Transition of Care in Hospitalized Patients with Hyperglycemia and Diabetes

Transition of Care in Hospitalized Patients with Hyperglycemia and Diabetes Critically ill patients in the ICU Hospital Non-ICU Settings Home Guillermo E Umpierrez, MD, FACP, FACE Professor of Medicine Emory University School of Medicine Director, Diabetes and Endocrinology Section Grady Health System Atlanta, Georgia Learning Objectives Describe strategies for treating and prevent rebound hyperglycemia during the transition from ICU to regulars floor Outline processes and procedures for an appropriate transition from the hospital to outpatient care 1

Diabetes Epidemic in the U.S. US Population Inpatient Diabetes 21.9 million people 5518 2778 Diabetes prevalence quadrupled, from 5.5 million to 21.9 million between 1980-2014 CDC s Division of Diabetes Translation. http://www.cdc.gov/diabetes/statistics. 23% of all discharges 8-9 million discharges Annual cost: $124 billion (2012) ADA. Diabetes Care. Mar 6 2013; HCUP Nationwide Inpatient Sample (NIS) 2012. http://hcupnet.ahrq.gov/hcupnet.jsp. What Glucose Level Predicts Hospital Complications? ADA, AACE, Endo N= 55,530 patients records in ICU and non-icu, Emory University Hospitals. Composite of complications: pneumonia, acute renal or respiratory failure, acute MI, bacteremia, and death. Umpierrez et al. Endocrine Society Annual Meeting, 2014 2

Mean BG (mg/dl) 5/19/2016 Hospital Mortality and Complications in Patients with Hyperglycemia and Diabetes ICU Non-ICU Mortality and Hospital Complications in General Surgery Patients >300 200-300 146-199 111-145 No History Diabetes Nondiabetics 153,910 History Diabetes Diabetics 62,868 % # Odd Ratio Odd Ratio 216,775 consecutive first admission 177 surgical, medical, cardiac ICUs 73 geographically diverse VAMC p = 0.1; p= 0.001 #p=0.017 A Frisch et al. Diabetes Care, 2010 Falciglia et al, Crit Care Med 2009 Insulin is the Preferred Treatment for Hyperglycemia in the Hospital Setting Critically ill patients in the ICU IV insulin infusion Non-critically ill patients Basal + prandial regimen ICU=intensive care unit; IV=intravenous Handelsman Y, et al. Endocrine Practice. 2015;21(suppl 1):1-87.; Moghissi ES et al; American Association of Clinical Endocrinologists; American Diabetes Association. Endocr Pract. 2009;15(4):353 369. American Diabetes Association. Diabetes Care. 2009;32(suppl 1):S1 S110. 3

Glucose (mg/dl) Blood Glucose (mmol/l) Blood Glucose (mg/dl) BG, mg/dl 5/19/2016 Why Insulin is the Most Appropriate Agent for Critically Ill Hospitalized Patients Most potent glucose-lowering agent Critically Ill Patients IV Insulin Rapidly effective Easily titratable (up or down) No real contraindications Moghissi ES et al; American Association of Clinical Endocrinologists; American Diabetes Association. Endocr Pract. 2009;15(4):353 369. American Diabetes Association. Diabetes Care. 2009;32(suppl 1):S1 S110. Strategies for Achieving Glycemic Targets in the ICU Leuven SICU Study 1 Yale Insulin Infusion Protocol 2 14 12 10 Intensive - Mean BG 103 mg/dl Conventional - Mean BG 153 mg/dl 450 400 350 8 300 6 250 200 4 150 2 0 Admission Day 1 Day 5 Day 15 Last day 100 50 0 0 12 24 36 48 60 72 Hours Glucommander 3 450 400 350 300 250 200 150 100 50 0 0 2 4 6 8 10 12 14 16 18 20 22 24 Hours NICE-SUGAR 4 180 160 CIT 140 IIT 120 108 100 80 0 Baseline 1 2 3 4 5 6 7 8 9 1011 121314 Days After Randomization 1. Van den Berghe et al. N Engl J Med. 2001;345:1359-1367. 2. Goldberg PA et al. Diabetes Care. 2004;27:461-467. 3. Davidson et al. Diabetes Care. 2005;28:2418-2423. 4. Finfer S, et al. N Engl J Med. 2009;360(13):1283-1297. 4

