Type II Diabetes

1. Type II Diabetes G. Michael Allan Associate Professor, Family Med, U of A

2. Conflict of Interest Disclosure • Family Doctor & Academic • Pay from University and Alberta Health • Research and Speaking Fees • Non-Profit Sources (CFPC, ENPCN, TOP, etc) • No funding from Industry

3. Objectives • Metabolic Syndrome (Deferred to Topic VIII) • Diagnosis • Monitoring Sugars (and potential outcomes) • Management of Sugars • Targets: • Early after Diagnosis • If Sugars are consistently high. • Oral Medications • Starting Insulin • Management of Additional risks • Risk Assessment: Is Diabetes a Coronary Heart Disease Equivalent. • Anti-Platelets • Hypertension • Lipids

4. Screening • BMI ≥25 plus one of,… • Physical inactivity, • Family History (first degree) • High risk Ethnic group (African American, native, etc) • PCOS, large (9lb) baby or Gestational DM • CVD risks (hypertension, lipids, or history of CVD) • History of Impaired glucose or ++ obesity • No risk factors: age ≥45 • Test every three years DIABETES CARE 2010; 33(1): S11-61

5. Diagnosis • AIC ≥6.5% • FPG ≥126mg/dl (7.0 mmol/l) post 8 hr fast • 2-h plasma glucose ≥200mg/dl (11.1 mmol/l) during an OGTT. Should follow a 75g glucose load. • If classic symptoms of hyperglycemic, a random plasma glucose ≥200mg/dl (11.1 mmol/l). • Tests 1-3 should be confirmed unless there is unequivocal hyperglycemia. DIABETES CARE 2010; 33(1): S11-61

6. Blood Sugar Monitoring • Clinical Question: What are the pros and cons of self-monitoring blood glucose for type 2 diabetics’ not using insulin?

7. Blood Sugar Monitoring • Evidence: Two recent well-designed randomized controlled trials compared self-monitoring to no self-monitoring in type 2 diabetics using diet or oral therapies only. • DiGEM1 included 453 diabetics (mean HgbA1c 7.5%) and ESMON2 184 new diabetics (mean HbA1c 8.7%). • After 12 months, there was no statistical difference in HbA1c, starting of new drugs, or weight/BMI in either study.1,2 • One third to a half of patients were not following the self-monitoring protocol by trial end.1 • There was a statistically significant worsening of depression scores (mean 6% worse/person) in the self-monitoring subjects.2

8. Blood Sugar Monitoring • Context: • Research preceding best summarized in a meta-analysis3: self-monitoring reduced HbA1c by 0.39%. • Questionable clinical significance and, • Likely exaggerated as higher quality studies (2 of 6 in this analysis) showed no difference.3 • Recent DiGEM1 and ESMON2 are higher quality, show no benefit in self-monitoring and identify some negatives. • Worsening of depression scores noted in ESMON2 • Others studies show worsening depressive symptoms4 • Negative impacts on quality of life.4,5 • 3 studies show regular self-monitoring is not cost-effective.5,6

9. Blood Sugar Monitoring • While regular self-monitoring in type 2 diabetics not on insulin appears unnecessary, this population should still know how to test their sugar in case it is low, they are feeling ill or they are interested in seeing the impacts of lifestyle behaviors. • Bottom-line: The present good quality evidence indicates that regular self-monitoring of blood glucose in type 2 diabetics not on insulin gives no benefits, is not cost-effective and may have negative consequences (like worsening on depression scales). 1. BMJ 2007;335;132-40. 2. BMJ 2008;336:1174–7. 3. Cochrane. 2005;2:CD005060 & Diabetes Care 2005;28:1510–7. 4. Diabetes Care. 2001;24(11):1870-7. 5. BMJ 2008;336:1177–80. 6. CMAJ 2010; 182(1): 28-34 & CMAJ 2010; 182(1): 35-38.

