2 diabetes patients [18]. Although the above mentioned studies were not comparable with ASDIAB in terms of duration of diabetes and patients lipid profiles, the findings highlight the need to screen for microalbuminuria and better manage type 2 diabetes. In this study, we selected hospital-based diabetes clinics in major cities in Asia with the necessary clinical management expertise, derived from previous clinical and epidemiological studies. This may lead to selection bias, due to differences in the socio-economic status or insurance system across the countries, the patients accessibility to the diabetes centre and only better managed patients (and thus clinically less severe)

5 228 Table 3 Metabolic syndrome and cardio-metabolic risk factors by autoantibody status Ab +ve Ab ve p-value n (% of 912 patients) 105 (11.5) 807 (88.5) Metabolic syndrome a, n (%) (NCEP ATP III definition) 28 (26.9) 430 (53.5) <0.001 * Body mass index (kg/m 2 ) ( ) ( ) <0.001 * Normal a (<25 kg/m 2 ) 90 (85.7) 412 (51.2) <0.001 * Overweight a (25 <30 kg/m 2 ) 11 (10.5) 273 (33.9) Obese a (30 kg/m 2 ) 4 (3.8) 120 (14.9) <0.001 * Normal b (<23 kg/m 2 ) 80 (76.2) 244 (30.3) Overweight b (23 <25 kg/m 2 ) 10 (9.5) 168 (20.9) Obese b (25 kg/m 2 ) 15 (14.3) 393 (48.8) Waist circumference ( ) ( ) <0.001 * Abdominal obesity a, n (%) 5 (4.8) 116 (14.5) * Waist a >102 cm (male) 4 (3.8) 56 (6.9) Waist a >88 cm (female) 1 (1.0) 60 (7.4) Abdominal obesity b, n (%) 13 (12.5) 360 (45.0) <0.001 * Waist b 90 cm (male) 10 (9.5) 223 (27.6) Waist b 80 cm (female) 3 (2.9) 137 (17.0) Systolic BP (mmhg) ( ) ( ) <0.001 * Diastolic BP (mmhg) ( ) ( ) <0.001 * n (%) with SBP 130 mm Hg and/or DBP 85 mm Hg 21 (20.0) 315 (39.1) <0.001 * n (%) on anti-hypertensive medication 2 (1.9%) 65 (8.1) * Hypertensive a, n (%) 21 (20.0) 331 (41.1) <0.001 * Triglycerides (mmol/l) 1.14 ( ) 1.58 ( ) <0.001 * n (%) with high triglycerides a 27 (26.0) 368 (45.7) <0.001 * HDL-cholesterol (mmol/l) 1.19 ( ) 1.05 ( ) <0.001 * n (%) with low HDL-cholesterol a 49 (47.1) 477 (59.3) * LDL-cholesterol (mmol/l) 3.21 ( ) 3.37 ( ) n (%) with high LDL-cholesterol a 73 (70.2) 607 (76.2) HOMA insulin resistance (mmol mu)/l ( ) 5.91 ( ) * Urine albumin/creatinine ratio (mg/g) ( ) ( ) Albuminuria, n (%) Normoalbuminuria c 83 (79.8) 528 (65.7) * Microalbuminuria c 18 (17.3) 232 (28.9) Macroalbuminuria c 3 (2.9) 44 (5.5) Note: The delineations for antibody (Ab) grouping are: Ab +ve = GADA and/or IA-2A positive, Ab ve = both GADA and IA-2A negative. Results are median (interquartile range). a Defined according to the NCEP ATP III guideline (see Section 2) [14]. b Defined according to the modified Asia-Pacific WHO guidelines (see Section 2) [15]. c Defined according to the criteria prescribed by the ADA (see Section 2) [16]. * Statistical significance at 5% level. are seen in the diabetes clinics. Nevertheless, it is noteworthy that our study provides further support to the findings of other studies which reported lower frequencies of GADA and ICA positivity in Asians [20,21,33,34] as compared to Caucasians [16,19,35]. The reason for this observation is still unclear. It could be Asians may be genetically less susceptible to type 1 diabetes [36,37], or there are other, yet to be identified autoantigens, which may be involved [21,38]. In addition, different genetic environmental interactions might operate in the