Managing Perioperative Diabetes What s new? Kathryn A. Myers MD FRCPC Chair Chief Division of GIM Professor of Medicine Western University

Managing Perioperative Diabetes What s new? Kathryn A. Myers MD FRCPC Chair Chief Division of GIM Professor of Medicine Western University

Objectives: By the end of this session, you will be able to: Identify the perioperative situations that are associated with better outcomes when glucose levels are optimal Recognize complications of diabetes that may result from the perioperative state Recall features of new classes of diabetes medications Evaluate a diabetes medication list in patients with chronic kidney disease

Why is this topic important? Diabetes is common and increasing in Canada and worldwide Individuals with diabetes are more likely to require surgery than those without diabetes Individuals with diabetes are at increased risk of postoperative surgical complications

Perioperative Goals Avoid hypoglycemia Prevent ketoacidosis Avoid marked hyperglycemia

Preoperative assessment Patient-related factors 1. Type 1 or Type 2; others 2. History of DKA/hyperosmolar states/hypoglycemia 3. Microvascular and macrovascular complications Procedure-related factors 1. Duration of surgery 2. Length of time NPO 3. Time of surgery

Does perioperative blood glucose control matter? Preoperative Intra Intraoperative Postoperative

Case You are seeing Deric Minors (DM), a 60 year old man with poorly controlled Type 2 diabetes mellitus who is scheduled for elective cholecystecomy in 2 weeks. He has a history of hypertension, longstanding diabetes and dyslipidemia. He does not have known microvascular or macrovascular complications. He has never had DKA or hypoglycemia. He takes metformin 1000 mg BID and sitagliptin 100 mg daily. His HbA1C is 12 % Should his surgery be delayed?

Does preoperative diabetes control matter? Acute hyperglycemia decreased immune function, increased surgical site infections in animal models Few trials of preoperative HbA1c, interventions Higher rate of complications in CABG with HbA1C above 7% compared with below 10.7 % surgical site infections versus 3.3 % (all patients) - Nicolini et al Int J Surg 2018 Conversation with patient and surgeon for consideration of delay for optimization

Which statement is true with regard to intraoperative glucose control? a) Maintenance of intraoperative blood glucose levels between 5-11 significantly lowers postoperative wound infections following total hip replacement b) Maintenance of intraoperative blood glucose levels between 5-11 significantly lowers postoperative wound infection following CABG c) Maintenance of intraoperative blood glucose levels between 5-11 significantly lowers postoperative for all procedures over 3 hours duration d) There is no high quality evidence that maintenance of intraoperative blood glucose levels between 5-11 significantly lowers postoperative wound infections

Boreland L, et al. Heart & Lung 2015; 44:430-440 CABG, coronary artery bypass grafting Diabetes Canada Guideline 2018; Chapter 16; Malcolm J, et al

2018 Diabetes Canada CPG Chapter 16. In-hospital Management of Diabetes Recommended glycemic targets for hospitalized people with diabetes Hospitalized population with diabetes Blood glucose targets (mmol/l) Non-critically ill preprandial: 5.0-8.0 random: <10.0 Critically ill 6.0-10.0 CABG intraoperatively 5.5-11.1 Perioperatively for other surgeries 5.0-10.0 Acute coronary syndrome 7.0-10.0 Labour and delivery 4.0-7.0 Less stringent targets may be appropriate in terminally ill patients or in people with severe comorbidities CABG, coronary artery bypass grafting NOT FOR COMMERCIAL USE

Does postoperative control of diabetes mellitus matter? RABBIT 2 trial (2011) Small single centre trial Showed better control of diabetes with basal-bolus insulin compared with SSI; 2/3 patients on oral regimen preop Lower rate of surgical site infections, other postoperative complications, higher risk of hypoglycemia

