Islet autoantibodies (IAs) precede the clinical onset of type 1 diabetes (T1D); however, the knowledge is limited about whether the prodrome affects complete blood counts (CBCs) in 4- to 12-year-old children with increased genetic risk for T1D. This study tested whether CBCs were altered in 4- to 12-year-old children without (n = 376) or with one or several IAs against insulin, GAD65, or IA-2 (n = 72). CBC was analyzed during longitudinal follow-up in 448 Swedish children enrolled in The Environmental Determinants of Diabetes in the Young (TEDDY) study. A linear mixed-effects model was used to assess potential association between IA and CBC measurements over time. The white blood cell and neutrophil counts were reduced in children with IAs, primarily in boys. In contrast, girls had lower levels of hemoglobin and hematocrit. Positivity for multiple IAs showed the lowest counts in white blood cells and neutrophils in boys and red blood cells, hemoglobin, and hematocrit in girls. These associations were primarily observed in children with the HLA-DR3-DQ2/DR4-DQ8 genotype. We conclude that the reduction in neutrophils and red blood cells in children with multiple IAs and HLA-DR3-DQ2/DR4-DQ8 genotype may signal a sex-dependent islet autoimmunity detected in longitudinal CBCs.

Autoantibodies against the β-cell autoantigens insulin (IAA), GAD antibody (GADA), or protein tyrosine phosphatase-like (IA-2A) precede the clinical onset of autoimmune type 1 diabetes (T1D). Children at genetic risk for T1D were monitored from birth in The Environmental Determinants of Diabetes in the Young (TEDDY) study for a first appearing islet autoantibody (IA), be it IAA or GADA (13). The IAA incidence rate was highest in the 1st year of life (18 months in Sweden) and declined over the following 5 years, whereas the incidence rate of GADA increased during the first 2 years (2.5 years in Sweden) of life and remained seemingly constant until the age of 6 years (13). The appearance of IAA or GADA as the first IA was related to the HLA-DR/DQ genotype (1,2). After an initial event that triggers autoimmunity reflected by a first appearing IAA or GADA, the pathogenesis progresses toward the clinical onset of disease more rapidly with an increasing number of IAs (4,5), which may be critical because islet β-cells are thought to be destroyed by autoreactive T cells, not by autoantibodies (6,7). So far, β-cell–specific autoantibodies are the best predictors of an ongoing autoimmune process resulting in clinical onset (8). Notwithstanding the critical role of the adaptive immune response in the etiology and pathogenesis of T1D (reviewed in [912]), there is growing evidence that the innate immune response also contributes to the pathogenesis (1214). Neutrophils are thought to contribute to the etiological triggering and the pathogenesis of T1D in mouse models (reviewed in [15]). A reduction in peripheral blood neutrophils has recently been reported in healthy IA-positive children all having a first-degree relative with the disease (16). Similarly, compared with healthy children, white blood cells, neutrophils, and lymphocytes were reduced in IA-positive children with a family history of T1D (17), and there is an unknown effect of the peripheral immune cell counts on the pathogenesis of T1D (18).

We have studied children in Sweden who are enrolled in the TEDDY study (19,20) and monitored from birth to determine the first appearing IA (1) as well as progression to multiple autoantibodies and clinical onset of diabetes (4,5). We specifically questioned whether complete blood count (CBC) was associated with the IA status in 4- to 12-year-old Swedish TEDDY children and whether the association differed by sex and HLA DQ-DR genotype.

TEDDY Design

The TEDDY study is a prospective cohort study funded by the U.S. National Institutes of Health with the primary goal to identify environmental causes of T1D. TEDDY includes six clinical research centers—three in the U.S.: Colorado, Georgia/Florida, and Washington, and three in Europe: Finland, Germany, and Sweden. Detailed study design and methods have been previously published (19,21). Written informed consents were obtained for all study participants from a parent or primary caretaker, separately, for genetic screening and participation in prospective follow-up. The current study was approved by the Regional Ethics Review Board in Lund and is monitored by an External Advisory Board formed by the National Institutes of Health.

CBC Measurements in TEDDY Children

The study cohort consists of 448 children, 4–12 years old, from the TEDDY clinic in Malmö, Sweden (Table 1). The CBC measurements were initiated in June 2014 and completed in February 2017. The samples were analyzed at their scheduled visits and within 8 h after sample draw. The TEDDY protocol requires visits every 3 months for children who were IA positive and every 6 months for children who remained autoantibody negative.

Table 1

Characteristics of TEDDY children (n = 448) investigated for CBC when negative or positive for one or several IAs

IA negative (n = 376)IA positive (n = 72)
Children, n (%) 376 (84) 72 (16) 
Sex, n   
 Girls 182 30 
 Boys 194 42 
Age at first CBC (months), median (min–max) 91 (52–145) 101.5 (59–139) n.s. 
 Girls 91 (53–144) 103 (59–139) 
 Boys 94 (52–145) 99 (59–137) 
CBC measures per child, min–max 1–6 1–9 
Months of follow-up, min–max 1–30 1–30 
IAs, n   
 1 25 
 2 16 
 3 31 
Change in IA status None None 
HLA DR-DQ, n (%)   
 DR3/4-DQ2/8 151 (40.2) 39 (54.1) 
 DR4/4-DQ 8/8 87 (23.1) 12 (16.7) 
 DR4/8-DQ 8/4 40 (10.6) 12 (16.7) 
 DR3/3-DQ 2/2 91 (24.2) 9 (12.5) 
 DR4/1-DQ 8/5 4 (1.1) 
 DR4/13-DQ 8/6 2 (0.5) 
 HLA ineligible 1 (0.3) 
  Total 376 (100) 72 (100) 
IA negative (n = 376)IA positive (n = 72)
Children, n (%) 376 (84) 72 (16) 
Sex, n   
 Girls 182 30 
 Boys 194 42 
Age at first CBC (months), median (min–max) 91 (52–145) 101.5 (59–139) n.s. 
 Girls 91 (53–144) 103 (59–139) 
 Boys 94 (52–145) 99 (59–137) 
CBC measures per child, min–max 1–6 1–9 
Months of follow-up, min–max 1–30 1–30 
IAs, n   
 1 25 
 2 16 
 3 31 
Change in IA status None None 
HLA DR-DQ, n (%)   
 DR3/4-DQ2/8 151 (40.2) 39 (54.1) 
 DR4/4-DQ 8/8 87 (23.1) 12 (16.7) 
 DR4/8-DQ 8/4 40 (10.6) 12 (16.7) 
 DR3/3-DQ 2/2 91 (24.2) 9 (12.5) 
 DR4/1-DQ 8/5 4 (1.1) 
 DR4/13-DQ 8/6 2 (0.5) 
 HLA ineligible 1 (0.3) 
  Total 376 (100) 72 (100) 

