Approximately 10%–15% of patients with type 1 diabetes (T1D) have affected first-degree relatives (1). The offspring of T1D mothers have a 2%–3% risk, whereas offspring of affected fathers have a 7% risk (2). The sibling of a child with T1D has a 3%–5% risk of developing T1D.
Few studies have evaluated glycemic control in familial and sporadic T1D patients in relation to demographic and familial characteristics (1, 3). O’Leary et al. (3) compared familial (parent and/or sibling) vs. sporadic T1D cases at clinical presentation and at 1-year post-diagnosis. Familial cases on average had less severe clinical presentation with lower mean HbA1c and lower insulin dose requirements at discharge. However, at 1-year follow-up, there was no difference in HbA1c and insulin doses in the two groups (3). Dahlquist and Mustonen (4) studied familial patients with and without a first-degree relative at clinical presentation and found similar characteristics for age of onset, sex ratio and seasonality. Unlike the O’Leary et al. (3) study, there was no follow-up of these patients post-diagnosis.
The study of Lebenthal et al. (1) on familial T1D showed that diabetic ketoacidosis (DKA) rate and HbA1c levels were lower in second affected family members in both parent-offspring and sib-pair groups at diagnosis. However, there was no follow-up glycemic control data post-diagnosis.
Other studies (5, 6) focused on parental knowledge and stressors that affect glycemic control in children with T1D. None of the previous studies evaluated the correlation of parents with T1D and their glycemic control in relation to the glycemic control of their affected offspring.
The primary purpose of our study was to (a) compare glycemic control in familial vs. sporadic patients with T1D at clinical presentation and at 1-year intervals over 5 years and (b) examine whether offspring’s HbA1c levels correlate with parental HbA1c levels over 5 years.
We hypothesized that:
Familial patients would present with less severe clinical manifestations at onset of T1D when compared to sporadic patients.
Familial patients would have better glycemic control, less severe hypoglycemia and lower readmission rates for ketosis and DKA at 5-year follow-up.
Glycemic control (HbA1c) of the second affected sibling would correlate with the glycemic control of the first affected sibling over 5 years post-diagnosis.
Glycemic control (HbA1c) of children with familial T1D would correlate with the glycemic control of their affected parent 5 years post-diagnosis.
We performed a retrospective chart review of the patients seen in our diabetes program (700 patients) at Hasbro Children’s Hospital. Our inclusion criteria were T1D patients ages 6 months to 18 years diagnosed between 1987 and 2010 who had either a biological parent or a biological sibling with T1D (familial cases) at Hasbro Children’s Hospital. Exclusion criteria include patients with chronic steroid use (prednisone), eating disorders and inflammatory bowel disease (Crohn’s disease or ulcerative colitis). Data were collected from charts and electronic medical records. Our cohort consisted of 33 parent-offspring and 19 sib-pairs. We matched one index offspring (familial) with one control (sporadic) patient by age at diagnosis, sex, ethnicity, insulin regimen and pubertal status (prepuberty: Tanner I; early: Tanner II–III; late: Tanner IV–V).
For familial and sporadic patients at clinical presentation with T1D, we collected data on age, sex, height, weight, body mass index (BMI), medical insurance, pH, hemoglobin A1c (HbA1c), glucose and HCO3 values. The primary outcome of interest is HbA1c. HbA1c values were measured using the Tosoh Automated Glycohemoglobin Analyzer HLC-723G8 (Tosoh Bioscience Inc, South San Francisco, CA, USA) which uses non-porous ion-exchange high-performance liquid chromatography. HbA1c values were measured quarterly in the outpatient clinic using a Siemens/Bayer DCA Analyzer WS-DCA2000+ (Bayer, Inc, Tarrytown, NY, USA). Follow-up HbA1c values were calculated at years 1 (HbA1c at diagnosis was excluded), 2, 3, 4 and 5 post-diagnosis by averaging values recorded at the quarterly clinic visits.