Blood glucose (mmol/l) 2018 Diabetes Canada CPG Chapter 16. In-hospital Management of Diabetes Basal-Bolus (BBI) Regimen Achieves Better Control than Sliding Scale (SSI) Alone RABBIT 2 RABBIT 2 Surgery 13.3 13.3 * 12.2 11.1 10.0 8.9 * * * SSI 11.1 10.0 8.9 * ŧ ŧ SSI 7.8 6.7 5.6 *p < 0.01; p < 0.05. Admit 1 2 3 4 5 6 7 8 9 10 Duration of treatment (days) BBI 7.8 6.7 Randomi -zation *p < 0.001, ŧp = 0.02, p = 0.01 1 2 3 4 5 6 7 8 9 Duration of treatment (days) BBI Adapted from: Umpierrez GE, et al. Diabetes Care 2007;30:2181-86. Adapted from: Umpierrez GE, et al. Diabetes Care 2011;34:256-61.

Does postoperative control of diabetes mellitus matter? RABBIT 2 trial (2011) BUT: Nurse dedicated to rounding on patients re: glucose levels daily 24 hour access to endocrinologist Hypoglycemia in real world practice likely to be more significant Requires team-based approach with hypoglycemia protocols

Back to the case DM decided to delay his surgery as he was not having symptoms, and concentrate on improving his diabetes He lost weight, started on basal insulin (60 units HS), and added an SGLT2 inhibitor to his metformin and sitagliptin His HbA1c is now 6% and he is scheduled for surgery What would you recommend regarding his diabetes regimen preoperatively?

A 60 year old man with Type 2 diabetes mellitus scheduled for cholecystectomy is on empagliflozin 25 mg, sitagliptan 100 mg and long-acting insulin 60 units HS.. He has optimal control (HbA1C 6.0%). It was recommended that he hold his insulin the night before surgery. On arrival, for surgery, he says he feels unwell, with nausea but not vomiting. He denies chest pain, dyspnea, abdominal or fever. The nurse calls you. His vitals are BP 120/60, HR 90, RR24, POC glucose 9 mmol/l. He has not taken any medications other than those for his diabetes.

. Bloodwork shows the following: Na 136, K 4.5 Bicarbonate 10 Urea 8 Creat 85 ABGs: ph 7.22 HCO3=8 pco2 20 po2=95 RA Lactate normal What is the most likely cause of the bloodwork abnormalities?

What is the most likely cause of the bloodwork abnormalities? a) Hyperosmolar hyperglycemic state b) Diabetic ketoacidosis due to unrecognized Type 1 diabetes mellitus c) Euglycemic diabetic ketoacidosis due to SGLT2 inhibitor d) Sepsis

Demonstration of the cascade of clinical events and metabolic changes that contribute sequentially to progressive clinical deterioration and development of full-blown episodes of eudka. Julio Rosenstock, and Ele Ferrannini Dia Care 2015;38:1638-1642 2015 by American Diabetes Association

Back to the case DM was on sitagliptin and metformin when his HbA1C was 12% He was started on an added SGLT2i was added without adequate control and then basal insulin (60 units HS) Were these appropriate, guideline-based choices?

HEALTHY BEHAVIOUR INTERVENTIONS 2018 Start healthy behaviour interventions (nutritional therapy, weight management, physical activity) +/- metformin A1C <1.5% above target AT DIAGNOSIS OF TYPE 2 DIABETES A1C 1.5% above target Symptomatic hyperglycemia and/or metabolic decompensation If not at glycemic target within 3 months, start/increase metformin Start metformin immediately Consider a second concurrent antihyperglycemic agent Initiate insulin +/- metformin If not at glycemic target If not at glycemic target Clinical CVD? YES NO Start antihyperglycemic agent with demonstrated CV benefit empagliflozin (Grade A, Level 1A) liraglutide (Grade A, Level 1A) canagliflozin* (Grade C, Level 2) If not at glycemic target * Avoid in people with prior lower extremity amputation See next page NOT FOR COMMERCIAL USE

Which one of the following is associated with the use of SGLT2 inhibitors? a) Weight gain b) Urogenital infections c) Hypertension d) Diarrhea