CBC

CBC was determined in a multiparameter automated hematology analyzer (CELL-Dyn Ruby; Abbott Laboratories, Diagnostic Division, Abbott Park, IL) (22). The instrument was operated according to the instructions by the manufacturer, including a daily calibration. Counts (cells ×109/L) of white blood cells, neutrophils, lymphocytes, monocytes, eosinophils, basophils, platelets, red blood cells (cells ×1012/L) and red blood cell parameters; hemoglobin (g/L), hematocrit (L/L), mean corpuscular volume (fL), mean corpuscular hemoglobin (MCH) (pg), MCH concentration (MCHC) (g/L), and red cell distribution width (% coefficient of variation) were obtained.

HLA-DR-DQ Typing

Genotype screening (20) was conducted using a dried blood spot punch or a small volume whole-blood lysate specimen format, as previously published (23). Infants from the general population were eligible for the study if they had any one of the following HLA genotypes (excluding those with DR4 subtype DRB1*04:03):

  1. DR4-DQA1*03:0X-DQB1*03:02/DR3-DQA1*05:01-DQB1*02:01

  2. DR4-DQA1*03:0X-DQB1*03:02/DR4-DQA1*03:0X-DQB1*03:021

  3. DR4-DQA1*03:0X-DQB1*03:02/DR8-DQA1*04:01-DQB1*04:02

  4. DR3-DQA1*05:01-DQB1*02:01/DR3-DQA1*05:01-DQB1*02:01

Note: 1Acceptable alleles in this haplotype include both DQB1*03:02 and *03:04.

Infants with a first-degree relative with T1D were eligible for enrollment if they had any of the following HLA genotypes:

  1. DR4-DQA1*03:0X-DQB1*03:021/DR3-DQA1*05:01-DQB1*02:01

  2. DR4-DQA1*03:0X-DQB1*03:021/DR4-DQA1*03:0X-DQB1*03:021

  3. DR4-DQA1*030X-DQB1*03021/DR8-DQA1*04:01-DQB1*04:02

  4. DR3-DQA1*05:01-DQB1*02:01/DR3-DQA1*05:01-DQB1*02:01 (24)

  5. DR4-DQA1*03:0X-DQB1*03:021/DR4-DQA1*03:0X-DQB1*02:0X

  6. DR4-DQA1*03:0X-DQB1*03:021/DR12-DQA1*0101-DQB1*05:012

  7. DR4-DQA1*030X-DQB1*03:021/DR13-DQA1*01:02-DQB1*06:04

  8. DR4-DQA1*03:0X-DQB1*03:02/DR4-DQA1*03:0X-DQB1*03:04

  9. DR4-DQA1*03:0X-DQB1*03:021/DR9-DQA1*03:0X-DQB1*03:03 (23)

  10. DR3-DQA1*05:01-DQB1*02:01/DR9-DQA1*03:0X-DQB1*03:03

Notes: 1Acceptable alleles in this haplotype included both DQB1*03:02 and *03:04. 2In this DQB1*05:01 haplotype, DR10 was excluded. Only DR1 was eligible.

Autoantibody Measurements

IAA, GADA, or IA-2A were measured in two laboratories by radiobinding assays (25,26). In Europe, all sera were assayed at the University of Bristol, Bristol, U.K. All samples positive for IA and 5% of negative samples were retested at the Barbara Davis Center for Childhood Diabetes at the University of Colorado, Denver, and deemed confirmed if concordant. Both laboratories have previously shown high sensitivity and specificity as well as concordance (27).

IA Analyses

Persistent IA positivity was defined as confirmed positive IAA, GADA, or IA-2A on at least two consecutive study visits. The first appearance of persistent confirmed IA in the follow-up was considered and counted.

Statistical Analysis

Considering correlations between measures from the same subject (within-subject correlation), a linear mixed-effects model was used to assess the association of islet autoimmunity on each CBC measurement. We first examined whether the association between the status of IA and each CBC measurement was different by sampling age, but no significant difference was noted in all CBC analyses. Hence, the model included random intercept and random slope as well as sampling age and the status of IAs as fixed effects. Unstructured within-subject correlation was assumed for the random error. The regression coefficient for the status of IAs was assessed to determine whether the associations of islet autoimmunity were ignorable or not. The association between age and each CBC measurement was assessed among IA-negative children. Age of initial measurement was compared using the Wilcoxon rank sum test, and the proportions of girls and HLA-DR3-DQ2/DR4-DQ8 subjects were compared using the Fisher exact test. Age-dependent effects were observed in most CBC parameters and later corrected for in the linear mixed-effects model. Due to this procedure, the observed effects of an increasing number of IAs on neutrophils and red blood cell parameters were all corrected for age.

Two-sided P values of less than 0.05 were considered for statistical significance. All analyses were performed using SAS 9.4 software (SAS Institute, Cary, NC).

CBC in Relation to Age in IA Negative Children

CBC was examined in 448 children participating in the longitudinal TEDDY study. Children were examined at one to six visits if negative (84%) or one to nine visits (16%) if positive for any IA. The number of CBC measurements was therefore higher in the IA-positive subjects (n = 72; median, 3) than in the IA-negative subjects (n = 376; median, 2), but the significant difference in the age when the CBC measurements started was not significant (Table 1). We examined whether CBC varied with age in the IA-negative children (Table 2). In agreement with earlier studies, older age was significantly associated with decreasing cell counts in the different white blood cell types, except for a nonsignificant association in eosinophils and basophils (Table 2). In contrast to the nucleated cells, red blood cell counts and hemoglobin levels increase with age, which was also confirmed in our study (28,29).