Transition from iv to sc insulin Protocols T1DM T2DM Stress hyperglycemia Umpierrez GE et al; Endocrine Society. J Clin Endocrinol Metab. 2012;97(1):16 38. Transition From IV Insulin to SC Insulin IV insulin should be transitioned to SC insulin therapy when patient begins to eat and BG levels are stable All patients with T1D All patients with T2D treated with insulin prior to admission Most patients with T2D treated with oral agents prior to admission Most patients with stress hyperglycemia requiring CII at a rate 2 units/hour Umpierrez GE et al; Endocrine Society. J Clin Endocrinol Metab. 2012;97(1):16 38. 5

Transition From IV Insulin to SC Insulin Because of short half-life of IV insulin, SC insulin should be administered prior to discontinuing the drip NPH: 1-2 hours Glargine and detemir: 2-4 hours If short-acting insulin also administered, IV insulin may be able to be stopped sooner, eg, after 1 hour Umpierrez GE et al; Endocrine Society. J Clin Endocrinol Metab. 2012;97(1):16 38. Calculating the SC Insulin Dose Establish the 24-hour insulin requirement by extrapolating from the average IV insulin dose required over the previous 6 8 hours (if stable) Take 80% of the total daily dose (TDD) and give one half as an intermediate-acting or long-acting insulin for basal coverage and one half as a short-acting or rapid-acting insulin in divided doses before meal (If patient is not eating, just give intermediate/longacting insulin.) Umpierrez GE et al; Endocrine Society. J Clin Endocrinol Metab. 2012;97(1):16 38. 6

Percent of Total Glucose Levels within the 80-140 Range on Glargine % Glucose levels within 80-140 70 60 50 40 30 20 10 0 40% 60% 80% Lantus Dose Group Schmeltz et al. Endocr Pract 12:641 650, 2006 Steps in the Initiation of a Basal-Bolus Insulin Regimen To estimate insulin Calculate dose estimated when no intravenous total daily insulin dose therapy of insulin has (type been given: 2 diabetes) 0.2 0.3 unit/kg/day in patients >70 years and/or GFR <60 ml/min Step 1 0.4 unit/kg/day if BG between 140 and 200 mg/dl and not meeting above criteria 0.5 unit/kg/day if BG between 201 and 400 mg/dl and not meeting above criteria Step 2 Step 3 Divide total daily dose (TDD) of insulin into 50% basal (long-acting insulin analog) and 50% nutritional (rapid-acting insulin analog) Give basal insulin (glargine/detemir) once daily or NPH twice daily at the same time each day Give nutritional insulin (rapid-acting insulin analog) in 3 divided doses before each meal, so long as consistent carbohydrate intake is ensured. Rapid-acting insulin dosing should be held if a patient is unable to eat Provide supplemental (correction) insulin in addition to basal and nutritional Adjust insulin dose(s) according to results of bedside BG measurements Initiation of insulin must be individualized, and elderly residents may require a lower starting dose. A diet with consistent carbohydrate intake should be emphasized in conjunction with a basal-bolus regimen For additional weight-based dosing options, please see: DeSantis AJ et al. Endocr Pract. 2006;12(5):491-505. Umpierrez GE et al. Diabetes Care. 2007;30(9):2181-2186. Lansang MC, Umpierrez GE. Diabetes Spectr. 2008;21(4):248-255. 7

