Diabetes overview Tom Archer MD MBA UCSD Anesthesia

Diabetes overview Tom Archer, MD, MBA UCSD Anesthesia

Diabetic outpatient case: • “Healthy” 48 y. o. diabetic for hernia repair / GA. • Elective, but patient has made plans. • Pt. stopped his metformin 48 hours before surgery, per instructions, and finger stick blood sugar = 357. • Surgeon wants to proceed with “minor procedure in a healthy patient”.

Do we…? • Proceed immediately and manage hyperglycemia “on the fly”? • Delay to later on the same day? • Delay to another day?

Other questions: • What is our blood glucose “cut-off” that makes us delay surgery? • Should we insist on other tests or evaluations before proceeding?

Would you proceed with the case immediately? UTHSCSA faculty response:

Your blood glucose “cut-off”, above which you would not immediately proceed with case. UTHSCSA faculty response:

Other considerations: • Evaluate for dehydration and acidosis. • Get EKG, Chem 7, ABG or urine for glucose / ketones. • Worry about infection and wound healing. • Worry about DKA, MI, stroke, hypotension.

My perspective: • Evidence-based safety, above all. • No UNNECESSARY delays or cancellations. • We have to JUSTIFY delays or cancellations. • Safety does NOT = more tests and delays.

Dangers of hyperglycemia • Long term dangers in asymptomatic patients. • Short term dangers in sick patients. • No studies of short term dangers of HG in healthy patients!

Enormous dangers of hyperglycemia in pregnant patients! • HG alters DNA transcription, causing: – “Diabetic embryopathy”-- birth defects. – “Diabetic fetopathy”– macrosomia and organ immaturity for gestational age (e. g. lung). – Placental vascular disease (IUGR, chronic malnutrition / hypoxia) – Non-specific inflammation, vasoconstriction, coagulation and fibrosis. – Decreased neutrophil / monocyte function. – Neonatal hypoglycemia

Type II DM in 2006 Genetic Obesity Inflammation Insulin resistance predisposition Hyperglycemia Atherosclerosis Decreased insulin output Nephropathy Retinopathy Neuropathy Pancreatic beta cell damage Immune dysfunction Poor wound healing

Hyperglycemia “cries wolf” to the innate immune system– activating it when it is not needed and weakening its capacity to respond to a real infection. portland. indymedia. org accessed on Google images

Inflammation as a cause of disease has entered the popular imagination. Diet (“macronutrients”) is rightly perceived as a factor in causing inflammation.

So, food kills! Calorie restricted mice live 30% longer than normally fed mice. Calorie restricted mice http: //www. lef. org/magazine/mag 2006/images/j an 2006_cover_lef_04. jpg

Insulin is a ANABOLIC hormone • Causes glucose uptake into muscle and liver, amino acid uptake into muscle and free fatty acid (FFA) uptake into adipose tissue. • Insulin affects gene transcription, allowing tissue growth and translocation of GLUT 4 transport protein to the cell membrane. • Insulin is NOT just about blood glucose control!

Insulin enables three distinct stages of glucose utilization Microvascular function—without insulin, microvasculature (capillaries and precapillary sphincters) do not supply blood appropriately to muscle cells. Uptake of glucose into muscle cell requires GLUT 4 transport protein, made in response to insulin. Phosphorylation of glucose to glucose-6 -phosphate by hexokinase inside the mitochondrion, a limiting step of glucose utilization.

European Journal of Endocrinology 150 97– 104

The hyperbolic function in diabetes Obese non-diabetic (insulin resistance, but compensated) B Pancreatic output of insulin C Obese, diabetic (no longer compensated) Peripheral tissue sensitivity to insulin Thin non-diabetic A

The Diabetes Control and Complications Trial • Definitive, landmark 1993 study. 1441 patients with type I DM followed for 9 years. • 35 -70% reduction in retinopathy, neuropathy and nephropathy with intensive blood glucose control.

Hgb A 1 c vs. Microvascular Disease and Myocardial Infarction in type II DM. Incidence per 1000 patient-years Microvascular disease Myocardial infarction Updated mean Hb. A 1 c (%) UKPDS 35. BMJ 2000; 321: 405 -12

DIGAMI study: Intensive blood glucose control after AMI lessens mortality. • Pts. had suspected MI and BG > 200 mg% • Both groups got thrombolysis, aspirin, beta blockers and ACE inhibitors as indicated. Malmberg K DIGAMI study 1999

Admission glucose predicts long term mortality after AMI. Intensive insulin therapy (over months) mitigates effect of admission blood glucose on mortality. <234 mg % 234 – 297 mg % > 297 mg % (Malmberg K DIGAMI study 1999)

Q: How does glycemic control compare with other post- AMI interventions? A: Extremely well! • Absolute reduction of death rate: – – – Thrombolysis: 3. 7% Acute beta blockade: 3. 5% Chronic beta blockade: 9. 3% Long term simvastatin: 10. 4% Aspirin: 3. 8% (CV events). – DIGAMI blood glucose control: 15% absolute reduction of death rate over 3. 4 years!

