Staging of Preclinical Type 1 Diabetes in Siblings of Affected Children
Samy Mrena, MB*; Kaisa Savola, MB‡; Petri Kulmala, MD‡; Hans K. Åkerblom, MD, PhD§; Mikael Knip, MD, PhD*; and the Childhood Diabetes in Finland Study Group
ABSTRACT. Objectives. To assess whether it is clini-cally relevant to classify siblings of children with recent-onset type 1 diabetes mellitus (T1DM) into various stages of preclinical diabetes, and to compare the risk of devel-oping clinical disease and the time to diagnosis between these stages.
Study Design. From a total of 801 families taking part in the Childhood Diabetes in Finland Study, 758 initially unaffected siblings were graded into four stages of pre-clinical T1DM based on the number of disease-associ-ated autoantibodies detectable close to the time of diag-nosis in the index case: no (no antibodies), early (one antibody specificity), advanced (two antibodies), and late prediabetes (more than three antibodies). Another clas-sification system, used with 712 siblings, was based on a combination of the number of antibodies and the first-phase insulin response (FPIR) to intravenous glucose: no (no antibodies), early (one antibody specificity, normal FPIR), advanced (two or more antibodies, normal FPIR), and late prediabetes (one or more antibodies, reduced FPIR).
Results. Six out of 661 siblings who initially pre-sented no signs of prediabetes (0.9%; 95% confidence interval [CI], 0.3%–2.0%) progressed to clinical T1DM. Based on the first set of criteria, 3 out of 49 individuals (6.1%; CI, 1.3%–16.9%; odds ratio [OR], 7.1; CI, 1.7–29.4) from the early prediabetes category, 3 out of 13 with advanced prediabetes (23.1%; CI, 5.0%–53.8%; OR, 32.8; CI, 7.2–150), and 23 out of 35 with late prediabetes (65.7%; CI, 47.8%– 80.9%; OR, 209; CI, 72.2– 607) presented with clinical signs of T1DM. According to the second set of criteria 1 out of 15 siblings with early prediabetes (6.7%; CI, 0.2%–32.0%; OR, 7.8; CI, 0.9 – 69.1), 6 out of 23 with advanced prediabetes (26.1%; CI, 10.2%– 48.4%; OR, 38.5; CI, 11.3–132), and 12 out of 13 with late prediabetes (92.3%; CI, 64.0%–99.8%; OR, 1310; CI, 146 –11 737) pre-sented with clinical signs of T1DM. The time to diagno-sis was significantly shorter in those with late prediabe-tes initially than in those with no signs of prediabeprediabe-tes.
Conclusions. Our observations indicate that it is pos-sible to grade siblings of children with newly diagnosed T1DM into categories with significant differences in the subsequent risk of clinical T1DM and time to diagnosis. Such a classification will become clinically relevant as soon as effective measures are available for preventing or delaying the manifestation of overt T1DM. Pediatrics
1999;104:925–930;autoantibodies, classification, prospec-tive, first-phase insulin response.
ABBREVIATIONS. T1DM, type 1 diabetes mellitus; ICA, islet cell antibodies; IAA, insulin autoantibodies; GADA, antibodies to the 65-kD isoform of glutamic acid decarboxylase; IA-2A, antibodies to the IA-2 protein; IVGTT, intravenous glucose tolerance test; FPIR, first-phase insulin response; JDFU, Juvenile Diabetes Foun-dation units; RU, relative units; Kg, glucose disappearance rate.
