Insulin resistance: targeting dyslipidemia beyond the LDL-cholesterol Rocky Mountain Metabolic Syndrome Symposium May 14 th 21 Mori Krantz MD FACC Associate Professor, University of Colorado Director of CV Prevention Programs, CPC Staff Cardiologist, Denver Health Disclosures: Speaking and grant support: GSK, Pfizer Major stock holder: none Acknowledgment: selected slides from AHA Spotlight series lipid group Presentation outline Describe all components of the metabolic syndrome How common is this syndrome? Understand pathogenesis of insulin resistance & atherogenic dyslipidemia What is our principal focus to reduce CVD in DM? Lipids vs. Glycemic control Treatment of LDL cholesterol Beyond LDL cholesterol Are there better predictors of CVD risk than LDL-C? Review the concept of residual risk Atherogenic dyslipidemia defined: lipid particles made simple Does size matter? Are Triglycerides (TG) an independent CVD risk factor? Non-HDL-C, ApoB, and the TC/HDL ratio What are the therapeutic strategies for atherogenic dyslipidemia? Fibrates, Omega-3 fatty acids, Niacin, Diet, Exercise Treatment targets Q & A
Components of the metabolic syndrome (a precursor to diabetes) Risk Factor ( 3) Defining Level Central Abdominal obesity Waist circumference >4 & 35 inches (men/women) Triglycerides 15 mg/dl HDL-C <4 mg/dl men <5 mg/dl women Blood pressure 13/ 85 mm Hg Fasting glucose 11 mg/dl NCEP Expert Panel. JAMA 21;285:2486-2497. The evolution of insulin resistance and the metabolic syndrome 85/6 95/65 11/7 155/98 The Economist, December 11, 23 Obesity & diabetes track in US adults Diabetes 25% increase 24.3 3 8 US adults (%) 5.3 Obesity (BMI 3 kg/m 2 ) 3% increase 6.9 19.5 15 4 1997 24* 1997 24* *Jan June CDC. 24 NHIS; www.cdc.gov/nchs/nhis.htm. Accessed August 25.
Incidence of CVD by the number of components of the Metabolic Syndrome Odds Ratio 7 6 5 4 3 2 1 5.86 2.7 1.95 2.5 1 1 2 3 4 No. of Components Diabetes Care 25:179, 22 Over half of patients referred to cardiologists have insulin resistance Patients with insulin resistance syndrome (%) 6 4 2 Cardiac rehabilitation Acute MI 58 59 5 N = 1912 Savage, 25 N = 235 Milani, 23 N = 85 Curran, 24 Savage PD et al. Am Heart J. 25;149:627-31. Milani RV, Lavie CJ. Am J Cardiol. 23;92:5-4. Curran PJ et al. J Am Coll Cardiol. 24;43(suppl A):249A. Insulin Resistance Etiology: Central Adiposity (not inert but metabolically active) C-Reactive Protein ^ (inflammatory marker) FREE FATTY ACIDS Adiponectin (insulin sensitizer) Kershaw EE, et al. J Clin Endocrinol Metab. 24;89:2548-2556. Lee YH, et al. Curr Diab Rep. 25;5:7-75. Boden G, et al. Eur J Clin Invest. 22;32:14-23. 9
Atherogenic Dyslipidemia: the hallmark of metabolic syndrome Elevated Triglycerides (>15 mg/dl) Reduced HDL-C (<4 mg/dl in men; <5 mg/dl in women) Elevated small dense LDL particles Highly likely in the setting of elevated TG, reduced HDL-c and borderline elevated LDL-c Elevated non-hdl-c and/or Apo B 1 NHANES: Room for improvement < 5% of Diabetics Achieve Risk-Factor Goals Adults With Diabetes (%) 6 5 4 3 2 1 NHANES III 1988 1994 (N=124) NHANES 1999 2 (N=37) HbA 1c <7% BP <13/8 mmhg TC <2 mg/dl Good Control * *Good control indicates HbA 1c, BP, and TC were at recommended levels. National Health and Nutrition Examination Study (NHANES): cross-sectional survey of a nationally representative sample of the noninstitutionalized civilian US population, aged 2 years and older, with previously diagnosed diabetes. Saydah S et al. JAMA. 24;291:335 342. We ve gotten worse in terms of glycemic control and better with regard to lipids. Should I focus on lipids or blood sugar?