Plasma Insulin (mu/ml) Serum Insulin Level 5/19/2016 SC Insulin Administration Scheduled (SSI only uses this component) Correction Basal Bolus (Nutritional) Correction Total daily insulin needs Basal Nutritional Long-acting insulin Rapid-acting insulin Clement S et al. Diabetes Care. 2004;27:553 591. Moghissi ES et al; American Association of Clinical Endocrinologists; American Diabetes Association. Endocr Pract. 2009;15(4):353 369. Transition from IV to SC Insulin Basal-bolus Therapy Basal Analog vs. NPH 75 50 Breakfast Lunch Dinner Bolus (nutritional) insulin Basal insulin Basal analog NPH 25 Correction insulin 0 0 8:00 12:00 16:00 20:00 24:00 4:00 8:00 Time 1. Moghissi ES et al; Endocr Pract. 2009;15(4):353 369. 2. Bray B. Consult Pharm. 2008;23(suppl B):17 23. 3. Roberts G et al. Med J Aust. 2012;196(4):266 269. 4. Umpierrez GE et al. Diabetes Care. 2011;34(2):256 261. 0 24 Time (hours) 1. Brunton S et al. J Fam Pract. 2005;54(5):445 452. 2. 2. Newton et al. Dean trial; JCEM 2009 8

Mean BG after surgery % BG 80-140 mg/dl BG < 60 mg/dl Yeldandi & Baldwin. DIABETES TECHNOLOGY & THERAPEUTICS 8 (6) 2006 Mean Daily Glucose and Hypoglycemia During Transition to SC Insulin after Resolution of DKA Umpierrez et al, Diabetes Care 32:1164 1169, 2009 NPH/ Regular Glargine/G lulisine P value Day 1 188 ± 61 213 ± 76 0.234 Day 2 206 ± 71 220 ± 61 0.370 Day 3 207 ± 86 180 ± 80 0.417 Day 4 211 ± 63 158 ± 44 0.068 Day 5 190 ± 45 124 ± 41 0.068 Hypoglycemia NPH/ Regular Glargine/G lulisine P value Patients with BG <70 mg/dl, n (%) 5 (15) 14 (41) 0.03 Episodes of BG <70 mg/dl, n 8 26 0.019 Patients with BG <40 mg/dl, n (%) 1 (3) 2 (6) NS Episodes of BG <40 mg/dl, n 1 2 NS Data for glucose levels are means ± SD. 9

Blood Glucose (mg/dl) Blood Glucose (mg/dl) 5/19/2016 Why Not Sliding Scale Insulin? Definition Use of a mealtime insulin, typically regular insulin, as the sole insulin for managing a patient s diabetes Potential problems Poor control of hyperglycemia (does not address basal insulin needs) Insulin stacking Hypoglycemia Not preferred method of SC insulin delivery Umpierrez GE et al; Endocrine Society. J Clin Endocrinol Metab. 2012;97(1):16 38. American Diabetes Association. Diabetes Care. 2014;37(suppl 1):S14 S80. Browning LA, Dumo P. Am J Health Syst Pharm. 2004;61(15):1611 1614. Hirsch IB. JAMA. 2009;301(2):213 214. Glucose Levels During Basal-bolus and SSI Treatment Changes in BG concentration after the first day of treatment with basal-bolus with glargine once daily plus glulisine before meals ( ) and with SSI 4 times daily ( ). P <0.001, P = 0.02, P = 0.01. A 220 200 180 160 140 B 120 Randomization 220 1 2 3 4 5 6 7 8 9 Duration of Treatment (days) Glucose levels before meals and bedtime. Premeal and bedtime glucose levels were higher throughout the day in the SSI group ( ) compared with basal-bolus regimen ( ). 200 180 160 140 120 Breakfast Lunch Dinner Bedtime Duration of Treatment (days) Umpierrez G E et al. Diabetes Care. 2011;34(2):256 261. Diabetes care by AMERICAN DIABETES ASSOCIATION Reproduced with permission of AMERICAN DIABETES ASSOCIATION. in the format Republish in continuing education materials via Copyright Clearance Center. 10