T
ype 1 diabetes mellitus (T1DM) is an autoim-mune disease with a long asymptomatic pre-clinical period characterized by gradualb-cell loss.1,2By the time clinical symptoms becomemani-fest, the vast majority of thebcells have been dam-aged previously. The duration of the preclinical phase apparently varies substantially because the disease can be diagnosed both in infancy and in old age.3,4The highest incidence is reported in children,
with an obvious peak in early puberty,5although the
latter is no longer detectable in Finland.6
In addition to signs of cell-mediated autoimmu-nity, the emergence of autoantibodies to various
b-cell structures, such as islet cell antibodies (ICA), insulin autoantibodies (IAA), glutamic acid decar-boxylase antibodies (GADA), and IA-2 antibodies (IA-2A) is associated with progression to clinical T1DM. The presence of ICA has for a long time formed the basis for predicting T1DM in first-degree relatives. Siblings of children with T1DM who are positive for ICA have a 40% to 50% cumulative risk of progressing to clinical disease within 10 years.7,8
The risk of T1DM has been reported to increase with the number of autoantibodies detectable, rising from ,10%, when ICA are detected alone to.80% when at least three antibodies are observed.8It is evident
that screening for all four essential antibodies (ICA, IAA, GADA, and IA-2A) is more informative than ICA screening alone.8The intravenous glucose
toler-ance test (IVGTT) is an efficient method for deter-mining the degree of b-cell dysfunction, and a re-duced first-phase insulin response (FPIR) has been shown to be highly predictive of progression to T1DM in ICA-positive first-degree relatives. The combination of two positive ICA results and two reduced FPIRs seems to be a reliable indicator of future disease progression in young relatives.9 –11
Increasing knowledge and intensive research on the pathogenesis of T1DM have produced hopes of finding an effective treatment to stop progressive
b-cell damage in individuals en route to overt T1DM. Combinations of antibodies and FPIR seem to pro-From the *Medical School, University of Tampere, Department of
Pediat-rics, Tampere University Hospital, and Tampere Diabetes Research Center, Tampere; the ‡Department of Pediatrics, University of Oulu, Oulu; and §Hospital for Children and Adolescents, University of Helsinki, Helsinki, Finland.
Received for publication Jan 4, 1999; accepted Apr 8, 1999.
Reprint requests to (M.K.) Medical School, University of Tampere, PO Box 607, FIN-33101 Tampere, Finland. E-mail: [email protected]
vide a promising tool for the identification of first-degree relatives at high risk of progression to the clinical disease. The present study was performed to evaluate whether it is clinically relevant to classify unaffected siblings of children with T1DM into var-ious stages of preclinical diabetes, and to subse-quently compare these stages with respect to the risk of developing the overt disease and the time to di-agnosis in the progressors.
METHODS Study Participants
The population was derived from the nationwide Childhood Diabetes in Finland study, which targeted children,15 years of age in whom T1DM was diagnosed between September 1986 to April 1989 and their families. The aim was to evaluate the genetic, immunologic, and environmental factors leading to the develop-ment of T1DM.12The follow-up was prospective and observation
of the siblings was initiated shortly after the proband had been diagnosed as having T1DM. Informed consent was obtained from the study participants or/and their parents. The study design was approved by the ethical committees of all 31 participating hospi-tals. Blood samples were taken at intervals of 3 to 6 months during the first 2 years and at 6- to 12-month intervals during the next 2 years. If the sibling was found to test negative for ICA and IAA on all occasions during the first 4 years, antibody surveillance was discontinued. Siblings positive for ICA and/or IAA on at least one occasion during the initial 4 years were subsequently observed at intervals of 12 months or less. Such siblings (n5108) were also invited for sequential IVGTTs at intervals of 6 to 12 months starting from the occasion when antibodies were detected for the first time. Eighty-three out of the 108 invited children (77%) un-derwent at least one IVGTT. All the siblings were observed for progression to T1DM up to the end of May 1997. Observation of the siblings progressing to T1DM ended at diagnosis, which was based on clinical symptoms and an increased random blood glu-cose concentration (.10 mmol/L) or elevated fasting (.6.7 mmol/L) or random blood glucose on two occasions in the ab-sence of symptoms.13The present study is based on the results
from the initial sample obtained for the analysis of autoantibodies and from the first IVGTT.
The 801 index cases of the Childhood Diabetes in Finland study had a mean age of 8.4 years (range, 0.8 –14.9 years), and the 758 unaffected siblings of the probands included in the present survey had a mean age of 9.9 years (range, 0.8 –19.7 years) at the initial sampling. The mean age of the 712 siblings who comprised the antibody-negative siblings and all antibody-positive ones with data available on both autoantibodies and FPIR was also 9.9 years (range, 0.8 –19.7). The mean age of the 83 cases with at least one IVGTT was 9.7 years (range, 2.1–18.9 years) at the time of diag-nosis of the index case and 11.2 years (range, 3.2–20.0 years) at the time of the first IVGTT. The median duration of follow-up for those who remained unaffected was 9.1 years (range, 7.7–10.7 years).
Disease-associated Autoantibodies
The presence of ICA was determined by a standard immuno-fluorescence assay performed on sections of frozen human pan-creas from a blood group O donor.14Fluorescein-conjugated
anti-human immunoglobulin G (Behringwerke AG, Marburg, FRG) was used to detect ICA. End-point dilution titers were identified and the results were expressed in Juvenile Diabetes Foundation units (JDFU) relative to an international reference standard.15The
detection limit was 2.5 JDFU. The sensitivity of the ICA assay was 100%, specificity 98%, validity 98%, and consistency 98% in the most recent international standardization round.
