Fulminant type 1 dm
7 likes3,142 views
1) Fulminant type 1 diabetes is a subtype of type 1 diabetes that accounts for around 20% of acute-onset type 1 diabetes cases in Japan. It is characterized by a very rapid onset typically within 1 week and associated with diabetic ketoacidosis. 2) Genetic factors like certain HLA genotypes are associated with increased risk of developing fulminant type 1 diabetes. Viral infections also seem to trigger onset in many cases. 3) Affected individuals tend to be adults with a mean age of around 40 years. Fulminant type 1 diabetes results in almost complete loss of insulin secretion and pancreatic beta cell mass within a short period.
![ Fulminant type 1 DMの診断クライテリア;
◦ 発症早期にDKAやケトン体陽性となり,
その際のHbA1cは低値(<8.5%)
◦ 尿中, 血中C-peptideは抑制されている.
3
Table 1. Criteria for definite diagnosis of fulminant type
1 diabetes
Fulminant type 1 diabetes mellitus is confirmed when all
the following 3 findings are present.
1. Occurrence of diabetic ketosis or ketoacidosis soon
(around 7 days) after the onset of hyperglycaemic
symptoms (elevation of urinary or serum ketone bodies
at first visit)
2. Plasma glucose level ≥288 mg/dL (16.0 mmol) and
haemoglobin A1c level <8.5%a
at first visit
3. Urinary C-peptide excretion <10 µg/day or fasting
serum C-peptide level <0.3 and <0.5 ng/mL after
intravenous glucagon (or meal) load at onset
aShown in JDS value [32].
Epidemiology
By using the criteria, 19.4% of acute-onset type 1 diabetes
belongs to fulminant type 1 diabetes in the national survey
in Japan [3]. This prevalence is concordant with our first
report in which 11 of the 56 patients (19.6%) with acute-
onset type 1 diabetes are fulminant type 1 diabetes.
In Ehime study, regional register study in southwestern
part of Japan, fulminant type 1 diabetes accounted for
0.2% of all new-onset diabetes and 14.8% of acute-onset
type 1 diabetes [10]. In Korea, 7.1% of type 1 diabetes
ratio: 2.9, 2.1). DRB1∗
01:01-DQB1∗
05:01
DQB1*06:01 and DRB1∗
15:02-DQB1∗
06:0
cantly lower (2.7 versus 7.7%, 3.5 versus
versus 11.5%, respectively). The frequenc
gotes with DRB1∗
04:05-DQB1∗
04:01 (13.7
odds ratio 6.7) and DRB1∗
09:01-DQB1∗
03
1.5%, odds ratio 7.0) were significantly hig
with fulminant type 1 diabetes.
These results were concordant with our p
in which DR4-DQ4 was the most frequen
fulminant type 1 diabetes [19] and adde
findings as regard to susceptible or resista
class II HLA in fulminant type 1 diabetes.
In addition, CTLA-4 CT60 polymorph
B∗
4002 are associated with fulminant ty
in recent reports [20,21] (Table 2).
Viral infection
We have several lines of evidence that ind
tion of viral infection to the aetiology of fu
diabetes. They are (1) preceding flu-like sy
were more frequently observed in fulmina
betes, in 71.7%, than in type 1A diabetic pat
[3]; (2) elevation of broad reacting an
Diabetes Metab Res Rev 2011; 27: 959–964.](https://image.slidesharecdn.com/fulminanttype1dm-131009225504-phpapp01/85/Fulminant-type-1-dm-3-320.jpg)
![10
Figure 1. Cellular infiltration in the serial sections of a patient with fulminant type 1 diabetes. (a) Insulin, (b) glucagon, (c) CD3 (T
cell) and (d) CD68 (macrophage) [29]
ntofcellularinfiltration
Viral infection
-several virus
+ Genetic Factors+ Genetic Factors
-Class I and II HLA, CTLA-4-Class I and II HLA, CTLA-4
Accelerated antiviral immune reaction
Diabetes Metab Res Rev 2011; 27: 959–964.