RABBIT-2 Surgery: Composite Hospital Complications and Outcomes: SSI vs Basal-bolus Insulin 30 25 20 26 Sliding Scale Insulin Basal Bolus Insulin 19.6 15 10 9 ICU length of stay 3.19 vs 1.23 days; P = 0.003 SSI vs BB 11 11 12.5 5 0 Number of patients with complications 3 Wound infections 3 0 Pneumonia 5 1 Acute respiratory failure 4 Acute renal failure 2 1 1 1 Bacteremia Mortality Postsurgery ICU admission P = 0.003; P = 0.050. Umpierrez G E et al. Diabetes Care. 2011;34(2):256 261. Transition From Hospital to Outpatient Care 11

Clinical Inertia on Discharge Planning Percentage of patient with uncontrolled diabetes discharged with no change in medications or follow-up HgbA1c within 60 d Griffith et al. JCEM, 97:2019 2026, 2006 12

Discontinuation of Anti-Hyperglycemic Therapy at discharge in Patients with Acute Myocardial Infarction Among 217 diabetic patients with AMI, 25 (11.5%) were DC off anti-hyperglycemic therapy No clear reason for stopping therapy in 88% of patients Hyperglycemia is a marker of poor outcome and mortality in patients with AMI, thus these findings may represent an opportunity to improve the quality of care Lovig et al. The Joint Commission Journal on Quality and Patient Safety 2012 Kosiborod et al, Diabetes Care 2012 Time to the first all-cause hospital readmission among patients with T2D who were taking insulin before and during hospitalization and who had either continued insulin therapy or disrupted insulin therapy after hospital discharge. Wu et al. Endocrine Pract 18:651-659, 2012 13

Transition From Hospital to Outpatient Care Preparation for transition to the outpatient setting should begin at the time of hospital admission Multidisciplinary team: bedside nurse, clinical pharmacist, registered dietitian, case manager Clear communication with outpatient providers is critical for ensuring safe and successful transition to outpatient management Umpierrez GE et al; Endocrine Society. J Clin Endocrinol Metab. 2012;97(1):16 38. Transition to Discharge Does patient have a glucose monitor for home use? Does patient know how to inject insulin and how to prevent and to treat hypoglycemia? Is patient clear about the diabetes therapy after discharge? Does patient have appropriate outpatient follow-up appointment with primary care or specialist? 14

Survival Skills to Be Taught Before Discharge Basic understanding of what diabetes is How and when to take diabetes medications Basic knowledge of effect of carbohydrates on glucose levels Recognition, treatment, and prevention of hypoglycemia Self-monitoring of BG and implication of results What to do during illness How to dispose of lancets and insulin syringes Moghissi ES et al; American Association of Clinical Endocrinologists; American Diabetes Association. Endocr Pract. 2009;15(4):353 369. Possibilities for Hospital Discharge Hyperglycemia Regimen Based on hemoglobin A1C: Home regimen Titration of home regimen Or new insulin regimen (if last option, simple regimen with aggressive patient education and prompt follow-up) 15

Recommendations for Managing Patients With Diabetes After Hospital Discharge Use admission A1C to adjust therapy at discharge ADD basal or REPLACE with basal/bolus ADD basal insulin therapy Adjust original therapy, ADD another agent or basal insulin Return to original therapy 10% 9% 8% 7% Umpierrez G et al, J Clin Endocrinol Metabol, 2012 Discharge Insulin Algorithm Discharge Treatment A1C < 7% A1C 7%-9% A1C >9% Re-start outpatient treatment regimen (OAD and/or insulin) Re-start outpatient oral agents and D/C on glargine once daily at 50% of hospital dose Umpierrez et al, Diabetes Care. 2014 Nov;37(11):2934-9. D/C on basal bolus at same hospital dose. Alternative: re-start oral agents and D/C on glargine once daily at 80% of hospital dose 16

