By Connor Emdin
January 30, 2018
Type 1 diabetes (T1D) is commonly thought of as a disease of children and young adults, with a peak age of diagnosis around 14 years.1 However, adults with T1D may be misdiagnosed as having type 2 diabetes (T2D) due to the much greater prevalence of T2D in older ages.2 Such misdiagnosis of T1D as T2D may have important clinical consequences. Individuals with undiagnosed T1D may be less likely to receive insulin therapy and may present with diabetic ketoacidosis, a life-threatening emergency characterized by elevated blood glucose and ketone levels.3
Researchers from the University of Exeter used human genetics to determine how frequently T1D develops among adults.4 They created a genetic risk score of 29 genetic variants previously identified as being associated with the development of T1D. These genetic variants are not associated with the development of T2D. Analyzing 379,511 white participants in a large UK biobank, they identified 13,250 patients who had diabetes mellitus (DM). Of these patients, they classified individuals as being of either high genetic risk for T1D (individuals in the top half of the T1D genetic risk score; 7,268 patients) or low genetic risk for T1D (individuals in the bottom half of the T1D genetic risk score; 5,982 patients). Prior studies have shown that individuals at low genetic risk for T1D are very unlikely to develop T1D; therefore, any excess of individuals with diabetes in the high-T1D genetic risk category as compared to the low-T1D genetic risk category were defined as individuals with “genetically defined T1D.”
They identified 1,286 more individuals with DM in the high genetic risk score group, who were labeled ‘genetically defined T1D.’ This was consistent with 9.6% of all diabetes cases in UK biobank being T1D. Of these 1286 individuals, 537 (58%) were diagnosed after age 30 and 749 (42%) were diagnosed before age 30. Individuals across the age spectrum, from 0 to 60, developed T1D. Further analysis suggested that T1D was present in 4% of DM cases diagnosed between ages 31-60 years, as compared to 74% of cases in patients under the age of 30 years.
The authors then attempted to derive summary characteristics for patients with genetically defined T1D. Categorical variables (e.g., admission with ketoacidosis) were calculated similarly to the incidence of T1D, i.e., the excess number of events between low- and high-risk groups were classified as affecting high-risk individuals. For continuous variables such as BMI, differences in mean values for individuals in the low- and high-risk groups were used to calculate values for T1D patients.
Individuals, aged 30-60, with genetically defined T1D were significantly more likely than individuals with T2D to develop diabetic ketoacidosis (11% vs. 0.3%, p<0.0001) and to require insulin therapy (89% vs. 6%, p<0.0001, Table). Furthermore, individuals with genetically defined T1D diagnosed after age 30 were less likely to be initiated on insulin one year after diagnosis compared to individuals diagnosed before age 30 (89% vs. 97%, p<0.0001), suggesting that individuals with T1D in adulthood may be less likely to be aggressively treated with insulin.
Table. Characteristics of patients diagnosed with genetically-defined T1D and T2D at age 30-60.
|Type 1 Diabetes (n=537)||Type 2 Diabetes (n=11 696)||p-value|
|Age at Diagnosis||42 (CI 41, 43)||52 (CI 52, 52)||<0.0001|
|Male||335 (62%)||7684 (66%)||0·84|
|Female||202 (38%)||4012 (34%)||0·84|
|BMI||27.4 (CI 26.7, 28.0)||32.4 (CI 32.2, 32.5)||<0.0001|
|Insulin treatment 1 year after diagnosis||476 (89%)||648 (6%)||<0·0001|
|Diabetic ketoacidosis as discharge diagnosis||61 (11%)||30 (0·3%)||<0·0001|
(Table adapted from Thomas et al. Lancet Diabetes and Endocrinology 2018)
The study has several important limitations. No supportive testing to confirm T1D versus T2D was available. All characteristics, including BMI, insulin use, and ketoacidosis were self-reported by patients and therefore subject to recall bias. However, the differences between groups are striking and suggest these findings are worthy of study in a prospective manner.
This interesting study demonstrates that a substantial proportion of patients diagnosed with DM in adulthood have may have T1D, contradicting the widely held belief of T1D as a disease of the young. These findings highlight the importance of developing reliable techniques to distinguish T1D, a rapidly progressive disease that can present with diabetic ketoacidosis, from T2D in adulthood. While autoantibodies to the islet antigen glutamic acid decarboxylase are sometimes used to identify T1D, these antibodies are only present in 70% of patients with T1D.5 Furthermore, interpretation of this testing is challenging in older patients due to the low prevalence of T1D and the presence of false positive results. A prospective study, incorporating factors such as autoantibody and C peptide testing, genetic susceptibility, and clinical characteristics may be informative in defining this population of late onset T1D more clearly.
Most importantly for now, these findings demonstrate that practicing clinicians must maintain a high degree of suspicion for development of T1D in adulthood, and consider a diagnosis of T1D among any adult with diabetes who rapidly progresses to insulin, regardless of the age of diabetes development.4
Study online for the obesity medicine board reviews with Harvard Medical School instructors: Blackburn Course in Obesity Medicine: Obesity Medicine Board Review
- Diaz-Valencia PA, Bougnères P, Valleron A-J. Global epidemiology of type 1 diabetes in young adults and adults: a systematic review. BMC Public Health. 2015;15(1):255. doi:10.1186/s12889-015-1591-y.
- Tuomilehto J. The emerging global epidemic of type 1 diabetes. Curr Diab Rep. 2013;13(6):795-804. doi:10.1007/s11892-013-0433-5.
- Diabetic ketoacidosis in adults. BMJ. 2015;351:h5866. doi:10.1136/bmj.h5866.
- Thomas NJ, Jones SE, Weedon MN, Shields BM, Oram RA, Hattersley AT. Frequency and phenotype of type 1 diabetes in the first six decades of life: a cross-sectional, genetically stratified survival analysis from UK Biobank. Lancet Diabetes Endocrinol. November 2017. doi:10.1016/S2213-8587(17)30362-5.
- Sabbah E, Savola K, Ebeling T, et al. Genetic, autoimmune, and clinical characteristics of childhood- and adult-onset type 1 diabetes. Diabetes care. 2000;23(9):1326-1332.
Connor Emdin is a post-doctoral research fellow in Sek Kathiresan’s lab at the Broad, specializing in the genetics of cardiovascular disease. He completed his doctorate in cardiovascular epidemiology at the University of Oxford from 2009-2013.
Follow Connor on Twitter: @connoremdin