Sage Journals: Discover world-class research

Abstract

Background

Fulminant type 1 diabetes mellitus (FT1DM), a subtype of type 1 diabetes mellitus, was first reported as a disease entity in 2000. Ketoacidosis at initial onset due to acute pancreatic cell destruction makes early diagnosis and treatment for FT1DM mandatory. In the early period of FT1DM, haemoglobin (Hb)A_1C levels are not markedly elevated. This study investigated serum glycated albumin (GA), which reflects acute short-term changes in plasma glucose, as a new clinical index for FT1DM at disease onset.

Methods

Subjects comprised 35 patients with FT1DM who had undergone measurement of HbA_1C and serum GA at initial visit and 42 patients with type 2 diabetes mellitus (T2DM) with HbA_1C <8.5% and no history of diabetes treatment as controls.

Results

HbA_1C was significantly lower in FT1DM than in T2DM, whereas serum GA was significantly higher. GA/HbA_1C ratio was thus significantly higher in FT1DM than in T2DM (3.9 ± 0.5 versus 2.8 ± 0.3; P < 0.0001). GA/HbA_1C ratio was >3.2 in 41 of 42 FT1DM patients (98%), compared with only one of 32 T2DM patients (3%).

Conclusions

Serum GA is significantly higher in FT1DM than in T2DM, whereas HbA_1C is significantly lower. FT1DM can thus be distinguished from untreated T2DM by GA/HbA_1C ratio at initial visit before treatment for diabetes.

Introduction

Fulminant type 1 diabetes mellitus (FT1DM) is a new subtype of type 1 diabetes first reported in 2000 by Imagawa et al. ¹ and characterized clinically by acute and almost complete pancreatic β-cell destruction without islet-related autoantibodies, such as glutamic acid decarboxylase antibody or insulinoma-associated antigen 2 antibody.^1–3 In Japanese patients, FT1DM reportedly accounts for about 19.4% of all cases of type 1 diabetes with abrupt onset.² In contrast to plasma glucose (PG) levels, which are markedly elevated at clinical onset, HbA_1C levels are normal or only slightly elevated in FT1DM.

Compared with non-diabetes, diabetes is known to be associated with higher glycation of several types of proteins, suggesting that some of these proteins play roles in the onset and progression of complications in chronic diabetes.⁴ Among these glycated proteins, haemoglobin (Hb)A_1C is widely used as a parameter of chronic glycaemic control.^5,6 Serum glycated albumin (GA) is also used as a parameter of glycaemic control. Since the half-life of albumin is shorter than that of erythrocytes, serum GA reflects PG excursions over a shorter period (about 14 d).^7,8 Serum GA thus more accurately reflects acute changes in PG. As FT1DM is characterized by abrupt elevation of PG, we hypothesized that high serum GA levels might offer a useful index in patients without markedly elevated HbA_1C.

FT1DM quickly results in acute ketoacidosis and requires urgent diagnosis and treatment, and misdiagnosis of FT1DM as type 2 diabetes mellitus (T2DM) can be life-threatening. However, HbA_1C is usually normal or only slightly elevated in both FT1DM and T2DM at disease onset. We therefore investigated the usefulness of serum GA or GA/HbA_1C ratio in FT1DM, particularly in comparison with T2DM.

Methods

Patients

We first enrolled 35 FT1DM patients (15 men, 20 women), including 31 patients identified from a nationwide survey² and four patients identified from case reports.^9–12 Clinical characteristics of all patients (age at onset, sex, body mass index [BMI] and date of onset of hyperglycaemia) were recorded. The following laboratory data were determined at first visit before treatment for diabetes in each hospital: PG concentration; HbA_1C levels and serum GA levels. In addition, 42 outpatients (33 men, nine women) initially diagnosed with T2DM (at initial visit) evaluated at Kinki Central Hospital between June 2006 and April 2009 who had not received treatment for diabetes were enrolled as controls. HbA_1C and serum GA levels were determined before starting diabetes treatment at first visit in these T2DM patients. The T2DM patients whose HbA_1C levels were more than 8.5% were excluded because inclusion criteria for FT1DM was HbA_1C <8.5% at first visit.^1,2 They showed HbA_1C <8.5% and no past history of diabetes treatment. Due to potential effects on HbA_1C and serum GA values, exclusion criteria for both groups were as follows: pregnancy; renal, hepatic, hematologic, malignant, or thyroid disease; and steroid therapy. Diagnosis of FT1DM and T2DM was based on previously reported criteria.^2,13