10. Getting Aggressive

11. Aggressive Blood Glucose • Question: What do recent meta-analysis and RCTs say about aggressive glucose control?

12. How does Aggressive Control Sound? • Look first at 2 high profile Meta-analysis • Lancet1: 5 RCTs (33,040 pts, mean age 62, 62% ♂) • Generally poorly done: Ex: Added PROactive, drug trial, not aggressive BG. • Non-fatal MI (4% vs 4.8%, NNT 131) • Combined CHD (6.1% vs 7.2%, NNT 93) • No diff in Mortality, CVD death, lots of hetero. • More wgt gain (2.5kg mean), 2.3% vs 1.2% severe hypo Lancet 2009; 373: 1765–72

13. How does Aggressive Control Sound? • Annals: 5 trials (27,802 pts), 3.4-10.7 years, • If new DM (UKPDS), age 53, • If later DM age low 60s • Non-fatal MI (AR 0.9%, NNT 112) x 5 years • Combined CVD (1.5%, NNT 67) driven by non-fatal MI • Note: looking at 3 most recent trials alone, never reach Stat Sign. • No diff in Mortality, CVD death, lots of hetero. • AR 3.9% severe hypo, NNT 26 over 5 years. • Bottom-Line: These data should not be combined. For patients with DM a while and higher HbA1c, uncertain risk/benefit balance • Ann Intern Med. 2009;151:394-403.

14. Example of Heterogeniety

15. Pooling hypoglycemia?

16. How does Aggressive Control Sound? • Other Authors, same data • “Our review and critique of recent large randomized trials in patients with type 2 diabetes suggest that tight glycemic control burdens patients with complex treatment programs, hypoglycemia, weight gain, and costs and offers uncertain benefits in return. We believe clinicians should prioritize supporting well-being and healthy lifestyles, preventive care, and cardiovascular risk reduction in these patients.”1 • Also Compare BP control in DM (NNT for CVD & mortality 8 to 30) Ann Intern Med 2008;148:855–68 or statin in DM NNT for CVD 23 Lancet 2008;371:117–25. Ann Intern Med. 2009;150:803-808.

17. Looking at the RCTs • ACCORD (10K pts, x 3.5 yrs) & ADVANCE (11K pts, x 5 yrs). Mean age 60’s, 40% ♀ • HgbA1C: 6.4/6.5 vs 7.5/7.3 (target ≤6/6.5) • CVD events & CVD mortality: No difference • Lots more doctors visits, phone calls, dose changes, drugs, wgt gain, hypoglycemia, etc. N Engl J Med 2008;358:2545-59. N Engl J Med 2008;358: 2560-72

18. Significant Outcomes of ACCORD & ADVANCE

19. Living Sugar Free & DM Health • RCT (5.6 yrs), ~1.8Kpts, 60y.o., 97%♂, DM x11.5 yrs, CVD 40%, mean HgbA1C = 9.4% • HgbA1C 8.4% reg vs 6.9% intense • No diff in macro OR micro-vascular outcomes, but slight increase in all albuminuria (6.6% vs 4.1%), NNT 40. • If look at trends: some better but some worse (like death) • Intense: ↑4kg, ↑1.5 BMI, Symptomatic Hypoglycemia (3.8/pt vs 13.3/pt) • Limits: males, later DM, • Agrees with other (but NOT guidelines!!!). NEJM 2009;360: (e-pub Dec 17, 2008).

20. A1C and Mortality: Is Lower Better • Cohort Study: GP Research Database, UK. • 47,970 pts (Type 2 DM ≥50, mean age 64, 55% male) • Followed ≤22 years, HbA1C mean over the study. • Outcome: Best Mortality Risk around A1C 7.5% • Hazard Ratio for A1C 6.4% = 1.52 • Hazard Ratio for A1C 10.5% =1.79 • Concerns: Cohort and funding? • Agrees with new RCTs Lancet 2010; 375: 481–89

21. Mortality by A1C Lancet 2010; 375: 481–89

22. What about new DM • 10 year post UKPDS trial1: 17 year follow up, 3277 patients (1525 completed) • tight glycemic control (irrespective of what use): ↓ death, MI • Metformin still > Intense > conventional • A1c increases over time • Initial UKPDS2 • Insulin or sulfonylurea tight glycemic control: • ↓ microvascular disease, trend ↓ in MI, no diff in death • ↑ weight gain and hypoglycemia • Metformin: ↓ micro, MI & death but less weight gain & hypoglycemia 1. NEJM 2008; 359: 2. Lancet 1998; 352: 837-53 and 854-65

23. Aggressive Blood Glucose Control • Bottom-line: • In long standing Type II DM, aggressive control risks may not be outweighed by benefits • In New Type II DM, aggressive control, particularly with Metoformin, benefits appear to out-weight the risk, particularly in Long term.

