Insulin sensitivity estimates and their longitudinal association with coronary artery disease in type 1 diabetes. Does it matter?

BACKGROUND: Insulin resistance and chronic kidney disease are both associated with increased coronary artery disease risk. Many formulae estimating glucose disposal rate in type 1 diabetes infer insulin sensitivity from clinical data. We compare associations and performance relative to traditional risk factors and kidney disease severity between three formulae estimating the glucose disposal rate and coronary artery disease in people with type 1 diabetes. METHODS: The baseline glucose disposal rate was estimated by three (Williams, Duca, and Januszewski) formulae in FinnDiane Study participants and related to subsequent incidence of coronary artery disease, by baseline kidney status. RESULTS: In 3517 adults with type 1 diabetes, during median (IQR) 19.3 (14.6, 21.4) years, 539 (15.3%) experienced a coronary artery disease event, with higher rates with worsening baseline kidney status. Correlations between the three formulae estimating the glucose disposal rate were weak, but the lowest quartile of each formula was associated with higher incidence of coronary artery disease. Importantly, only the glucose disposal rate estimation by Williams showed a linear association with coronary artery disease risk in all analyses. Of the three formulae, Williams was the strongest predictor of coronary artery disease. Only age and diabetes duration were stronger predictors. The strength of associations between estimated glucose disposal rate and CAD incidence varied by formula and kidney status. CONCLUSIONS: In type 1 diabetes, estimated glucose disposal rates are associated with subsequent coronary artery disease, modulated by kidney disease severity. Future research is merited regarding the clinical usefulness of estimating the glucose disposal rate as a coronary artery disease risk factor and potential therapeutic target.

Review and comparison of retinal vessel calibre and geometry software and their application to diabetes, cardiovascular disease, and dementia.

Developments in retinal imaging technologies have enabled the quantitative evaluation of the retinal vasculature. Changes in retinal calibre and/or geometry have been reported in systemic vascular diseases, including diabetes mellitus (DM), cardiovascular disease (CVD), and more recently in neurodegenerative diseases, such as dementia. Several retinal vessel analysis softwares exist, some being disease-specific, others for a broader context. In the research setting, retinal vasculature analysis using semi-automated software has identified associations between retinal vessel calibre and geometry and the presence of or risk of DM and its chronic complications, and of CVD and dementia, including in the general population. In this article, we review and compare the most widely used semi-automated retinal vessel analysis softwares and their associations with ocular imaging findings in common systemic diseases, including DM and its chronic complications, CVD, and dementia. We also provide original data comparing retinal calibre grading in people with Type 1 DM using two softwares, with good concordance.

No relationship between socioeconomic status, education level and development and progression of diabetic retinopathy in type 2 diabetes: a FIELD trial substudy.

In 6002 Australian adults with type 2 diabetes and a median 5-year follow-up in the FIELD (Fenofibrate Intervention and Event Lowering in Diabetes) trial, baseline socioeconomic status (SES) and self-reported education level were not related to development of on-trial sight-threatening diabetic retinopathy. Similarly, in a retinal photography substudy (n = 549), two-step diabetic retinopathy progression was not related to SES or education.

Fenofibrate, which reduces risk of sight-threatening diabetic retinopathy in type 2 diabetes, is associated with early narrowing of retinal venules: a FIELD trial substudy.

Retinal vessel calibre metrics were evaluated at baseline and 2 years in a FIELD substudy (n = 208). Central retinal venule calibre was significantly reduced by fenofibrate and unchanged by placebo. Arteriole metrics did not change. Larger studies relating retinal vessel calibre to future diabetes complications and response to therapy are merited.

Insulin antibodies are prevalent in adults with type 1 diabetes referred for islet cell transplantation and are modified by islet transplantation and immunosuppression: an Australian experience.

We have analysed insulin antibodies in 149 adults with type 1 diabetes and 2859 people without diabetes. We have determined that insulin antibody levels are higher in adults with, versus without, diabetes and that the levels are falling, and more patients are becoming antibody-negative post islet cell transplantation.

