Management of Diabetes and Hyperglycemia in the Hospital

Management of Diabetes and Hyperglycemia in the Hospital

Stephen Clement M.D.

Associate Professor

Georgetown University Hospital

Insulin Rx is important

Hyperglycemia in hospital is common Majority treated with insulin Insulin is one of five “high alert

meds” with greatest risk for causing medication error injuries.

Current Rx practices varied & commonly do not enable targeted glucose control

Scope of Problem at GU Hospital

Multiple cases of errors in insulin orders/administration causing: DKA (lack of basal insulin) Severe hypoglycemia (insulin stacking, wrong dose)

Emergence of glucose control as a target to improve hospital outcomes (reduced mortality and reduced length of stay)

New AACE/ADA targets

Post-op Order

Hyperglycemia*: A Common Comorbidity in Medical-Surgical Patients in a Community Hospital

64%64%

12%12%

26%26%

NormoglycemiaNormoglycemia

Known DiabetesKnown Diabetes

New HyperglycemiaNew Hyperglycemia

Umpierrez G et al, J Clin Endocrinol Metabol 87:978, 2002Umpierrez G et al, J Clin Endocrinol Metabol 87:978, 2002

n = 2,020n = 2,020

* Hyperglycemia: Fasting BG * Hyperglycemia: Fasting BG 126 mg/dl 126 mg/dl or Random BG or Random BG 200 mg/dl X 2 200 mg/dl X 2

Obstacles to In-Hospital Glucose Control

Infection Severe Stress Illness Procedures NPO Status Fear of Hypoglycemia Lack of Activity Meals

Hyperglycemia in the Hospital

Nuisance

or Opportunity?

Hyperglycemia: An Independent Marker of In-Hospital Mortality in Patients with Undiagnosed Diabetes

0

10

20

30

NormoglycemiaNormoglycemia Known Known New New DiabetesDiabetes Hyperglycemia Hyperglycemia

1.7%1.7% 3%3%

16% 16% **

Mort

alit

y (

%)

Mort

alit

y (

%)

* P < 0.01* P < 0.01

Umpierrez G et al, J Clin Endocrinol Metabol 87:978, 2002Umpierrez G et al, J Clin Endocrinol Metabol 87:978, 2002

Hyperglycemia: An Independent Marker of In-Hospital Mortality in Patients with Undiagnosed Diabetes

0

10

20

30Non ICU MortalityNon ICU Mortality

NormoglycemiaNormoglycemia Known Known New New DiabetesDiabetes Hyperglycemia Hyperglycemia

0.8%0.8% 1.7%1.7%

10.0% 10.0% **

Mort

alit

y (

%)

Mort

alit

y (

%)

* P < 0.01* P < 0.01

Umpierrez G et al, J Clin Endocrinol Metabol 87:978, 2002Umpierrez G et al, J Clin Endocrinol Metabol 87:978, 2002

Hyperglycemia: Effect on Length of Stay and Disposition at Discharge

NewNew Known NormoglycemiaKnown Normoglycemia HyperglycemiaHyperglycemia DiabetesDiabetes

Length of stay (d)Length of stay (d) 9 ± 0.7 9 ± 0.7a, ba, b 5.5 ± 0.2 5.5 ± 0.2 4.5 ± 0.1 4.5 ± 0.1

ICU admission (%)ICU admission (%) 29 29a, ba, b 14 14aa 9 9

Disposition at dischargeDisposition at discharge

Home (%)Home (%) 56 56a, ba, b 7474aa 84 84

TCU (%) TCU (%) 20 20aa 15 15aa 10 10

Nursing home (%)Nursing home (%) 8 8cc 9 9aa 4 4Results are Results are ±± SEM. TCU, Transitional Care Unit SEM. TCU, Transitional Care Unit

a a P < 0.01 vs. normoglycemia; P < 0.01 vs. normoglycemia; b b P < 0.01 vs. Known diabetesP < 0.01 vs. Known diabetescc P < 0.02 vs. normoglycemiaP < 0.02 vs. normoglycemia

Umpierrez GE et al, J Clin Endocrinol Metabol 87:978, 2002Umpierrez GE et al, J Clin Endocrinol Metabol 87:978, 2002

In-Hospital Glucose and Acute Renal Graft Rejection

Glycemic control was assessed during the first 100 postoperative hours

Only 3 of 27 patients (11%) with mean BG < 200 mg/dL had rejection episodes

58% with mean BG > 200 mg/dl had rejection episodes

Thomas. Early peri-operative glycaemic control and allograft rejection in patients with diabetes mellitus: a pilot study. Transplantation 2001;72:1321.

