| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrine Care |
Departments of Endocrinology (H.B., G.S.), Vascular Diseases (A.G.), and Clinical Chemistry (P.F.), University of Lund, Malmö University Hospital, SE-205 02 Malmö; and Department of Medicine (M.L.-O.), Lund University, Lund University Hospital, SE-221 85 Lund, Sweden
Address correspondence and requests for reprints to: Dr. H. Borg, Wallenberg Laboratory, Entrance 46 2nd floor, Malmö University Hospital, SE-205 02 Malmö, Sweden. E-mail: Henrik.Borg{at}endo.mas.lu.se
Abstract
It is unclear whether high levels of antigen-specific islet antibodies [GADA (glutamic acid decarboxylase 65 antibodies) and IA2-ab (protein tyrosine phosphatase-like protein antibodies)] predict ß-cell failure in patients with onset of diabetes in adult age. Therefore, GADA and IA2-ab levels at the diagnosis of diabetes were related to fasting plasma C-peptide levels 5 yr later in 148 patients with diabetes onset in adult age (age at onset, 20–77 yr; median, 57 yr). Classical islet cell antibodies (ICA) were also determined.
Complete ß-cell failure (undetectable fasting plasma C-peptide) was only present in 4 patients at diagnosis of diabetes, but in 21 patients 5 yr thereafter. At diagnosis, ICA were detected in 20 of 21 (95%) patients with ß-cell failure after 5 yr and in only 7 of 127 (5%) without, whereas GADA and/or IA2-ab (>97.5 percentile of healthy controls) were detected in all 21 (100%) with but also in 23 of 127 (18%) patients without ß-cell failure after 5 yr. Thus, ICA had a higher positive predictive value (74%) than GADA and/or IA2-ab (47%; P < 0.05). With high cutoff values for GADA and IA2-ab, however, GADA and/or IA2-ab were detected in 19 of 21 (90%) patients with ß-cell failure vs. only in 5 of 127 (4%) without, giving a positive predictive value of 79%. Slightly elevated GADA levels in IA2-ab-negative patients were associated with progressive but not complete ß-cell failure within the study period. Hence, high GADA and/or IA2-ab levels predict a future complete ß-cell failure, whereas low GADA levels predict slowly progressive ß-cell insufficiency.
TYPE 1 (INSULIN-DEPENDENT) diabetes is an autoimmune disease. Islet antibodies [i.e. islet cell antibodies (ICA) (1) and their major components, i.e. the antigen-specific GADA (glutamic acid decarboxylase antibodies) (2) and IA2-ab (protein tyrosine phosphatase-like protein antibodies) (3)] are detected in 95% of children at the time of diagnosis of diabetes (4). Accordingly, almost all children with type 1 diabetes demonstrate islet antibodies. The incidence of type 2 (noninsulin-dependent) diabetes is increasing both in the United States (5) and in Europe (6), but type 1 diabetes is still the predominant type of diabetes in children (5). Among patients with diabetes onset in adult age the situation is different. Although nonautoimmune type 2 diabetes is the major form of diabetes in adult patients (7), autoimmune type 1 diabetes is not uncommon. Actually, as many as 44% of type 1 diabetic patients develop their disease after the age of 30 yr (8). Moreover, islet autoimmunity is frequent in adult patients considered to have type 2 diabetes. Indeed, 10–25% of patients with the type 2 diabetes phenotype demonstrate islet autoimmunity at the diagnosis of diabetes (9, 10, 11, 12). Islet antibodies identify patients with progressive ß-cell insufficiency (13, 14, 15, 16, 17), and most of them will later develop ß-cell failure and overt type 1 diabetes (18). In this context, ICA seem to have some advantages compared with GADA (18), possibly due to the fact that determination of ICA also includes measurements of IA2-ab. In the current study, ICA, GADA, and IA2-ab were determined at the diagnosis of diabetes in an unselected material of 148 patients with diabetes onset in adult age and related to ß-cell failure vs. nonfailure 5 yr after diagnosis of diabetes. The aim of this study was to evaluate the value of the levels of GADA and IA2-ab in the prediction of future ß-cell failure.