Intensive insulin therapy in critically ill patients (Van den Berghe et. al. 2001) • Conventional therapy: Average BG = 173. • Intensive therapy: Average BG = 103. • Intensive therapy decreased: – – – mortality (4. 6% vs. 8%) renal failure sepsis polyneuropathy prolonged mechanical ventilation red cell transfusions

Trauma patients– thresholds for damage from HG: • Yendamuri 2003: HG > 135 mg / d. L increased ICU LOS, infection and mortality. • Laird 2004: HG > 200 mg / d. L increased mortality and infection.

Cardiac surgery and hyperglycemia • CPB damages endothelium (apart from its role in causing hyperglycemia). • CPB and hypothermia cause extreme hyperglycemia– which damages endothelium. • CPB is a “double insult” for endothelium.

Tight glycemic control in diabetic CABG patients improves outcomes. Lazar 2004, Ouattara 2005 • Tight blood glucose control (target 125 -200 mg %) significantly reduced: – – – – – Atrial fibrillation Sternal and leg wound infections and pneumonia Time on ventilator Maximum weight gain (edema? ) ICU stay duration Post-op hospital stay Recurrent ischemia Angina class Mortality in first two years

Increasing mean intraoperative blood glucose is associated with increasing morbidity in cardiac surgery Gandhi GY 2005

Furnary 2003 • Best results with average post-CABG glucose < 150 mg %. • Insulin infusion needed until POD #3

Changing practice: downward trend over time in average post. CABG glucose. Furnary 2002

Pre-op Hgb A 1 c > 7% associated with increased infections in surgical patients Dronge AS 2006 • 647 known diabetic VA patients for non-cardiac surgery. • Infectious outcomes followed were: pneumonia, wound infection, urinary tract infection or sepsis. • Retrospective analysis and statistical association only– no proven causation.

So should we delay elective surgery if Hgb A 1 c > 7%? • Discussion of this option at UTHSCSA. “Be careful what you wish for. ” • Be wary of setting unnecessarily stringent standards. • You may get sued if you don’t adhere to your “standard” and there is a problem. • It’s hard to abolish a standard, once set.

Paying (and suing? ) for glycemic control in cardiac and general surgery? • Will glycemic control soon be monitored and rewarded / punished by CMMS and other payers? • “… perioperative maintenance of normoglycemia will become a valid performance measure for practicing surgical specialists. ” • CMMS “target” in Turina study was 200 mg %. “ 150 mg % is better, ” say commenting surgeons. Turina M 2006

Hyperglycemia and the brain: • Cardiac arrest, stroke, neurotrauma and neurosurgery: – Both animal and human studies show that hyperglycemia during or after brain injury causes worse outcomes. Capes SE 2001, Jeremitsky E 2005, Wass CT 1996

Hyperglycemia damages (“activates”) the endothelium. Hyperglycemia causes inflammation. Reinhart K 2002, Dandona P 2005 J Clin Invest, Dandona P 2003 Curr Drug Targets

Hyperglycemia, sepsis and pre-eclampsia all “activate” (damage) endothelium, white cells and platelets, leading to white cell adhesion and infiltration, thrombosis and edema (inflammation). WBC Platelet Hyperglycemia, sepsis or pre-eclampsia Platelets Protein (edema) Archer TL 2006 unpublished

HG damages mitochondria • HG causes excessive entry of electrons (as NADH) into mitochondrial electron transport chain. • Excess electrons create reactive oxygen species, which damage mitochondria.

HG produces advanced glycation end-products (AGE) • Mechanically cross link and stiffen collagen and elastin fibers, decreasing tissue elasticity (e. g. arteries). • Activate AGE receptors on macrophages to produce inflammatory mediators.

Cooper ME 2004

Advanced Glycation Endproducts (AGE) • Stiffen tissues, ? Causing • Atherosclerosis • Diastolic dysfunction • Stiff joints Cooper ME 2004

Can we reverse AGE? • Alagebrium chloride is in phase II trials– breaks AGE cross-links and restores vascular flexibility. • Pimagedine appears to block cross-link formation. • Reversal / prevention of atherosclerosis, systolic hypertension and diastolic dysfunction? Cooper ME 2004

Endothelial cells send molecular signals to surrounding smooth muscle Glucose makes endothelium produce Insulin makes endothelium produce Vessel lumen vasoconstrictive signals (thromboxane, endothelin) Archer TL 2006 unpublished, Idea from Dandona P 2004 vasodilatory signals (NO, prostacyclin)

How rapidly can endothelial “activation” occur? • Evidence from 300 cal of oral glucose vs. 300 cal of vodka. • Oral glucose increases inflammatory markers within 2 hours. • Equal calories as ethanol do not. • Is this relevant to our outpatient scenario? We don’t know. Dhindsa S 2004

Back to the outpatient case… • Why is blood glucose = 357 so bad in ICU patient, yet well-tolerated in many ambulatory patients? • Should we delay the case, since blood glucose in critically ill patients appears to be increasingly dangerous above the 150 mg % range? • Or should we proceed, since no one has shown that short-term hyperglycemia harms asymptomatic patients?

Critical illness and hyperglycemia work synergistically to damage endothelium. • The significance of a blood glucose level depends on what else is going on with the patient’s endothelium. In other words, we have to look at a given blood sugar value in context.