Point-of-care testing of HbA1c, renal function and lipids in remote or disadvantaged regions.

For optimal diabetes management HbA1c, lipid and renal function measurements are key. Suitable point-of-care systems can provide such data in remote areas, diabetes camps, disaster relief, small community hospitals and in community screening days in disadvantaged regions. Some systems are currently used in Australian rural, remote and indigenous health services.

Associations between circulating inflammatory markers, diabetes type and complications in youth.

BACKGROUND: Inflammation is implicated in the pathogenesis of diabetes and its complications in adults. Little is known about the relative contribution of inflammation in common types of diabetes in youth: type 1 diabetes (T1D), type 2 diabetes (T2D), and cystic fibrosis-related diabetes (CFRD). This study investigates inflammatory markers by diabetes type and complication status, and assesses indicators of inflammation and complications. METHODS: A cross-sectional study of 134 T1D, 32 T2D, 32 CFRD and 48 subjects without diabetes (including 11 with CF and normal glucose tolerance) was undertaken. Inflammation was assessed by sE-selectin by ELISA, hsCRP by turbidimetry, WCC and ESR. Nephropathy was defined by albuminuria, autonomic neuropathy by heart rate variability, and peripheral neuropathy by vibration and thermal threshold testing and retinopathy by seven-field stereoscopic fundus photography. Descriptive statistics, parametric and non-parametric ANOVA and regression analyses were performed, with significance at P < .05. RESULTS: Of 198 diabetic participants; 49% female, mean (SD) age, median diabetes duration and median HbA1c were 16 (2.5) and 6 (3-9) years, and 8.1 (6.9-9.3)%, respectively. All inflammatory markers were lower in T1D than in other diabetes groups (P < .05) but higher than in non-diabetic controls. T2D (n = 32) and CFRD (n = 32) subjects had comparable elevated levels of inflammation. Body mass index (BMI) was a strong independent explanatory variable of inflammation. In multivariate analysis, hsCRP and ESR were associated with complications in addition to HbA1c, BMI, and diastolic BP. CONCLUSIONS: Circulating inflammatory markers are elevated in adolescents with diabetes, being higher and comparable in T2D and CFRD than in T1D. Inflammation is independently associated with diabetes complications, consistent with inflammation driving vascular pathology in diabetes.

Attractions and barriers to Australian physician-researcher careers.

BACKGROUND: There is a global concern that physician-researchers are 'a dying breed'. Recent studies of clinical career choices of Australian medical students and doctors have signalled the rising age of medical graduates, generational shifts in work-life attitudes and increased proportion of female graduates. There are scant data regarding Australian physician-researchers. AIMS: To develop and utilise a questionnaire determining respondent characteristics and 'push' and 'pull' factors for medical graduates to incorporate research into their careers. METHODS: We developed and administered an 88-item online survey, including quantitative and qualitative questions, to medical students, faculty and alumni of Sydney Medical School, The University of Sydney, asking about their medical career, research experience and interest and reasons for doing or not doing medical research. Responses to all 74 quantitative questions are reported here. RESULTS: Data from 427 respondents (44% female; mean ± standard deviation age 38 ± 13 years; 56% completed or undertaking a PhD) were analysed. Attractions of research included a desire to improve human health, intellectual stimulation and career diversity. Barriers included low funding rates, job insecurity and low salaries. Although few were prepared to undertake or recommend full-time research, 71% would recommend part-time research. Respondents perceived a smaller-than-actual gap between clinical and research salaries, and if comparable (75-100% of a clinician's) salaries were available, 89% would like to spend 21-60% of their work time undertaking research. CONCLUSION: Many Australian medical students and doctors are interested in research, especially part time. Perceived obstacles include job insecurity, low funding rates and salary. Respondents underestimated clinical and research salary differences.