Table 2

Effects of age (years) on each CBC measurement in IA-negative subjects (n = 376)

CBCEstimateSEP value
White blood cells (109 cells/L) −0.123 0.044 0.005 
Neutrophils (109 cells/L) −0.060 0.031 0.053 
Lymphocytes (109 cells/L) −0.036 0.013 0.006 
Monocytes (109 cells/L) −0.010 0.004 0.015 
Eosinophils (109 cells/L) −0.002 0.008 0.766 
Basophils (109 cells/L) −0.001 0.001 0.306 
Red blood cells (1012 cells/L) 0.040 0.015 0.010 
Hemoglobin (g/L) 0.167 0.040 <0.0001 
CBCEstimateSEP value
White blood cells (109 cells/L) −0.123 0.044 0.005 
Neutrophils (109 cells/L) −0.060 0.031 0.053 
Lymphocytes (109 cells/L) −0.036 0.013 0.006 
Monocytes (109 cells/L) −0.010 0.004 0.015 
Eosinophils (109 cells/L) −0.002 0.008 0.766 
Basophils (109 cells/L) −0.001 0.001 0.306 
Red blood cells (1012 cells/L) 0.040 0.015 0.010 
Hemoglobin (g/L) 0.167 0.040 <0.0001 

Values in boldface type are statistically significant (P < 0.05).

CBC in Children With and Without IAs

In the next step, differences in the peripheral blood cell counts were examined between children with or without IAs (Table 3). Children with IAs had reduced white blood cell counts (P = 0.046) as a result from a reduction in neutrophil cell counts (P = 0.017). The two red blood cell parameters, hemoglobin (P = 0.026) and hematocrit (P = 0.031), were also reduced in children with IA. The neutrophil-to-lymphocyte ratio did not differ between the two groups.

Table 3

The association between IA and each CBC measurement and association to sex

CBCEstimateSEP value
White blood cells (109 cells/L)    
 All subjects −0.315 0.157 0.046 
 Girls −0.025 0.212 0.906 
 Boys −0.540 0.229 0.019 
Neutrophils (109 cells/L)    
 All subjects −0.239 0.099 0.017 
 Girls −0.031 0.128 0.807 
 Boys −0.386 0.151 0.012 
Lymphocytes (109 cells/L)    
 All subjects −0.034 0.057 0.553 
 NLR (all) −0.115 0.067 0.090 
 Girls 0.048 0.082 0.555 
 Boys −0.084 0.080 0.295 
Monocytes (109 cells/L)    
 All subjects −0.024 0.015 0.108 
 Girls −0.024 0.022 0.281 
 Boys −0.024 0.021 0.248 
Eosinophils (109 cells/L)    
 All subjects −0.023 0.030 0.437 
 Girls −0.031 0.045 0.490 
 Boys −0.028 0.040 0.486 
Basophils (109 cells/L)    
 All subjects −0.004 0.002 0.064 
 Girls −0.001 0.003 0.702 
 Boys −0.007 0.003 0.029 
Platelets (109 cells/L)    
 All subjects −8.190 6.519 0.210 
 Girls 0.995 8.731 0.909 
 Boys −16.260 9.542 0.090 
Red blood cells (1012 cells/L)    
 All subjects −0.089 0.050 0.077 
 Girls −0.134 0.077 0.084 
 Boys −0.065 0.065 0.322 
Hemoglobin (g/L)    
 All subjects −0.297 0.133 0.026 
 Girls −0.491 0.192 0.012 
 Boys −0.180 0.185 0.337 
Hematocrit (L/L)    
 All subjects −0.800 0.366 0.031 
 Girls −1.170 0.581 0.047 
 Boys −0.518 0.521 0.328 
Mean corpuscular volume (fL)    
 All subjects 0.042 0.374 0.911 
 Girls −0.089 0.544 0.870 
 Boys 0.330 0.500 0.510 
Red cell distribution width (%CV)    
 All subjects 0.109 0.075 0.146 
 Girls 0.094 0.121 0.440 
 Boys 0.130 0.095 0.171 
CBCEstimateSEP value
White blood cells (109 cells/L)    
 All subjects −0.315 0.157 0.046 
 Girls −0.025 0.212 0.906 
 Boys −0.540 0.229 0.019 
Neutrophils (109 cells/L)    
 All subjects −0.239 0.099 0.017 
 Girls −0.031 0.128 0.807 
 Boys −0.386 0.151 0.012 
Lymphocytes (109 cells/L)    
 All subjects −0.034 0.057 0.553 
 NLR (all) −0.115 0.067 0.090 
 Girls 0.048 0.082 0.555 
 Boys −0.084 0.080 0.295 
Monocytes (109 cells/L)    
 All subjects −0.024 0.015 0.108 
 Girls −0.024 0.022 0.281 
 Boys −0.024 0.021 0.248 
Eosinophils (109 cells/L)    
 All subjects −0.023 0.030 0.437 
 Girls −0.031 0.045 0.490 
 Boys −0.028 0.040 0.486 
Basophils (109 cells/L)    
 All subjects −0.004 0.002 0.064 
 Girls −0.001 0.003 0.702 
 Boys −0.007 0.003 0.029 
Platelets (109 cells/L)    
 All subjects −8.190 6.519 0.210 
 Girls 0.995 8.731 0.909 
 Boys −16.260 9.542 0.090 
Red blood cells (1012 cells/L)    
 All subjects −0.089 0.050 0.077 
 Girls −0.134 0.077 0.084 
 Boys −0.065 0.065 0.322 
Hemoglobin (g/L)    
 All subjects −0.297 0.133 0.026 
 Girls −0.491 0.192 0.012 
 Boys −0.180 0.185 0.337 
Hematocrit (L/L)    
 All subjects −0.800 0.366 0.031 
 Girls −1.170 0.581 0.047 
 Boys −0.518 0.521 0.328 
Mean corpuscular volume (fL)    
 All subjects 0.042 0.374 0.911 
 Girls −0.089 0.544 0.870 
 Boys 0.330 0.500 0.510 
Red cell distribution width (%CV)    
 All subjects 0.109 0.075 0.146 
 Girls 0.094 0.121 0.440 
 Boys 0.130 0.095 0.171 

Values in boldface type are statistically significant (P < 0.05). CV, coefficient of variation; NLR, neutrophil-to-lymphocyte ratio.