etiology of type 1 diabetes between Asians and Caucasians [36]. Further investigations need to be carried out to explain the differences. Further, our study also concurs with previous studies which reported that Asian type 2 diabetes patients were less obese [7,17,18,39] than Caucasians [12,15,16]. These show that Asians tend to develop diabetes with a lesser degree of obesity compared to Caucasians. With childhood obesity increasing substantially [5,6,40 42] and the prevalence of type 2 diabetes reaching epidemic levels in Asia [43], urgent action on lifestyle changes such as weight control and exercise are needed in this young Asian population to reduce the impact of this epidemic on healthcare resources in the region. Comparing the proportion of our patient cohort classified as obese by the NCEP ATP III and Asian cut-offs, our data showed that the lower Asian cut-off seemed more appropriate for our cohort. This suggests that the use NCEP ATP III criteria for the definition of metabolic syndrome may have underestimated the population at risk and the frequency of metabolic syndrome could be more appropriately estimated in our Asian cohort using the ethnic-specific, Asian cut-off for defining central obesity [28] in the diagnosis of metabolic syndrome.

6 229 Currently, there are no formally defined criteria for early diagnosis of metabolic syndrome in children, therefore in this study, the NCEP ATP III guideline was used for defining metabolic syndrome in our patient cohort which consisted of about 10% of children aged years. The establishment of a consensus guideline for diagnosis of metabolic syndrome in children in future would allow more accurate diagnosis in this young population. In this study, we evaluated the islet autoimmunity status and relation to clinical characteristics, beta cell function and cardio-metabolic risk factors in young-onset Asian diabetic patients aged years. However, the clinical features may differ between the second decade and the fourth decade of age. Exploring these features according to age in future paper would provide useful information on the characteristics of Asian diabetic youths in different age groups. In conclusion, 1 in 10 of our young Asian patient cohort with recent-onset diabetes had evidence of islet autoimmunity. At least 60% of those with autoantibodies and 50% of those without autoantibodies had typical features of type 1 and type 2 diabetes respectively. Our study showed that the risks of developing long-term complications especially cardiovascular disease and nephropathy in Asian youth with young-onset diabetes may be high. This is of socioeconomic importance as the long-term management of hyperglycaemia, hypertension, dyslipidaemia, obesity and albuminuria presenting at this young age will add a considerable burden to the overall healthcare system in the terms of cost, manpower and infrastructure requirements. Acknowledgements This study was supported by a research grant from Novo Nordisk Asia Pacific Pte Ltd, Singapore. The authors wish to thank the study site coordinators for their participation in the study. We also like to acknowledge Choo Hui Hwa, Lee Peak Yuen, Ray Spark, Sharon Goh, Victor Ong and Wong Hung Chew for their expert technical assistance. Appendix A A.1. Other members of the ASDIAB Study Group Prof Zhu Xi Xing, Huashan Hospital, Shanghai, China; Dr