Endocrine Abstracts

Background: Cardiovascular disease (CVD) is the leading cause of death in persons with type 1 diabetes (T1DM). However, there are no CVD-risk prediction models for this population in widespread use and those developed from general population and type 2 diabetes have been shown to underestimate CVD-risk in T1DM. Recently, the Steno Type 1 Risk Engine (ST1RE) was developed for predicting CVD in a cohort of T1DM persons without clinical CVD. We investigate the relationship between the scores obtained from that engine and preclinical atherosclerosis measured as arterial stiffness (AS), in order to identify potential cut-off points of AS of interest in clinical practice.

Methods: One-hundred and seventy-nine persons (18–65 years old) with T1DM of >1-year duration and without clinical CVD were consecutively evaluated for: i) clinical and anthropometric data (including classical cardiovascular risk factors), ii) microvascular complications and iii) aortic pulse-wave velocity (aPWV) determined by applanation tonometry, the gold-standard for measuring AS. The ST1RE was used to estimate the 10-year CVD-risk and classify these persons into three groups: low-risk (<10%; n=105), moderate-risk (10–20%; n=53) and high-risk (>20%; n=21).

Results: One hundred and seventy nine persons were included (men: 50.8%, age: 41.2±13.1 years, duration of T1DM: 16 (12–23) years). As compared with subjects in the low- and moderate-risk group, those in the high-risk were older (32.5±8.3, 50.8±6.0, 60.7±6.6 years; P<0.001), had higher prevalence of hypertension (14.3, 37.7, 66.7%; P<0.001) and dyslipidaemia (36.4, 77.8, 89.5%; P<0.001), higher BMI (24.3±3.2, 26.6±3.8, 27.8±4.4 Kg/m²; P<0.001), higher insulin resistance (eGDR: 9.2±1.8, 7.0±2.1, 5.5±1.8 mg·kg⁻¹·min⁻¹; P<0.001) and worst glycaemic control (HbA1c: 7.6, 8.0, 8.5%; P<0.001). They also had higher prevalence of microvascular complications (27.2, 43.4, 81.0%; P<0.001) and higher aPWV (6.4±1.0, 8.4±1.3, 10.3±2.6 m/s; P<0.001). aPWV was positively associated with the ST1RE score (r=0.777; P<0.001). The best cut-off point of AS for identifying subjects in the moderate-high risk group was >7.3 m/s (Se: 86%, Sp: 83%; C-statistic: 0.914 (95CI:0.873–0.95)), and for identifying those in the high-risk, > 8.7 m/s (Se: 76%, Sp: 86%; C-statistic: 0.879 (95%CI:0.809–0.948)).

Conclusions: AS was highly and positively correlated with the ST1RE score. We provide cut-off points of AS that discriminate T1DM subjects with moderate-high and high CVD-risk that could be of great interest in clinical practice.