Suboptimal behaviour and knowledge regarding overnight glycaemia in adults with type 1 diabetes is common.

BACKGROUND: In people with type 1 diabetes (T1D), nocturnal hypoglycaemia (NH) can be slept through and can cause seizures, arrhythmias and death. Hypoglycaemia avoidance can induce hyperglycaemia and ketosis. Patient behaviour impacts clinical outcomes and may be changed by education. AIM: To develop and utilise a survey to evaluate patient self-management of overnight glycaemia in adults with T1D. METHODS: Adults with T1D attending two Australian tertiary referral diabetes clinics completed a survey about their diabetes self-management and glycaemic control, including responses to hypothetical pre-bed blood glucose (BG) levels (4-20 mmol/L). Statistical analyses included t-tests, Chi square tests and ANOVA with significance considered at P < 0.05. RESULTS: There were 205 participants (103 females), with a mean (SD) age of 41 (17) years, T1D duration of 20 (16) years, HbA1c of 7.8(1.4)%, (61.3(8.2) mmol/mol), 38% on insulin pump therapy (CSII) and 36% with impaired hypoglycaemia awareness (IHA). Mean (SD) number of BG tests/day was 5.4 (2.7). Patients set higher BG target levels at bedtime and overnight: 7.5(1.4) and 7.1(1.3) mmol/L, respectively, compared to daytime (6.9(1.0); P < 0.0001 and P = 0.002 respectively). Only 36% of participants reported treating nocturnal hypoglycaemia (NH) with the recommended refined, then complex, carbohydrate. Only 28% of patients made safe choices in all bedtime BG scenarios, with higher rates for CSII users, P = 0.0005. Further education was desired by 32% of respondents, with higher rates in those with (44%) versus without IHA (25%), P = 0.006. CONCLUSIONS: Many adults with T1D have suboptimal knowledge and behaviour regarding overnight BG self-management. A survey, piloted herein, may facilitate the identification of patients who could benefit from further education.

Baseline Circulating FGF21 Concentrations and Increase after Fenofibrate Treatment Predict More Rapid Glycemic Progression in Type 2 Diabetes: Results from the FIELD Study.

BACKGROUND: It is not known whether circulating fibroblast growth factor 21 (FGF21) concentrations are associated with glycemic progression in patients with established type 2 diabetes. This study reports this relationship in type 2 diabetes patients participating in the Fenofibrate Intervention and Event Lowering in Diabetes (FIELD) trial. METHODS: Plasma FGF21 was quantified in 9697 study participants. Among patients with lifestyle-only glucose control measures at baseline, glycemic progression was defined as the initiation of oral hypoglycemic agents or insulin therapy. We assessed the relationship of FGF21 concentrations with glycohemoglobin (Hb A1c), the homeostasis model assessment of ß-cell function (HOMA-B) and insulin resistance (HOMA-IR), and glycemic progression. RESULTS: Among 2584 patients with lifestyle-only glycemic therapy at baseline, plasma FGF21 concentrations were positively associated with HOMA-IR (5.1% increase per 100% increase in FGF21 concentrations). Patients with higher baseline plasma FGF21 concentrations had higher risk of glycemic progression over a 5-year period (P = 0.02), but the association was not significant after further adjusting for alanine aminotransferase (ALT) enzyme activity. During the fenofibrate active run-in phase, higher tertiles of fenofibrate-induced increase in FGF21 concentrations were associated with higher risk of glycemic progression (adjusted hazards ratio = 1.09 and 1.18 for tertiles 2 and 3, respectively, P for trend = 0.01), even after adjusting for ALT enzyme activity. This association was statistically significant in the fenofibrate group only (P = 0.01). CONCLUSIONS: Higher baseline and fenofibrate-induced increase in FGF21 concentrations predict more rapid glycemic progression in type 2 diabetes patients. This association may be partly explained by hepatic function.

The association between total phthalate concentration and non-communicable diseases and chronic inflammation in South Australian urban dwelling men.