24. But wait,… what about Microvascular outcomes

25. Micro-vascular: No small deal? You are seeing a 65 year old ♀ with diabetes for 10 years. Her A1C is now at 8.0%. She has microalbuminuria for the first time. Dr Allan has been going on and on about lack of CVD benefit for A1C less than 7 but you wonder if there are benefits in microvascular disease (as suggested by UKPDS). Question: What are the micro-vascular benefits of tight glucose control?

26. Micro-vascular: No small deal? In patients with Diabetes, the micro-vascular effect of tight control include, Reduced microalbuminuria Reduced cataract surgery Retained light touch Retained visual acuity Visual acuity so good, you can see the increase in mortality.

27. Micro-vascular: No small deal? If you have 100 patients with DM x 10 yrs, & 50% with micro or macroalbuminuria, followed 7 years How many will need dialysis & how many will die? Dialysis: Maybe 1 (0.7%) & death: 40% (1/2 cardiac)

28. Micro-vascular: No small deal? If you have 100 patients with DM x 10 yrs, & 50% with micro or macroalbuminuria, followed 7 years How many will need dialysis & how many will die? Dialysis: Maybe 1 (0.7%) & death: 40% (1/2 cardiac) Diabetes Care 2003;26:2353-8 UKPDS 10 year data

29. Micro-vascular: No small deal? ACCORD1: 10251 DM pts with CVD or 2+ risk mean age 62, DM x 10 years, 38% ♀ Intensive (<6%) 6.4% vs Standard (7-8%) 7.5% 1) ACCORD: Lancet published on-line: June 29, 2010 DOI:10.1016/S0140-6736(10)60576-4.

30. Results: Primary: Development of renal failure (initiation of dialysis or ESRD, renal transplantation, or rise of serum creatinine >291・72 μmol/L), or retinal photocoagulation or vitrectomy to treat retinopathy. Secondary adds neuropathy

31. Micro-vascular: No small deal? Results show some benefits in some patient oriented outcomes (like visual loss, light touch loss, cataracts, etc). Bigger outcomes no difference, because rare. Bottom-line: “The observed benefits associated with intensive glycaemia management should be weighed against higher total and cardiovascular-related mortality, weight gain, and severe hypoglycaemia in patients at high risk of cardiovascular disease. Hence, caution should be exercised in pursuit of a strategy of intensive glycaemic control for prevention of microvascular complications in patients with established type 2 diabetes”

32. The Best Medicine

33. Oral Medications • Metformin first: UKPDS showed 35% reduction (above) • Meta1 showed Metformin is superior to other drugs • NNT 56 for CV mortality (vs other oral meds) • After that, things become more difficult,… 1. Arch Intern Med. 2008;168:2070-80 2. N Engl J Med. 2007;357:28-38. N Engl J Med. 2007;356:2457-71. JAMA. 2007;298(10):1189-1195 3. Lancet 2009;373:2125–35. 4. JAMA. 2007;298(10):1180-1188

34. Oral Medications • Limit use of Rosiglitazone. • 3 meta-analysis show increased harms1-3 • MI: Odds ratio=1.28-1.8, RR=43%, NNH=52 (x 5 yrs) • Additional harms • CHF: Odds Ratio=1.93 (CI 1.30-2.93)3, (NNH=71)4 • Fracture (absolute 3%, NNH 34)4 • Authors views = relationship with manufacturer5 • American Diabetes Association and the European Association 2009: recommend against use of Rosi.6 1. N Engl J Med 2007;356: 2457-71. 2. Arch Intern Med 2010; 170(14):1191-1201. 3. FDA analysis (Accessed August 6, 2010, 4. Lancet 2009;373:2125–35. 5. BMJ 2010;340:c1344. doi: 10.1136/bmj.c1344. 6. Diabetes Care. 2009;32(1):193-203

35. Interpreting MI outcome: Study/Industry vs FDA at the same reports Food and Drug Administration. Briefing document: July 13-14, 2010 meeting of the Endocrinologic and Metabolic Drugs Advisory Committee. (Accessed August 6, 2010)

36. Starting Insulin • Question: In patients with Type II Diabetes inadequately controlled on oral therapy, what is the optimal regimen for initiating insulin in type 2 DM? • How good are family practitioners at doing this?

