Epidemiological, Phenotypic Characteristics And Etiological Investigations Of Fulminant Type 1 Diabetes

Part 1 Epidemiological Features of Fulminant Type 1 DiabetesObjective:1.To investigate the prevalence of fulminant type 1 diabetes (F1D) in the Second Xiangya Hospital.2.To investigate the temporal, spatial, population distribution and predisposing factors of F1D.Design:Hospital-based cross-sectional investigation.Subjects and Methods:With the diagnostic criteria proposed by Hanafusa, fulminant type 1 diabetes were screened from the admitted diabetic patients in the Second Xiangya Hospital from 2001 to 2008. And the temporal, spatial and population distribution and the possible predisposing factors of F1D were analyzed in the patients mentioned above and that from F1 China Participating Hospitals.Results:1.Fulminant type 1 diabetes accounts for 1.24‰(11/8801) in the consecutive hospitalized diabetes,1.5%(11/729) in type 1 diabetes and 10.3%(11/107) in ketosis-or ketoacidosis-onset type 1 diabetes in the Second Xiangya hospital.2.No regional difference or significant year and month preference of F1D were observed.3.The average age of onset of fulminant type 1 diabetic patients were 32±13 year, and no sex difference was observed. Female patients were more prone to suffering coma (P=0.041) and having lower HbAlc levels (P=0.018) than the male patients.4.Three (12.5%) fulminant type 1 diabetic patients were developing after drug hypersensitivity syndrome, and another three (12.5%) patients associated with pregnancy were observed.Conclusion:1.Fulminant type 1 diabetes accounts for about 0.1% of the hospitalized diabetes and 10% of the ketosis-or ketoacidosis-onset type 1 diabetes in the Second Xiangya hospital.2.Pathogenesis of fulminant type 1 diabetes were possibly associated with pregnancy and DHS. Part 2 Clinical Features of Fulminant Type 1 DiabetesObjective:To investigate the clinical features of fulminant type 1 diabetic patients at onset and the changes of the pancreatic beta-cell function during a three-year follow-up.Design:Case-control study.Subjects and Methods:Firstly, using retrospective review of patient records, clinical features on admission were compared between fulminant (F1D1) (n=11) and non-fulminant (nonF1D) (n=96) type 1 diabetes in Second Xiangya Hospital. Secondly, Clinial features were compared among autoimmune type 1 diabetes (T1A), idiopathic type 1 diabetes (T1B) and F1D2 patients (11 from F1D1 and another 13 patients from the F1 China participating hospitals) after matching with age and sex. The following data were collected and compared among the three groups at 6,12,24,36 months after the onset, including fasting and 2-hour-postgrandial plasma glucose concentrations, and C-peptide levels, BMI and HbAlc levels.Result:1.Compared with patients of nonF1D, patients of F1D1 had remarkably shorter duration from the onset of hyperglycemic symptoms to ketosis-or ketoacidosis, and showed higher plasma glucose concentrations and lower HbAlc levels (P< 0.001, respectively) at onset. In addition, flu-like symptoms and abdominal symptoms were more frequently observed in F1D1 than that in nonF1D (P< 0.01, respectively). There was no significant difference of GADA positive rate between F1D1 and nonF1D patients, however the different GADA titers existed [0.0078 (-0.2590, 0.3430) vs.0.1176 (-0.0700,2.3400),P<0.01].2.Compared with patients of T1A and T1B, patients of F1D2 had higher levels of plasma glucose, serum P-hydroxybutyric acid, serum amylase, glutamic-pyruvic transaminase, glutamic-pyruvic transaminas, urea nitrogen and serum creatinine, and lower levels of HbAlc, arterial pH, base excess and CO2CP at onset (all P<0.01). Patients of F1D2 had significantly lower serum sodium and chloride levels and higher serum potassium levels than the patients with T1A and TIB at onset (all P <0.01).3. Both FCP and PCP levels of fulminant type 1 diabetes were not significantly different at 6,12,24 and 36 months after the onset of overt diabetes (all P>0.05). Both FCP and PCP levels of fulminant type 1 diabetes were significantly lower than the patients with T1A and T1B at 6 and 12 months after the onset (all P< 0.01).Conclusion:1.Metabolic derangement is more severe in fulminant type 1 diabetic patients than that in autoimmune and idiopathic type 1 diabetic patients at onset. 