ACCORD: Study design N = 1,251 with T2DM and existing CVD or additional CV risk factors BP trial (n) Lipid trial* (n) SBP <12 mg Hg SBP <14 mg Hg Group A Group B Glycemia trial A1C <6% A1C 7.%-7.9% 1178 1184 1193 1178 1383 137 1374 1391 5128 5123 2362 2371 2753 2765 4733 5518 *Statin + fibrate vs statin, treatment group assignment blinded until end of trial Primary outcome: CV death, MI, stroke ACCORD Study Group. Am J Cardiol. 27;99(suppl):21i-33i. www.accordtrial.org ACCORD: Treatment effects on glucose control 9. 8.5 8. Standard therapy A1C (%) 7.5 7. 6.5 6. Intensive therapy 1 2 3 4 5 6 Time (years) ACCORD Study Group. N Engl J Med. 28;358:2545-59. ACCORD: significant increase in death with intensive therapy Patients with events (%) 25 2 15 1 5 HR 1.22 (1.1-1.46) P =.4 1 2 3 4 5 6 Time (years) CVD Death: HR 1.35 (1.4 1.76) P=.2 Intensive therapy Standard therapy ACCORD Study Group. N Engl J Med. 28;358:2545-59.
ADVANCE: Study design N = 11,14 with T2DM and high risk for CV events Perindopril + Indapamide Placebo Intensive glucose control Intensive glucose control + Perindopril/Indapamide Intensive glucose control + Placebo Standard glucose control Standard glucose control + Perindopril/Indapamide Standard glucose control + Placebo n = 5569 n = 5571 Primary outcome: Macro (CV death, MI, stroke), micro (new/worsening nephropathy, retinopathy) ADVANCE Collaborative Group. J Hypertens. 21;19(suppl 4):S21-S28. www.advance-trial.com ADVANCE Treatment effect on glucose control 1. 9. Mean A1C (%) 8. 7. Standard control P <.1 6. Intensive control 5.. 6 12 18 24 3 36 42 48 54 6 66 Follow-up (months) ADVANCE Collaborative Group. N Engl J Med. 28;358:256-72. ADVANCE: Effect on macrovascular outcomes CV death, MI stroke 25 Cumulative incidence (%) 2 15 1 5 HR.94 (.84-1.6) P =.32 Standard control Intensive control 6 12 18 24 3 36 42 48 54 6 66 Follow-up (months) ADVANCE Collaborative Group. N Engl J Med. 28;358:256-72.
NCEP: approach to lipids Step one - No surprise - lower LDL-C HPS and CARDS: Benefits of lowering LDL-C in diabetes HPS All diabetes Event rate (%) Statin Placebo 9.4 12.6 Statin better.73 Placebo better P <.1 _ LDL-C (mg/dl)* 34.8 Diabetes, no CVD 9.3 13.5.67.3 34.8 CARDS 5.8 9..63.1 46.4.5.7 1. Relative risk 1.7 *Statin vs placebo HPS = Heart Protection Study CARDS = Collaborative Atorvastatin Diabetes Study HPS Collaborative Group. Lancet. 23;361:25-16. Colhoun HM et al. Lancet. 24;364:685-96. ASCOT-LLA: Statin reduces CV events in DM & HTN N = 2532, baseline LDL-C 128 mg/dl 14. % 12. 1. 8. 6. 4. 23% Risk reduction P =.36 2. HR =.77 (.61.98)...5 1. 1.5 2. 2.5 3. 3.5 Years Number at risk Placebo 1258 1231 129 1191 1171 165 699 37 Atorvastatin 1274 1237 1219 12 1175 158 714 375 Nonfatal MI, CV mortality, UA, stable angina, arrhythmias, stroke, TIA, PAD, retinal vascular thrombosis, revascularization ASCOT-LLA = Anglo-Scandinavian Cardiac Outcomes Trial Lipid Lowering Arm Placebo Atorvastatin 1 mg Sever PS et al. Diabetes Care. 25;28:1151-7.