Shi Hong Li, Huashan Hospital, Shanghai, China; Dr Lu Juming, Chinese PLA General Hospital, Beijing, China; Dr Wan Nazaimoon Wan Mohamud, Institute for Medical Research (Ministry of Health), Kuala Lumpur, Malaysia; A/Prof Dr Ikram Shah Ismail, University Malaya Medical Centre, Kuala Lumpur, Malaysia; Dr Noraini Nadior, Damansara Specialist Hospital, Kuala Lumpur, Malaysia; Dr G R Letchuman, Ipoh Hospital (Ministry of Health), Perak, Malaysia; Dato Dr Singaraveloo, Johor Bahru Hospital (Ministry of Health), Johor, Malaysia; Prof Dr Wan Mohamad Wan Bebakar, Hospital Universiti Sains Malaysia, Kelantan, Malaysia; Dr Rema Mohan, Madras Diabetes Research Foundation, Chennai, India; Dr Soon Puay Cheow, Tan Tock Seng Hospital, Singapore; Dr Loh Keh Chuan, Tan Tock Seng Hospital, Singapore; Dr Lee Choon Horn, Changi General Hospital, Singapore; Dr Tavintharan Subramaniam, Changi General Hospital, Singapore; Dr Chionh Siok Bee, National University Hospital, Singapore. Conflict of Interest The authors state that they have no conflict of interest. references [1] P. Onkamo, S. Väänänen, M. Karvonen, J. Tuomilehto, Worldwide increase in incidence of type 1 diabetes-the analysis of the data on published incidence trends, Diabetologia 42 (1999) [2] M. Karvonen, J. Tuomilehto, I. Libman, R. LaPorte, World Health Organization DIAMOND Project Group, A review of the recent epidemiological data on the worldwide incidence of type 1 (insulin-dependant) diabetes mellitus, Diabetologia 36 (1993) [3] S. Wild, G. Roglic, A. Green, R. Sicree, H. King, Global prevalence of diabetes-estimates for the year 2000 and projections for 2030, Diabetes Care 27 (2004) [4] T. Kitagawa, M. Owada, T. Urakami, K. Yamauchi, Increased incidence of non-insulin dependent diabetes mellitus in Japanese school children correlates with an increased intake of animal protein and fat, Clin. Pediatr. 37 (1998) [5] S. Likitmaskul, P. Kiattisathavee, K. Chaichanwatanakul, L. Punnakanta, K. Angsusingha, C. Tuchinda, Increasing prevalence of type 2 diabetes mellitus in Thai children and adolescents associated with increasing prevalence of obesity, J. Pediatr. Endocrinol. Metab. 16 (2003) [6] T. Urakami, S. Kubota, Y. Nitadori, K. Harada, M. Owada, T. Kitagawa, Annual incidence and clinical characteristics of type 2 diabetes in children as detected by urine glucose screening in the Tokyo metropolitan area, Diabetes Care 28 (2005) [7] A. Ramachandran, C. Snehalatha, K. Satyavani, S. Sivasankari, V. Vijay, Type 2 diabetes in Asian-Indian urban children, Diabetes Care 26 (2003) [8] T. Otani, H. Yokoyama, Y. Higami, T. Kasahara, Y. Uchigata, Y. Hirata, Age of onset and type of Japanese younger diabetics in Tokyo, Diabetes Res. Clin. Pract. 10 (1990) [9] C.S. Cockram, Diabetes mellitus: perspective from the Asia- Pacific region, Diabetes Res. Clin. Pract. 50 (Suppl. 2) (2000) S3 S7. [10] H.J. Arnqvist, B. Littorin, L. Nyström, B. Schersten, J. Ostman, J. Blohme, et al., Difficulty in classifying diabetes at presentation in the young adult, Diabet. Med. 10 (1993) [11] American Diabetes Association, Type 2 diabetes in children and adolescents, Diabetes Care 23 (2000) [12] A. Fagot-Campagna, D.J. Pettitt, M.M. Engelgau, N.R. Burrows, L.S. Geiss, R. Valdez, et al., Type 2 diabetes among North American children and adolescents: an epidemiologic review and a public health perspective, J. Pediatr. 136 (2000) [13] A. Ferrara, G. Imperatore, C.P. Quesenberry, A.J. Karter, A. Fagot-Campagna, I.J. Ergas, et al., Characteristics and quality of health care of type 1 and type 2 diabetic youths: the Northern California Kaiser Performance Diabetes Registry, Diabetes 51 (Suppl. 2) (2002) A26. [14] T.G. Barrett, S. Ehtisham, A. Smith, A.T. Hattersley, UK Diabetes survey shows type 2 diabetes present in 0.4% of