OBJECTIVE: To investigate associations between urinary total phthalate concentration, chronic low-grade inflammation and non-communicable diseases in a cohort of South Australian men. METHODS: 1504 men aged 39-84 years who provided a urinary sample at the follow-up visit of the Men Androgen Inflammation Lifestyle Environment and Stress (MAILES) study, a randomly-selected group of urban-dwelling, community-based men from Adelaide, Australia (n = 2038; study participation rate: 78.1%). Total phthalate concentration was quantified in fasting morning urine samples. Chronic diseases were assessed through self-report questionnaire or directly measured using standardised clinical and laboratory procedures. Inflammatory biomarkers were assayed by ELISA or spectroscopy. Multivariable linear and logistic regression models were applied to determine associations of log-transformed urinary phthalate concentration with inflammation and chronic disease. RESULTS: Total phthalates were detected in 99.6% of urinary samples; geometric mean (95% CI) was 114.1 (109.5-118.9)µg/g creatinine. Higher total phthalate levels were associated with higher levels of hs-CRP, IL-6 (all p < 0.05) and TNF-α but not MPO. Urinary total phthalate concentrations were positively associated with cardiovascular disease, type-2-diabetes and hypertension. Comparing extreme quartiles of total phthalate, prevalence ratios were 1.78 (95% CI 1.17 - 2.71, p-trend = 0.001) for cardiovascular disease and 1.84 (95%CI 1.34 - 2.51, p-trend = 0.001) for type-2-diabetes and 1.14 (95%CI 1.01 - 1.29, p-trend = 0.013) for hypertension. Total phthalates and asthma and depression were not significantly associated. CONCLUSION: A positive association between total phthalates and cardiovascular disease, type-2-diabetes, hypertension and increased levels of chronic low-grade inflammatory biomarkers was observed in urban-dwelling Australian men.

Liberal Glycemic Control in Critically Ill Patients With Type 2 Diabetes: An Exploratory Study.

OBJECTIVES: The optimal blood glucose target in critically ill patients with preexisting diabetes and chronic hyperglycemia is unknown. In such patients, we aimed to determine whether a " liberal" approach to glycemic control would reduce hypoglycemia and glycemic variability and appear safe. DESIGN: Prospective, open-label, sequential-period exploratory study. SETTING: Medical-surgical ICU. PATIENTS: During sequential 6-month periods, we studied 83 patients with preexisting type 2 diabetes and chronic hyperglycemia (glycated hemoglobin, ≥ 7.0% at ICU admission). INTERVENTION: During the "standard care" period, 52 patients received insulin to treat blood glucose concentrations greater than 10 mmol/L whereas during the "liberal" period, 31 patients received insulin to treat blood glucose concentrations greater than 14 mmol/L. MEASUREMENTS AND MAIN RESULTS: Time-weighted mean glucose concentrations and the number and duration of moderate (< 4.0 mmol/L) and severe (≤ 2.2 mmol/L) hypoglycemic episodes were recorded, with moderate and severe hypoglycemic episodes grouped together. Glycemic variability was assessed by calculating the coefficient of variability for each patient. Safety was evaluated using clinical outcomes and plasma concentrations of markers of inflammation, glucose-turnover, and oxidative stress. Mean glucose (TWglucoseday 0-7, standard care: 9.3 [1.8] vs liberal: 10.3 [2.1] mmol/L; p = 0.02) and nadir blood glucose (4.4 [1.5] vs 5.5 [1.6] mmol/L; p < 0.01) were increased during the liberal period. There was a signal toward reduced risk of moderate-severe hypoglycemia (relative risk: liberal compared with standard care: 0.47 [95% CI, 0.19-1.13]; p = 0.09). Ten patients (19%) during the standard period and one patient (3%) during the liberal period had recurrent episodes of moderate-severe hypoglycemia. Liberal therapy reduced glycemic variability (coefficient of variability, 33.2% [12.9%] vs 23.8% [7.7%]; p < 0.01). Biomarker data and clinical outcomes were similar. CONCLUSIONS: In critically ill patients with type 2 diabetes and chronic hyperglycaemia, liberal glycemic control appears to attenuate glycemic variability and may reduce the prevalence of moderate-severe hypoglycemia.