NCEP approach: Step 2, beyond LDL The concept of residual dyslipidemia risk Residual Risk in Patients Treated With Intensive Statin Therapy N LDL-C (mg/dl)* 4162 8888 1 1 95 62 14 81 11 77 Cannon CP, et al. N Engl J Med. 24;35:1495-154 Pedersen TR, et al. JAMA. 25;294:2437-2445 LaRosa JC, et al. N Engl J Med. 25;352:1425-1435 Superko HR. Br J Cardiol. 26;13:131-136 *Mean or median LDL-C after treatment 23 Heart Protection Study Among Simvastatin-treated patients who entered the trial with LDL-C levels < 116 mg/dl and who had mean LDL-C levels of ~ 7 mg/dl during the study: The 5-year risk of a major vascular event was still 18%. This would equate to a 1-year risk of 35%. Heart Protection Study Collaborative Group. Lancet. 22;36:7-22 24
Shortcomings of LDL: both calculated and measured Friedewald equation (LDL-C = TC HDL-C - TG/5) Cholesterol content in LDL varies between individuals Influenced by metabolic factors (insulin resistance) Changes with statin treatment Less accurate at LDL < 1 mg/dl Less accurate as triglycerides >15 mg/dl Does not represent all atherogenic ApoB-lipoproteins Does not accurately represent the number of LDL particles Direct LDL cholesterol measures Poor standardization Does not represent all atherogenic ApoB-containing lipoproteins 25 Atherogenic lipoproteins: non-hdl-c Non-HDL-C Apo B-containing lipoproteins VLDL IDL LDL Very low-density lipoprotein (VLDL) Made in the liver 1 TG-rich 1 Atherogenic 2 Intermediate-density lipoprotein (IDL) Formed from VLDL by TG lipolysis 1 Also known as a VLDL remnant particle 1 Atherogenic 2 Low-density lipoprotein (LDL) JUST ONE COMPONENT OF RISK Formed from IDL due to further TG lipolysis 1 Taken up by the liver and other organs via the LDL receptor 1 Major atherogenic particle 2 HDL High-density lipoprotein (HDL) Anti-atherogenic: carries cholesterol away from artery wall for metabolism or excretion 1 Other possible antiatherogenic effects include anti-inflammatory and antioxidant actions, and improvement in endothelial function 1 *Formed in the liver and cleared by LDL receptor. Apo=apolipoprotein; sdldl=small, dense low-density lipoprotein; TG=triglyceride. 1. Stone NJ, Blum CB. Management of Lipids in Clinical Practice. 25:24-27, 33, 34, 12. 2. Garg R et al. Prev Cardiol. 25;8:173-177. Are there stronger CVD risk predictors beyond LDL-C?
LDL-C a weaker predictor of CVD vs. non-hdl or ApoB Combined Analysis of >18, patients in TNT & IDEAL trials Measure Hazard Ratio Model 1* Hazard Ratio Model 1 + LDL-C Hazard Ratio Model 1 + Non-HDL Hazard Ratio Model 1 + Apo B LDL-C 1.15 (1.1-1.2) P = <.1.9 (.82-.99) P =.4.95 (.87-1.5) P =.33 Non-HDL-C 1.19 (1.14-1.25) P = <.1 1.31 (1.19-1.44) P = <.1 1.14 (1.-1.3) P =.6 ApoB 1.19 (1.14-1.24) P = <.1 1.24 (1.13-1.36) P = <.1 1.5 (.92-1.2) P =.47 * Model 1: age, sex and study (TNT / IDEAL) ÅòIncreasing LDL inversely associated with CVD events Kastelein JJ et al. Circulation. 28;117:32-39 28 Do I need to measure ApoB/Apo A ratio or is TC/HDL adequate? Answer: Minimal incremental predictive value in measuring Apolipoproteins ApoB better statistical agreement vs. TG and decreased variability Large economic cost What is perhaps the most common metabolic manifestations of obesity?