Subjects and Methods
Subjects
Between September 1985 and August 1987, all new, consecutively diagnosed adult diabetic patients (>20 yr, n = 233) in the city of Malmö, Sweden, were included in this prospective study (9, 19, 20). Besides one black woman from Sudan, all patients were Caucasians. In the study, a first blood sample was taken after an overnight fast close to the diagnosis of diabetes (within 3 months after the diagnosis and within 6 months after the first hyperglycemic symptoms) (19). After 5 yr, 35 of the 233 patients had died, and 16 had moved from the city. Accordingly, 182 patients could be invited to a follow-up study conducted 5 yr after diagnosis of diabetes, and 156 of the 182 (83%) eligible patients accepted. A fasting plasma C-peptide (P-C-peptide) value below the detection limit of the assay used (<0.10 nmol/L) 5 yr after diagnosis of diabetes was predetermined as the end point (i.e. a sign of complete ß-cell failure). Only patients with a blood sample volume sufficient for measurements of all the antibodies at the diagnosis and a blood sample for fasting P-C-peptide measurement at the 5-yr follow-up were included in the final study. This group comprised 148 of the 156 patients, 20–77 yr [median, 57 (interquartile range, 21)] at the diagnosis; 70 of the 148 patients (47%) were women. Blood samples 3 yr after diagnosis were obtained in 129 of these 148 patients. The 26 patients who did not participate in the follow-up study and the 8 lacking blood sample at the diagnosis of diabetes were younger at the diagnosis of diabetes than the 148 who did participate [51 (17) yr vs. 57 (21) yr, P = 0.01]. There were no significant differences, however, with regard to body mass index [27.6 (8.8) kg/m2 vs. 27.3 (6.3) kg/m2], HbA1c [8.9 (4.0) % vs. 8.3 (3.8) %], or fasting P-C-peptide [0.57 (0.53) nmol/L vs. 0.65 (0.72) nmol/L] at diagnosis between participants and nonparticipants. Control samples for fasting P-C-peptide were taken from 34 nondiabetic subjects examined by an iv glucose-glucagon infusion test [median age, 43 (36) yr; range, 19–78; 19 women] (21).
An iv glucose-glucagon infusion test was performed in subgroups of the patients. When this prospective study started in 1985 through 1987, GADA and IA2-ab had not been described and the antigen-unspecific ICA test was the only autoantibody test that was clinically available. Therefore, only ICA and not GADA and IA2-ab were originally used to define autoimmune diabetes. Accordingly, to evaluate ß-cell function in detail, 38 patients with ICA at diagnosis were invited to an iv glucose and glucagon infusion test (21), 22 of whom accepted (16 with ICA, GADA, and IA2-ab; 4 with ICA and GADA; and 2 with ICA and IA2-ab). The 22 antibody-positive patients were examined within 8 months after the diagnosis of diabetes together with 34 nondiabetic age-matched control subjects. After 3 yr (32–36 months), 20 of the 38 antibody-positive patients (18 of 20 tested at diagnosis) were examined. Finally, after 5–7 yr (60–85 months) 13 antibody-positive patients (8 of 13 tested at diagnosis) were examined together with 23 of 34 the nondiabetic control subjects.
Control samples for GADA and IA2-ab were taken from 199 nondiabetic subjects (age, 44 (30) yr; range, 18–70 yr; 100 women).