IAA were analyzed by a modification of the liquid phase ra-dioimmunoassay described by Palmer et al.16The samples were
treated with acid charcoal to remove insulin before the assay. Eighty mL of serum were incubated for 20 hours with
mono-125I(TyrA14)-human insulin (Novo Research Institute, Bagsvaerd,
Denmark). Free and bound insulin fractions were separated using polyethylene glycol. The results were expressed in nU/mL, in which 1 nU/mL corresponds to a specific binding of 0.01%. The
interassay coefficient of variation was,8%. If the specific insulin binding exceeded 54 nU/mL (representing the 99th percentile in 105 nondiabetic siblings), the sibling was considered to be IAA positive. The sensitivity of the IAA assay was 78%, specificity 100%, validity 92%, and consistency 100% in the proficiency test-ing program.
An immunoprecipitation radioligand assay was used to detect GADA.17,18The labeled GAD65 antigen was obtained from
recom-binant human islet GAD65 cDNA transcribed and translated ac-cording to the manufacturer’s instructions (Promega, Madison, WI) in the presence of [35S]methionine (Amersham; Amersham,
Bucks, UK). Sera (2mL) were incubated overnight at14°C with ;30 000 cpm human GAD65 in a total volume of 50mL of Tris-buffered saline with Tween-20. To isolate the immunocomplexes, 25mL of Protein A-Sepharose CL-4B (Pharmacia, Uppsala, Swe-den) in a total volume of 100 mL of Tris-buffered saline with Tween-20 was added. A scintillation counter was used to count the immunocomplexes precipitated. All samples were analyzed in quadruplicate with and without competition from excess amounts of unlabeled recombinant GAD 65 (1 mg/well) produced in an expression system using infant hamster kidney cells and purified with Triton X-114. The results were expressed in relative units (RU), representing the specific binding as a percentage of that obtained with a positive standard serum. The limit for GADA positivity was set at 6.5 RU, which represents the 99th percentile in a series of 372 healthy control siblings. The disease sensitivity of this assay was 76% and the specificity 99%, based on the 1995 Multiple Autoantibody Workshop.19
A radiobinding assay was used to analyze the IA-2A.20The
radioactive IA-2 protein was produced in the TNT Coupled Re-ticulocyte Lysate System (Promega) in the presence of [35
S]methi-onine. Sera were incubated overnight at14°C with 10 000 cpm of labeled IA-2 protein. Protein-A Sepharose (Pharmacia) was used to isolate the immune complexes on the next day. After thorough washing, the radioactivity of the samples was measured using a liquid scintillation counter (1450 Microbeta Trilux; Wallac, Turku, Finland). The results were expressed in RU based on a standard curve constructed from the dilution of a pool of strongly positive samples and a pool of negative samples. The standard curve was run on each plate. A sibling was considered IA-2A positive if the serum antibody levels exceeded 0.429 RU, which represents the 99th percentile in 374 healthy Finnish children and adolescents. The disease sensitivity of this assay was 62% and the specificity 97% based on 140 samples included in the 1995 Multiple Autoan-tibody Workshop.19
IVGTT
Siblings participating in the IVGTT were given a glucose dose of 0.5 g/kg in;3 minutes (615 s) after overnight fasting for 10 to 16 hours. Blood samples were taken before the glucose infusion and at 1, 3, 6, 10, 20, 30, 40, 50, and 60 minutes thereafter. Serum insulin concentrations were measured radioimmunologically.21
The glucose oxidase method was used to measure the blood glucose levels.22The sum of the insulin concentrations at 1 and 3
minutes was defined as the FPIR to glucose. To evaluate the degree of glucose tolerance, the glucose disappearance rate (Kg)
was assessed. The disappearance rate was expressed as the per-centage decrease in blood glucose per minute (%/min). FPIR levels ,45 mU/L, which represents the 3rd percentile of FPIR values in healthy controls23after adjustment for the insulin assay
used based on serum exchange of insulin samples between Oulu and Boston and Kgvalues ,1.30%/min, were considered to be
abnormal.
Data Handling and Statistical Analyses
The second set of criteria was based on a combination of autoantibodies and FPIR. The siblings with no antibodies were still placed in the first group (no prediabetes), those with one antibody specificity detectable but a normal FPIR comprised the second group (early prediabetes), those with two or more anti-bodies but still with a normal FPIR the third group (advanced prediabetes), and those with an abnormal FPIR and at least one antibody specificity the fourth group (late prediabetes). Initially antibody-positive siblings with no available FPIR were excluded from this classification, leaving 712 out of 758 in the analyses.