Hypertriglyceridemia Prevalence TG >5mg/dL 2_%-3% 5-6 MM Patients US Adult Population Total = 217 million Clinical Relevance Risk for pancreatitis Risk for CVD TG 2-499mg/dL ~13% 25-27 MM Patients *This census is outdated as there are roughly 35 million in the US *Numbers exceeding metabolic syndrome cut-point (>15 mg/dl) are much greater Estimated from NCEP ATPIII guidelines data; US Census 23 Does high TG matter or just LDL-C? Incidence of CHD Events According to Serum LDL-C and TG Concentration* CHD Cases / 1 in 8 Years 3 25 2 15 1 5 Baseline TG < 2 mg/dl Baseline TG 2 mg/dl 18 43 47 38 56 112 <13 13-159 16-189 19 Baseline LDL-C (mg/dl) 17 255 * Lipids from 4849 middle-aged men who were followed up for 8 years to record incidence of CHD. Study demonstrated that fasting levels of TGs were an independent risk factor for CHD events, irrespective of serum levels of LDL-C. CHD=coronary heart disease; LDL-C=low-density lipoprotein cholesterol; TG=triglyceride. Assmann G et al. Eur Heart J. 1998;19(suppl M):M8-M14. Hypertriglyceridemia Increases CHD Risk In Patients with good versus bad Ratios * * Bar represents 5% of subjects in which 25% of CHD events occurred Assmann G, Schulte H. Am J Cardiol. 1992;7:733-737 Copyright 1992, with permission from Excerpta Medica Inc. 33
Do high triglycerides matter? Independent predictor in post-acs patients Miller et al J. Am. Coll. Cardiol. 28;51;724-73 Triglycerides summary Independent Risk Factor for CHD Meta-analysis of 17 prospective studies that evaluated fasting TG levels and CVD endpoints 8 studies adjusted for HDL-C Results when adjusting for HDL-C Relative risk (RR) = 1.14 for men, RR = 1.37 for women Higher CVD risk for women Conclusion: elevated TG is a risk factor for CVD independent of HDL-C and the risk is greater for women Note: Triglyceride goals (< 15 mg/dl) are rarely achieved Hokanson JE, Austin MA. J Cardiovasc Risk. 1996;3:213-219 35 Impact of LDL Size Differences At the Same LDL Cholesterol Concentration Up to 7% More Particles 1 mg/dl Large LDL 1 mg/dl Small LDL Cholesterol Balance 36
High TG lead to a decrease in Particle Size & an increase in Particle Number Apo B Fewer Particles More Particles LDL= 13 mg/dl More Apo B Cholesterol Ester Correlates with: TC 198 mg/dl LDL-C 13 mg/dl TG 9 mg/dl HDL-C 5 mg/dl Non-HDL-C148 mg/dl Apo=apolipoprotein; HDL-C=high-density lipoprotein cholesterol; LDL-C=low-density lipoprotein cholesterol; TC=total cholesterol; TG=triglyceride. Otvos JD et al. Am J Cardiol. 22;9:22i-29i. Correlates with: TC 21 mg/dl LDL-C 13 mg/dl TG 25 mg/dl HDL-C 3 mg/dl Non-HDL-C18 mg/dl High Triglycerides Are Associated With LDL Subclass Pattern B = BAD Cumulative Percent 1 8 6 4 2 Pattern A (large, fluffy) Pattern B (small, dense) 4 8 12 16 2 24 28 TGs (mg/dl) LDL=low-density lipoprotein; TG=triglyceride. Austin MA, King MC, Vranizan KM, Krauss RM. Circulation. 199;82:495-56. Small, Dense LDL susceptibility to oxidation vascular permeability Conformational change in ApoB component affinity for LDL receptor for hepatic clearance Association with insulin resistance = high TG and low HDL-C (atherogenic dyslipidemia phenotype) Austin MA et al. Curr Opin Lipidol. 1996;7:167-171 39
Outcomes: Technically, Size doesn t matter particle number does The relationship between small LDL particle size and CHD is intertwined with a complex physiologic syndrome atherogenic dyslipidemia which includes high TG, low HDL-C, and increased LDL particle number Following multivariate analysis that includes TGs, HDL- C, and LDL-particle number among 17 cross-sectional, 8 prospective and 6 interventional trials: LDL particle size is rarely a significant, independent predictor of CHD events Cromwell WC, Otvos JD. Curr Atheroscler Rep. 24;6:381-387 4 Women s Health Study CVD Relative Risk* and Lipid Quartiles P<.1 P<.5 P<.1 P=.8 P=.2 Quartile of Lipid *Adjusted for aspirin or vitamin E use Blake GJ et al. Circulation. 22;16:193-1937 41 Cholesterol (mg/dl) Why use Non-HDL-C Goals? Treatment approach case 1 vs. case 2 225 2 15 1 5 Case 1 Case 2 VLDL LDL HDL VLDL TC= 29 mg/dl TG 12 mg/dl 4 mg/dl VLDL-C 24 mg/dl 88 mg/dl LDL-C 145 mg/dl 89 mg/dl (may not trigger Rx) HDL-C 4 mg/dl 32 mg/dl Non-HDL-C 169 mg/dl 177 mg/dl (should Non-HDL-C trigger goal = LDL-C Rx) goal + 3 TG/HDL-C ratio 3. 12.5 TC/HDL-C ratio 5. 6.5 TC 29 mg/dl 29 mg/dl LDL HDL HDL-C=high-density lipoprotein cholesterol; LDL-C=low-density lipoprotein cholesterol; TC=total cholesterol; TG=triglyceride; VLDL=very low-density lipoprotein. Non-HDL Cholesterol = 177 mg/dl
