e32 Diabetes Care Volume 40, March 2017

Higher Parathyroid Level Is Emilie H. Zobel,1 Simone Theilade,1 Bernt Johan von Scholten,1 Associated With Increased Arterial Frederik Persson,1 Lise Tarnow,2,3 Stiffness in Type 1 Diabetes Maria Lajer,1 Tine W. Hansen,1 and Peter Rossing1,2,4 Diabetes Care 2017;40:e32–e33 | DOI: 10.2337/dc16-2428

Vascular calcification is a common con- between cfPWV and mineral from the is triggered sequence of aging, hypercholesterol- density, clinical bone markers (PTH, by low serum calcium. PTH secretion re- emia, chronic renal insufficiency, and 25-hydroxyvitamin D, calcium, and phos- sults in raised serum calcium through its diabetes (1). Stiffening of the large ar- phorus), and markers of bone min- release from the , reduced renal teries is a result of vascular calcification. eral metabolism (endostatin, sclerostin, excretion, and increased small intestine Carotid–femoral pulse wave velocity Dickkopf 1, and osteoprotegerin). In un- absorption (3). Aside from its well- (cfPWV) is considered the gold standard adjusted analyses, bone mineral density, established role in calcium , measure of arterial stiffness (2) and has all clinical bone markers, and markers of elevated PTH has been linked to pres- been shown to be a strong predictor of mineral metabolismdexcept calcium, ence of hypertension and cardiac hy- mortality and cardiovascular outcome. phosphorus, and Dickkopf 1dwere asso- pertrophy (4), and PTH excess may be We evaluated the association be- ciated with cfPWV (P # 0.041). After ad- related to development of cardiovascu- tween arterial stiffness (evaluated by justment for age, sex, and mean arterial lar disease (5). cfPWV) and bone mass density (evalu- pressure, the level of bone mineral den- Our study consisted of a well- ated with dual-energy X-ray absorpti- sity, PTH, and sclerostin remained associ- characterized group of persons with ometry at the femoral neck), as well ated with cfPWV (P # 0.027). After type 1 diabetes. Arterial stiffness was as a comprehensive panel of bone adjustment for additional risk factors evaluated using the gold standard markers, in a well-characterized cohort (HbA1c, total cholesterol, BMI, antihyper- method, analyzed as a continuous vari- of 347 persons with type 1 diabetes. We tensive treatment, urinary albumin ex- able and with proper adjustment. Bone hypothesized that vascular calcification cretion rate, estimated glomerular mineral density was evaluated with ro- is linked to decalcification of the bones. filtration rate, and smoking), PTH re- bust methods, and all bone markers Theparticipantswererecruitedfrom mained positively associated with were analyzed as continuous variables; the outpatient clinic at Steno Diabetes cfPWV (P = 0.014). no arbitrary cutoffs were applied. Center, Gentofte, Denmark. Written in- In the search for a link between bone Our findings highlight PTH as a poten- formed patient consent and ethical and vascular disease, we demonstrated tial mediator for the cross talk between approval of the study were obtained. an association between arterial stiffness bone and vascular disease. However, A total of 164 (47%) participants were and bone mineral density, clinical bone our findings need validation, and pro- women, mean 6 SD age was 55.8 6 9.6 markers, and markers of mineral metab- spective studies investigating the

OBSERVATIONS years and cfPWV was 11.0 6 3.4 m/s, olism. These associations lost significance relationship between PTH and cardio- – and median (interquartile range) para- after comprehensive adjustment, except vascular outcome in type 1 diabetes hormone (PTH) was 39.1 (29.1, for the relationship between higher PTH are warranted. Depending on the results 57.3) pg/mL. and increased arterial stiffness. of such studies, therapies known to re- e-LETTERS Table 1 shows the unadjusted PTH is one of the main regulators of duce PTH (e.g., cinacalcet) could poten- and stepwise adjusted associations calcium homeostasis. Secretion of PTH tially reduce the cardiovascular risk in

1Steno Diabetes Center, Gentofte, Denmark 2Department of Clinical Medicine, Faculty of Health, Aarhus University, Aarhus, Denmark 3Department of Clinical Research, Nordsjællands Hospital, University of Copenhagen, Hillerød, Denmark 4Department of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark Corresponding author: Emilie H. Zobel, [email protected]. Received 14 November 2016 and accepted 20 December 2016. © 2017 by the American Diabetes Association. 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. care.diabetesjournals.org Zobel and Associates e33