Hospital Discharge Algorithm Based on Admission HbA1C for the Management of Patients with T2DM 8.75% 7.9% % 7.35% Umpierrez et al, Diabetes Care. 2014 Nov;37(11):2934-9. Hospital Discharge Algorithm Based on Admission HbA1C for the Management of Patients with T2DM Primary outcome: - change in A1C at 4 wks and 12 wks after discharge All Patients OAD OAD + Glargine Glargine+ Glulisine Glargine # patients, n (%) 224 81 (36) 61 (27) 54 (24) 20 (9) A1C Admission, % 8.7±2.5 6.9±1.5 9.2±1.9 11.1±2.3 8.2±2.2 A1C 4 Wks F/U, % 7.9±1.7 7.0±1.4 8.0±1.4ψ 8.8±1.8ψ 7.7±1.7 A1C 12 Wks F/U, % 7.3±1.5 6.6±1.1 7.5±1.6 8.0±1.6 6.7±0.8 BG<70 mg/dl, n (%) 62 (29) 17 (22) 17 (30) 23 (44) 5 (25) BG<40 mg/dl, n (%) 7 (3) 3 (4) 0 (0) 3 (6) 0 (0) p< 0.001 vs. Admission A1C; ψp=0.08 Umpierrez et al, Diabetes Care. 2014 Nov;37(11):2934-9. 17

Revised Discharge Insulin Algorithm Discharge Treatment A1C < 7% A1C 7%-9% A1C >9% A1C <8% A1C 8%-10% A1C >10% Re-start outpatient treatment regimen (OAD and/or insulin) Re-start outpatient oral agents and D/C on glargine once daily at 50% of hospital dose Umpierrez et al, Diabetes Care. 2014 Nov;37(11):2934-9. D/C on basal bolus at same hospital dose. Alternative: re-start oral agents and D/C on glargine once daily at 80% of hospital dose Non-Insulin Therapies in the Hospital Which agents for which patients? Dipeptidyl peptidase 4 inhibitor alone or in combination with basal insulin Well tolerated with similar glucose control and frequency of hypoglycemia compared with a basal bolus regimen in general medicine and surgery patients Incretin agents do not cause hypoglycemia Need RCT evidence of safety and efficacy compared with standard therapies 1. American Diabetes Association. Diabetes Care. 2016;39(Suppl. 1):S99 S104. 2. Umpierrez GE et al. Diabetes Care. 2013;36:3430-3435. 3. Schwartz SS et al. Postgrad Med. 2015;127:251-257. 4. Umpierrez GE et al. Diabetes Care. 2013;36:2112-2117. 18

Mean Daily Blood Glucose (mg/dl) 5/19/2016 Mean Daily BG During Treatment Randomization Umpierrez et al. Diabetes Care. 2013 Nov;36(11):3430-5. Randomization Blood Glucose (<180 mg/dl and >180 mg/dl) and Mean Daily Glucose concentration p= 0.08 Sita + Basal vs. Basal Bolus p= 0.91 Umpierrez et al. Diabetes Care. 2013 Nov;36(11):3430-5. 19

B l o o d g l u c o s e ( m g / d L ) 5/19/2016 Mean Daily BG During Treatment S i t a g l i p t i n + B a s a l (n= 140) 2 4 0 B a s a l B o l u s (n= 140) 2 0 0 1 6 0 1 2 0 8 0 R a n d 1 2 3 4 5 6 7 8 9 1 0 Data are mean ± SE D u r a t i o n o f t r e a t m e n t ( d a y s ) Umpierrez et al. Unpublished, Preliminary information Summary 1. Diabetes is a common diagnosis in the hospital setting; hospitalization provides an opportunity to identify and improve glycemic control 2. The many transitions of care during hospitalization and back to the outpatient setting can create challenges to glycemic control 3. A team approach, medication reconciliation, and policies to manage hyperglycemia and insulin therapy can improve diabetes care 4. Patients with diagnosed diabetes or newly diagnosed diabetes may require changes to or intensification of therapy and appropriate education 20