Because Non-HDL contains Apo-B, all of these particles are involved in Atherogenesis All ApoB lipoproteins: LDL, IDL, VLDL, RLP 21 Modified from Crawford MH, DiMarco JP, editors: Cardiology, London, 21, Mosby. Mosby items and derived items copyright 24, 2 by Mosby, Inc. 43 An Unmet Clinical Need for TG Reduction in Insulin resistant patients: Rx options Niacin Omega 3 s Fibrates Statins Fibrates Primary and Secondary Prevention: limited older data: more recent trials no benefit in FIELD and ACCORD 42% 22% 9% *** 34% 66% *** Primary Prevention Secondary Prevention Frick MH et al. N Engl J Med. 1987;317:1237-1245 Manninen V et al. Circulation. 1992;85:37-45 The BIP Study Group. Circulation. 2;12:21-27 Rubins HB et al. N Engl J Med. 1999;341:41-418 *Post hoc analysis of subgroup with TG > 2 mg/dl and HDL-C < 42 mg/dl **Post hoc analysis of subgroup with TG 2 mg/dl and HDL-C < 35 mg/dl ***Difference between placebo and Rx for primary endpoint was statistically significant (P <.5) 45
Niacin vs. Placebo among Insulin Resistance Patients: older data 6-yr AMI rate by baseline FPG Hazard Ratio.7.76.75.43 2 Placebo Niacin 15 Event Rate, % 1 5 <95 95 14 15 125 FBG, mg/dl n=1119 for niacin; n=2787 for placebo. * z for interaction =.44 (p=.66); indicates homogeneity. CORONARY DRUG PROJECT (CDP) Canner PL et al. Am J Cardiol. 25;95:254-257. 126 Addition of Omega-3 FA esters 4 g/day in Patients Taking Simvastatin with Triglycerides 2-499 mg/dl Parameter Non-HDL-C TG TC VLDL-C Apo B HDL-C LDL-C 137 268 184 52 86 46 91 Simvastatin + Omega-3 (n=122) 123 182 172 37 8 48 88-9. -29.5-4.8-27.5-4.2 +3.4 +.7 141 271 184 52 87 43 88 Simvastatin + Placebo (n=132) BL EOT Median BL EOT % Chg* 134 26 178 49 85 44 85 Median % Chg* -2.2-6.8 <.1-6.3-1.7-7.2-1.9-1.2-2.8 Difference -23.2-3.1-2.3-2.3 +4.6 +3.5 P value <.1 <.5 <.5 <.5 <.5 =.5 *Response to the addition of Omega-3 fatty acids, 4 g/day after 8 weeks of simvastatin 4 mg therapy during lead-in Davidson et al. Apo=apolipoprotein; BL=baseline; EOT=end of treatment; HDL-C=high-density lipoprotein cholesterol; LDL-C=low-density lipoprotein cholesterol; TC=total cholesterol; TG=triglyceride; VLDL-C=very low-density lipoprotein cholesterol. Suggested Treatment Goals In Dyslipidemia Patients at High Risk LDL-C goal (mg/dl) Non-HDL-C goal (mg/dl ApoB goal (mg/dl) Highest-risk patients: Known CVD; or <7 <1 <8 (or 7 per Dr. Eckel) Diabetes + > 1 major CVD risk factor* High-risk patients: No diabetes/known CVD AND >2 major CVD risk factors*; or <1 <13 <9 Diabetes without major CVD risk factors* Adapted from Lipoprotein Management in Patients with Cardiometabolic Risk (Diabetes Care 28; 31:811-822. CVD = cardiovascular disease. * Major CVD risk factors = smoking, HTN, family history of premature coronary artery disease 48
49 Don t forget Lifestyle Lyon Heart Study 24 25-56% P=.3 14 19-65% P=.1 6 17-7% P<.1-61% 8 P<.5 7 Numbers in bars are # of events De Lorgeril M, et al. Arch Intern Med. 1998;158:1161. Circulation. 1999;99:779. Guess my syndrome (X)? Conclusions Insulin resistance (metabolic syndrome) is a growing epidemic BP & lipid control more important than glycemic control or aspirin to lower CVD risk in this primary prevention population Atherogenic dyslipidemia mediates the increased risk for CVD The strongest predictors of CVD include: TC/HDL ratio (or ApoB/Apo A ratio) Non-HDL cholesterol (VLDL, LDL, IDL) LDL is actually a weaker CVD risk predictor Nonetheless, achieving LDL goals is the first step and we have the best prospective data for this paradigm Next, manage high TG and low HDL Exercise Dietary modification Omega-3s, Niacin, Fibrates Though further outcome studies are warranted