Laboratory measurements

PG was determined using the glucose-oxidase method. HbA_1C was measured by high-performance liquid chromatography, with calibration using Japan Diabetes Society Lot 2.¹⁴ Serum GA was determined by enzymatic methods using albumin-specific protease, ketoamine oxidase and albumin assay reagent (Lucica GA-L; Asahi Kasei Pharma, Tokyo, Japan).^15,16 GA was hydrolysed to amino acids by an albumin-specific proteinase, and then oxidized by ketoamine oxidase to produce hydrogen peroxide, which was measured quantitatively. GA level was calculated as the percentage of GA relative to total albumin, which was measured in the same serum sample using a new bromocresol purple method.¹⁵

Statistical analyses

Data are presented as mean ± standard deviation (SD). Correlation coefficients were calculated by simple regression analysis, and differences in means between groups were analysed by unpaired Student's t-test. Correlation coefficients were calculated by simple regression analysis. The stepwise multivariate analysis was performed to identify explanatory variables for FT1DM. The explanatory variables included for the analysis were age, sex, BMI and GA/HbA_1C ratio.

Sensitivities and specificities of each of the three methods (HbA_1C, serum GA and GA/HbA_1C ratio) were calculated and used to generate receiver operating characteristic (ROC) curves.^17,18 Evaluation of diagnostic accuracy was accomplished by ROC analyses.

All analyses were performed using StatView software (version 5.0 for Windows; Abacus Concepts, Berkeley, CA, USA). Values of P < 0.05 were considered statistically significant.

Results

Mean age was significantly younger in FT1DM patients (42.2 ± 14.4 y) than in T2DM patients (56.1 ± 13.1 y; P < 0.0001). Mean BMI was significantly lower in FT1DM patients (20.9 ± 2.7 kg/m²) than in T2DM patients (25.0 ± 4.0 kg/m²; P < 0.0001). In the FT1DM patients, mean duration of diabetes was 3.8 ± 2.3 d (Table 1).

Table 1

Clinical characteristics of fulminant type 1 diabetes mellitus and type 2 diabetes mellitus

	FT1DM	T2DM	P
N	35	42
Male/female	15/20	33/9	0.0013
Age (years)	42.2 ± 14.4	56.1 ± 13.1	<0.0001
BMI (kg/m²)	20.9 ± 2.7	25.0 ± 4.0	<0.0001
Duration from onset (day)	3.8 ± 2.3	–	–
Plasma glucose (mg/dl)	818 ± 340* (421–2250)	150 ± 40^† (113–341)	–
HbA_1C (%)	6.0 ± 0.8	7.0 ± 0.7	<0.0001
GA (%)	23.6 ± 4.3	19.5 ± 2.5	<0.0001
GA/HbA_1C ratio	3.9 ± 0.5	2.8 ± 0.3	<0.0001

FT1DM, fulminant type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus; BMI, body mass index; Hb, haemoglobin; GA, glycated albumin

*Randomly measured plasma glucose

^†Fasting plasma glucose, parentheses denote range

In patients with FT1DM at initial diagnosis, randomly measured PG was markedly elevated (818 ± 340 mg/dL; range, 421–2250 mg/dL). In patients with T2DM, fasting PG was 150 ± 40 mg/dL (range, 113–341 mg/dL). HbA_1C was significantly lower in FT1DM patients (6.0 ± 0.8%) than in T2DM patients (7.0 ± 0.7%; P = 0.0002). By contrast, serum GA was significantly higher in FT1DM patients (23.6 ± 4.3%) than in T2DM patients (19.5 ± 2.5%; P < 0.0001) (Figure 1).