膵臓の組織所見ではインスリン分泌細胞の減少,
T-cellの浸潤が認められる.](https://image.slidesharecdn.com/fulminanttype1dm-131009225504-phpapp01/85/Fulminant-type-1-dm-10-320.jpg)
Fulminant type 1 dm
- 1. Fulminant Type 1 DM 劇症型1型糖尿病 1
- 2. Fulminant type 1 DM 1型糖尿病の亜型で, 劇症発症するタイプ. ◦ アジアで多く報告されており, 日本国内の急性1型糖尿病の20%がこのタイプ. ◦ ウイルス感染症や妊娠等を契機とし, 1週間以内に急激に発症し, その際膵酵素上昇が認められることが多い. 膵臓の組織検査では, 炎症性細胞(主にT cell)の浸潤が認められる. ◦ 通常type 1 DMで見られる様な自己抗体は陰性. (GAD抗体, Islet-cell Ab, 抗インスリン抗体, Insulinoma-associated antigen 2 Ab). ◦ 症例報告レベルでは陽性例もある.(Internal Medicine 2003;42:517-520) ◦ 遺伝子が関連していると考えられており, HLA-DR-DQ, HLA-B, CTLA-4とFulminant type 1 DMの関連が示唆 2 Diabetes Metab Res Rev 2011; 27: 959–964.
- 3. Fulminant type 1 DMの診断クライテリア; ◦ 発症早期にDKAやケトン体陽性となり, その際のHbA1cは低値(<8.5%) ◦ 尿中, 血中C-peptideは抑制されている. 3 Table 1. Criteria for definite diagnosis of fulminant type 1 diabetes Fulminant type 1 diabetes mellitus is confirmed when all the following 3 findings are present. 1. Occurrence of diabetic ketosis or ketoacidosis soon (around 7 days) after the onset of hyperglycaemic symptoms (elevation of urinary or serum ketone bodies at first visit) 2. Plasma glucose level ≥288 mg/dL (16.0 mmol) and haemoglobin A1c level <8.5%a at first visit 3. Urinary C-peptide excretion <10 µg/day or fasting serum C-peptide level <0.3 and <0.5 ng/mL after intravenous glucagon (or meal) load at onset aShown in JDS value [32]. Epidemiology By using the criteria, 19.4% of acute-onset type 1 diabetes belongs to fulminant type 1 diabetes in the national survey in Japan [3]. This prevalence is concordant with our first report in which 11 of the 56 patients (19.6%) with acute- onset type 1 diabetes are fulminant type 1 diabetes. In Ehime study, regional register study in southwestern part of Japan, fulminant type 1 diabetes accounted for 0.2% of all new-onset diabetes and 14.8% of acute-onset type 1 diabetes [10]. In Korea, 7.1% of type 1 diabetes ratio: 2.9, 2.1). DRB1∗ 01:01-DQB1∗ 05:01 DQB1*06:01 and DRB1∗ 15:02-DQB1∗ 06:0 cantly lower (2.7 versus 7.7%, 3.5 versus versus 11.5%, respectively). The frequenc gotes with DRB1∗ 04:05-DQB1∗ 04:01 (13.7 odds ratio 6.7) and DRB1∗ 09:01-DQB1∗ 03 1.5%, odds ratio 7.0) were significantly hig with fulminant type 1 diabetes. These results were concordant with our p in which DR4-DQ4 was the most frequen fulminant type 1 diabetes [19] and adde findings as regard to susceptible or resista class II HLA in fulminant type 1 diabetes. In addition, CTLA-4 CT60 polymorph B∗ 4002 are associated with fulminant ty in recent reports [20,21] (Table 2). Viral infection We have several lines of evidence that ind tion of viral infection to the aetiology of fu diabetes. They are (1) preceding flu-like sy were more frequently observed in fulmina betes, in 71.7%, than in type 1A diabetic pat [3]; (2) elevation of broad reacting an Diabetes Metab Res Rev 2011; 27: 959–964.
