Independent Associations Between a Metabolic Syndrome Severity Score and Future Diabetes By Sex and Race:  the Atherosclerosis Risk in Communities Study and Jackson Heart Study

Presentation Number: SH04-11
Date of Presentation: April 3rd, 2017

Mark Daniel DeBoer*1, Matthew James Gurka2, Sherita Hill Golden3, Solomon Musani4, Mario Sims4, Abhishek Vishnu5, Yi Guo2, Michelle Cardel2 and Thomas A. Pearson2
1Univ of Virginia, Charlottesville, VA, 2University of Florida, Gainesville, FL, 3Johns Hopkins University, Baltimore, MD, 4University of Mississippi Medical Center, Jackson, MS, 5Icahn School of Medicine at Mt. Sinai, New York, NY


Background: Traditional metabolic syndrome (MetS) criteria (e.g. ATP-III) have been criticized as 1) not providing information on risk for Type 2 diabetes mellitus (T2DM) beyond the individual MetS components, 2) having racial/ethnic discrepancies and 3) being unable to track for incremental changes in MetS status over time. We previously developed a sex- and race/ethnicity-specific MetS severity score. Our objective in the current project was to assess for association between the degree of MetS severity and risk for Type 2 diabetes mellitus (T2DM) beyond that conferred by the individual MetS components and whether there was additional risk conferred by changes in MetS severity over time.

Methods: We assessed hazard ratios (HR) for ATP-III-MetS and a sex- and race-specific continuous MetS-severity Z-score with incident T2DM over a median of 7.8 years of follow-up among participants of the Atherosclerosis Risk in Communities study (n=10,957) and Jackson Heart Study (n=2,137). We further assessed for associations between the change in MetS severity score between Visits 1 and 2 (4 year interval) and future diabetes, in models that included the baseline MetS severity score.

Results: ATP-III-MetS had a HR for incident T2DM of 4.36 (95% confidence interval 3.83, 4.97) that was attenuated in models that included the individual MetS components. By contrast, participants in the 4th quartile of MetS severity (compared to the 1st quartile), had a HR of 17.4 (12.6, 24.1) for future T2DM; in models that also included the individual MetS components, this remained significant: HR 3.69 (2.42, 5.64). There was a race x MetS interaction in these models such that the HR was greater for black participants (HR=5.30) compared to whites (HR=2.24). When the change in MetS-severity Z-score between Visits 1 and 2 was included in the hazard models, this conferred further association with later diabetes, with changes in MetS-severity score of >0.5 having a HR of 2.66 compared to those with changes <0.

Conclusions: Use of a continuous sex- and race-specific MetS-severity Z-score provided additional risk prediction beyond that of the individual MetS components, suggesting added risk conferred by the processes underlying MetS. Increases in this score over time were associated with further risk, supporting potential clinical utility in following MetS severity over time.


Nothing to Disclose: MDD, MJG, SH, SM, MS, AV, YG, MC, TAP