Patients with event (%) 2018 Diabetes Canada CPG Chapter 13. Pharmacologic Glycemic Management of Type 2 Diabetes Empagliflozin reduced CV events CV death, non-fatal MI, or non-fatal stroke PBO EMP A HR P NNT3 20 15 CV death, MI, stroke (%) 12.1 10.5 0.86 0.04 63 CV deaths (%) 5.9 3.7 0.62 <0.001 46 Nonfatal MI (%) 5.2 4.5 0.87 0.22 Nonfatal stroke (%) 2.6 3.2 1.24 0.16 Hosp. heart failure (%) 4.1 2.7 0.65 0.002 72 All-cause mortality (%) 8.3 5.7 0.68 <0.001 39 Placebo Empagliflozin 10 5 HR 0.86 95.02% CI (0.74, 0.99) P < 0.001 for non-inferiority p=0.04 for superiority 0 0 6 12 18 24 30 36 42 48 Months No. of patients Empagliflozin 4687 4580 4455 4328 3851 2821 2359 1534 370 Placebo 2333 2256 2194 2112 1875 1380 1161 741 166 Zinman B et al. N Engl J Med 2015; DOI: 10.1056/NEJMoa1504720 NOT FOR COMMERCIAL USE

2018 Diabetes Canada CPG Chapter 13. Pharmacologic Glycemic Management of Type 2 Diabetes Canagliflozin reduced CV events CV death, non-fatal MI, or non-fatal stroke Outcome (per 1000 pt-y) PBO CANA HR P or 95% CI NNT 5 CV death, MI, stroke 31.5 26.9 0.86 0.02 44 CV deaths 12.8 11.6 0.87 (0.72-1.06) Nonfatal MI 11.6 9.7 0.85 (0.69-1.05) Nonfatal stroke 8.4 7.1 0.90 (0.71-1.15) Hosp. heart failure 8.7 5.5 0.67 (0.52-0.87) 63 All-cause mortality 19.5 17.3 0.87 (0.74-1.01) Placebo Canagliflozin HR 0.86 95% CI (0.75, 0.97) P < 0.001 for non-inferiority p=0.02 for superiority No. of patients Canagliflozin 5795 5566 4343 2555 2460 2363 1661 Placebo 4347 4153 2942 1240 1187 1120 789 Neal B et al. N Engl J Med 2017; DOI:10.1056/NEJMoa1611925 NOT FOR COMMERCIAL USE

Patients with event (%) 2018 Diabetes Canada CPG Chapter 13. Pharmacologic Glycemic Management of Type 2 Diabetes Liraglutide reduced CV events CV death, non-fatal MI, or non-fatal stroke PBO LIRA HR P NNT4 CV death, MI, stroke (%) 14.9 13.0 0.87 0.01 53 20 CV death (%) 6.0 4.7 0.78 0.007 77 Nonfatal MI (%) 6.8 6.0 0.88 0.11 Nonfatal stroke (%) 3.8 3.4 0.89 0.30 15 10 Hosp. heart failure (%) 5.3 4.7 0.87 0.14 All-cause mortality (%) 9.6 8.2 0.85 0.02 72 Placebo Liraglutide 5 HR 0.87 95.02% CI (0.78, 0.97) P < 0.001 for non-inferiority p=0.01 for superiority Patients at risk 0 0 6 12 18 24 30 36 42 48 54 Time from randomization (months) Liraglutide 4668 4593 4496 4400 4280 4172 4072 3982 1562 424 Placebo 4672 4588 4473 4352 4237 4123 4010 3914 1543 407 Marso SP et al. N Engl J Med 2016;375(4):311-22. NOT FOR COMMERCIAL USE