Table 1—Unadjusted and stepwise-adjusted associations between cfPWV and measures of bone/mineral metabolism Adjusted for age, sex, Adjusted for age, sex, mean Unadjusted and mean arterial pressure arterial pressure, and other risk factors b P b P b P Bone mineral density Femoral neck bone mineral density 20.23 ,0.001 20.15 0.027 20.06 0.43 Clinical bone markers PTH* 0.23 ,0.001 0.28 ,0.001 0.17 0.014 25-Hydroxyvitamin D 0.12 0.041 20.12 0.09 20.10 0.17 Ionized calcium 20.01 0.08 20.06 0.37 20.02 0.77 Phosphorus 0.07 0.19 0.11 0.13 0.001 0.98 Markers of bone mineral metabolism Endostatin* 0.16 0.003 0.12 0.09 0.05 0.49 Sclerostin* 0.14 0.011 0.16 0.022 0.01 0.85 Dickkopf 1* 0.007 0.90 0.06 0.43 0.12 0.15 Osteoprotegerin* 0.36 ,0.001 0.13 0.08 0.02 0.77

Other risk factors included HbA1c, total cholesterol, BMI, antihypertensive treatment, urinary albumin excretion rate, estimated glomerular filtration rate, and smoking. cfPWV was measured with the SphygmoCor (AtCor Medical, Sydney, Australia). Plasma vitamin D [25(OH)D3] levels were determined by high-performance liquid chromatography–tandem mass spectrometry. Plasma PTH levels were analyzed using a second-generation electrochemiluminescence immunoassay (Cobas e601, Roche Diagnostics). Serum endostatin, sclerostin, Dickkopf 1, and osteoprotegerin were measured by sandwich ELISA (Biomedica Medizinprodukte, Austria). *Log2 transformed for analyses. The b estimates represent standardized effect.

subjects with type 1 diabetes with ele- revision of the manuscript for key intellectual References vated PTH. content. B.J.v.S. and T.W.H. analyzed and in- 1. Staessen J, Amery A, Fagard R. Isolated sys- terpreted the data, performed statistical anal- tolic hypertension in the elderly. J Hypertens ysis, and critically revised the manuscript for 1990;8:393–405 key intellectual content. F.P. analyzed and Acknowledgments. The authors thank all par- 2. Laurent S, Cockcroft J, Van Bortel L, et al.; Eu- interpreted the data and made critical re- ticipants and acknowledge the work of labora- ropean Network for Non-invasive Investigation of vision of the manuscript for key intellectual tory technicians Tina R. Juhl, Anne G. Lundgaard, Large Arteries. Expert consensus document on content. L.T. conceived and designed the re- Berit R. Jensen, and Ulla M. Smidt at Steno arterial stiffness: methodological issues and clini- Diabetes Center. search, analyzed and interpreted the data, and cal applications. Eur Heart J 2006;27:2588–2605 Funding. The work leading to this article re- made critical revision of the manuscript for key 3. Fraser WD. . Lancet ceived funding from the European Commission’s intellectual content. P.R. conceived and de- 2009;374:145–158 Seventh Framework Programme under SysKID signed the research, analyzed and interpreted 4. Saleh FN, Schirmer H, Sundsfjord J, Jorde R. consortium grant agreement no. HEALTH-F2- the data, handled funding and supervision, 2009-241544. made critical revision of the manuscript for and left ventricular hyper- – Duality of Interest. No potential conflicts of key intellectual content, and supervised the trophy. Eur Heart J 2003;24:2054 2060 interest relevant to this article were reported. study. E.H.Z. is the guarantor of this work 5. Folsom AR, Alonso A, Misialek JR, et al. Para- Author Contributions. E.H.Z. analyzed and and, as such, had full access to all the data in thyroid hormone concentration and risk of car- interpreted the data and drafted the manu- the study and takes responsibility for the in- diovascular diseases: the Atherosclerosis Risk in script. S.T. and M.L. researched data, analyzed tegrity of the data and the accuracy of the data Communities (ARIC) study. Am Heart J 2014; and interpreted the data, and made critical analysis. 168:296–302