Figure 1

Haemoglobin (Hb)A_1C (a) and serum glycated albumin (GA) (b) in patients with fulminant type 1 diabetes mellitus (open bar) and type 2 diabetes mellitus (solid bar)

In T2DM patients fasting PG was significantly correlated with both HbA_1C (R = 0.510, P = 0.0015) and serum GA (R = 0.404, P = 0.0146), while in FT1DM patients, randomly measured PG was significantly correlated with serum GA (R = 0.495, P = 0.003), but not with HbA_1C (R = 0.220, P = 0.204). There was no significant relationship between BMI and serum GA in all the patients. No significant association was observed between BMI and HbA_1C.

A significant positive correlation existed between HbA_1C and serum GA in T2DM patients (R = 0.683, P < 0.0001) (Figure 2). FT1DM patients also showed a significant positive correlation between HbA_1C and serum GA (R = 0.610, P = 0.0001) (Figure 2). The regression line for serum GA levels to HbA_1C levels was shifted upwards as compared with the T2DM group (y = 3.36x + 3.45 in FT1DM patients versus y = 2.30x + 3.40 in T2DM patients), indicating relatively high elevation of serum GA levels compared with that of HbA_1C levels (Figure 2).

Figure 2

Correlation between haemoglobin (Hb)A_1C and serum glycated albumin (GA) in patients with fulminant type 1 diabetes mellitus (open circles) and type 2 diabetes mellitus (solid circles). The regression line for each is also shown

GA/HbA_1C ratio was significantly higher in FT1DM patients (3.9 ± 0.5) than in T2DM patients (2.8 ± 0.3; P < 0.0001) (Figure 3). GA/HbA_1C ratio was >3.2 in 41 of 42 FT1DM patients (98%), compared with only one of 32 T2DM patients (3%) (Figure 3). ROC analyses showed that although both specificity and sensitivity of HbA_1C and those of serum GA for differentiating FT1DM and FT2DM were low, a cut-off value of 3.2 for GA/HbA_1C ratio yielded 97% sensitivity and 98% specificity for differentiating FT1DM from T2DM (Figure 4).

Figure 3

Comparison of glycated albumin (GA)/haemoglobin (Hb)A_1C ratio in patients with fulminant type 1 diabetes mellitus (open circles) and type 2 diabetes mellitus (solid circles). GA/HbA_1C ratio in each patient and mean (±standard deviation) for each group are shown. The dotted line represents a GA/HbA_1C ratio of 3.2

Figure 4

Receiver operating characteristic (ROC) curves for HbA_1C (open squares), serum glycated albumin (GA) (open circles) and GA/HbA_1C ratio (closed circles) for differentiating fulminant type 1 diabetes mellitus from type 2 diabetes mellitus

Stepwise multivariate analysis using age, sex, BMI and GA/HbA_1C ratio as explanatory variables showed that GA/HbA_1C ratio (β = −0.773, F = 138.1, P < 0.0001) and age (β = 0.186, F = 8.0, P = 0.0124) were found to be independent explanatory variables for FT1DM. Sex and BMI were not independent risks for FT1DM.

Discussion

In this study, HbA_1C was significantly lower in FT1DM patients than in T2DM patients, whereas serum GA was significantly higher. Thus, in contrast to only a slightly elevated HbA_1C with an abrupt rise in PG levels, serum GA was already elevated. As a result, GA/HbA_1C ratio was significantly higher in FT1DM patients than in T2DM patients at initial visit before treatment for diabetes, suggesting the utility of this ratio in differentiating between FT1DM and T2DM.