37. Starting Insulin • 4-T study1: 708 patients x 3 years on either: long-acting basal insulin once per day, biphasic mixed insulin twice a day or prandial insulin with meals.1 • HbA1c levels not significantly different between 3 groups • Sign. more patients in the basal & prandial groups attained a HbA1c ≤7.0% (63% & 67% vs 49% biphasic). • Basal insulin had significantly • Less weight gain (3.6kg) vs prandial (6.4kg) or biphasic insulin (5.7kg), • Fewer hypoglycemic events/person/yr (1.7 basal vs 3.0 biphasic, vs 5.7 prandial) • More patients requiring a second type of insulin (82% basal, 74% prandial and 68% biphasic) • Higher total dose of insulin (by weight). • Size, duration and three comparator arms makes this primary study.

38. Starting Insulin • APOLLO2: 418 pts x44 wks: basal insulin or prandial insulin. • No difference in HbA1C and number to HbA1c ≤7.0% not compared • Confirmed symptomatic hypoglycemia: prandial (7.4/pt) vs basal (1.73 /pt) ss. • No stat differ in weight gain (3.01kg prandial vs 3.54kg basal). • Treatment satisfaction slightly more for basal insulin (over prandial). • INITIATE3: 233 pts x 28 wks: basal insulin or biphasic (BID) • HbA1c was ss lower for biphasic insulin (6.9%) vs basal insulin (7.4%) • 66% biphasic vs 40% basal reached a HbA1c ≤7.0% ss (NNT 4). • Biphasic group (vs basal) had ss higher • Total insulin dose (78 vs 51 units), weight gain (5.4kg vs 3.5 kg) and minor hypoglycemic episodes (3.4/patient vs 0.7/patient). • JDDM 114: 160 pts x6 months: biphasic (BID) insulin or intense insulin (prandial +/- basal) • HbA1C similar & no difference in number attaining HbA1c ≤7.0%. • 63% of the intense group used basal insulin. • Biphasic group greater increase in BMI. • No difference in major hypoglycemic events.

39. Starting Insulin • INSIGHT5 found initiating basal insulin in poorly controlled type 2 diabetes resulted in significantly lower HbA1C than continued oral hypoglycemic agents. • Mean HbA1c and hypoglycemic rates were not different between Family practitioners and diabetic experts patients.6 • Specialists are five times more likely to initiate insulin than family practitioners.7 • Bottom-line: In type 2 diabetes poorly controlled with oral agents, complex insulin regimens are no better than basal insulin and can cause poorer excess weight gain and hypoglycemia. Family practitioners who start insulin are as effective as specialist. 1. NEJM 2009;361(18):1736-47. 2. Lancet 2008; 371(9640): 1073–84. 3. Diabetes Care 2005; 28 (2): 260-5. 4. Diabetes Res Clin Pract. 2008;79(1):171-6. 5. Diabet Med 2006; 23: 736-42. 6. Can Fam Phys, April 2008; 54: 550 - 58. 7. Diabet Care 2005;28(3):600-6.

40. Adjunctive Therapies

41. Managing Diabetes • A 58 year old woman was diagnosed with DM 6 years ago. Recently she has developed micoralbuminuria (2 repeat readings). • She has been reluctant to embrace her DM in the past. • You start into the DM talk again and she seems unconvinced. What can you tell her?

42. Managing Diabetes • Intensive management of DM over 10+ years can result in: • Decreased mortality, NNT 5 • Decreased CVD events, NNT 4 • Decreased severe Retinopathy, NNT 7 • Decreased Dialysis, NNT 16 • No change important changes • All of the above

43. Managing Diabetes • STENO: Small RCT (160 pts x7 yrs then 130 x6) • All DM + microalbuminuria (Danish, white) • Standard vs intense (only 1 hit all 5 targets) • Interventions: ACE, Diuretic, Statin, ASA (+ metformin), & twice the exercise. • Mortality: Tx 30% vs 50%, NNT 5 • Macro: ≥1 CVD event 31% vs 60% NNT 4 • If CVD event: Tx had 2 average, Control had 3.3 • Micro: Laser Tx Retinopathy 18% vs 34%, NNT 7 • Dialysis 1% vs 7.5%, NNT 16. N Engl J Med 2008;358:580-91.