- 4. 4 0 Hypothetical β-cell mass (%) Insulin- requiring state 100 20 <1 week Time Years Fulminant Autoimmune Figure 1 Schematic process showing the destruction of the hypothetical β-cell mass in fulminant type 1 diabetes and type 1A (autoimmune) diabetes. 25 NATURE CLINICAL PRACTICE ENDOCRINOLOGY & METABOLISM 2007; 3:37-45
- 5. Fulminant Type 1 DMの疫学 日本国内では急性発症の1型DMの19.4%を占める ◦ 新規発症のDMの0.2%の頻度. ◦ 男女比は1:1と差は無く, 男性例では平均43歳, 女性では35歳が発症年齢. 成人で多く, <20歳での発症は8.7%のみ. ◦ 遺伝子が関連しており, HLA-DR-DQ, HLA-B, CTLA-4が発症に関連している. 5 Diabetes Metab Res Rev 2011; 27: 959–964. Fulminant Type 1 Diabetes – an Important Subtype in East Asia 961 Table 2. Genetic factors in Japanese type 1 diabetes Fulminant Type 1A HLA DRB1-DQB1 (frequent) ∗04:05-∗04:01, ∗09:01-∗03:03 ∗09:01-∗03:03, ∗04:05-∗04:01, ∗08:02-∗03:02 HLA DRB1-DQB1 (Rare) ∗ 15:02-∗ 06:01, ∗ 01:01-∗ 05:01, ∗ 08:03-∗ 06:01, ∗ 15:01-∗ 06:02, ∗ 15:02-∗ 06:01 CTLA-4 CT60 AA GG HLA-B ∗ 4002 None
- 6. 遺伝子型とFulminant type発症のOR 6 FIGURE 1. OR for the genotypic combination of DR4-DQ4 and DR9-DQ3 in fulmi- nant type 1 diabetes.12 X, neither DR4-DQ4 nor DR9-DQ3. DR9-DQ3 is strongly associated with the susceptibility to fulminant type 1 diabetes (OR = 10.0, 95% CI: 2.0–49.0, P = 0.0007), even though the clin- ical phenotypes of these patients are almost the same as those whose onset were not associated with pregnancy. These results suggest that a differential Ann. N.Y. Acad. Sci. 1079: 24–30 (2006).
- 7. 発症年齢分布. ◦ 白; 男性例 灰色; 女性例 明灰; 妊娠例 7 is 39.1±15.7 years (mean±SD; Table 1).Among 161 patients, only 14 (8.7%) were under 20 years of age,indicating that more than 90% of patients Numberofpatients Age (years) 25 20 15 10 5 0 0–9 10–19 20–29 30–39 40–49 50–59 60–69 70–79 80+Figure 2 Distribution of ages at onset of fulminant type 1 diabetes. The white columns depict males and the gray columns depict females. The light gray bars show pregnant patients or those who recently gave birth. Adapted, with permission, from Hanafusa et al.16 Permission obtained from the Japan Diabetes Society © Hanafusa T et al. (2005) J Japan Diab Soc 48 (Suppl 1): A1–A13. NATURE CLINICAL PRACTICE ENDOCRINOLOGY & METABOLISM 2007; 3:37-45
- 8. 8 発症の季節性; 夏, 冬に多い傾向 (ウイルス感染に関連？) REVIEW www.nature.com/clinicalpractice/endmet Numberofpatients Month 20 15 10 5 0 Jan Feb M ar Apr M ay Jun Jul Aug Sep O ct N ov D ec Figure 4 Seasonal variation of fulminant type 1 diabetes. Numbers of patients include both patients seen in 10 diabetes centers and those reported from other hospitals. Adapted, with permission, from Hanafusa et al.16 Permission obtained from the Japan Diabetes Society © Hanafusa T et al. (2005) J Japan Diab Soc 48 (Suppl 1): A1–A13. Box 1 The human leukocyte antigen system. NATURE CLINICAL PRACTICE ENDOCRINOLOGY & METABOLISM 2007; 3:37-45
- 9. 臨床経過 急性ウイルス感染症を契機とした発症が有名で, ◦ Fulminant type 1 DMの71.7%で 孔するFlu-like symptomがある. (通常のType 1 DMでは26.9%) ◦ 抗Enterovirus IgAの上昇も認められることが多い. 他にはHHV-6, Influenza B, Mumps, Coxackie, HAV, CMV, EBVに対する 抗体が上昇している報告例があり, 発症に関連している可能性がある. 9 Diabetes Metab Res Rev 2011; 27: 959–964.
- 10. 10 Figure 1. Cellular infiltration in the serial sections of a patient with fulminant type 1 diabetes. (a) Insulin, (b) glucagon, (c) CD3 (T cell) and (d) CD68 (macrophage) [29] ntofcellularinfiltration Viral infection -several virus + Genetic Factors+ Genetic Factors -Class I and II HLA, CTLA-4-Class I and II HLA, CTLA-4 Accelerated antiviral immune reaction Diabetes Metab Res Rev 2011; 27: 959–964. 膵臓の組織所見ではインスリン分泌細胞の減少, T-cellの浸潤が認められる.