7 230 newly diagnosed children, associated with overweight, female, and ethnic minority status, Diabetes 51 (Suppl. 2) (2002) A25. [15] M. Ganz (Ed.), Prevention of Type 2 Diabetes, Wiley, New York, NY, 2005, pp [16] H. Borg, H.J. Arnqvist, E. Björk, J. Bolinder, J.W. Eriksson, L. Nyström, et al., Evaluation of the new ADA and WHO criteria for classification of diabetes mellitus in young adult people (15 34 years) in the Diabetes Incidence Study in Sweden (DISS), Diabetologia 46 (2003) [17] H. Yokoyama, M. Okudaira, T. Otani, A. Sato, J. Miura, H. Takaike, et al., Higher incidence of diabetic nephropathy in type 2 than in type 1 diabetes in early-onset diabetes in Japan, Kidney Int. 58 (2000) [18] H. Yokoyama, M. Okudaira, T. Otani, C. Watanabe, H. Takaike, J. Miuira, et al., High incidence of diabetic nephropathy in early-onset Japanese NIDDM patients: risk analysis, Diabetes Care 21 (1998) [19] C.F. Verge, N.J. Howard, M.J. Rowley, I.R. Mackay, P.Z. Zimmet, M. Egan, et al.,a population-based study, Antiglutamate decarboxylase and other antibodies at the onset of childhood IDDM, Diabetologia 37 (1994) [20] P.Z. Zimmet, M.J. Rowley, I.R. Mackay, W.J. Knowles, Q.Y. Chen, L.H. Chapman, et al., The ethnic distribution of antibodies to glutamic acid decarboxylase: presence and levels of insulin-dependent diabetes mellitus in Europid and Asian subjects, J. Diabetes Complications 7 (1993) 1 7. [21] W.M. Wan Nazaimoon, I. Faridah, M. Singaraveloo, I.S. Ismail, W.B. Wan Mohamad, R. Letchuman, et al., Prevalence of glutamic acid decarboxylase antibodies amongst young Malaysian diabetics, Diabetes Res. Clin. Pract. 43 (1999) [22] C.Y. Pan, W.Y. So, B.A.K. Khalid, V. Mohan, A.C. Thai, P. Zimmet, et al., Metabolic, immunological and clinical characteristics in newly diagnosed Asian diabetes patients aged years, Diabet Med. 21 (2004) [23] Q.Y. Chen, M.J. Rowley, G.C. Byrne, T.W. Jones, T. Tuomi, W.J. Knowles, et al., Antibodies to glutamic acid decarboxylase in Australian children with insulindependent diabetes mellitus and their first-degree relatives, Pediatr. Res. 34 (1993) [24] R.H. Law, M.J. Rowley, I.R. Mackay, B. Corner, Expression in Saccharomyces cerevisiae of antigenically and enzymatically active recombinant glutamic acid decarboxylase, J. Biotechnol. 61 (1998) [25] C.F. Verge, D. Stenger, E. Bonifacio, P.G. Colman, C. Pilcher, P.J. Bingley, et al., Combined use of autoantibodies (IA-2 autoantibody, GAD autoantibody, insulin autoantibody, cytoplasmic islet cell antibodies) in type 1 diabetes: combinatorial islet autoantibody workshop, Diabetes 47 (1998) [26] D.R. Matthews, J.P. Hosker, A.S. Rudenski, B.A. Naylor, D.F. Treacher, R.C. Turner, Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man, Diabetologia 28 (1985) [27] National Cholesterol Education Program, Third report of the expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III), NIH Publication No , National Heart, Lung, and Blood Institute, Bethesda, MD, 