44. Diabetes is Coronary Heart Disease Risk Equivalent! 1. Arch Intern Med. 2003 Jul 28;163(14):1735-40 2. Diabet. Med 2009; 26: 142–148. Diabetes Care. 2006;29(2):391-7. Arch Intern Med. 2004;164(13) :1438-43 • DM is likely not a CVD risk equivalent. • Although some have studies suggest possible CVD equivalent risks with DM1 • Multiple studies have show it is not a CVD risk equivalent2 • To assess your patients CVD risk, the UKPDS risk engine is generally recommended: http://www.dtu.ox.ac.uk/index.php?maindoc=/riskengine/download.php • Also University of Edinburgh Cardiovascular Risk Calculator http://cvrisk.mvm.ed.ac.uk/calculator/calc.asp

45. Hypertension • Targets1 of 130/80 • Some good data to support to BP <140 & <80 • Decision for 130/80 on epi-data (>115/75 increased risk) • Regardless of target, Hypertension is the most important risk factor in DM. Relative risk reductions with different interventions in DM2 1) Diabetes Care 2010; 33(1): S11-61 2) Ann Intern Med. 2008;148:846-854. Lancet 2009;373:1765–72. Lancet 2008; 371: 117–25. Ann Intern Med. 2003;138:587-592.

46. Anti-platelets

47. Primary Prevention: ASA in DM 1. JAMA. 2008;300(18):2134-2141. 2. BMJ. 2008 Oct 16;337:a1840. 3. Diabetes Care. 2003 Dec;26(12):3264-72 • 2 recent RCTs focusing on ASA in Diabetics have failed to show a benefit. • 2,500 DM pts, 4.4 yrs f/u, low dose ASA or placebo • No diff in CVD1 (Limited by lower events than expected) • 1270 DM pts (& asymptomatic peripheral artery disease), 6.7 yrs f/u, low dose ASA vs placebo • No Difference in CVD2 • DM pts in PPP study3 of ASA: unexpected findings • Diabetics (1000 pt) had no difference in CVD outcomes • While non Non-diabetics (3000 pts) with ≥1 risks had statistically important benefit with ASA

48. Primary Prevention: ASA in DM 1. Diabetes Metab. 2006 Sep;32 Spec No2:2S52-6. 2. 1 Int J Cardiol. 2008 Sep 26;129(2):172-9. Others have suggested that cause is due to Anti-platelet resistance being more common in DM but we are unsure why diabetics don’t seem to get the expected results from ASA.1 This has led some to say: ASA should not be a standard in DM2

49. Primary Prevention: ASA in DM 2. Can J Diabet 2008; 32: S102-06. 3. Diabetes Care 2010; 33(1): S11-61 4. Diabetes Metab. 2006 Sep;32 Spec No2:2S52-6. 5. Diabet Care 2008; 31 (S2):S222-25 • New CDA Guidelines2: ASA for diabetics at high risk for CVD • ♂ ≥ 45 years, ♀ ≥ 50 years OR • Microvascular or macrovascular disease, family history, diabetes > 15 years or extreme level of a single risk factor (LDL > 5.0) • US guidelines3: Consider aspirin is ↑ CVD risk (10-year risk >10%). • Most men >50 or women >60 years of age plus • One additional major risk factor.

50. Primary Prevention: ASA in DM 2. Can J Diabet 2008; 32: S102-06. 3. Diabetes Care 2010; 33(1): S11-61 4. Diabetes Metab. 2006 Sep;32 Spec No2:2S52-6. 5. Diabet Care 2008; 31 (S2):S222-25 6. BMJ 2009;339:b4531 • However, the previous studies didn’t find benefit in some high risk (PAD) groups. • Others suggest that • That it be based on individual patient risks • Or schemes such as if: 10 yr risk ≥15% then ASA, 7%-14% consider ASA or ≤6% no ASA. • Just in: recent meta-analysis6 confirms limited evidence to support use in DM • Maybe men (not in bigger RCT) but lots of heterogeneity