Assay methods
GADA and IA2-ab were determined by radioligand binding assays. The GADA assay was based on 125I-labeled human recombinant GAD 65 (22). Both the sensitivity and the specificity of this GADA assay were 100% when compared with a 35S-GADA assay evaluated in the Diabetes Autoantibody Proficiency Testing Program (Prof. Noel Maclaren, Louisiana State University, New Orleans, LA) for GADA (no. 2, 24 samples tested). In the current study, a value above 1.9 U/mL (97.5% percentile of the 199 nondiabetic controls) was considered abnormal. The IA2-ab assay was based on 35S-methionine labeled human recombinant in vitro transcribed-translated IA2 (4). The IA2-ab results are presented as an index. In the latest Diabetes Autoantibody Proficiency Testing Program for IA2-ab (no. 3, 24 samples tested), this IA2-ab assay performed with 100% sensitivity and 100% specificity. In the current study, an IA2-ab index above 1.1 (>97.5% percentile of the 199 nondiabetic controls) was considered abnormal. ICA were determined by a prolonged immunofluorescence assay (23). In the latest Diabetes Autoantibody Proficiency Program (no. 13, 20 samples tested), this ICA assay performed with 100% sensitivity and 100% specificity. In the current study, the detection limit was 3 JDF units for the pancreas used; i.e. an ICA value of 3 JDF units or more was considered abnormal. The levels of GADA and IA2-ab [0.4 (0.4) U/ml vs. 0.3 (0.2) U/mL; P = 0.04, and index 0.1 (0.3) vs. index 0.0 (0.2); P = 0.01, respectively], but not ICA, were slightly higher in the control subjects above 45 yr of age than in those below. One of the controls, a 61-yr-old man, was positive both for GADA and ICA.
Fasting P-C-peptide was used as a measure of endogenous ß-cell function in the complete material. The detection limit was 0.10 nmol/L for the assay used (21), and the fasting P-C-peptide level of the controls was 0.49 (0.28) nmol/L. Fasting P-C-peptide was measured in 135 of the 148 patients at the diagnosis, in 127 of the 129 patients at the 3 yr follow-up, and in all 148 patients at the 5 yr follow-up. Whole venous blood glucose was determined by a routine hexokinase method. HbA1c was determined by a high-performance liquid chromatography method (24). Reference values for healthy individuals were 3.90–5.30%.
Intravenous glucose-glucagon infusion test
In the ß-cell function test, blood samples for fasting P-C-peptide and glucose were taken 10 min before and immediately before a 3-min iv infusion of 0.5 g glucose x kg-1 body weight. After the infusion, blood samples were taken at 1, 3, 5, 10, 20, 30, 40, 50, 60, 70, 80, and 90 min, when an iv injection of 1 mg glucagon was given, and a final blood sample taken after 6 min (96 min after glucose infusion) (21).
Statistics
Nonparametric Kruskal-Wallis and Mann-Whitney tests were used to evaluate differences between groups, and Friedman and Wilcoxon Signed Rank tests were used to evaluate paired differences. Fishers’ exact and McNemar’s tests were used to evaluate frequency differences between groups. Probability less than 0.05 was considered significant. Receiver Operating Characteristic (ROC) curves were constructed with SAS 6.10/OS/2 Warp Connect software (SAS Institute, Inc., Cary, NC). Data are presented as median (interquartile range), unless stated otherwise.
Results
Antibody status and fasting P-C-peptide
At the diagnosis of diabetes, all three antibodies were found in 19 patients (13%), two antibodies in 8 patients (5%) (ICA and GADA in 6 of 8, ICA and IA2-ab in 1 of 8, and GADA and IA2-ab in 1 of 8), and one antibody in 18 patients (12%) (ICA in 1 of 18, GADA in 6 of 18, and IA2-ab in 11 of 18).