CBC in Relation to Sex in Children With and Without IAs

Stratified analyses by sex were performed to evaluate sex differences (Table 3). Reduction of white blood cell counts (P = 0.02) caused by a reduction of neutrophil (P = 0.012) and basophil (P = 0.029) counts showed a significant difference between children with and without IAs in boys, whereas a reduction of hemoglobin (P = 0.012) and hematocrit (P = 0.047) was found in girls with IA.

CBC in Relation to the Number of IAs and HLA Genotype

We investigated whether the number of IAs was associated with CBC by comparing the different cell counts (Table 4). The number of IAs was counted as the appearance of any of the following three IAs—IAA, GADA, or IA-2A—in the follow-up. Of the 72 children with IA, all 3 IAs appeared in 31 children, 2 IAs appeared in 16 children, and 1 IA appeared in 25 children. The number of IAs was inversely correlated with the number of white blood cells, neutrophils, and lymphocytes. Sex differences were also identified in this analysis. Children with three IAs and therefore at the greatest risk for T1D showed a reduction in white blood cell counts (P = 0.007), primarily in boys (P = 0.019) and in children with HLA-DR3-DQ2/DR4-DQ8 (P = 0.045). The reduction of white blood cells was mainly caused by the reduction in neutrophil counts (P = 0.003), primarily in boys (P = 0.004) and in children with HLA-DR3-DQ2/DR4-DQ8 (P = 0.010), but also the reduction of lymphocyte counts (P = 0.038) in all children.

Table 4

Association between the number of IA and each CBC measurement in 72 subjects (boys n = 42)

CBC n IAs EstimateSEP value
White blood cells (109 cells/L)     
 All subjects 0.142 0.257 0.582 
 −0.379 0.293 0.197 
 3 −0.613 0.225 0.007 
 Boys −0.001 0.392 0.998 
 −0.701 0.404 0.086 
 3 −0.789 0.332 0.019 
 HLA-DR3-DQ2/DR4-DQ8 0.512 0.401 0.203 
 −0.392 0.415 0.347 
 3 −0.729 0.360 0.045 
Neutrophils (109 cells/L)     
 All subjects 0.081 0.161 0.616 
 −0.328 0.178 0.068 
 3 −0.427 0.141 0.003 
 Boys 0.033 0.251 0.897 
 −0.479 0.259 0.068 
 3 −0.634 0.216 0.004 
 HLA-DR3-DQ2/DR4-DQ8 0.270 0.245 0.275 
 −0.420 0.246 0.091 
 3 −0.572 0.216 0.010 
Lymphocytes (109 cells/L)     
 All subjects 0.092 0.091 0.315 
 0.025 0.109 0.822 
 3 −0.175 0.0840 0.038 
 Girls 0.080 0.124 0.519 
 2 0.368 0.175 0.038 
 −0.115 0.118 0.332 
Monocytes (109 cells/L)     
 All subjects 0.003 0.0244 0.885 
 2 −0.057 0.0281 0.044 
 −0.026 0.0214 0.231 
Red blood cells (1012 cells/L)     
 All subjects 0.019 0.080 0.813 
 2 −0.200 0.089 0.026 
 −0.112 0.073 0.125 
 Girls 0.060 0.119 0.616 
 2 −0.471 0.142 0.002 
 −0.109 0.104 0.295 
 HLA-DR3-DQ2/DR4-DQ8 0.029 0.129 0.826 
 −0.211 0.136 0.125 
 3 −0.320 0.113 0.006 
Hemoglobin (g/L)     
 All subjects −0.031 0.210 0.882 
 2 −0.635 0.235 0.008 
 −0.282 0.193 0.147 
 Girls −0.038 0.300 0.900 
 2 −1.248 0.335 0.0007 
 −0.407 0.255 0.116 
 HLA DR3/4-DQ2/8 −0.255 0.372 0.495 
 −0.614 0.388 0.116 
 3 −0.710 0.326 0.031 
Hematocrit (L/L)     
 All subjects 0.147 0.577 0.799 
 2 −1.801 0.640 0.006 
 3 −1.034 0.525 0.050 
 Girls 0.395 0.907 0.664 
 2 −3.635 0.990 0.0007 
 −1.011 0.776 0.200 
 HLA-DR3-DQ2/DR4-DQ8 −0.375 0.996 0.707 
 −1.922 1.029 0.065 
 3 −2.514 0.853 0.004 
Mean corpuscular volume (fL)     
 All subjects 0.854 0.597 0.153 
 −0.481 0.734 0.513 
 −0.351 0.544 0.519 
 Boys 1 1.728 0.850 0.044 
 −0.505 0.900 0.576 
 −0.118 0.716 0.869 
 Not HLA-DR3-DQ2/DR4-DQ8 1 2.210 0.826 0.008 
 −0.605 1.246 0.628 
 −0.192 0.722 0.791 
MCH (pg)     
 HLA-DR3-DQ2/DR4-DQ8 1 0.861 0.418 0.042 
 0.034 0.433 0.938 
 0.562 0.354 0.116 
CBC n IAs EstimateSEP value
White blood cells (109 cells/L)     
 All subjects 0.142 0.257 0.582 
 −0.379 0.293 0.197 
 3 −0.613 0.225 0.007 
 Boys −0.001 0.392 0.998 
 −0.701 0.404 0.086 
 3 −0.789 0.332 0.019 
 HLA-DR3-DQ2/DR4-DQ8 0.512 0.401 0.203 
 −0.392 0.415 0.347 
 3 −0.729 0.360 0.045 
Neutrophils (109 cells/L)     
 All subjects 0.081 0.161 0.616 
 −0.328 0.178 0.068 
 3 −0.427 0.141 0.003 
 Boys 0.033 0.251 0.897 
 −0.479 0.259 0.068 
 3 −0.634 0.216 0.004 
 HLA-DR3-DQ2/DR4-DQ8 0.270 0.245 0.275 
 −0.420 0.246 0.091 
 3 −0.572 0.216 0.010 
Lymphocytes (109 cells/L)     
 All subjects 0.092 0.091 0.315 
 0.025 0.109 0.822 
 3 −0.175 0.0840 0.038 
 Girls 0.080 0.124 0.519 
 2 0.368 0.175 0.038 
 −0.115 0.118 0.332 
Monocytes (109 cells/L)     
 All subjects 0.003 0.0244 0.885 
 2 −0.057 0.0281 0.044 
 −0.026 0.0214 0.231 
Red blood cells (1012 cells/L)     
 All subjects 0.019 0.080 0.813 
 2 −0.200 0.089 0.026 
 −0.112 0.073 0.125 
 Girls 0.060 0.119 0.616 
 2 −0.471 0.142 0.002 
 −0.109 0.104 0.295 
 HLA-DR3-DQ2/DR4-DQ8 0.029 0.129 0.826 
 −0.211 0.136 0.125 
 3 −0.320 0.113 0.006 
Hemoglobin (g/L)     
 All subjects −0.031 0.210 0.882 
 2 −0.635 0.235 0.008 
 −0.282 0.193 0.147 
 Girls −0.038 0.300 0.900 
 2 −1.248 0.335 0.0007 
 −0.407 0.255 0.116 
 HLA DR3/4-DQ2/8 −0.255 0.372 0.495 
 −0.614 0.388 0.116 
 3 −0.710 0.326 0.031 
Hematocrit (L/L)     
 All subjects 0.147 0.577 0.799 
 2 −1.801 0.640 0.006 
 3 −1.034 0.525 0.050 
 Girls 0.395 0.907 0.664 
 2 −3.635 0.990 0.0007 
 −1.011 0.776 0.200 
 HLA-DR3-DQ2/DR4-DQ8 −0.375 0.996 0.707 
 −1.922 1.029 0.065 
 3 −2.514 0.853 0.004 
Mean corpuscular volume (fL)     
 All subjects 0.854 0.597 0.153 
 −0.481 0.734 0.513 
 −0.351 0.544 0.519 
 Boys 1 1.728 0.850 0.044 
 −0.505 0.900 0.576 
 −0.118 0.716 0.869 
 Not HLA-DR3-DQ2/DR4-DQ8 1 2.210 0.826 0.008 
 −0.605 1.246 0.628 
 −0.192 0.722 0.791 
MCH (pg)     
 HLA-DR3-DQ2/DR4-DQ8 1 0.861 0.418 0.042 
 0.034 0.433 0.938 
 0.562 0.354 0.116 