Add additional antihyperglycemic agent best suited to the individual by prioritizing patient characteristics (agents listed in alphabetical order by CV outcome data): Class Effect on CVD Outcomes Hypoglycemia Weight Relative A1C Lowering when added to metformin Other therapeutic considerations Cost GLP-1R agonists lira: Superiority in T2DM with clinical CVD exenatide LAR & lixi: Neutral Rare to GI side-effects, Gallstone disease Contraindicated with personal / family history of medullary thyroid cancer or MEN 2 Requires subcutaneous injection $$$$ SGLT2 inhibitors Cana & empa: Superiority in T2DM patients with clinical CVD Rare to Genital infections, UTI, hypotension, dose-related changes in LDL-C. Caution with renal dysfunction, loop diuretics, in the elderly. Dapagliflozin not to be used if bladder cancer. Rare diabetic ketoacidosis (may occur with no hyperglycemia). Increased risk of fractures and amputations with canagliflozin. Reduced progression of nephropathy & CHF hospitalizations with empagliflozin and canagliflozin in those with clinical CVD $$$ DPP-4 Inhibitors alo, saxa, sita: Neutral Rare Neutral Caution with saxagliptin in heart failure Rare joint pain $$$ Insulin glar: Neutral degludec: noninferior to glar Yes No dose ceiling, flexible regimens Requires subcutaneous injection $- $$$$ Thiazolidinediones Neutral Rare CHF, edema, fractures, rare bladder cancer (pioglitazone), cardiovascular controversy (rosiglitazone), 6-12 weeks for maximal effect $$ -glucosidase inhibitor (acarbose) Rare Neutral GI side-effects common Requires 3 times daily dosing $$ Insulin secretagogue: Meglitinide Sulfonylurea Yes Yes More rapid BG-lowering response Reduced postprandial glycemia with meglitinides but usually requires 3 to 4 times daily dosing. Gliclazide and glimepiride associated with less hypoglycemia than glyburide. Poor durability $$ $ Weight loss agent (orlistat) None GI side effects Requires 3 times daily dosing $$$

Add additional antihyperglycemic agent best suited to the individual by prioritizing patient characteristics (agents listed in alphabetical order by CV outcome data): Class Effect on CVD Outcomes Hypoglycemia Weight Relative A1C Lowering when added to metformin Other therapeutic considerations Cost GLP-1R agonists lira: Superiority in T2DM with clinical CVD exenatide LAR & lixi: Neutral Rare to GI side-effects, Gallstone disease Contraindicated with personal / family history of medullary thyroid cancer or MEN 2 Requires subcutaneous injection $$$$ SGLT2 inhibitors Cana & empa: Superiority in T2DM patients with clinical CVD Rare to Genital infections, UTI, hypotension, dose-related changes in LDL-C. Caution with renal dysfunction, loop diuretics, in the elderly. Dapagliflozin not to be used if bladder cancer. Rare diabetic ketoacidosis (may occur with no hyperglycemia). Increased risk of fractures and amputations with canagliflozin. Reduced progression of nephropathy & CHF hospitalizations with empagliflozin and canagliflozin in those with clinical CVD $$$ DPP-4 Inhibitors alo, saxa, sita: Neutral Rare Neutral Caution with saxagliptin in heart failure Rare joint pain $$$ Insulin glar: Neutral degludec: noninferior to glar Yes No dose ceiling, flexible regimens Requires subcutaneous injection $- $$$$ Thiazolidinediones Neutral Rare CHF, edema, fractures, rare bladder cancer (pioglitazone), cardiovascular controversy (rosiglitazone), 6-12 weeks for maximal effect $$ -glucosidase inhibitor (acarbose) Rare Neutral GI side-effects common Requires 3 times daily dosing $$ Insulin secretagogue: Meglitinide Sulfonylurea Yes Yes More rapid BG-lowering response Reduced postprandial glycemia with meglitinides but usually requires 3 to 4 times daily dosing. Gliclazide and glimepiride associated with less hypoglycemia than glyburide. Poor durability $$ $ Weight loss agent (orlistat) None GI side effects Requires 3 times daily dosing $$$