For familial and sporadic patients, we also collected number of emergency room (ER) visits, routine clinic visits and admissions for hypoglycemia (lethargy, seizures and requiring glucagon) or DKA (pH <7.3 or HCO3 <15) episodes over the 5-year period through chart review at Hasbro Children’s Hospital.
In sib-pairs, we correlated HbA1c at diagnosis and 5 years post-diagnosis. It is the correlation of the mean HbA1c every year over 5 years.
In parent-offspring group, we correlated offspring HbA1c and parent HbA1c at time of child’s diagnosis and 5 years thereafter.
In quantifying the relationships of patient characteristics between familial and sporadic patients, and sib-pairs, we used the paired t-test (pairing on variable-matched pairs and sib-pairs, respectively), the Pearson correlation and the χ2-test. All t-test results were checked for distribution influences with the non-parametric Wilcoxon signed-rank test. We used an alpha probability of 0.05 as the threshold for statistical significance in two-tailed comparisons. Means are presented with standard deviations throughout. Statistical analyses were performed with Stata v. 10 (Stata Corp., College Station, TX, USA).
Clinical and metabolic characteristics at presentation and 5-year follow-up in familial vs. sporadic T1D patients
Our cohort included 33 familial (parent-offspring group) vs. 33 sporadic patients. Male preponderance was seen in our cohort, and most children were pre-pubertal. The majority were Caucasians. The height, weight and BMI standard deviation score were similar in both groups (Table 1).
|Sex (M/F), %||75/25||75/25|
|Race (C/non-C), %||94/6||94/6|
|Puberty stage (pre/early/late)||24/3/6||24/3/6|
|IR (intensive/conventional), %||25/75||25/75|
|Insurance (private/Medicaid), %||70/30||76/24||0.34|
The metabolic parameters (HbA1c, HCO3, pH and glucose) and DKA rates were not significantly different for the two subgroups at presentation (Table 2).
|Mean HbA1c, %||9.6±1.5||10.7±2.5||0.17|
|Mean glucose, mg/dL||428±140||463±181||0.44|
|Mean HCO3, meq/L||21±7.5||18±7.0||0.25|
There was no difference in glycemic control (mean HbA1c) (Table 3) or number of clinic visits (12 vs. 12.5, p=0.68) (Table 4) between both subgroups over a 5-year period. The number of ER visits was not significantly different between the two groups (0.48 vs. 0.24, p=0.10).
|Years post-diagnosis||Familial||Sporadic||na||Familial/sporadic, na||p-Value|
|Years post-diagnosis||Familial||Sporadic||Familial/sporadic, n||p-Value|
Clinical and metabolic characteristics at presentation and at 5-year follow-up in sib-pairs
Our cohort contained 19 sib-pairs. The mean glucose at diagnosis was not significantly different between the sib-pair groups; however, there was a significant difference in HCO3 values (p<0.01) (Table 5). None of the second affected siblings presented with DKA. Mean HbA1c for both SP1 and SP2 over a 5-year follow-up was 9.2% (Table 6). There was a statistically significant difference between HbA1c values only at year 4. The mean HbA1c values for SP2 (n=17) correlated positively with that of SP1 (n=18) when HbA1c values were averaged for each patient yearly over 5 years (r=0.67, p<0.01) (Figure 1). Two sib-pairs were excluded because of unavailable data.
|Gender (M/F), %||53/47||63/37|
|Mean HbA1c, %||N/A||9.7±3.4|
|Mean glucose, mg/dL||433±464||347±122||0.489|
|Mean HCO3, meq/L||17.9±7.3||24.1±3.3||0.004|
Clinical and metabolic characteristics at presentation in parent-offspring pairs
Our cohort had seven parent-offspring pairs. Twenty-six pairs were excluded since we could not access parent’s charts from the time of their child’s diagnosis. Over 5 years, the mean HbA1c values of offspring did not differ significantly from parents (parent’s vs. offspring’s HbA1c: 8.6% vs. 9.3%, p=0.34). HbA1c values for offspring correlated positively with parents but were not significant (r=0.57, p=0.18) (Figure 2).