Elucidating the Biological Mechanisms Linking Depressive Symptoms With Type 2 Diabetes in Men: The Longitudinal Effects of Inflammation, Microvascular Dysfunction, and Testosterone.

OBJECTIVE: This prospective cohort study sought to examine key biological measures linking depressive symptoms with Type 2 diabetes, specifically inflammation, microvascular dysfunction, and androgens. METHODS: A cohort of 688 men without diabetes who were 35 years or older were followed up for 5 years. Venous interleukin-6, high-sensitivity C-reactive protein, sE-selectin, endogenous total testosterone, fasting glucose, and glycated hemoglobin (HbA1c) were quantified at baseline and 5 years later. Depressive symptoms were assessed using the Beck Depression Inventory-I, and men were categorized into persistent, remitted, incident, and nondepressed groups (reference). Logistic regression was used to determine odds ratios (ORs) for diabetes adjusted for propensity score calculated from 18 established risk factors. RESULTS: Diabetes developed in 112 men (16.3% of sample). Persistent depressive symptoms were associated with diabetes (adjusted OR = 2.45, 95% confidence interval [CI] = 1.16-5.20, p = .019). Baseline testosterone (OR = 0.43, 95% CI = 0.22-0.81, p = .01) and follow-up testosterone (OR = 0.51, 95% CI = 0.31-0.84, p = .008) were inversely associated with Type 2 diabetes. Annualized HbA1c was positively associated with annualized change in cognitive Beck Depression Inventory symptoms (ß = 0.14, p = .001) and inversely associated with annualized change in testosterone (ß = -0.10, p = .014). Annualized change in fasting glucose was associated with sE-selectin (ß = 0.12, p < .001) and somatic depressive symptoms (ß = -0.12, p = .002). CONCLUSIONS: The findings suggest that lower endogenous total testosterone levels and persistent depressive symptoms were associated with Type 2 diabetes risk and HbA1c in men over a 5-year period.

The relationship of fibroblast growth factor 21 with cardiovascular outcome events in the Fenofibrate Intervention and Event Lowering in Diabetes study.

AIMS/HYPOTHESIS: Circulating fibroblast growth factor 21 (FGF21) levels are often elevated in obesity, dyslipidaemia, insulin resistance and type 2 diabetes. This study investigated the relationship of plasma FGF21 levels with cardiovascular events in patients with type 2 diabetes. METHODS: Plasma FGF21 levels were measured by ELISA at baseline in 9,697 individuals with type 2 diabetes participating in the Fenofibrate Intervention and Event Lowering in Diabetes (FIELD) study. We assessed the association of FGF21 levels with the incidence of different cardiovascular outcomes over 5 years. The primary outcome was total cardiovascular disease (CVD) events and the secondary outcomes were the four individual components: coronary heart disease events, total stroke, CVD mortality and coronary and carotid revascularisation. The tertiary outcome was hospitalisation for angina pectoris. RESULTS: Higher baseline FGF21 levels were associated with higher risks of all cardiovascular outcome events after adjusting for the study treatment allocation (all p < 0.01). The associations remained significant for total CVD events and for coronary and carotid revascularisation after further adjusting for confounding factors, with the HR (95% CI) being 1.28 (1.10, 1.50) and 1.26 (1.01, 1.56), respectively, for the highest tertile compared with the lowest tertile (overall effect p = 0.002 and 0.007, respectively). The addition of FGF21 levels to a model including established CVD risk factors predicting total CVD events led to a non-significant increase in the C-statistic but there was a significant improvement in integrated discrimination and net reclassification. CONCLUSIONS/INTERPRETATION: Higher baseline plasma FGF21 levels were associated with higher risk of cardiovascular events in patients with type 2 diabetes. TRIAL REGISTRATION: ISRCTN64783481.