- 11. 日本国内のSurvey 222名のType 1 DM患者の解析. ◦ 43名(19.4%)がFulminant type 1 DM患者であった. それ以外に日本国内で報告された118名のFulminant type 1 DM患者を加えて 161例を解析. 11 Diabetes Care 26:2345–2352, 2003 Table 1—Profiles of fulminant and autoimmune type 1 diabetic patients Fulminant Autoimmune n 161 137 Clinical characteristics Duration of symptoms (days) 4.4 Ϯ 3.1 36.4 Ϯ 25.1 Age (years) 39.1 Ϯ 15.7 30.1 Ϯ 16.2 BMI (kg/m2 ) 20.7 Ϯ 3.9 18.8 Ϯ 2.8 Family history of type 1 diabetes 1/160 2/135 Family history of type 2 diabetes 25/119 33/102 Other autoimmune disease 9/85 17/51 Laboratory data 1 HbAlc (%) 6.4 Ϯ 0.9 12.2 Ϯ 2.2 Urinary C-peptide (g/day) 4.3 Ϯ 4.0 21.0 Ϯ 14.8 Fasting serum C-peptide (nmol/l) 0.10 Ϯ 0.07 0.23 Ϯ 0.13 Peak serum C-peptide (nmol/l) 0.10 Ϯ 0.10 0.40 Ϯ 0.26 Increment of serum C-peptide (nmol/l) 0.03 Ϯ 0.03 0.20 Ϯ 0.20 Fulminant type 1 diabetes in Japan
- 12. 12 tients (35.1 Ϯ 15.8) at onset of fulmin from 1 year to 80 adolescent onset o (onset before age 20 patients (3 male and adult onset was obse 147 patients. The mean BMI a betes was 20.7 in ful patients, which wa than that in autoim patients. There was ence in the presenc type 1 and type 2 d relatives between fu mune diabetes, but other than type 1 di quently observed in tes than in fulminan Thirst was the m tom accompanying and was observed i minant and autoim patients. Lower bod served in fulminant immune diabetes. Flu-like sympto toms, and drowsin Table 1—Profiles of fulminant and autoimmune type 1 diabetic patients Fulminant Autoimmune n 161 137 Clinical characteristics Duration of symptoms (days) 4.4 Ϯ 3.1 36.4 Ϯ 25.1 Age (years) 39.1 Ϯ 15.7 30.1 Ϯ 16.2 BMI (kg/m2 ) 20.7 Ϯ 3.9 18.8 Ϯ 2.8 Family history of type 1 diabetes 1/160 2/135 Family history of type 2 diabetes 25/119 33/102 Other autoimmune disease 9/85 17/51 Laboratory data 1 HbAlc (%) 6.4 Ϯ 0.9 12.2 Ϯ 2.2 Urinary C-peptide (g/day) 4.3 Ϯ 4.0 21.0 Ϯ 14.8 Fasting serum C-peptide (nmol/l) 0.10 Ϯ 0.07 0.23 Ϯ 0.13 Peak serum C-peptide (nmol/l) 0.10 Ϯ 0.10 0.40 Ϯ 0.26 Increment of serum C-peptide (nmol/l) 0.03 Ϯ 0.03 0.20 Ϯ 0.20 Accompanying symptoms Thirst 93.7 93.3 Body weight loss (kg) 3.5 Ϯ 2.7 5.5 Ϯ 3.7 Flu-like symptoms (total) 71.7 26.9 Fever 60.0 ND Headache 11.5 ND Sore throat 25.2 ND Cough 12.0 ND Rhinorrhea 7.9 ND Joint pain 5.5 ND Abdominal symptoms (total) 72.5 7.5 patients. There was n ence in the presence type 1 and type 2 dia relatives between fulm mune diabetes, but au other than type 1 diab quently observed in tes than in fulminant Thirst was the m tom accompanying th and was observed in minant and autoimm patients. Lower body served in fulminant d immune diabetes. Flu-like symptom toms, and drowsine quently observed in t than in the autoimmu onset (P Ͻ 0.0001) (T flu-like symptoms, f common and was ob patients. Nausea, vom also frequent and wer of patients. Of the 14 patien diabetes during or aft study, 13 belonged to (Table 1). Their onse in one patient, the 22 Laboratory data 1 HbAlc (%) 6.4 Ϯ 0.9 12.2 Ϯ 2.2 Urinary C-peptide (g/day) 4.3 Ϯ 4.0 21.0 Ϯ 14.8 Fasting serum C-peptide (nmol/l) 0.10 Ϯ 0.07 0.23 Ϯ 0.13 Peak serum C-peptide (nmol/l) 0.10 Ϯ 0.10 0.40 Ϯ 0.26 Increment of serum C-peptide (nmol/l) 0.03 Ϯ 0.03 0.20 Ϯ 0.20 