Author Interviews, Beth Israel Deaconess, Diabetes, Heart Disease, University of Michigan / 18.07.2015 Interview with: Venkatesh L. Murthy, MD, PhD, FACC, FASNC University of MichiganVenkatesh L. Murthy, MD, PhD, FACC, FASNC University of Michigan Dr. Ravi Shah MD Beth Israel Deaconess Medical Centerand Dr. Ravi Shah MD Beth Israel Deaconess Medical Center   MedicalResearch: What is the background for this study? Response: Recent changes recommend statin therapy for cardiovascular risk reduction in an increasingly large number of Americans. Conversely, a number of studies have identified an increased risk of diabetes with statin treatment. Thus, there is increasing need for tools to target statin therapy to those with a favorable risk-benefit profile. MedicalResearch: What are the main findings? Response: In our study, we analyzed data from 3,153 individuals from the Multi-Ethnic Study of Atherosclerosis who underwent CT scanning at baseline for assessment of calcium score. The CT scans were analyzed to assess liver attenuation as a measure of the amount of liver fat. We demonstrated that high liver fat doubled the risk of diabetes over a median of 9 years of follow-up. Importantly, statin therapy also doubled the risk of diabetes. The two together had an additive effect, even after adjusting for BMI, age, gender, family history of diabetes, waist circumference, lipids, hsCRP and exercise habits. As in prior studies, the risk of cardiovascular disease (CVD) events increased with increasing calcium score, as has previously been shown in MESA and in other studies. We then divided the cohort into six groups based on calcium score (0, 1-100 and >100) and liver fat (low/high). Using published data from meta-analyses of statin trials, we computed the number needed to treat to prevent one hard CVD event for statin therapy. Using data from our study, we computed the number needed to harm to cause one additional case of diabetes from statin therapy. The numbers needed to treat with ranged from 29-40 for calcium score of >100 to 218-252 for calcium score of 0. Conversely, the numbers needed to harm were approximately 63-68 for those with low liver fat versus 22-24 for those with high liver fat. Thus the combination of calcium score and liver fat assessment, from a single standard calcium score scan, allows for physicians to provide better assessment of risk and benefit of statins in discussion with their patients.