Our study showed that (a) there was no difference in glycemic control at clinical presentation and over 5 years in familial compared to sporadic patients; (b) in sib-pairs, the second affected sibling had lower DKA rates than the first affected sibling at onset; (c) HbA1c was correlated in sib-pairs; and (d) there existed a trend in the correlation of glycemic control between parents and offspring.
Our finding of similar metabolic control at diagnosis between familial and sporadic T1D patients differs from previous findings (1, 3). Lebenthal et al. (1) found lower DKA rates (23% vs. 38% p=0.006) and HbA1c (11.1% vs. 11.8% p=0.025) between familial and sporadic patients, and O’Leary et al. (3) showed lower mean HbA1c and higher HCO3 in familial patients. We found higher HCO3 values in familial patients, and lower HbA1c values, although not statistically significant, which could be due to our small sample size.
Our study had a longer follow-up of glycemic control compared to that of O’Leary et al. (3) (5 years vs. 1 year). However, consistent with their findings from 1 year post-diagnosis, glycemic control was similar between familial vs. sporadic patients. In our study, as in the Lebenthal et al. study (1), the second affected sibling exhibited lower DKA rates at presentation compared to the first sibling. This could be due to increased parental awareness of initial diabetes symptoms from experience related to their first affected child.
In our small sample of parent-offspring pairs with T1D, we found a positive, but not significant, correlation between their HbA1c values.
Limitations of our study include the following: (a) small sample size, (b) the change in insulin regimen over 5 years was not accounted for when analyzing glycemic control results, and (c) parental diabetes knowledge was not assessed directly in either group, which may explain the differences in glycemic control.
The results from this retrospective study indicate that metabolic control at diagnosis of type 1 diabetes, and over 5 years, is similar between familial and sporadic patients. In sib-pairs, the second affected sibling had milder clinical presentation compared to the first affected sibling. Glycemic control correlated well between sib-pairs. We also found a trend in the correlation of glycemic control between parents and offspring. Therefore, based on our results, parental experience from their own diabetes did not always positively influence glycemic control in their children. However, the experience related to their first affected child had a significant impact with early recognition of symptoms resulting in decreased DKA rates in the second affected child at onset.
Based on our results, we recommend that diabetes education be uniformly provided regardless of whether parents have diabetes themselves or have another child with diabetes. However, with these families, it may be helpful to inquire about the challenges they have faced in managing their own or their child’s diabetes. Indeed, this may provide a built-in opportunity to help improve future glycemic control for all affected family members. Diabetes remains a challenging disease to manage, and reinforcing parental knowledge periodically can only help all involved. Future studies should investigate parental diabetes knowledge and stress-related factors in both familial and sporadic groups.
Lebenthal Y, de Vries L, Phillip M, Lazar L. Familial type 1 diabetes mellitus – gender distribution and age at onset of diabetes distinguish between parent-offspring and sib-pair subgroups. Pediatr Diabetes 2010;11:403–11.[Web of Science]
Maahs DM, West NA, Lawrence JM, Mayer-Davis EJ. Epidemiology of type 1 diabetes. Endocrinol Metab Clin North Am 2010;39:481–97.
O’Leary LA, Dorman JS, LaPorte RE, Orchard TJ, Becker DJ, et al. Familial and sporadic insulin-dependent diabetes: evidence for heterogeneous etiologies? Diabetes Res Clin Pract 1991;14:183–90.[CrossRef]
Dahlquist GG, Mustonen LR. Clinical onset characteristics of familial versus nonfamilial cases in a large population-based cohort of childhood-onset diabetes patients. Diabetes Care 1995;18:852–4.[CrossRef]
Jacobson AM, Hauser ST, Lavori P, Willett JB, Cole CF, et al. Family environment and glycemic control: a four-year prospective study of children and adolescents with insulin-dependent diabetes mellitus. Psychosom Med 1994;56:401–9.[PubMed]
Forsander GA, Sundelin J, Persson B. Influence of the initial management regimen and family social situation on glycemic control and medical care in children with type I diabetes mellitus. Acta Paediatr 2000;89:1462–8.[PubMed] [CrossRef]