Accompanying symptoms Thirst 93.7 93.3 Body weight loss (kg) 3.5 Ϯ 2.7 5.5 Ϯ 3.7 Flu-like symptoms (total) 71.7 26.9 Fever 60.0 ND Headache 11.5 ND Sore throat 25.2 ND Cough 12.0 ND Rhinorrhea 7.9 ND Joint pain 5.5 ND Abdominal symptoms (total) 72.5 7.5 Nausea, Vomiting 65.4 ND Upper abdominal pain 39.2 ND Lower abdominal pain 11.0 ND Diarrhea 5.5 ND Drowsiness 45.2 5.3 Pregnancy† 21.0 1.5 Laboratory data 2 Plasma glucose level (mmol/l) 44.4 Ϯ 20.0 24.1 Ϯ 11.8 Arterial pH 7.125 Ϯ 0.125 7.309 Ϯ 0.124 Serum exocrine pancreatic enzyme level 98/2 17/26 Amylase 74/54 11/81 Elastase-1 54/9 1/37 Lipase 50/9 5/38 Diabetes Care 26:2345–2352, 2003
- 13. tients (35.1 Ϯ 15.8) at onset of fulmin from 1 year to 80 adolescent onset o (onset before age 20 patients (3 male and adult onset was obse 147 patients. The mean BMI a betes was 20.7 in ful patients, which wa than that in autoim patients. There was ence in the presenc type 1 and type 2 d relatives between fu mune diabetes, but other than type 1 di quently observed in tes than in fulminan Thirst was the m tom accompanying and was observed i minant and autoim patients. Lower bod served in fulminant immune diabetes. Flu-like sympto toms, and drowsin Table 1—Profiles of fulminant and autoimmune type 1 diabetic patients Fulminant Autoimmune n 161 137 Clinical characteristics Duration of symptoms (days) 4.4 Ϯ 3.1 36.4 Ϯ 25.1 Age (years) 39.1 Ϯ 15.7 30.1 Ϯ 16.2 BMI (kg/m2 ) 20.7 Ϯ 3.9 18.8 Ϯ 2.8 Family history of type 1 diabetes 1/160 2/135 Family history of type 2 diabetes 25/119 33/102 Other autoimmune disease 9/85 17/51 Laboratory data 1 HbAlc (%) 6.4 Ϯ 0.9 12.2 Ϯ 2.2 Urinary C-peptide (g/day) 4.3 Ϯ 4.0 21.0 Ϯ 14.8 Fasting serum C-peptide (nmol/l) 0.10 Ϯ 0.07 0.23 Ϯ 0.13 Peak serum C-peptide (nmol/l) 0.10 Ϯ 0.10 0.40 Ϯ 0.26 Increment of serum C-peptide (nmol/l) 0.03 Ϯ 0.03 0.20 Ϯ 0.20 Accompanying symptoms Thirst 93.7 93.3 Body weight loss (kg) 3.5 Ϯ 2.7 5.5 Ϯ 3.7 Flu-like symptoms (total) 71.7 26.9 Fever 60.0 ND Headache 11.5 ND Sore throat 25.2 ND Cough 12.0 ND Rhinorrhea 7.9 ND Joint pain 5.5 ND Abdominal symptoms (total) 72.5 7.5 quently observed in than in the autoimm onset (P Ͻ 0.0001) ( flu-like symptoms, common and was o patients. Nausea, vo also frequent and we of patients. Of the 14 patie diabetes during or af study, 13 belonged (Table 1). Their onse in one patient, the 2 tient, the 26th week 29th week in one pa in three patients, th patient, the 35th we 36th week in two p week after delivery i Compared to au tients, fulminant typ markedly higher pla trations and signific pH at the first visit (T of three serum exo zyme levels—amy lipase—was increas range in 98 patient diabetes group, only mal levels of all these Nausea, Vomiting 65.4 ND Upper abdominal pain 39.2 ND Lower abdominal pain 11.0 ND Diarrhea 5.5 ND Drowsiness 45.2 5.3 Pregnancy† 21.0 1.5 Laboratory data 2 Plasma glucose level (mmol/l) 44.4 Ϯ 20.0 24.1 Ϯ 11.8 Arterial pH 7.125 Ϯ 0.125 7.309 Ϯ 0.124 Serum exocrine pancreatic enzyme level 98/2 17/26 Amylase 74/54 11/81 Elastase-1 54/9 1/37 Lipase 50/9 5/38 Serum sodium level (mEq/l) 131 Ϯ 9 137 Ϯ 4 Serum potassium level (mEq/l) 5.5 Ϯ 1.2 4.3 Ϯ 0.8 Serum chloride level (mEq/l) 94 Ϯ 10 101 