Add additional antihyperglycemic agent best suited to the individual by prioritizing patient characteristics (agents listed in alphabetical order by CV outcome data): Class Effect on CVD Outcomes Hypoglycemia Weight Relative A1C lowering when added to metformin Other therapeutic considerations Cost GLP-1R agonists lira: Superiority in T2DM with clinical CVD exenatide LAR & lixi: Neutral Rare to GI side-effects, Gallstone disease Contraindicated with personal / family history of medullary thyroid cancer or MEN 2 Requires subcutaneous injection $$$$ SGLT2 inhibitors Cana & empa: Superiority in T2DM patients with clinical CVD Rare to Genital infections, UTI, hypotension, dose-related changes in LDL-C. Caution with renal dysfunction, loop diuretics, in the elderly. Dapagliflozin not to be used if bladder cancer. Rare diabetic ketoacidosis (may occur with no hyperglycemia). Increased risk of fractures and amputations with canagliflozin. Reduced progression of nephropathy & CHF hospitalizations with empagliflozin and canagliflozin in those with clinical CVD $$$ DPP-4 Inhibitors alo, saxa, sita: Neutral Rare Neutral Caution with saxagliptin in heart failure Rare joint pain $$$ Insulin glar: Neutral degludec: noninferior to glar Yes No dose ceiling, flexible regimens Requires subcutaneous injection $- $$$$ Thiazolidinediones Neutral Rare CHF, edema, fractures, rare bladder cancer (pioglitazone), cardiovascular controversy (rosiglitazone), 6-12 weeks for maximal effect $$ -glucosidase inhibitor (acarbose) Rare Neutral GI side-effects common Requires 3 times daily dosing $$ Insulin secretagogue: Meglitinide Sulfonylurea Yes Yes More rapid BG-lowering response Reduced postprandial glycemia with meglitinides but usually requires 3 to 4 times daily dosing. Gliclazide and glimepiride associated with less hypoglycemia than glyburide. Poor durability $$ $ Weight loss agent (orlistat) None GI side effects Requires 3 times daily dosing $$$

A 63 year old man with chronic kidney disease (creatinine 190 umol/l, egfr 35) is having a total hip arthroplasty. His HbA1c is 10%. His current medications include metformin 500 mg BID, gliclazide extended release 60 mg daily, sitagliptan 100 mg OD and canagliflozin 100 mg OD. Which of the following of his current diabetes medications can be continued without dosage adjustment at his current creatinine clearance? a) Metformin, gliclazide b) Metformin, sitagliptan c) Sitagliptan, canaglifozin d) Gliclazide, canagliflozin

2018 Diabetes Canada CPG Chapter 13. Pharmacologic Glycemic Management of Type 2 Diabetes Antihyperglycemic Agents and Renal Function Alpha-glucosidase Inhibitors CKD Stage 5 4 3b 3a 1 or 2 egfr (ml/min/1.73 m 2 ): <15 15 29 30 44 60 Biguanides DPP-4 Inhibitors GLP-1 Receptor Agonists Insulin Secretagogues SGLT2 Inhibitors Thiazolidinediones Acarbose Canagliflozin* 25 100 mg daily Dapagliflozin Empagliflozin Pioglitazone Use alternative agent Metformin 500-1000 mg daily Alogliptin 6.25 mg daily 30 12.5 mg daily Linagliptin 15 Saxagliptin 15 2.5 mg daily 50 Sitagliptin 25 mg daily 50 mg daily 50 Dulaglutide Exenatide Exenatide QW 30 50 Liraglutide Lixisenatide Gliclazide Glimepiride Glyburide Repaglinide Rosiglitazone Insulins 15 15 Dose adjustment required Fluid retention 30 30 30 30 30 Caution 45 Do not initiate 45-59 50 30 60 30 60 60 60 30 45 45 60 60 60 30 60 Dose adjustment not required * May be used for cardiorenal benefits in those with clinical CVD, A1C above target and egfr >30 ml/min/1.73m 2 60

Take home points Consider optimization of diabetes mellitus prior to elective surgery in patients with poor control; intraoperative in CABG 5-11 Preoperative and postoperative control likely matters; use more basal insulin and less reactive protocols such as QID SSI as long as hypoglycemia can be avoided New medications (some) SGLT2i and GLP-1 agonists reduce cardiovascular outcomes, renal outcomes Most diabetes medications require dose reductions with CKD use the Diabetes Canada graphic (Appendix )