Figure 1A
shows that at the diagnosis of
diabetes, fasting P-C-peptide was significantly lower in patients with
two or three antibodies (n = 25) than in patients with one
antibody or patients without antibodies (n = 113;
P < 0.0001). Complete ß-cell failure at diagnosis
(Fig. 1B) was, however, only found in 4 of 137 (3%) patients (in 2
patients with three antibodies, in 1 with ICA and GADA, and in 1 with
only GADA). After the diagnosis, fasting P-C-peptide levels profoundly
decreased in those with two or three antibodies (P =
0.006 after 3 yr and P = 0.02 between 3 and 5 yr after
diagnosis). Moreover, in patients with only ICA or only GADA fasting
P-C-peptide levels showed a decrease 5 yr after diagnosis
(P = 0.04), whereas fasting P-C-peptide levels did not
decrease after diagnosis in patients with only IA2-ab or in patients
without antibodies (Fig. 1A). After 5 yr, complete ß-cell failure was
seen in 21 of 148 (14%) patients [most with two or three antibodies
(20 of 21, 95%)] but in none of the 103 (0%) patients without
antibodies nor in any of the 11 (0%) patients with only IA2-ab (Fig. 1B). Among the patients with multiple antibodies, the only patient with
the combination of GADA and IA2-ab (35.7 U/mL and index 1.2,
respectively) had a high fasting P-C-peptide level, both at diagnosis
and at follow-up after 5 yr (1.70 nmol/L and 1.47 nmol/L,
respectively). Among the patients with one antibody (GADA or ICA), only
one (GADA positive) patient had a complete ß-cell failure. In
patients with only GADA or only ICA, fasting P-C-peptide levels after 5
yr were significantly lower than in the antibody negative patients
[0.26 (0.18) nmol/L vs. 0.89 (0.50) nmol/L,
P = 0.0006], but significantly higher than
in patients with two or three antibodies [0.26 (0.18) nmol/L
vs. 0.00 (0.00) nmol/L, P = 0.001]. The
patients with only GADA (n = 6) had significantly lower GADA
levels compared with those with GADA in combination with ICA [4.6
(22.8) U/mL vs. 72.3 (129.2) U/mL, P =
0.0009]. All patients with only IA2-ab had low levels of the antibody
at the diagnosis (IA2-ab index range, 1.2–2.7), and their fasting
P-C-peptide values after 5 yr did not deviate from the levels in the
antibody-negative patients.
|
shows that high levels of the
different antibodies were clearly associated with complete ß-cell
failure 5 yr after diagnosis. Interestingly, however, in patients
without complete ß-cell failure, those positive for ICA [0.22 (0.13)
nmol/L vs. 0.90 (0.53) nmol/L, P <
0.0001], positive for GADA [0.24 (0.22) nmol/L vs. 0.90
(0.54) nmol/L, P < 0.0001], and positive for high
IA2-ab [index >2.7; 0.16 (0.11) nmol/L vs. 0.89 (0.51)
nmol/L, P = 0.001], respectively, had significantly
lower fasting P-C-peptide levels 5 yr after diagnosis than those
without antibodies.
|
To clarify the validity of fasting P-C-peptide as a marker of
ß-cell function, a glucose-glucagon infusion test was conducted in
patients with two or three antibodies and in controls. Figure 3 shows that in the nondiabetic control
subjects, there was a rapid increase in P-C-peptide levels 1 min after
glucose infusion not seen at the time of diagnosis in the diabetic
patients with two or three antibodies, who only showed a marginal
increase in P-C-peptide 0–90 min after the glucose infusion. The
increment in P-C-peptide after the glucagon injection was also much
lower at diagnosis in the antibody positive patients compared with the
controls [
-C-peptide 90–96 min 0.31 (0.18) nmol/L vs.
1.51 (0.94) nmol/L; P < 0.0001]. Figure 3 then shows
that after the diagnosis of diabetes there was a deterioration in
ß-cell function in the patients with two or three antibodies. Five
years after diagnosis, 10 of 11 patients with fasting P-C-peptide less
than 0.10 nmol/L showed no increase in P-C-peptide after the glucagon
injection (1 patient had a marginal increase to 0.18 nmol/L). Hence, a
nondetectable fasting P-C-peptide level was associated with complete or
almost complete ß-cell failure in all cases.
|
Antibody cutoff based on healthy controls. Figure 2
shows that
ICA were detected in 20 of 21 (95%) patients with complete ß-cell
failure after 5 yr and only in 7 of 127 (6%) without. GADA higher than
1.9 U/mL was also detected in 20 of 21 (95%) with ß-cell failure
but, on the other hand, in 12 of 127 (9%) without. As for GADA, the
frequency of IA2-ab index higher than 1.1 was high in patients with
ß-cell failure but also quite high in patients without [15 of 21
(71%) with vs. in 17 of 127 (13%) patients without].