Relationship of fibroblast growth factor 21 with baseline and new on-study microvascular disease in the Fenofibrate Intervention and Event Lowering in Diabetes study.

AIMS/HYPOTHESIS: Baseline circulating fibroblast growth factor 21 (FGF21) levels can predict total cardiovascular disease events in the Fenofibrate Intervention and Event Lowering in Diabetes (FIELD) study. This paper describes the relationship of baseline FGF21 levels and new on-study microvascular disease in patients with type 2 diabetes from the FIELD study. METHODS: Baseline FGF21 levels were measured in plasma by enzyme-linked immunosorbent assay in 9697 study participants. Total microvascular disease was defined as the presence of any nephropathy, retinopathy, neuropathy and/or microvascular amputation. The relationship between FGF21 levels and microvascular disease was assessed by multivariable logistic regression. RESULTS: Higher baseline FGF21 levels were found in patients with baseline total microvascular disease (p<0.001). The association remained significant after adjusting for potential confounding factors (OR [95% CI] 1.13 [1.08, 1.19] per SD increase in log e -transformed FGF21 levels, p<0.001). Of 6465 patients without baseline total microvascular disease, 1517 developed new on-study total microvascular disease over 5 years of follow-up. Higher baseline FGF21 levels were associated with a higher risk of new on-study total microvascular disease after adjusting for potential confounding factors (OR [95% CI] 1.09 [1.02, 1.16] per SD increase in log e -transformed FGF21 levels, p=0.01). Addition of FGF21 levels in a model of new on-study total microvascular disease with established risk factors significantly, but modestly, increased the integrated discrimination improvement and the net reclassification improvement (both p < 0.01). CONCLUSIONS/INTERPRETATION: Higher baseline FGF21 levels are seen in patients with type 2 diabetes and established microvascular disease, and predict the future development of new microvascular disease.

Interactive calculator to estimate insulin sensitivity in type 1 diabetes.

The gold standard for measuring insulin sensitivity (IS) is the hyperinsulinemic-euglycemic clamp, a time, costly, and labor-intensive research tool. A low insulin sensitivity is associated with a complication-risk in type 1 diabetes. Various formulae using clinical data have been developed and correlated with measured IS in type 1 diabetes. We consolidated multiple formulae into an online calculator (bit.ly/estimated-GDR), enabling comparison of IS and its probability of IS <4.45 mg/kg/min (low) or >6.50 mg/kg/min (high), as measured in a validation set of clamps in 104 adults with type 1 diabetes. Insulin sensitivity calculations using different formulae varied significantly, with correlations (R2) ranging 0.005-0.87 with agreement in detecting low and high glucose disposal rates in the range 49-93% and 89-100%, respectively. We demonstrate that although the calculated IS varies between formulae, their interpretation remains consistent. Our free online calculator offers a user-friendly tool for individual IS calculations and also offers efficient batch processing of data for research.

Increased Diabetes Complications in a Mouse Model of Oxidative Stress Due to 'Mismatched' Mitochondrial DNA.

Associations between chronic diabetes complications and mitochondrial dysfunction represent a subject of major importance, given the diabetes pandemic and high personal and socioeconomic costs of diabetes and its complications. Modelling diabetes complications in inbred laboratory animals is challenging due to incomplete recapitulation of human features, but offer mechanistic insights and preclinical testing. As mitochondrial-based oxidative stress is implicated in human diabetic complications, herein we evaluate diabetes in a unique mouse model that harbors a mitochondrial DNA from a divergent mouse species (the 'xenomitochondrial mouse'), which has mild mitochondrial dysfunction and increased oxidative stress. We use the streptozotocin-induced diabetes model with insulin supplementation, with 20-weeks diabetes. We compare C57BL/6 mice and the 'xenomitochondrial' mouse, with measures of heart and kidney function, histology, and skin oxidative stress markers. Compared to C57BL/6 mice, the xenomitochondrial mouse has increased diabetic heart and kidney damage, with cardiac dysfunction, and increased cardiac and renal fibrosis. Our results show that mitochondrial oxidative stress consequent to divergent mtDNA can worsen diabetes complications. This has implications for novel therapeutics to counter diabetes complications, and for genetic studies of risk, as mtDNA genotypes may contribute to clinical outcomes.