Cross-tabulation and x2 statistics were used to analyze the
distributions. A parametric one-way analysis of variance was used when normally distributed variables were compared between groups and the nonparametric Kruskall-Wallis analysis of vari-ance when analyzing variables with a skewed distribution. Thet
test was used for comparisons between two groups in the case of normally distributed variables. and the Mann-WhitneyUtest in the case of variables with skewed distributions. The 95% confi-dence intervals were calculated by the exact method.
RESULTS
Classification Based on the Number of Antibodies Six hundred and sixty-one siblings (87.2%) had no prediabetes, 49 (6.5%) had early prediabetes, 13 (1.7%) had advanced prediabetes, and 35 (4.6%) had late prediabetes. Altogether there were 97 siblings (12.8%) who tested positive for at least one antibody specificity, 60 (7.9%) for ICA ($2.5 JDFU), 28 (3.7%) for IAA, 57 (7.5%) for GADA, and 40 (5.3%) for IA-2A.
The distribution of various antibody specificities and combinations in early, advanced, and late pre-diabetes is shown in Table 1. ICA were the most frequent antibody specificity among those with early prediabetes, whereas the ICA plus GADA combina-tion was most common in those with advanced pre-diabetes. The combination ICA plus GADA plus IA-2A was most prevalent in the siblings with late prediabetes. Significantly lower levels of FPIR were
observed in the late prediabetes category than in those with no signs of prediabetes or in those at the early stage (Fig 1). There was also a significant dif-ference between the early and advanced prediabetes categories. Those with late prediabetes had a signif-icantly lower Kg (median, 1.38%/min; range, 1.01–
2.62%/min) than those with no prediabetes (median, 1.61%/min; range, 1.16 – 4.08%/min; P , .05) and those with early prediabetes (median, 1.67%/min; range, 1.11–3.65%/min;P ,.05).
The 758 siblings included 35 (4.6%) who pro-gressed to clinical T1DM during prospective obser-vation for an average of 9 years (range, 0.02–10.3 years), leaving 723 (95.4%) unaffected. In the group with no signs of prediabetes initially, only 6 siblings (0.9%; CI, 0.3%–2.0%) developed T1DM, whereas 3 (6.1%; CI, 1.3%–16.9%) did so in the category of early prediabetes, 3 (23.1%; CI, 5.0%–53.8%) in the cate-gory of advanced prediabetes and as many as 23 (65.7%; CI, 47.8%– 80.9%) among those with late pre-diabetes. Accordingly the siblings with early abetes had a sevenfold, those with advanced predi-abetes a 33-fold, and those with late predipredi-abetes a 209-fold risk for progressing to overt T1DM when compared with the siblings without any signs of prediabetes (Table 1).
There were significant differences in the time to diagnosis between the categories of preclinical T1DM in the 35 children and adolescents who pro-gressed to clinical disease (Fig 2), the time being significantly shorter in the late prediabetes category than in those children with no signs of prediabetes or those in the early stage of prediabetes. There was also a significant difference between the cases with early and advanced prediabetes.
Classification Based on a Combination of Antibodies and FPIR
According to this classification 661 (92.8%) still had no signs of prediabetes, whereas 15 (2.1%) had signs of early prediabetes, 23 (3.2%) advanced pre-diabetes, and 13 (1.8%) late prediabetes. Forty-three siblings (6.0%) had detectable ICA, whereas IAA positivity was observed in 13 cases (1.8%), GAD antibodies in 35 (4.9%), and IA-2 antibodies in 26 (3.7%).
Table 2 shows the frequency of various antibody specificities and combinations in relation to the stage of prediabetes. ICA were the most frequent single antibody specificity in the siblings with early predi-abetes, whereas the combination ICA plus GADA plus IA-2A was most prevalent both in those with advanced prediabetes and late prediabetes. The chil-dren in the late prediabetes category had signifi-cantly lower Kg (median, 1.32%/min; range, 0.66 –
1.93%/min) than those in the early prediabetes category (median, 1.69%/min; range, 1.11–3.65%/ min;P, .05).