2002, pp [28] The Asia-Pacific perspective: redefining obesity and its treatment, Regional office for the Western Pacific of the World Health Organization, World Health Organization, International Association for the Study of Obesity and International Obesity Task Force, Health Communications Australia Pty Limited, Sydney, 2000, pp [29] American Diabetes Association, Standards of medical care in diabetes 2006, Diabetes Care 29 (2006) S4 S42. [30] D. Dabelea, R.B. D Agostino Jr., E.J. Mayer-Davis, D.J. Pettitt, G. Imperatore, L.M. Dolan, et al., Testing the accelerator hypothesis: body size, beta-cell function, and age at onset of type 1 (autoimmune) diabetes, Diabetes Care 29 (2006) [31] B.L. Rodriguez, W.Y. Fujimoto, E.J. Mayer-Davis, G. Imperatore, D.E. Williams, R.A. Bell, et al., Prevalence of cardiovascular disease risk factors in U.S. children and adolescents with diabetes: the SEARCH for diabetes in youth study, Diabetes Care 29 (2006) [32] M.C. Ng, S.C. Lee, G.T. Ko, J.K. Li, W.Y. So, Y. Hashim, et al., Familial early-onset type 2 diabetes in Chinese patients: obesity and genetics have more significant roles than autoimmunity, Diabetes Care 24 (2001) [33] V. Mohan, R. Deepa, E. Bhatia, A.K. Singh, G.A. Hitman, P.Z. Zimmet, et al., Antibodies to pancreatic islet cell antigens in diabetes seen in Southern India with particular reference to fibrocalculous pancreatic diabetes, Diabet. Med. 15 (1998) [34] J.K. Li, J.C. Chan, P.Z. Zimmet, M.J. Rowley, I.R. Mackay, C.S. Cockram, Young Chinese adults with new onset of diabetic ketoacidosis clinical course, autoimmune status and progression of pancreatic beta-cell function, Diabet. Med. 17 (2000) [35] M. Landin-Olsson, A. Karlsson, G. Dahlquist, L. Blom, A. Lernmark, G. Sundkvist, Islet cell and other organ-specific autoantibodies in all children developing type 1 (insulindependent) diabetes mellitus in Sweden during one year and in matched control children, Diabetologia 32 (1989) [36] Y. Park, Why is type 1 diabetes uncommon in Asia? Ann. N.Y. Acad. Sci (2006) [37] I. Kockum, A. Lernmark, G. Dahlquist, A. Falorni, W.A. Hagopian, M. Landin-Olsson, et al.,the Swedish Childhood Diabetes Study Group and The Diabetes Incidence in Sweden Study (DISS) Group, Genetic and immunological findings in patients with newly diagnosed insulindependent diabetes mellitus, Horm. Metab. Res. 28 (1996) [38] A.C. Thai, W.Y. Ng, K.Y. Loke, W.R. Lee, K.F. Lui, J.S. Cheah, Anti-GAD antibodies in Chinese patients with youth and adult-onset IDDM and NIDDM, Diabetologia 40 (1997) [39] W. Kiess, C. Marcus, M. Wabitsch (Eds.), Obesity in Childhood and Adolescence, vol. 9, Pediatr. Adolesc. Med., Basel, Karger, 2004, pp [40] M.C. Eppens, M.E. Craig, T.W. Jones, M. Silink, S. Ong, J.P. Yeo, on behalf of the International Diabetes Federation Western Pacific Region Steering Committee, Type 2 diabetes in youth from the Western Pacific region: glycaemic control, diabetes care and complications, Curr. Med. Res. Opin. 22 (2006) [41] Y. Wu, Overweight and obesity in China, BMJ 333 (2006) [42] Y. Matsushita, N. Yoshiike, F. Kaneda, K. Yoshita, H. Takimoto, Trends in childhood obesity in Japan over the last 25 years from the national nutrition survey, Obes. Res. 12 (2004) [43] K.H. Yoon, J.H. Lee, J.W. Kim, J.H. Cho, Y.H. Choi, S.H. Ko, et al., Epidemic obesity and type 2 diabetes in Asia, Lancet 368 (2006)