GA/HbA_1C ratio in T2DM patients has been reported as 2.9 ± 0.4,^8,19 consistent with the present findings. In diabetic patients with poor glycaemic control, after a short-term improvement in PG levels, the magnitude of the decrease in serum GA exceeds that in HbA_1C. GA/HbA_1C ratio thus decreases.^7,8 Conversely, with an abrupt deterioration in glycaemic control, the magnitude of the increase in serum GA would be expected to be greater than that in HbA_1C. In the present study, randomly measured PG levels correlated significantly with serum GA, but not with HbA_1C in patients with FT1DM. A higher GA/HbA_1C ratio in F1TDM may reflect abrupt deterioration in PG levels. Although we recently reported that BMI negatively regulated serum GA levels in diabetic patients whose HbA_1C levels had been stable,¹⁹ there was no relationship between BMI and serum GA in all the study patients. Furthermore, the stepwise multivariate analysis showed that GA/HbA_1c ratio, but not BMI, was independent explanatory variable for FT1DM. This finding suggests that a higher GA/HbA_1C ratio in F1TDM patients may not be attributable to low BMI.

The diagnosis of FT1DM is clinically important, as initiation of insulin treatment is crucial to avoiding abrupt deterioration and poor prognosis.^2,3 HbA_1C levels are normal or only slightly elevated in FT1DM patients, which is critically distinguishable from patients with autoimmune type 1 diabetes mellitus (AT1DM). Most patients with FT1DM display markedly elevated glucose levels, and excluding a diagnosis of T2DM is not difficult in such cases. However, about 10% of FT1DM patients show randomly measured PG levels ≤400 mg/dL.²⁰ By contrast, randomly measured PG in patients with T2DM sometimes exceeds 288 mg/dL (16 mmol/L), which is a diagnostic level for FT1DM.³ In these cases, FT1DM is not easily distinguishable from T2DM. Clinical indicators other than PG levels are thus needed for differentiating FT1DM from T2DM.

In the present study, a cut-off value of 3.2 for GA/HbA_1C ratio yielded 97% sensitivity and 98% specificity for differentiating FT1DM from T2DM, far superior to the sensitivity and specificity of HbA_1C or serum GA. Measurement of serum GA in addition to PG and HbA_1C is thus recommended on initial evaluation in patients with suspected FT1DM, as calculation of GA/HbA_1C ratio can help to distinguish FT1DM from T2DM. The development of a GA enzyme assay now permits rapid testing using an autoanalyser, although measurements of serum ketone bodies, pH in arterial blood gas and serum or urinary C-peptide immunoreactivity remain essential to distinguish FT1DM from T2DM.

FT1DM cannot be distinguishable from AT1DM by GA/HbA_1C ratio, because GA/HbA_1C ratio in AT1DM is reported to be higher than that in T2DM²¹ and sometimes to be nearly 3.5.²² Instead, HbA_1C is useful for the differential diagnosis of FT1DM and AT1DM, because HbA_1C levels are nearly normal or slightly elevated in FT1DM patients and, on the other hand, they are usually extremely high in AT1DM patients.^1–3

To the best of our knowledge, this is the first study to demonstrate that FT1DM can be distinguished from untreated T2DM using GA/HbA_1C ratio.

DECLARATIONS

Competing interests: All authors had no conflicts of interest.

Funding: No funding was required in this research.

Ethical approval: The ethics committee of Kinki Central Hospital approved this study (No. 85).

Guarantor: MK.

Contributorship: MK, SK, TH and TK are the senior authors; JM, HS and AI contributed equally.

References

Imagawa

, Hanafusa

, Miyagawa

, Matsuzawa

. A novel subtype of type 1 diabetes mellitus characterized by a rapid onset and an absence of diabetes-related antibodies. Osaka IDDM Study Group. N Engl J Med 2000;342:301–7

Imagawa

, Hanafusa

, Uchigata

, Fulminant type 1 diabetes: a nationwide survey in Japan. Diabetes Care 2003;26:2345–52

Hanafusa

, Imagawa

. Fulminant type 1 diabetes: a novel clinical entity requiring special attention by all medical practitioners. Nat Clin Pract Endocrinol Metab 2007;3:36–45

Cohen

. Nonenzymatic glycation: a central mechanism in diabetic microvasculopathy? J Diabet Complications 1998;2:214–7