There were 25 (3.5%) siblings who manifested clin-ical signs of T1DM during the observation period, comprising 6 out of the 661 children classified as having no prediabetes (0.9%; CI, 0.3%–2.0%), 1 (6.7%; CI, 0.2%–32.0%) with early prediabetes, 6 (26.1%; CI, 10.2%– 48.4%) with advanced prediabetes, and 12
TABLE 1. Combinations of Antibodies in the Initial Sample of 758 Siblings Classified According to the Number of Antibodies, Their Relation to Type 1 Diabetes Mellitus, and to Odds Ratios for Clinical Disease
Antibody Status n Progression to Type 1
Diabetes Mellitusn
(%)
Odds Ratio (95% Confidence
Interval)
No prediabetes 661 6 (0.9) 1
Early prediabetes 49 3 (6) 7.1 (1.7–29.4)
ICA 17 1 (6)
IAA 12 0
GADA 15 2 (13)
IA-2A 5 0
Advanced prediabetes 13 3 (23) 32.8 (7.2–150)
ICA1GADA 6 1 (17)
IAA1GADA 3 0
ICA1IA-2A 2 1 (50)
ICA1IAA 1 1 (100)
IA-2A1GADA 1 0
Late prediabetes 35 23 (66) 209 (72.2–607) ICA1GADA1IA-2A 23 16 (70)
ICA1IAA1IA-2A 3 3 (100) ICA1IAA1GADA 3 2 (67) IAA1GADA1IA-2A 1 0
All four 5 2 (40)
(92.3%; CI, 64.0%–99.8%) with late prediabetes. The siblings with early prediabetes had an eightfold higher risk of developing T1DM than those with no signs of prediabetes, whereas the risk was 39-fold in those with advanced prediabetes and 1310-fold in those with late prediabetes (Table 2). There were significant differences in the time to diagnosis be-tween the categories of prediabetes, a significantly shorter time being observed in those with late pre-diabetes than in those with no signs of prepre-diabetes (Fig 3).
DISCUSSION
The risk of progressing to T1DM is a product of many factors, including genetic susceptibility, age, and the presence of various disease-associated auto-antibodies. In this prospective study, we analyzed
the risk of T1DM in siblings of children with newly diagnosed T1DM in relation to detectable autoanti-bodies and insulin secretory capacity. The staging was primarily based on a single blood sample ob-tained near to the time of diagnosis of diabetes in the affected sibling. There is an obvious pressure for evaluating the risk of future T1DM in siblings at the time when the first child in the family presents with the disease. Among the 758 siblings observed here, 97 (12.8%) tested positive for at least one antibody
Fig 1. First phase insulin response (FPIR) in 82 siblings of children with type 1 diabetes mellitus classified according to the stage of preclinical diabetes based on the number of autoantibodies. Siblings testing positive for islet cell antibodies and/or insulin autoantibodies on one or more occasions during the first 4 years (n5108) were invited to an intravenous glucose tolerance test. Eighty-three of these siblings (77%) underwent an intravenous glucose tolerance test at least once. The results are presented as box plots rep-resenting the median (——) and the 25th and 75th centiles. The error bars represent the small-est and the largsmall-est observed values that are not outliers (F).
Fig 2. Time to diagnosis (years) in 35 siblings of children with type 1 diabetes mellitus classified according to the stage of pre-clinical diabetes based on the number of autoantibodies. The dashed line indicates the median.
TABLE 2. Combinations of Antibodies in the Initial Sample of 712 Siblings Classified According to the Number of Antibodies and First-phase Insulin Response, Their Relation to Progression to Type 1 Diabetes Mellitus and to Odds Ratios for Clinical Disease
Antibody Status n Progression to Type 1 Diabetes Mellitusn
(%)
Odds Ratio (95% Confidence
Interval)
No prediabetes 661 6 (0.9) 1
Early prediabetes 15 1 (7) 7.8 (0.9–69.1)
ICA 10 0
IAA 3 0
GADA 2 1 (50)
Advanced prediabetes 23 6 (26) 38.5 (11.3–132)
ICA1IAA 1 1 (100)
ICA1GADA 4 0
ICA1IA-2A 1 0
IAA1GADA 1 0
ICA1GADA1IA-2A 9 3 (33) ICA1IAA1GADA 2 1 (50) IAA1GADA1IA-2A 1 0
All four 4 1 (25)
Late prediabetes 13 12 (92) 1310 (146–11 737)
GADA 1 0
ICA1GADA 1 1 (100)
ICA1IA-2A 1 1 (100)
ICA1GADA1IA-2A 9 9 (100)
All four 1 1 (100)
specificity at a time close to the diagnosis in the index case, which is a relatively high proportion in com-parison with the general population but is in agree-ment with the results of a recent cross-sectional American survey of first-degree relatives.24Of the 35
children who progressed to clinical T1DM during prospective follow-up, 29 (82.9%) had been positive for at least one antibody in the first sample available. This is a somewhat smaller proportion than that observed in children with T1DM at diagnosis.19,25,26It
should be noted, however, that 4 out of the 6 initially antibody-negative progressors seroconverted to an-tibody positivity later in the preclinical period (Knip M, unpublished observation).
