Table 1.
Demographic, metabolic, immunological and genetic markers in children with familial and sporadic type 1 diabetes
| Variable | n | Familial, n=519 (10.4%) | Sporadic, n=4474 (89.6%) | p value | Adjusted p valuea |
|---|---|---|---|---|---|
| Demographics | |||||
| Age at diagnosis, years, mean ± SD | 4993 | 7.84 ± 3.88 | 8.05 ± 3.89 | 0.229 | |
| Sex, male, % (95% CI) | 4993 | 55.7 (51.4, 60.0) | 56.7 (55.2, 58.1) | 0.705 | |
| Pubertal, % (95% CI) | 3764 | 16.9 (13.2, 20.6) | 17.2 (15.9, 18.4) | 0.941 | 0.445 |
| Metabolic decompensation at diagnosis | |||||
| Duration of symptoms, % | 4614 | <0.001 | <0.001 | ||
| No symptoms | 2.1 | 0.8 | |||
| < 1 week | 38.1 | 20.8 | |||
| 1–4 weeks | 46.6 | 58.7 | |||
| > 4 weeks | 13.1 | 19.7 | |||
| Impaired consciousness, % (95% CI) | 4784 | 1.6 (0.5, 2.7) | 5.9 (5.2, 6.6) | <0.001 | <0.001 |
| Ketoacidosis, % (95% CI) | 4817 | 7.8 (5.4, 10.1) | 19.1 (18.0, 20.3) | <0.001 | <0.001 |
| Severe ketoacidosis, % (95% CI) | 4817 | 2.4 (1.1, 3.7) | 4.9 (4.2, 5.5) | 0.017 | 0.018 |
| Weight loss, %, median (range) | 4610 | 2.0 (0–25.3) | 5.6 (0–40.0) | <0.001 | <0.001 |
| pH, median (range) | 4817 | 7.40 (6.93–7.57) | 7.38 (6.72–7.54) | <0.001 | <0.001 |
| β-Hydroxybutyrate, mmol/l, median (range) | 4384 | 0.50 (0–18.0) | 1.90 (0–27.0) | <0.001 | <0.001 |
| Plasma glucose, mmol/l, median (range) | 4869 | 20.8 (3.6–63.7) | 24.2 (3.2–97.6) | <0.001 | <0.001 |
| HbA1c, mmol/mol, median (range) | 841 | 76.0 (38.0–141.5) | 92.0 (36.0–189.0) | <0.001 | <0.001 |
| HbA1c, %, median (range) | 841 | 9.1 (5.6–15.1) | 10.6 (5.4–19.4) | <0.001 | <0.001 |
| Autoantibodies | |||||
| ICA, % (95% CI) | 4738 | 90.9 (88.3, 93.5) | 91.8 (91.0, 92.7) | 0.535 | 0.400 |
| ICA, JDFU, median (range) | 4347 | 64.0 (3.0–2620.0) | 64.0 (3.0–5120.0) | 0.894 | 0.585 |
| IAA, % (95% CI) | 4738 | 49.8 (45.3, 54.2) | 42.2 (40.7, 43.7) | 0.002 | 0.004 |
| IAA, RU, median (range) | 2037 | 10.5 (2.9–484.9) | 10.2 (2.8–7809.0) | 0.771 | 0.329 |
| IA-2A, % (95% CI) | 4738 | 75.4 (71.6, 79.2) | 75.0 (73.7, 76.3) | 0.890 | 0.806 |
| IA-2A, RU, median (range) | 3556 | 104.2 (0.8–223.2) | 105.8 (0.8–553.3) | 0.768 | 0.811 |
| GADA, % (95% CI) | 4738 | 67.1 (63.0, 71.3) | 66.3 (64.8, 67.7) | 0.735 | 0.679 |
| GADA, RU, median (range) | 3144 | 43.3 (5.4–3800.0) | 35.7 (5.4–24,849.0) | 0.245 | 0.080 |
| ZnT8A, % (95% CI) | 4738 | 66.3 (62.1, 70.5) | 69.8 (68.4, 71.1) | 0.132 | 0.155 |
| ZnT8A, RU, median (range) | 3289 | 12.0 (0.5–186.7) | 12.1 (0.5–1201.9) | 0.753 | 0.767 |
| Number of positive antibodies, median (mean) | 4738 | 4 (3.50) | 4 (3.45) | 0.252 | 0.183 |
| Number of positive biochemical antibodies, median (mean) | 4738 | 3 (2.59) | 3 (2.53) | 0.182 | 0.211 |
| Autoantibody negative, % (95% CI) | 4738 | 3.5 (1.9, 5.2) | 2.1 (1.7, 2.6) | 0.079 | 0.041 |
| Positivity for multiple (≥2) autoantibodies, % (95% CI) | 4738 | 92.8 (90.5, 95.1) | 92.5 (91.7, 93.2) | 0.875 | 0.911 |
| Genetics | 4993 | ||||
| DR3-DQ2/DR4-DQ8, % (95% CI) | 23.7 (20.0, 27.4) | 21.1 (19.9, 22.2) | 0.182 | 0.186 | |
| DR3-DQ2/xb, % (95% CI) | 13.1 (10.2, 16.0) | 15.6 (14.5, 16.6) | 0.156 | 0.143 | |
| DR4-DQ8/yc, % (95% CI) | 52.0 (47.7, 56.3) | 47.5 (46.0, 48.9) | 0.055 | 0.049 | |
| xb/yc, % (95% CI) | 11.2 (8.5, 13.9) | 15.9 (14.8, 17.0) | 0.006 | 0.006 | |
| DR3-DQ2, % (95% CI) | 36.8 (32.7, 41.0) | 36.6 (35.2, 38.0) | 0.979 | 0.975 | |
| DR4-DQ8, % (95% CI) | 75.7 (72.0, 79.4) | 68.5 (67.1, 69.9) | 0.001 | 0.001 | |
| Risk group, % | 0.157 | 0.162 | |||
| 0 | 0.4 | 0.8 | |||
| 1 | 1.7 | 2.1 | |||
| 2 | 13.5 | 16.2 | |||
| 3 | 20.4 | 23.0 | |||
| 4 | 40.3 | 36.8 | |||
| 5 | 23.7 | 21.1 | |||
Categorical variables are presented as % (95% CI) and continuous variables as mean (SD) or median (range)
For comparing frequencies, cross tabulation and χ2 statistics with continuity correction or Fisher’s exact test when appropriate were used. Differences in levels of parametric variables were analysed with Student’s t test, and Kruskal–Wallis test or Mann–Whitney U test for non-parametric variables
Adjustment for confounding factors was performed using logistic/ordinal/multinomial regression for dichotomous/ordinal/categorical variables and quantile regression in R for skewed variables
aAdjusted for sex and age at diagnosis
bx ≠ DR4-DQ8
cy ≠ DR3-DQ2
JDFU, Juvenile Diabetes Foundation units; RU, relative units