2.The pancreatic beta-cell function of fulminant type 1 diabetic patients were almost completely destroyed at onset and no recovery was observed during a three-year follow up. Part 3 Etiological Investigation of Fulminant Type 1 DiabetesObjective:1. To investigate the HLA-DQ haplotypes of fulminant type 1 diabetes.2. To investigate the IgM of Coxsackie and Mumps viruses of fulminant type 1 diabetes at onset.3. To investigate the immunological features of fulminant type 1 diabetes.Design:Case-control study.Subjects and Methods:1. Nineteen fulminant type 1 diabetic patients were recruited. Genomic DNA was extracted from whole blood using phenol C chloroform method and the alleles of HLA-DQ were tested by PCR and sequence-based genotyping (SBT). Frequencies of HLA-DQ haplotypes were compared to those of autoimmune type 1 diabetes and normal controls based on the data of our previous work.2. Twenty-one fulminant type 1 diabetics,25 autoimmune type 1 diabetic patients and 30 normal controls were recruited. Serum were collected in the first 3 months after the onset. IgM of Coxsackie and Mumps viruses were tested with ELISA.3. Twenty fulminant type 1 diabetics and forty age-and sex-matched autoimmune type 1 diabetics were recruited. Serum were collected in the first 3 months after the onset. And GADA(glutamic acid decarboxylase autoantibody), IA2A (protein tyrosine phosphatase antibody) and ZnT8A(zinc transporter 8 autoantibody) were tested with radioligand assay, and the GAD A were reexamed at 6,12,24,36 months after the onset. Clinical features were compared between autoantibody-positive and autoantibody-negative fulminant type 1 diabetic patients.4. Six fulminant type 1 diabetic,6 autoimmune type 1 diabetic patients and 7 normal controls were recruited. PBMCs were isolated and the GAD-reactive T cells were detected by the enzyme-linked immunospot (ELISPOT) assay.Results:1. The most frequent haplotype found in F1D patients was DQA1*03-DQB1*0303 (18.4%), and the second was DQA1*0102-DQB1*0601 (15.8%). The frequency of DQA1*03-DQB 1*0303 haplotype was lower in F1D patients than that in autoimmune type 1 diabetes patients (P=0.02). However, the frequency of DQA1*0102-DQB1*0601 was higher in F1D patients than that in autoimmune type 1 diabetes patients and normal controls. (P=0.004, P =0.007, respectively)2. One of 21 F1D patients were Coxsackie IgM positive, and 5 of 21 F1D patients were Mumps IgM positive.The positive rate of Mumps virus were higher in fulminant type 1 diabetic patients than that in autoimmune type 1 diabetic patients and normal controls(P<0.05, respectively).3. Eight of 20 F1D patients were autoantibody-positive, including 7 patients GADA positive,4 patients ZnT8A positive and 3 patients both GADA and ZnT8A positive.Both the positive rate and the titers of GADA and IA2A in F1D patients were lower than that in autoimmune type 1 diabetes(P<0.01, respectively). However, no significant differences in clinical features were observed in the study, including age at onset, BMI, arterial pH, serum amylase, CO2CP and insulin requirement.4. Among subjects with fulminant type 1 diabetes (3/6), autoimmune type 1 diabetes (3/6) were recorded significant GAD-stimulated responses by ELISPOT assay, in contrast to none in normal controls.5. Two patients with GADA positive at onset turned to negative after two or three years, while the index of another patient increased three years later (from 0.343 to 1.467).Conclusion:1. HLA DQA1*0102-DQB 1*0601 maybe a susceptible haplotype of F1D patients.2. Parts of fulminant type 1 diabetic patients were infected with Coxsackie or Mumps virus before onset or at early stage.3. Autoimmunity maybe involved in the pathogenesis of parts of fulminant type 1 diabetes.Part 4 Genome-wide Expression Profiling of PBMCs in Fulminant Type 1 DiabetesObjective:To investigate differentially expressed genes of PBMCs between fulminant type 1 diabetic patients and normal controls.Design:Case-control study.Subjects and Methods:Six fulminant type 1 dibetic patients and six age-and sex-matched normal controles were recruited. PBMCs were separated by Ficoll centrifugation and total mRNA were extracted with trizol reagent. Quality and integrity of total RNA were assessed by spectrophotometry and 1% agarose gel electrophoresis. RNA was reverse-transcribed and then labled with Cy3 in the course of in vitro transcription. Labled cDNAs were then purifed and hybridized to Agilent Whole Human Genome Microarray 4 X 44k. Slides were scaned with Agilent Scanner and data were normalized and analysed with GeneSpring. Comfirmation tests were determined by Real-Time PCR in four genesResults:1. Among 44,000 genes and ESTs,759 genes expressed differentially between PBMCs from fulminant type 1 diabetic patients and normal controls when P<0.05 (307 overexpressed and 452 underexpressed), and 113 genes expressed differentially when fold changes> 1.1 and P< 0.05 (94 overexpressed and 19 underexpressed).2. These genes encode proteins with cytokines,chemokines and recepters, signal transduction, transcription factors, apoptosis and cytoskeleton.3. The results of comfirmation tests were in accordance with the results of gene chip.Conclusion:Gene expression profiling of PBMCs can distiguish fulminant type 1 diabetic patients from healthy controls. The candidate genes provided important implications for etiology and pathogenesis of fulminant type 1 diabetes.