Background: Clinical characteristics such as HbA1c, systolic blood pressure (SBP), albuminuria and estimated glomerular filtration rate (eGFR) are important when treating type 1 diabetes (T1D). We investigated the variability in these measures as risk markers for micro- and macrovascular complications. Methods: This prospective study included 1062 individuals with T1D. Visit-to-visit variability (VVV) of HbA1c, SBP, albuminuria, and eGFR was calculated as the SD of the residuals in individual linear regression models using all available measures in a specified period of 3 years (SVVV) as well as the entire follow-up (CVVV). Endpoints included: composite cardiovascular events (CVE) defined as myocardial infarction, non-fatal stroke, coronary arterial intervention, or peripheral arterial intervention; end-stage kidney disease (ESKD) defined as chronic kidney disease stage 5, chronic dialysis, or kidney transplantation; eGFR-decline ≥30%; and mortality. Adjustment included age, sex, cholesterol, HbA1c, SBP, body mass index, smoking, albuminuria, eGFR, and intercept and slope of respective regression models. Findings: SVVV and CVVV for every exposure associated significantly with each other (p<0·05) (R=0·42-0·88 depending on exposure). SBP SVVV was significantly associated with CVE (adjusted hazard ratio per doubling, CI 95%: 1·38, 1·09-1·76), ESKD (2·00, 1·27-3·13) and mortality (1·47, 1·16-1·89). HbA1c SVVV was significantly associated with development of ESKD (1·76, 1·02-1·51) and mortality (2·01, 1·56-2·59); albuminuria SVVV with eGFR-decline (1·26, 1·15-1·38) and ESKD (1·24, 1·02-1·51), but neither CVE nor mortality. Adjusted eGFR SVVV was only associated with eGFR-decline (1·12, 1·03-1·21). CVVV of all exposures were significantly associated with risk of all endpoints, except for eGFR CVVV with the endpoints CVE and mortality. Interpretation: In T1D, higher short- and long-term VVV of basic clinical risk markers add important risk stratification information for the development of micro- and macrovascular complications. Funding Statement: No external funding to declare. Declaration of Interests: P.R. reports giving lectures for AstraZeneca, Bayer, Eli Lilly, and Novo Nordisk; serving as a consultant for AstraZeneca, Bayer, Gilead, Sanofi Aventis, and Novo Nordisk, with all fees given to Steno Diabetes Center Copenhagen; and having equity interest in Novo Nordisk. All other authors declare no competing interests. Ethics Approval Statement: All participants gave informed written consent and the study complied with the Declaration of Helsinki. The research protocol, as well as all further activities, were reviewed and approved by the local ethics committee.
Read full abstract