Koenig

, Peterson

, Jones

, Saudek

, Lehrman

, Cerami

. Correlation of glucose regulation and hemoglobin AIc in diabetes mellitus. N Engl J Med 1976;295:417–20

Bunn

, Gabbay

, Gallop

. The glycosylation of hemoglobin: relevance to diabetes mellitus. Science 1978;20:21–7

Tahara

, Shima

. Kinetics of HbA1c, glycated albumin, and fructosamine and analysis of their weight functions against preceding plasma glucose level. Diabetes Care 1995;18:440–7

Takahashi

, Uchino

, Shimizu

, Comparison of glycated albumin (GA) and glycated hemoglobin (HbA_1C) in type 2 diabetic patients: usefulness of GA for evaluation of short-term changes in glycemic control. Endocr J 2007;54:139–44

Matsunaga

, Fujita

, Koyano

, A case suspected of diabetes mellitus characterized by rapid onset and absence of diabetes-related antibodies. Juntendo Igaku 2003;48:516–9 (in Japanese)

10.

Iwaoka

. A case of fulminant type 1 diabetes with transiently positive anti-GAD antibodies. Endocr J 2003;50:225–31

11.

Nakao

, Okaji

, Wakasaki

, A case of fulminant type 1 diabetes who showed elevation of serum pancreatic exocrine enzymes late after onset of diabetes. J Japan Diabet Soc 2005;48:627–31 (In Japanese)

12.

Ogawa

, Katabami

, Kondo

, Kawata

, Saito

. Three cases of fulminant type 1 diabetes. J Japan Diabet Soc 2005;48(Suppl. 1):A45–7 (in Japanese)

13.

The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 2003;26:5–20

14.

Tominaga

, Makino

, Yoshino

, Japanese standard reference material for JDS Lot 2 haemoglobin A1c. I: comparison of Japan Diabetes Society-assigned values to those obtained by the Japanese and USA domestic standardization programmes and by the International Federation of Clinical Chemistry reference laboratories. Ann Clin Biochem 2005;42:41–6

15.

Kouzuma

, Usami

, Yamakoshi

, Takahashi

, Imamura

. An enzymatic method for the measurement of glycated albumin in biological samples. Clin Chim Acta 2004;324:61–71

16.

Kouzuma

, Uemastu

, Usami

, Imamura

. Study of glycated amino acid elimination reaction for an improved enzymatic glycated albumin measurement method. Clin Chim Acta 2004;346:135–43

17.

Hanley

, McNeil

. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982;143:29–36

18.

Hanley

, McNeil

. A method of comparing the areas under a receiver operating characteristic curves derived from the same case. Radiology 1983;148:839–43

19.

Koga

, Matsumoto

, Saito

, Kasayama

. Body mass index negatively influences glycated albumin, but not glycated hemoglobin, in diabetic patients. Endocr J 2006;53:387–91

20.

Hanafusa

, Imagawa

, Iwahashi

, Report of Japan Diabetes Society Committee on fulminant type 1 diabetes mellitus research: epidemiological and clinical analysis and proposal of diagnostic criteria. J Japan Diabetes Soc 2005;48(Suppl. 1):A1–13 (in Japanese)

21.

Yoshiuchi

, Matsuhisa

, Katakami

, Glycated albumin is a better indicator for glucose excursion than glycated hemoglobin in type 1 and type 2 diabetes. Endocr J 2008;55:503–7

22.

Imai

, Oikawa

, Shimada

. Improved monitoring of the hyperglycemic state in type 1 diabetes patients by use of the glycoalbumin/HbA_1C ratio. Rev Diabet Stud 2007;4:44–8

Serum glycated albumin to haemoglobin A 1C ratio can distinguish fulminant type 1 diabetes mellitus from type 2 diabetes mellitus

Abstract

Background

Methods

Results

Conclusions

Introduction

Methods

Patients

Laboratory measurements

Statistical analyses

Results

Discussion

DECLARATIONS

References