Indeed, although GADA higher than 1.9 U/mL and/or IA2-ab index higher
than 1.1 were detected in all 21 (100%) with ß-cell failure, this
was also detected in 23 of 127 (18%) patients without ß-cell failure
after 5 yr. Sensitivity, specificity, and positive predictive value in
the prediction of future ß-cell failure for the different antibodies
are presented in Table 1.
|
). Inspection of the ROC-curves
indicated that the optimal cutoffs were higher both for GADA and IA2-ab
(GADA >4.3 U/mL and IA2-ab index >2.7, respectively) compared with
the cutoffs based on healthy controls (GADA >1.9 U/mL and/or IA2-ab
index >1.1). With these cutoffs, GADA (>4.3 U/mL) were detected in 20
of 21 (95%) patients with complete ß-cell failure after 5 yr
vs. in 9 of 127 (7%) without, whereas IA2-ab (index >2.7)
were detected in 12 of 21 (57%) with vs. only in 4 of the
127 (3%) patients without. GADA and/or IA2-ab were detected in all 21
(100%) with vs. in only 9 of 127 (7%) without ß-cell
failure after 5 yr (P < 0.0001) (Table 1).
|
, a GADA levels above 41.4
U/mL was detected in 14 of the 21 (67%) patients with ß-cell failure
vs. only in 3 of the 127 (2%) without (P <
0.0001). Indeed, a GADA level higher than 41.4 U/mL and/or an IA2-ab
index higher than 2.7 were found at the diagnosis of diabetes in 19 of
the 21 (95%) patients with ß-cell failure after 5 yr vs.
only in 5 of 127 (4%) without (P < 0.0001), giving
high specificity and high positive predictive value (Table 1). Figure 2
shows that of the 19 patients with ß-cell failure after 5 yr, 7
(37%) had high levels of both GADA and IA2-ab, 7 (37%) had only high
levels of GADA, and 5 (26%) had only high levels of IA2-ab.
Noteworthy, the five patients with high GADA and/or IA2-ab levels (A,
B, C, D, and E, Fig 2) without complete ß-cell failure had
significantly lower fasting P-C-peptide levels 5 yr after the diagnosis
compared with the patients without antibodies [0.20 (0.17) nmol/L
vs. 0.89 (0.51) nmol/L; P = 0.0004; range,
0.10–0.43 nmol/L]. Among the ICA-positive patients with GADA and/or IA2-ab, 24 of 26 (92%) had a GADA level higher than 41.4 U/mL and/or an IA2-ab index higher than 2.7 as compared with none of the 18 (0%) ICA-negative GADA and/or IA2-ab-positive patients (P < 0.0001). Hence, the presence of ICA indicated high levels of GADA and/or IA2-ab.
Clinical features
Compared with patients without ß-cell failure, those with had significantly higher HbA1c values [9.5 (2.4) % vs. 7.6 (3.6) %, P = 0.005] and lower body mass index values [23.0 (3.6) kg/m2 vs. 27.9 (6.2) kg/m2, P < 0.0001] at the diagnosis; however, there was no significant difference in age [56 (36) years vs. 58 (15) yr]. Among patients with two or three antibodies, only half (13 of 27; 48%) were on insulin treatment at diagnosis in comparison with almost all (25 of 27; 93%) after 5 yr (P = 0.0007). At diagnosis there was no significant differences in fasting P-C-peptide level between the 13 patients on insulin treatment vs. the 14 without [0.19 (0.11) nmol/L vs. 0.21 (0.11) nmol/L, P = 0.72]. Among patients with only GADA or only ICA, one of seven (14%) was on insulin treatment at diagnosis as compared with three of seven (43%) after 5 yr. Among the 103 antibody-negative patients the frequency of insulin treatment was low both at the diagnosis (5 of 103; 5%) and after 5 yr (10 of 103; 10%; P > 0.2). Among the 11 patients with only IA2-ab, none (0%) were on insulin treatment at diagnosis or after 5 yr.
Among antibody-positive patients with preserved ß-cell function at
diagnosis of diabetes, there was no significant difference in the
deterioration in ß-cell function between those treated with insulin
soon after diagnosis (n = 11) vs. those who were
started on insulin treatment later (n = 10) [-fasting
P-C-peptide diagnosis–5 yr, 0.19 (0.14) nmol/L vs. 0.21
(0.21) nmol/L, P = 0.34].