For any given blood glucose value, end organ damage will depend on degree of non-hyperglycemic endothelial damage. Zone of severe end -organ damage Nonhyperglycemic endothelial damage (sepsis, etc. ) Zone of moderate end- limited organ damage end-organ Zone of no end-organ damage Archer TL 2006 unpublished damage Blood glucose

Hyperglycemia may cause different degrees of end-organ damage, depending on the degree of pre-existing endothelial damage. * Patient B Nonhyperglycemic endothelial damage (sepsis, etc. ) Zone of severe end -organ damage Zone of moderate end- limited organ damage end-organ Zone of no end-organ damage * Patient A Archer TL 2006 unpublished Blood glucose

For any degree of non-hyperglycemic endothelial damage, end organ damage will depend on level of hyperglycemia. Zone of severe end -organ damage Nonhyperglycemic endothelial damage (sepsis, etc. ) Zone of moderate end- limited organ damage end-organ Zone of no end-organ damage Archer TL 2006 unpublished damage Blood glucose

So, what do we do with our outpatient with blood glucose = 357? • Discuss benefits of tight glucose control with patient and surgeon. • Use the occasion as a “teachable moment” (Roizen) to stimulate patient toward better health. • Be aware that surgeons themselves are starting to talk about BG = 150 -200 mg% as a reasonable perioperative target for major surgery.

What else should we do? • Consider NOT stopping metformin-- to help keep blood glucose down. – Evidence against perioperative metformin is weak to non-existent. Misbin RI 2004, Holstein A 2005

In summary • Both HG and serious illness “activate” endothelium, causing inflammation, vasoconstriction, coagulation and fibrosis.

In summary • Long term HG increases CV disease, retinopathy, neuropathy and nephropathy in asymptomatic diabetics. • Short term HG causes multiple and serious complications in sick patients-- in whom the endothelium is already compromised.

In summary • However, there is no evidence at present that short term hyperglycemia in an asymptomatic patient confers increased risk of surgical or anesthetic complications. • Until such evidence is forthcoming, caution, “clinical judgment” and common sense must guide our practice.

The End

References (1) • • • • Ahmed N. Diabetes Research & Clinical Practice. 67(1): 3 -21, 2005 Jan. Capes SE et al Stroke 2001; 32: 2426 -2432. Cooper ME American Journal of Hypertension 2004; 17: 31 S-38 S Coutinho M et al Diabetes Care 1999; 22: 233 -240. Dandona P Current Drug Targets 2003; 4, 487 -492. Dandona P et al Med Clin N Am 88 2004, 911 -931 Dandona P et al J Clin Invest 2005, 115: 2069 -2072 Dhindsa S et al Metabolism 2004; 53: 330 -334. Diabetes Control and Complications Trial (DCCT) Research Group. New England Journal of Medicine September 30, 1993 Volume 329: 977986. No 14. Dronge AS et al Arch Surg 2006; 141: 375 -380 Finney SJ et al JAMA 2003; 290: 2041 -2047 Freire AX et al Chest 2005; 128: 3109 -3116 Furnary AP et al J Thorac Cardiovasc Surg 2003; 125: 1007 -21 Gandhi GY et al Mayo Clin Proc July 2005; 80(7): 862 -866

References (2) • • • • Holstein A et al Diabetologia (2005) 48: 2454 -2459. Jeremitsky E et al J Trauma 2005; 58: 47 -50. Khaw KT et al BMJ 2001; 322: 1 -6 Krinsley JS et al Mayo Clin Proc 2004; 79(8): 992 -1000. Laird AM et al J Trauma 2004; 56: 1058 -1062 Langouche L et al J of Clin Invest 2005; 115(8): 2277 -2286. Lazar HL et al Circulation 2004; 109: 1497 -1502 Leiter LA et al AJH 2005; 18: 121 -128 Malmberg K et al Circulation 1999; 99: 2626 -2632 Misbin RI Diabetes Care Volume 2004; 27(7): 1791 -1793 Ouattara A et al Anesthesiology 2005 V 103, No 4, pp. 687 -694 Reinhart K et al Crit Care Med 2002; 30[Suppl. ]: S 302 -312 Turina M et al Crit Care Med 2005 Vol. 33, No. 7, pp. 1624 -1633. Turina M et al Annals of Surgery vol 243 number 6 June 2006 UKPDS Lancet 1998; 352: 854 -865.

References (3) • Umpierrez GE et al J Clin Endocrinol Metab 2002 87: 978 -982. • Viberti G New England Journal of Medicine Vol 332: 1293 -1294, May 11, 1995, Number 19. • Van den Berghe G et al N Engl J Med 2001; 345: 1359 -67 • Vanhorebeek I et al Lancet 2005; 365: 53 -59. • Vanhorebeek I et al Curr Opin Crit Care 2005, 11: 304311 • Wass CT et al Mayo Clin Proc 1996 Vol 71(8): 801 -812. • Yendamuri S et al J Trauma 2003; 55: 33 -38. • Whitcomb BW Crit Care Med 2005; 33: 2772 -2777