Background/Purpose: Diffuse Idiopathic Skeletal Hyperostosis (DISH) is a poorly understood condition characterized by new bone formation mainly affecting the spine. An association of DISH with Type 2 DM (T2D) has been reported for years, but whether the relationship is genetic is unknown. To investigate whether genetic loci for T2D and related quantitative traits play a role in DISH pathogenesis, we constructed genetic risk scores (GRSs) for T2D, fasting insulin, and fasting glucose to assess the association of these GRSs with DISH.

Methods: A convenience sample of 3,117 patients from the COPDGene Study was used to investigate the association between T2D and DISH. COPDGene is a multicenter study that enrolled 10,129 smokers between 2008-2011 to define subtypes of smoking related lung disease and their genetic associations. HRCT of the chest, DNA, medical history, and diabetes status (self-report or taking antidiabetic medications) were obtained. Two readers visually scored spine imaging on HRCT for DISH, based on Resnick criteria. Genotyping was performed using the Illumina Omni-Express Chip. An 83 SNP T2D GRS, a 43 SNP fasting glucose GRS, and a 22 SNP fasting insulin GRS based on validated GWAS loci, alone and in combination, were calculated. In the combined GRS, each SNP was used once. Univariate analyses between age, sex, race, BMI and DISH were performed. A logistic regression model adjusted for age, sex, race, and BMI was used to estimate the association of each GRS with DISH.

Results: We analyzed 437 DISH cases and 2,680 controls among men and women in the COPDGene study who had available HRCTs of the chest. See Table 1 for odds ratios (OR) for the univariate and multivariate models. In the univariate analysis, DISH was associated with male sex, increasing age, and the T2D phenotype (Table 1). The T2D GRS was associated with diabetes (OR 1.05, 95% CI 1.03-1.07, P<0.0001). None of the four GRSs tested (T2D, fasting insulin, fasting glucose, combination) were associated with DISH (Table 1).

Table 1. Association between DISH, clinical features and GRSs

*Adjusted for age, gender, race, BMI  

Conclusion: This is the first study to use GRS to examine the link between DISH and diabetes related traits. T2D genes are not associated with DISH despite a correlation to T2D, arguing against a genetic relationship. The previously reported associations of DISH with T2D may be due to the effects of increased levels of circulating proteins produced in diabetics, such as insulin and osteocalcin, on the bone formation pathway. The role of smoking remains to be explored. Further studies are warranted to better elucidate the mechanisms that associate diabetes with DISH.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/diffuse-idiopathic-skeletal-hyperostosis-and-diabetes-using-genetic-risk-scores-to-explore-the-association/