Ϯ 5 Serum AST level (IU/l) 49 Ϯ 83 22 Ϯ 17 Serum ALT level (IU/l) 44 Ϯ 51 26 Ϯ 33 Serum total cholesterol level (mmol/l) 5.1 Ϯ 1.6 5.5 Ϯ 1.7 Serum triglycerides (mmol/l) 2.0 Ϯ 1.8 1.3 Ϯ 1.0 Autoantibodies Anti-GAD antibody 7/138 114/14 Anti–IA-2 antibody 0/43 31/24 Anti-thyroglobulin antibody 5/63 13/64 Anti–thyroid microsomal antibody 6/59 24/64 Data are means Ϯ SD, %, or n (positive/negative). Family history of type 1 or type 2 diabetes given for first-degree relatives. Pregnancy indicates percent female subjects ages 13–49 years who had type 1 diabetes during or after pregnancy in this study (13 of 62 fulminant diabetic subjects and 1 of 68 autoimmune diabetic subjects). Data for serum enzyme levels give number of patients with elevated value of at least one of the three enzymes/patients with no elevated value of the three enzymes. ND, not determined. Diabetes Care 26:2345–2352, 2003
- 14. Fulminant typeでは, インスリン必要量は 初期より多い. ◦ 0.7 U/kg/d程必要となる 14 tients showed elevated serum levels of at least one of these enzymes and 26 patients showed normal levels of all these enzymes. As shown in Table 1, fulminant pa- reported previously (16). None of the au- toimmune type 1 diabetic patients suf- fered from such episodes. The other fulminant and autoimmune type 1 diabe- tes patients survived the metabolic disor- sis at onset; 2) few patients were children or adolescents, and Ͼ90% of patients were adults; 3) male patients accounted for ϳ50% of all cases; 4) flu-like symp- toms, especially fever, were frequently observed at onset; and 5) almost all female patients who developed type 1 diabetes during pregnancy had the fulminant type. This nationwide survey indicated that fulminant type 1 diabetes is a major sub- type in Japan. Several cases similar to ful- minant type 1 diabetes have been reported in Japan after (16–23) and even before (8–12) Imagawa et al. proposed this novel subtype. However, few cases have been reported from Western coun- tries (24–27). The number of fulminant type 1 diabetes in Caucasians might be relatively small, and therefore this sub- type might have been disregarded among the majority of autoimmune type diabetes cases; however, the fulminant subtype does not appear to be rare, at least in the Japanese population. Second, Ͼ90% of fulminant type 1 diabetic patients were adults and no sex difference was observed among cases. The incidence of type 1 diabetes in Japanese children is 0.8 people/10,000 person- years and lower than the incidence in Eu- ropean countries (28); however, type 1 diabetes is not a rare disease. Nonetheless, few cases of fulminant diabetes were re- ported in the Japanese children or adoles- cents in this study. These findings suggested that fulminant type 1 diabetes is most likely an adult-onset disease. Fe- male sex was predominant in autoim- Figure 3—Changes in HbA1c levels and insulin injection dosages in fulminant (F) and autoim- mune (A) type 1 diabetes during 12 months after the onset. f, F, fulminant patients; Ⅺ, E, autoimmune patients. P Ͻ 0.05 for a vs. b, c, d, and e; b vs. d; e vs. f, g, and h; f vs. g; bЈ vs. dЈ and fЈ; cЈ vs. dЈ and gЈ; dЈ vs. hЈ; and fЈ vs. hЈ. Diabetes Care 26:2345–2352, 2003