Discussion
In this prospective study of patients with diabetes onset in adult age, a comparison between fasting P-C-peptide and the P-C-peptide responses to iv glucose and glucagon infusions demonstrated that undetectable fasting P-C-peptide was a reliable sign of severe ß-cell dysfunction. Nondetectable fasting P-C-peptide 5 yr after diagnosis was associated with no P-C-peptide response to iv glucose and a lacking or marginal response of P-C-peptide to glucagon. Our study shows that the presence of at least two of three antibodies (ICA, GADA, and IA2-ab) at the diagnosis predicted a complete ß-cell failure 5 yr thereafter in patients with diabetes onset in adult age. Among the three antibodies, ICA had the highest positive predictive value with regard to future ß-cell failure (74%). High levels of GADA and/or IA2-ab at diagnosis, however, had also a high positive predictive value; in fact slightly higher (79%) than ICA. Although less severe than in patients with multiple antibodies, isolated GADA positivity at low levels was associated with future deterioration in ß-cell function.
As recently reported by Decochez et al. (25), ICA were the single marker that most strongly predicted future ß-cell failure. The close correlation, however, between high levels of GADA and/or IA2-ab at diagnosis and future complete ß-cell failure suggests that GADA and IA2-ab measurements are an attractive alternative to ICA determinations. Indeed, our study showed that patients with ICA had high levels of GADA and IA2-ab, which may explain why ICA predicted future ß-cell failure with a high specificity.
The association between isolated GADA positivity at low levels and less severe but progressive ß-cell deterioration is an argument for GADA measurements. In a previous report on subgroups of this cohort we also found a correlation between the levels of GADA and the degree of decline ß-cell function (26). A recent observation of a correlation between GADA levels and future insulin requirement in adult onset diabetes (27) is also in agreement with the current study. ICA unrelated to GADA and IA2-ab have been suggested to be a feature of a special slowly progressive form of type 1 diabetes (28). In the current study, only one patient with isolated ICA positivity at the diagnosis of diabetes was found and he showed slowly progressive ß-cell dysfunction.
In children, IA2-ab are detected without GADA in 20% of patients with recent onset diabetes (29). In the current study on adult patients, IA2-ab were associated with future ß-cell failure only if combined with GADA or, as in one case, with ICA. Our study, therefore, infers that GADA rather than IA2-ab are engaged in the primary autoimmune process involved in the development of type 1 diabetes in adults. It has been suggested that IA2-ab are more closely associated with insulin secretion and its loss than GADA (30, 31). Hence, IA2-ab may be a late occurring phenomenon, associated with major ß-cell damages and insulin release, not necessarily secretion, from disintegrating ß-cells. Our study shows, however, that it is important to measure IA2-ab in GADA- positive patients, at least in those with low GADA levels, to define those who will develop complete ß-cell failure within a few years.
Islet antibodies are markers for autoimmune diabetes, and most likely reflect T-cell-mediated autoreactive destruction of the ß-cells (32). The reappearance of ICA and GADA is associated with lymphocytic infiltration of the islets of Langerhans and recurrence of type 1 diabetes after pancreas transplantation (33). In keeping with this, GADA indicate insulitis in patients with recent onset type 1 diabetes (34). The levels of islet antibodies at the diagnosis of diabetes, as in our patients, may, therefore, reflect the magnitude of the T-cell-mediated islet destruction.
Insulin treatment has been shown to be ß-cell protective (35, 36); an early correct diagnosis of type 1 diabetes leading to institution of insulin may preserve ß-cell function. We could not detect any beneficial effects of early insulin treatment as reflected by fasting P-C-peptide levels. Our study was, however, not designed to assess the putative protective effects of insulin treatment on ß-cells. Our patients were treated by their clinicians following current clinical practice and the antibody results were not provided to the clinicians. To be ß-cell protective, insulin administration may have to be started before major ß-cell deterioration has occurred. In this study, all patients with future complete ß-cell failure had a certain degree of impairment in ß-cell function already at diagnosis. Patients positive only for GADA had normal fasting P-C-peptide levels at diagnosis but thereafter displayed slowly progressive ß-cell dysfunction. We postulate that patients with GADA but without ICA may be the group of patients that would gain most from early insulin administration.