Not enough known about fenofibrate's kidney effects in people with Type 2 diabetes.

Globally ≈10% of adults have diabetes, with 80% in disadvantaged regions, hence low-cost renoprotective agents are desirable. Fenofibrate demonstrated microvascular benefits in several cardiovascular end-point diabetes trials, but knowledge of effects in late-stage kidney disease is limited. We report new FIELD substudy data and call for further kidney outcomes data.

Applications of SGLT2 inhibitors beyond glycaemic control.

Sodium-glucose cotransporter 2 (SGLT2) inhibitors were initially developed for their glucose-lowering effects and have shown a modest glycaemic benefit in people with type 2 diabetes mellitus (T2DM). In the past decade, a series of large, robust clinical trials of these therapies have demonstrated striking beneficial effects for various care goals, transforming the chronic disease therapeutic landscape. Cardiovascular safety studies in people with T2DM demonstrated that SGLT2 inhibitors reduce cardiovascular death and hospitalization for heart failure. Subsequent trials in participants with heart failure with reduced or preserved left ventricular ejection fraction demonstrated that SGLT2 inhibitors have beneficial effects on heart failure outcomes. In dedicated kidney outcome studies, SGLT2 inhibitors reduced the incidence of kidney failure among participants with or without diabetes. Post hoc analyses have suggested a range of other benefits of these drugs in conditions as diverse as metabolic dysfunction-associated steatotic liver disease, kidney stone prevention and anaemia. SGLT2 inhibitors have a generally favourable adverse effect profile, although patient selection and medication counselling remain important. Concerted efforts are needed to better integrate these agents into routine care and support long-term medication adherence to close the gap between clinical trial outcomes and those achieved in the real world.

Association of circulating fibroblast growth factor 21 levels with all-cause and cardiovascular mortality: The multi-ethnic study of atherosclerosis.

BACKGROUND: Fibroblast growth factor 21 (FGF21) levels are often elevated in cardiovascular disease (CVD). However, no study has assessed its association with cardiovascular and all-cause mortality in a population free of clinically evident CVD. METHODS: A total of 5543 Multi-Ethnic Study of Atherosclerosis (MESA) participants (mean age 62.7 years, 47.5 % male), free of clinically evident CVD at baseline, were studied. From baseline (2000-2002), 1606 deaths (including 387 CVD deaths) were observed over a median follow-up of 17.7 years. Multivariable Cox regression analysis was performed to assess the association of plasma FGF21 levels with mortality. RESULTS: FGF21 levels at baseline were associated with all-cause mortality, even after adjustment for traditional risk factors, including demographic, socioeconomic and cardiovascular risk factors (adjusted hazard ratio 1.08 [95% confidence interval 1.01, 1.16] per 1 SD increase in ln-transformed levels; 1.27 for the highest vs, lowest quartile). Baseline FGF21 levels were significantly associated with both CVD and non-CVD mortality in unadjusted models. However, the association with non-CVD mortality, but not CVD mortality, remained statistically significant after adjusting for covariates. Similar results were obtained in FGF21 quartile analyses and also when using competing risk regression or matched case-control cohort in sensitivity analyses. CONCLUSIONS: In subjects without clinically-evident CVD at baseline, over 17.7 years follow-up there is a modest association of baseline FGF21 levels with all-cause mortality. The finding that this is driven primarily by a significant association with non-CVD mortality over almost two decades merits further investigation.