Values in boldface type are statistically significant (P < 0.05).

Red cell parameters were also associated with the number of IA, especially when all three IAs appeared. The red blood cell count was reduced in all children with two IAs (P = 0.026), particularly in girls (P = 0.002) and in children with the HLA-DR3/4-DQ2/8 genotype and three IAs (P = 0.006) (Table 4). These reductions in red blood cell counts were reflected in reduced concentrations of both hemoglobin and hematocrit in the presence of two or more IAs in girls and in children with the HLA-DR3-DQ2/DR4-DQ8 genotype (Table 4). Furthermore, in children with the HLA-DR3-DQ2/DR4-DQ8 genotype and one IA, the MCH was also decreased (P = 0.042). In contrast, boys with one positive IA had increased mean corpuscular volume levels (P = 0.044), particularly in children not carrying the HLA-DR3-DQ2/DR4-DQ8 genotype (P = 0.008).

The major findings of CBC in TEDDY children were lower numbers of white blood cells, primarily of neutrophils in boys and in children with HLA-DR3-DQ2/DR4-DQ8 with an increasing number of IAs. Also, an increasing number of IAs was related to lower numbers of hemoglobin and hematocrit in girls and in children with HLA-DR3-DQ2/DR4-DQ8. These findings may be related to an interaction between HLA risk and development of chronic islet autoimmunity. In TEDDY, chronic islet autoimmunity is defined as the length of being persistent confirmed IA positive after seroconversion and also in relation to the number of different IAs (30,31). The major finding that levels of neutrophils, primarily in boys, as well as in children with HLA-DR3-DQ2/DR4-DQ8, decreased with an increasing number of IAs is a novel finding, which to our knowledge has not been reported before. Our data are otherwise consistent with a reduction in peripheral blood neutrophils as recently reported in healthy IA-positive children all with a first degree relative with the disease (16,17). However, in contrast to one of these reports, a reduction in platelet counts in children positive for one or several IAs was not detected. The difference may be explained by the fact that our study was longitudinal, including a relatively larger number of children. Our results take these previous reports to a next step, indicating that the larger the number of IA, the lower the number of neutrophils. Reduction of red blood cells, hemoglobin, and hematocrit in girls positive for two or more IAs has not been reported previously.

The strength of the current study is that we have been able to determine CBC during almost 3 years in this TEDDY subset of 448 children who visit the Malmö clinic. They represent ∼15% of all children in TEDDY. During this period of investigation, children with IAs underwent CBC measurements more than three times and children with no IA one to two times because they only visit every 6 months. Another strength is that the CBC was done at random because our TEDDY laboratory could only perform the CBC in a maximum of eight children per day with blood samples collected in the morning. During the current period of investigating these 4- to 12-year-olds, we did not expect that we would come across a child converting to IA because of small numbers of IA-positive children monitored in this study. Although TEDDY aims to identify the environmental factors behind seroconversion, the strength of the current study was to contribute to the second end point in TEDDY, which is to identify factors that predispose to T1D, to determine the pathogenic mechanisms that eventually result in clinical onset of T1D.

The weakness was that none of the children changed IA status during the follow–up, and therefore, we cannot test whether the observed changes in CBC may precede seroconversion. Therefore, our data underscore the importance to investigate CBC when children at increased genetic risk for T1D are monitored from birth in the future. Another limitation is the number of IA-positive children monitored for CBC. Therefore, we continue the CBC follow-up in children with and without IAs to understand underlying mechanisms of altered CBC. Validation of these results outside of the TEDDY cohort may be accomplished in studies monitoring newborns, such as in the recently initiated Primary Oral Insulin Trial (POInT) study (32).

Pediatric blood reference intervals are mainly based on retrospective data from hospitalized individuals (28,33). However, the data in this study are prospective from healthy children with a genetic risk for T1D. This premise makes our data more reliable and useful when IA-positive children are compared with those negative for any IA. HLA risk eligibility for TEDDY in Sweden represented 7.5% of all newborns (20). Further CBC analyses are needed because there is an apparent lack of information of CBC neutrophil and red blood cell parameters in relation to HLA-DR-DQ risk not only for T1D but also celiac disease, thyroiditis, multiple sclerosis, and other HLA-associated organ-specific autoimmune disorders.