It has been observed earlier that the risk of pro-gressing to T1DM increases with the number of au-toantibodies detectable.8,24 –28The present results
pro-vide further epro-vidence for the claim that combined analysis of T1DM-associated autoantibodies offers an effective means of assessing the disease risk in first-degree relatives of children with T1DM. We set out here to consider whether it is clinically relevant to classify siblings into stages of preclinical diabetes based on their antibody status and FPIR, and to determine the risk of progression to T1DM based on the stage of prediabetes at a time close to the diag-nosis in the index case. The two sets of classification criteria produced similar results, with the risk of T1DM increasing according to the progression in the stage of prediabetes. This suggests that such a clas-sification is definitely informative and may be useful for assessing the risk of T1DM in healthy siblings at the time of diagnosis in the first affected child. The classification will become clinically relevant as soon as the first effective means of preventing or delaying clinical T1DM have been developed.
There were some differences in the risk of progres-sion in relation to the stage of prediabetes depending on the type of classification. In the classification based exclusively on the number of antibodies the risk of those with late prediabetes developing T1DM was 66%, although grading according to both
anti-bodies and FPIR resulted in a risk as high as 92% at this late stage. Similar, but smaller differences were observed at the earlier stages of preclinical T1DM. Thus, the analysis of FPIR improves the predictive power of the staging strategy, and an IVGTT is there-fore highly recommendable for siblings testing pos-itive for one or more autoantibodies. FPIR neverthe-less remains an indirect indicator of the remaining
b-cell mass, which declines toward the clinical onset of T1DM.8When comparing the outcome of the two
classifications used here, it should be born in mind that there is a difference in the number of siblings included, because not all the antibody-positive sib-lings wished to attend for an IVGTT. Accordingly, the classification based on the FPIR as well covered fewer siblings than the staging based exclusively on the number of antibodies. If we classify the 712 sib-lings covered by the second set of criteria according to the first set of criteria, we obtain, however, results very similar to those seen in all 758 siblings with a risk for T1DM of 6% in early, 33% in advanced, and 58% in late prediabetes. Although we observed sig-nificant differences in the time to clinical diagnosis in relation to the stage of preclinical diabetes, this time period also varied greatly within the same stage, eg, from 0.02 to 7.7 years among those with late predia-betes. This makes it difficult to predict the time of diagnosis on an individual basis.
CONCLUSION
In summary, the analysis of four diabetes-associ-ated autoantibodies facilitates estimation of the T1DM risk in unaffected siblings, and classification into stages of preclinical diabetes based on a combi-nation of the number of antibodies and the FPIR is an effective tool for grading the risk of T1DM in this context. It will obviously not be possible to identify all siblings who will progress to clinical T1DM at the time of diagnosis in the first affected child, however, because some progressors have no signs of preclini-cal diabetes at that point; eg, 6 (0.9%) out of the present 661 children with no signs of preclinical di-abetes at the time of diagnosis of the index case in the family subsequently presented with clinical T1DM. The majority of these siblings seroconverted to anti-body positivity later in the prediabetic process. The staging of preclinical T1DM could become an impor-tant tool in the near future for the identification of those siblings who should be treated with effective preventive modalities as soon as such treatment is available.
APPENDIX
Members of the Childhood Diabetes in Finland study group are the following: Principal Investigators: H. K. Åkerblom and J. Tuomilehto; Coordinators: R. Lounamaa and L. Toivanen; Data Management: E. Virtala and J. Pitka¨niemi; Local Investigators: A. Fagerlund, M. Flittner, B. Gustafsson, C. Ha¨ggqvist, A. Hakulinen, L. Herva, P. Hiltunen, T. Huhtama¨ki, N.-P. Huttunen, T. Huup-ponen, M. Hyttinen, T. Joki, R. Jokisalo, M.-L. Ka¨a¨r, S. Kallio, E. A. Kaprio, U. Kaski, M. Knip, L. Laine, J. Lappalainen, J. Ma¨enpa¨a¨, A.-L. Ma¨kela¨, K. Niemi, A. Niiranen, A. Nuuja, P. Ojaja¨rvi, T. Otonkoski, K. Pihlajama¨ki, S. Po¨ntynen, J. Rajantie, J. Sankala, J. Schumacher, M. Sillanpa¨a¨, M.-R. Ståhlberg, C.-H. Stråhlman, T. Uotila, M. Va¨re, P. Varimo, and G. Wetterstrand; Special Investi-gators: A. Aro, H. Hurme, M. Hiltunen, H. Hyo¨ty, J. Ilonen, J.