In conclusion, in patients with onset of diabetes in adult age, high levels of GADA and/or IA2-ab at the diagnosis predict a future complete ß-cell failure, whereas isolated GADA positivity in lower levels, are associated with a more slowly developing ß-cell insufficiency.
Acknowledgments
We thank Ulrika Gustavsson, Ingegerd Larsson, Ann Radelius, and Christina Rosborn for excellent technical assistance.
Footnotes
1 Supported by grants from Albert Påhlsson Foundation, the Child
Diabetes Fund, the Juvenile Diabetes Foundation–Wallenberg Diabetes
Research Program (K98-99JD-128B), Lundström Foundation,
Malmö Diabetes Association, Novo-Nordic Foundation, Research
Funds Malmö University Hospital, Swedish Diabetes Association,
Swedish Medical Research Council (7507 and 5913), and University Funds
Lund University. 
Received June 30, 2000.
Revised March 1, 2001.
Accepted March 14, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  A. Goel, H. Chiu, J. Felton, J. P. Palmer, and B. Brooks-Worrell T-Cell Responses to Islet Antigens Improves Detection of Autoimmune Diabetes and Identifies Patients With More Severe {beta}-Cell Lesions in Phenotypic Type 2 Diabetes Diabetes, August 1, 2007; 56(8): 2110 - 2115. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. D. G. Leslie, R. Williams, and P. Pozzilli Type 1 Diabetes and Latent Autoimmune Diabetes in Adults: One End of the Rainbow J. Clin. Endocrinol. Metab., May 1, 2006; 91(5): 1654 - 1659. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Stenstrom, A. Gottsater, E. Bakhtadze, B. Berger, and G. Sundkvist Latent Autoimmune Diabetes in Adults: Definition, Prevalence, {beta}-Cell Function, and Treatment Diabetes, December 1, 2005; 54(suppl_2): S68 - S72. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Biesenbach, M. Auinger, M. Clodi, F. Prischl, R. Kramar, and Diabetic Nephropathy Study Group of the Austrian S Prevalence of LADA and frequency of GAD antibodies in diabetic patients with end-stage renal disease and dialysis treatment in Austria Nephrol. Dial. Transplant., March 1, 2005; 20(3): 559 - 565. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Aguilera, R. Casamitjana, G. Ercilla, J. Oriola, R. Gomis, and I. Conget Adult-Onset Atypical (Type 1) Diabetes: Additional insights and differences with type 1A diabetes in a European Mediterranean population Diabetes Care, May 1, 2004; 27(5): 1108 - 1114. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Hanifi-Moghaddam, N. C. Schloot, S. Kappler, J. Seissler, and H. Kolb An Association of Autoantibody Status and Serum Cytokine Levels in Type 1 Diabetes Diabetes, May 1, 2003; 52(5): 1137 - 1142. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Stenstrom, B. Berger, H. Borg, P. Fernlund, J. S. Dorman, and G. Sundkvist HLA-DQ Genotypes in Classic Type 1 Diabetes and in Latent Autoimmune Diabetes of the Adult Am. J. Epidemiol., November 1, 2002; 156(9): 787 - 796. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Borg, A. Gottsater, P. Fernlund, and G. Sundkvist A 12-Year Prospective Study of the Relationship Between Islet Antibodies and {beta}-Cell Function At and After the Diagnosis in Patients With Adult-Onset Diabetes Diabetes, June 1, 2002; 51(6): 1754 - 1762. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) (complete ß-cell
failure in 0 of 11 at diagnosis, 0 of 9 after 3 yr, and 0 of 11 after 5
yr), patients with only GADA or only ICA (
) (complete ß-cell
failure in 1 of 5 at diagnosis, 0 of 5 after 3 yr, and 1 of 7 after 5
yr), patients with two or three antibodies (
) (complete ß-cell
failure in 3 of 25 at diagnosis, 17 of 26 after 3 yr, and 20 of 27
after 5 yr), and patients without antibodies (
) (complete ß-cell
failure in 0 of 96 at diagnosis, 0 of 88 after 3 yr, and 0 of 103 after
5 yr). At the 3-yr follow-up, a patient with only GADA at diagnosis,
and ß-cell failure at diagnosis and after 5 yr, was not assessed.