In agreement with numerous studies, white blood cell counts varied during the 1st years of life and decreased slightly thereafter. Moreover, the red blood cell count is known to increase by age in children, which was found also in this study (28,33). Furthermore, we examined the association between age and CBC in the IA-negative children considered as the general population with the above limitations. As revealed in Table 2, age-dependent effects were observed in most CBC parameters and later corrected for in the linear mixed-effects model. Due to this procedure, the observed associations between the increasing number of IAs and the reduction in neutrophils and red blood cell parameters were all corrected for age.

A diminished number of white blood cells was reported in several autoimmune diseases (17,18,34). Even though neutrophils are innate immune cells, they are involved in the activation and recruitment of both innate and adaptive immune cells (35,36). An impaired neutrophil response is thought to cause or initiate several autoimmune diseases, including systemic lupus erythematosus, vasculitis, and multiple sclerosis (35,37). Our results demonstrate a reduction in neutrophils in boys and in HLA-DR3-DQ2/DR4-DQ8 children positive for three IAs. Recent studies have also reported a reduction in circulating neutrophils in patients newly diagnosed with T1D (who did not have diabetic ketoacidosis at onset) and in IA-positive healthy individuals, suggesting β‐cell specific autoimmunity (16,17,38). Neutrophils are thought to be recruited to infiltrate pancreatic islets by the physiological death of β-cells and by a signaling cross talk with other innate immune cells activating the autoreactive T cells (36). Recent studies have also suggested that the reduction of neutrophil numbers may be related to an accumulation of neutrophils in the pancreas (16,17), perhaps associated with insulitis (39). These data would be consistent with the observations that the larger the number of IAs, the higher the risk for insulitis (40,41). However, further studies need to dissect the neutrophil count reduction in relation to pathophysiological changes that occur in the pancreas before the clinical onset of T1D.

Mild neutropenia is a common finding in children with viral infections (42). The reduction in the neutrophil counts in healthy Swedish TEDDY boys and children with the HLA-DR3-DQ2/DR4-DQ8 genotype could be due to an impaired hematopoietic cell production in the bone marrow, impaired maturation, apoptosis, peripheral consumption or damage, pooling to other organs such as the pancreas, and perhaps tissue detention (43,44). The reduction in neutrophil counts associated with a presence of three IAs may contribute to an accelerated pathogenesis by potentiating the autoimmune attack on β-cells, by increasing the risk for infection, or by some other processes. However, it can also be a secondary phenomenon due to detention of neutrophils in the pancreas when the β-cell destruction is more pronounced. Alterations in the CBC among autoantibody-positive children may be caused by an impaired hematopoiesis caused by infections or toxins (42,45).

Under normal circumstances, no sex differences in neutrophil, lymphocyte, or basophil counts at the age of 4–12 years have been reported (28). Boys in this study had reduced basophil and neutrophil counts and increased MCHC counts associated with the status of being positive for three IAs. The MCHC count is normally decreasing by age (28). Viral and bacterial infections may alter neutrophil, basophil, and lymphocyte counts; for instance, whooping cough is known to decrease basophil numbers. Basophils have been shown to migrate to secondary lymphoid organs where they cross talk with T- and B-lymphocytes thereby linking the T-helper cell environment as a contributor to the development of autoimmune systemic lupus erythematosus, and as a consequence, the basophils in the circulation decrease (24). Girls with two IAs had increased levels of lymphocytes which could be associated with viral infections because many viruses, such as rotavirus, mumps virus, and others, have been associated with the pathogenesis of T1D (46,47).

The reduction of red blood cells, hemoglobin and hematocrit associated with two IAs in girls may be explained by a peripheral destruction of red blood cells, impaired hematopoiesis, or infections. Previous studies have associated rotavirus and enterovirus with islet β-cell autoimmunity and T1D (48,49). Hence, whether a reduction in CBC is caused by or contributes to the development of the second or third IA is unclear. Our results in TEDDY girls suggest that an alteration of CBC was observed with a second IA.

We concluded from the current study that deviations in CBC after seroconversion were common primarily in boys and in children with the HLA-DR3-DQ2/DR4-DQ8 genotype. Our statistical models indicate that the reduction of neutrophil counts in boys and HLA-DR3-DQ2/DR4-DQ8 children positive for three autoantibodies may be consequences of islet β-cell autoimmunity, because the reduction in neutrophil levels correlated with the number of IAs. Additional factors influencing the CBC in children at genetic risk for T1D and positive for one to three IAs were also affected by sex, HLA genotype, and number of IAs. The cellular and molecular mechanisms behind the aggravating CBC alterations different in boys or girls with an increasing number of IAs need to be further explored.

Acknowledgments. The authors acknowledge the participation of the TEDDY children and the work of the TEDDY study group (see Supplementary Data).

Funding. The TEDDY study is funded by the National Institute of Diabetes and Digestive and Kidney Diseases (U01-DK-63829, U01-DK-63861, U01-DK-63821, U01-DK-63865, U01-DK-63863, U01-DK-63836, U01-DK-63790, UC4-DK-63829, UC4-DK63861, UC4-DK-63821, UC4-DK-63865, UC4-DK-63863, UC4-DK-63836, UC4-DK-95300, UC4-DK-100238, UC4-DK-106955, UC4-DK-112243, UC4-DK-117483, and contract no. HHSN267200700014C), National Institute of Allergy and Infectious Diseases, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institute of Environmental Health Sciences, JDRF, Centers for Disease Control and Prevention, and National Institutes of Health Clinical Center. This work was supported in part by the National Institutes of Health/National Center for Advancing Translational Sciences Clinical and Translational Science Awards to the University of Florida (UL1-TR-000064) and the University of Colorado (UL1-TR-001082). The funders had no impact on the design, implementation, analysis, and interpretation of the data.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. F.S. performed CBC analysis, interpreted data, and wrote the manuscript. H.-S.L. and E.F. performed statistical analyses and reviewed and edited the manuscript. H.E.L. and C.T. reviewed and edited the manuscript. Å.L. conceived the study, contributed to study design, and reviewed and edited the manuscript. Å.L. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