Karjalainen, M. Knip, P. Leinikki, A. Miettinen, T. Peta¨ys, H. Reijonen, A. Reunanen, L. Ra¨sa¨nen, T. Saukkonen, E. Savilahti, E. Tuomilehto-Wolf, P. Va¨ha¨salo, and S. M. Virtanen.
ACKNOWLEDGMENTS
This study was supported by grants from the Juvenile Diabetes Foundation International (Grant 197032), the Finnish Diabetes Research Foundation, the Medical Research Council, Academy of Finland (Grant 26109), and the Novo Nordisk Foundation. The Childhood Diabetes in Finland study has also been supported by grants from the Association of Finnish Life Insurance Companies, the Sigrid Juse´lius Foundation, the National Institutes of Health (Grant DK 37 957), and the University of Helsinki.
We thank Sirpa Anttila, Susanna Heikkila¨, Pa¨ivi Koramo, Erik Mrena, and Riitta Pa¨kkila¨ for technical assistance.
REFERENCES
1. Eisenbarth GS. Type I diabetes mellitus. A chronic autoimmune disease. N Engl J Med. 1986;314:1360 –1368
2. Bonifacio E, Bottazzo GF. Immunology of IDDM (type 1 diabetes)— entering the ’90s. In: Alberti KGMM, Krall LP, eds.Diabetes Annual/6. Amsterdam: Elsevier; 1990:20 – 47
3. Krolewski AS, Warram JH, Rand LI, Kahn CL. Epidemiologic approach to the etiology of Type 1 diabetes and its complications.N Engl J Med. 1987;317:1390 –1397
4. Bennett PH. Epidemiology of diabetes mellitus. In: Rifkin H, Porte D Jr, eds.Diabetes Mellitus, Theory and Practice. New York, NY: Elsevier; 1990:357–377
5. Reunanen A, Åkerblom HK, Ka¨a¨r M-L. Prevalence and ten year (1970 –79) incidence of insulin-dependent diabetes mellitus in children and adolescents in Finland.Acta Paediatr Scand. 1982;71:893– 899 6. Tuomilehto J, Virtala E, Karvonen M, et al. Increase in incidence of
insulin-dependent diabetes mellitus among children in Finland.Int J Epidemiol. 1995;24:984 –992
7. Bonifacio E, Bingley PJ, Shattock M, et al. Quantification of islet-cell antibodies and prediction of insulin-dependent diabetes.Lancet. 1990; 335:147–149
8. Bingley PJ, Christie MR, Bonifacio E, et al. Combined analysis of auto-antibodies improves prediction of IDDM in islet cell antibody-positive relatives.Diabetes. 1994;43:1304 –1310
9. Srikanta S, Ganda OP, Rabizadeh A, Soeldner JS, Eisenbarth GS. First-degree relatives of patients with type 1 diabetes mellitus. Islet cell antibodies and abnormal insulin secretion.N Engl J Med. 1985;313: 461– 464
10. Vardi P, Crisa L, Jackson RA, et al. Predictive value of intravenous glucose tolerance test insulin secretion less than or greater than the first percentile in islet cell antibody positive relatives of type 1 (insulin-dependent) diabetic patients.Diabetologia. 1991;34:93–102
11. Chase HP, Garg SK, Butler-Simon N, et al. Prediction of the course of pre-type 1 diabetes.J Pediatr. 1991;118:838 – 841
12. Tuomilehto J, Lounamaa R, Tuomilehto-Wolf E, et al. Epidemiology of
childhood diabetes mellitus in Finland— background of a nationwide study of type I (insulin-dependent) diabetes mellitus.Diabetologia. 1992; 35:70 –76
13. WHO Study Group.Diabetes Mellitus. Technical Report Series. Geneva, Switzerland: World Health Organization; 1985;727:10 –12
14. Bottazzo GF, Florin-Christensen A, Doniach D. Islet-cell antibodies in diabetes mellitus with autoimmune polyendocrine deficiencies.Lancet. 1974;2:1279 –1282
15. Lernmark Å, Molenaar JL, van Beers WA, et al. The fourth international serum exchange workshop to standardize cytoplasmic islet cell anti-bodies.Diabetologia. 1991;34:534 –535
16. Palmer JP, Asplin CM, Clemons P, et al. Insulin antibodies in insulin-dependent diabetics before insulin treatment. Science. 1983;222: 1337–1339
17. Petersen JS, Hejnaes KR, Moody A, et al. Detection of GAD(65) anti-bodies in diabetes and other autoimmune diseases using a simple radioligand assay.Diabetes. 1994;43:459 – 467
18. Sabbah E, Kulmala P, Veijola R, et al. Glutamic acid decarboxylase antibodies in relation to other autoantibodies and genetic risk markers in children with newly diagnosed insulin-dependent diabetes.J Clin Endocrinol Metab. 1996;81:2455–2459
19. Verge CF, Stenger D, Bonifacio E, et al. Combined use of autoantibodies (IA-2 autoantibody, GAD autoantibody, insulin autoantibody, cytoplas-mic islet cell antibodies) in type 1 diabetes. Combinatorial islet autoan-tibody workshop.Diabetes. 1998;47:1857–1866
20. Savola K, Bonifacio E, Sabbah E, et al. IA-2 antibodies—a sensitive marker of IDDM with clinical onset in childhood and adolescence. Diabetologia. 1998;41:424 – 429
21. Herbert V, Lau KS, Gottlieb CW, Bleicher SW. Coated charcoal immu-noassay of insulin.J Clin Endocrinol. 1965;25:1375–1384
22. Hjelm M. Enzymatic determination of hexoses in blood and urine.Scand J Clin Lab Invest Suppl. 1966;18:85–98
23. Srikanta S, Ganda OP, Gleason RE, Jackson RA, Soeldner JS, Eisenbarth GS. Pre-type I diabetes. Linear loss of beta cell response to intravenous glucose.Diabetes. 1984;33:717–720
24. Verge CF, Gianani R, Kawasaki E, et al. Prediction of type 1 diabetes in first-degree relatives using a combination of insulin, GAD, and ICA512bdc/IA-2 autoantibodies.Diabetes. 1996;45:926 –932
25. Bingley PJ, Bonifacio E, Williams AJK, Genovese S, Bottazzo GF, Gale EAM. Prediction of IDDM in the general population. Strategies based on combinations of autoantibody markers.Diabetes. 1997;46:1701–1710 26. Bonifacio E, Genovese S, Braghi S, et al. Islet autoantibody markers in IDDM: risk assessment strategies yielding high sensitivity.Diabetologia. 1996;38:816 – 822
27. Christie MR, Roll U, Payton MA, Hatfield ECI, Ziegler AG. Validity of screening for individuals at risk for type 1 diabetes by combined anal-ysis of antibodies to recombinant proteins. Diabetes Care. 1997;20: 965–970
DOI: 10.1542/peds.104.4.925
1999;104;925
Pediatrics
Childhood Diabetes in Finland Study Group
Samy Mrena, Kaisa Savola, Petri Kulmala, Hans K. Åkerblom, Mikael Knip and the
Staging of Preclinical Type 1 Diabetes in Siblings of Affected Children
Services
Updated Information &
http://pediatrics.aappublications.org/content/104/4/925
including high resolution figures, can be found at:
References
http://pediatrics.aappublications.org/content/104/4/925#BIBL
This article cites 25 articles, 8 of which you can access for free at:
Subspecialty Collections
http://www.aappublications.org/cgi/collection/diabetes_mellitus_sub
Diabetes Mellitus
http://www.aappublications.org/cgi/collection/endocrinology_sub
Endocrinology
http://www.aappublications.org/cgi/collection/pediatrician_wellness
Pediatrician Wellness following collection(s):
This article, along with others on similar topics, appears in the
Permissions & Licensing
http://www.aappublications.org/site/misc/Permissions.xhtml
in its entirety can be found online at:
Information about reproducing this article in parts (figures, tables) or
Reprints
http://www.aappublications.org/site/misc/reprints.xhtml
DOI: 10.1542/peds.104.4.925
1999;104;925
Pediatrics
Childhood Diabetes in Finland Study Group
Samy Mrena, Kaisa Savola, Petri Kulmala, Hans K. Åkerblom, Mikael Knip and the
Staging of Preclinical Type 1 Diabetes in Siblings of Affected Children
http://pediatrics.aappublications.org/content/104/4/925
located on the World Wide Web at:
The online version of this article, along with updated information and services, is
by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.