1.
Krischer
JP
,
Lynch
KF
,
Schatz
DA
, et al.;
TEDDY Study Group
.
The 6 year incidence of diabetes-associated autoantibodies in genetically at-risk children: the TEDDY study
.
Diabetologia
2015
;
58
:
980
987
[PubMed]
2.
Krischer
JP
,
Lynch
KF
,
Lernmark
Å
, et al.;
TEDDY Study Group
.
Genetic and environmental interactions modify the risk of diabetes-related autoimmunity by 6 years of age: The TEDDY Study
.
Diabetes Care
2017
;
40
:
1194
1202
[PubMed]
3.
Krischer
JP
,
Liu
X
,
Lernmark
Å
, et al.;
TEDDY Study Group
.
The influence of type 1 diabetes genetic susceptibility regions, age, sex, and family history on the progression from multiple autoantibodies to type 1 diabetes: a TEDDY Study report
.
Diabetes
2017
;
66
:
3122
3129
[PubMed]
4.
Elding Larsson
H
,
Vehik
K
,
Gesualdo
P
, et al.;
TEDDY Study Group
.
Children followed in the TEDDY study are diagnosed with type 1 diabetes at an early stage of disease
.
Pediatr Diabetes
2014
;
15
:
118
126
[PubMed]
5.
Vehik
K
,
Lynch
KF
,
Schatz
DA
, et al.;
TEDDY Study Group
.
Reversion of β-cell autoimmunity changes risk of type 1 diabetes: TEDDY Study
.
Diabetes Care
2016
;
39
:
1535
1542
[PubMed]
6.
Bulek
AM
,
Cole
DK
,
Skowera
A
, et al
.
Structural basis for the killing of human beta cells by CD8(+) T cells in type 1 diabetes
.
Nat Immunol
2012
;
13
:
283
289
[PubMed]
7.
Coppieters
KT
,
Dotta
F
,
Amirian
N
, et al
.
Demonstration of islet-autoreactive CD8 T cells in insulitic lesions from recent onset and long-term type 1 diabetes patients
.
J Exp Med
2012
;
209
:
51
60
[PubMed]
8.
Ziegler
AG
,
Rewers
M
,
Simell
O
, et al
.
Seroconversion to multiple islet autoantibodies and risk of progression to diabetes in children
.
JAMA
2013
;
309
:
2473
2479
[PubMed]
9.
Katsarou
A
,
Gudbjörnsdottir
S
,
Rawshani
A
, et al
.
Type 1 diabetes mellitus
.
Nat Rev Dis Primers
2017
;
3
:
17016
[PubMed]
10.
Rewers
M
,
Ludvigsson
J
.
Environmental risk factors for type 1 diabetes
.
Lancet
2016
;
387
:
2340
2348
[PubMed]
11.
Gomez-Tourino
I
,
Arif
S
,
Eichmann
M
,
Peakman
M
.
T cells in type 1 diabetes: instructors, regulators and effectors: a comprehensive review
.
J Autoimmun
2016
;
66
:
7
16
[PubMed]
12.
Price
JD
,
Tarbell
KV
.
The role of dendritic cell subsets and innate immunity in the pathogenesis of type 1 diabetes and other autoimmune diseases
.
Front Immunol
2015
;
6
:
288
[PubMed]
13.
Dotta
F
,
Fondelli
C
,
Falorni
A
.
Can NK cells be a therapeutic target in human type 1 diabetes
?
Eur J Immunol
2008
;
38
:
2961
2963
[PubMed]
14.
Wentworth
JM
,
Fourlanos
S
,
Harrison
LC
.
Reappraising the stereotypes of diabetes in the modern diabetogenic environment
.
Nat Rev Endocrinol
2009
;
5
:
483
489
[PubMed]
15.
Battaglia
M
.
Neutrophils and type 1 autoimmune diabetes
.
Curr Opin Hematol
2014
;
21
:
8
15
[PubMed]
16.
Valle
A
,
Giamporcaro
GM
,
Scavini
M
, et al
.
Reduction of circulating neutrophils precedes and accompanies type 1 diabetes
.
Diabetes
2013
;
62
:
2072
2077
[PubMed]
17.
Harsunen
MH
,
Puff
R
,
D’Orlando
O
, et al
.
Reduced blood leukocyte and neutrophil numbers in the pathogenesis of type 1 diabetes
.
Horm Metab Res
2013
;
45
:
467
470
[PubMed]
18.
Qin
J
,
Fu
S
,
Speake
C
,
Greenbaum
CJ
,
Odegard
JM
.
NETosis-associated serum biomarkers are reduced in type 1 diabetes in association with neutrophil count
.
Clin Exp Immunol
2016
;
184
:
318
322
[PubMed]
19.
TEDDY Study Group
.
The Environmental Determinants of Diabetes in the Young (TEDDY) study: study design
.
Pediatr Diabetes
2007
;
8
:
286
298
[PubMed]
20.
Hagopian
WA
,
Erlich
H
,
Lernmark
A
, et al.;
TEDDY Study Group
.
The Environmental Determinants of Diabetes in the Young (TEDDY): genetic criteria and international diabetes risk screening of 421 000 infants
.
Pediatr Diabetes
2011
;
12
:
733
743
[PubMed]
21.
TEDDY Study Group
.
The Environmental Determinants of Diabetes in the Young (TEDDY) Study
.
Ann N Y Acad Sci
2008
;
1150
:
1
13
[PubMed]
22.
Lehto
T
,
Hedberg
P
.
Performance evaluation of Abbott CELL-DYN Ruby for routine use
.
Int J Lab Hematol
2008
;
30
:
400
407
[PubMed]
23.
Dantonio
P
,
Meredith-Molloy
N
,
Hagopian
WA
, et al
.
Proficiency testing of human leukocyte antigen-DR and human leukocyte antigen-DQ genetic risk assessment for type 1 diabetes using dried blood spots
.
J Diabetes Sci Technol
2010
;
4
:
929
941
[PubMed]
24.
Charles
N
,
Rivera
J
.
Basophils and autoreactive IgE in the pathogenesis of systemic lupus erythematosus
.
Curr Allergy Asthma Rep
2011
;
11
:
378
387
[PubMed]
25.
Bonifacio
E
,
Yu
L
,
Williams
AK
, et al
.
Harmonization of glutamic acid decarboxylase and islet antigen-2 autoantibody assays for national institute of diabetes and digestive and kidney diseases consortia
.
J Clin Endocrinol Metab
2010
;
95
:
3360
3367
[PubMed]
26.
Babaya
N
,
Yu
L
,
Miao
D
, et al
.
Comparison of insulin autoantibody: polyethylene glycol and micro-IAA 1-day and 7-day assays
.
Diabetes Metab Res Rev
2009
;
25
:
665
670
[PubMed]
27.
Törn
C
,
Mueller
PW
,
Schlosser
M
,
Bonifacio
E
,
Bingley
PJ
;
Participating Laboratories
.
Diabetes Antibody Standardization Program: evaluation of assays for autoantibodies to glutamic acid decarboxylase and islet antigen-2
.
Diabetologia
2008
;
51
:
846
852
[PubMed]
28.
Aldrimer
M
,
Ridefelt
P
,
Rödöö
P
,
Niklasson
F
,
Gustafsson
J
,
Hellberg
D
.
Population-based pediatric reference intervals for hematology, iron and transferrin
.
Scand J Clin Lab Invest
2013
;
73
:
253
261
[PubMed]
29.
Taylor
MR
,
Holland
CV
,
Spencer
R
,
Jackson
JF
,
O’Connor
GI
,
O’Donnell
JR
.
Haematological reference ranges for schoolchildren
.
Clin Lab Haematol
1997
;
19
:
1
15
[PubMed]
30.
Steck
AK
,
Vehik
K
,
Bonifacio
E
, et al.;
TEDDY Study Group
.
Predictors of progression from the appearance of islet autoantibodies to early childhood diabetes: The Environmental Determinants of Diabetes in the Young (TEDDY)
.
Diabetes Care
2015
;
38
:
808
813
[PubMed]
31.
Köhler
M
,
Beyerlein
A
,
Vehik
K
, et al.;
TEDDY study group
.
Joint modeling of longitudinal autoantibody patterns and progression to type 1 diabetes: results from the TEDDY study
.
Acta Diabetol
2017
;
54
:
1009
1017
[PubMed]
32.
Ziegler
AG
,
Danne
T
,
Dunger
DB
, et al
.
Primary prevention of beta-cell autoimmunity and type 1 diabetes - The Global Platform for the Prevention of Autoimmune Diabetes (GPPAD) perspectives
.
Mol Metab
2016
;
5
:
255
262
[PubMed]
33.
Mcpherson
RA
,
Pincus
MR
.
Henry’s Clinical Diagnosis and Management by Laboratory Methods
. 22nd ed.
Philadelphia
,
Elsevier Science Health Science
,
2011
34.
Starkebaum
G
.
Chronic neutropenia associated with autoimmune disease
.
Semin Hematol
2002
;
39
:
121
127
[PubMed]
35.
Amulic
B
,
Cazalet
C
,
Hayes
GL
,
Metzler
KD
,
Zychlinsky
A
.
Neutrophil function: from mechanisms to disease
.
Annu Rev Immunol
2012
;
30
:
459
489
[PubMed]
36.
Diana
J
,
Simoni
Y
,
Furio
L
, et al
.
Crosstalk between neutrophils, B-1a cells and plasmacytoid dendritic cells initiates autoimmune diabetes
.
Nat Med
2013
;
19
:
65
73
[PubMed]
37.
Casserly
CS
,
Nantes
JC
,
Whittaker Hawkins
RF
,
Vallières
L
.
Neutrophil perversion in demyelinating autoimmune diseases: mechanisms to medicine
.
Autoimmun Rev
2017
;
16
:
294
307
[PubMed]
38.
Wang
Y
,
Xiao
Y
,
Zhong
L
, et al
.
Increased neutrophil elastase and proteinase 3 and augmented NETosis are closely associated with β-cell autoimmunity in patients with type 1 diabetes
.
Diabetes
2014
;
63
:
4239
4248
[PubMed]
39.
Lundberg
M
,
Seiron
P
,
Ingvast
S
,
Korsgren
O
,
Skog
O
.
Insulitis in human diabetes: a histological evaluation of donor pancreases
.
Diabetologia
2017
;
60
:
346
353
[PubMed]
40.
In’t Veld
P
,
Lievens
D
,
De Grijse
J
, et al
.
Screening for insulitis in adult autoantibody-positive organ donors
.
Diabetes
2007
;
56
:
2400
2404
[PubMed]
41.
Atkinson
MA
,
Gianani
R
.
The pancreas in human type 1 diabetes: providing new answers to age-old questions
.
Curr Opin Endocrinol Diabetes Obes
2009
;
16
:
279
285
[PubMed]
42.
Dale
DC
.
How I manage children with neutropenia
.
Br J Haematol
2017
;
178
:
351
363
[PubMed]
43.
Huang
J
,
Xiao
Y
,
Xu
A
,
Zhou
Z
.
Neutrophils in type 1 diabetes
.
J Diabetes Investig
2016
;
7
:
652
663
[PubMed]
44.
Kaplan
MJ
.
Role of neutrophils in systemic autoimmune diseases
.
Arthritis Res Ther
2013
;
15
:
219
[PubMed]
45.
Kolb-Mäurer
A
,
Goebel
W
.
Susceptibility of hematopoietic stem cells to pathogens: role in virus/bacteria tropism and pathogenesis
.
FEMS Microbiol Lett
2003
;
226
:
203
207
[PubMed]
46.
Op de Beeck
A
,
Eizirik
DL
.
Viral infections in type 1 diabetes mellitus--why the β cells
?
Nat Rev Endocrinol
2016
;
12
:
263
273
[PubMed]
47.
Coppieters
KT
,
Boettler
T
,
von Herrath
M
.
Virus infections in type 1 diabetes
.
Cold Spring Harb Perspect Med
2012
;
2
:
a007682
[PubMed]
48.
Honeyman
MC
,
Coulson
BS
,
Stone
NL
, et al
.
Association between rotavirus infection and pancreatic islet autoimmunity in children at risk of developing type 1 diabetes
.
Diabetes
2000
;
49
:
1319
1324
[PubMed]
49.
Honeyman
MC
,
Stone
NL
,
Falk
BA
,
Nepom
G
,
Harrison
LC
.
Evidence for molecular mimicry between human T cell epitopes in rotavirus and pancreatic islet autoantigens
.
J Immunol
2010
;
184
:
2204
2210
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.

Supplementary data