Mean perioperative glucose patients with diabetes undergoing their first cadaveric renal transplantation

200 mg/dl

300 mg/dl

Post-operative Infection and Blood glucose

11.5

31.3

0

5

10

15

20

25

30

35

Preop POD1 POD2

Percent Developing Infection

≤ 220

> 220

Pomposelli. Early postoperative glucose control predicts nosocomial infection rate in diabetic patients. J Parenteral and Enteral Nutrition; 1998; 22: 77.

Portland Diabetic Project: Insulin Infusion Reduces DSWI

DSWI = deep sternal wound infection; CII = continuous insulin infusion.

4.0

3.0

2.0

1.0

0.0

DSWI(%)

87 88 89 90 91 92 93 94 95 96 97

Year

Patients with diabetes

Patients withoutdiabetes

Furnary AP, et al. Ann Thorac Surg. 1999;67:352–362

CIICII

(N = 3,554)

SCISCI

Mortality Post-CABG Patients

Furnary AP, et al. J Thoracic Cardiovasc Surg. 2003;125:1007-1021

0

5

10

15

<150 150–175 175–200 200–225 225–250 >250

Average postoperative glucose (mg/dL)

Mortality

Cardiac-related mortality

Noncardiac-related mortality

0.9%1.3%

2.3%

4.1%

6.0%

14.5%

Risk Reduction by Meticulous Risk Reduction by Meticulous Blood Glucose Control in an ICUBlood Glucose Control in an ICURisk Reduction by Meticulous Risk Reduction by Meticulous Blood Glucose Control in an ICUBlood Glucose Control in an ICU

Van Den Berghe: NEJM 345: 1359, 2001

103 Vs 153 mg%

0 10 20 30 40 50 60

ICU motality

Sepsis

Dialysis

Blood Transfusion

requiring > 14 d vent support

Glucose

Insulin

Immune dysfunction

Infection dissemination

Reactive O2 species

Transcription factors

Secondary mediators

(i.e., nFB)

Metabolic stress response

FFA

Ketones

Lactate

Stress hormones and peptides

Cellular injury/apoptosis

Inflammation

Tissue damage

Altered tissue/wound repair

Acidosis

Thrombosis

Infarction/ischemia

Prolonged hospital stay

DisabilityDeath

Shechter et al, 1999

= 79 mg/dl= 110 mg/dl

Platelet-Derived Thrombosis

Diabetes Care 27:553-90, 2004

http//care.diabetesjournals.org

Society of Hospital Medicine AADE

Endocrine Practice 2004

Glycemic Targetsfor Hospitalized Patients

Non-ICU: Fasting BG < 110 mg/dl Peak BG < 180 mg/dl

ICU:

< 110 mg/dl

Diabetes Care 27:553-91, 2004 Endocrine Practice 2004

Basal Insulin Requirement

Amount of exogenous insulin per unit time necessary to maintain blood sugars in between meals and when not eating.

In absence of basal insulin, BG’s increase 45 (mg*dL-1hr-1) after insulin withdrawal in insulin-deficient patients.

Clement et al. Diab Tech Therapeutics 4:459-466, 2002

Example of Poor Outcome from

Lack of Basal Insulin

BG at midnight: 248 mg/dl, HC03 27, AG 10

BG at 8 AM: 616 mg/dl HCO3 11, AG 24

46 mg*dL-1hr-1

Patient Arrested and Died

Glucose and Ketone Rise After Insulin Withdrawal

050

100150200250300350400450

0 1 2 3 4

(hours)

Blo

od

Glu

co

se

(m

g/d

l)

Glucose

3-OHButyrate(mmol/l)0

2

1

Husband et al. Diab Res 3:193-98, 1986

Characteristics ofInsulin Deficient Patient*

Known type 1 DM H/O pancreatectomy or pancreatic

dysfunction H/O wide BG fluctuations H/O Ketoacidosis H/O insulin use > 5 years

*If “Yes,” then Always provide basal insulin

Nutritional insulin requirement

Amount of insulin necessary to cover intravenous dextrose, TPN, enteral tube feedings, nutritional supplements &/or discrete meals

Illness or Stress-Related Insulin

The increase in daily insulin requirement attributed to illness, stress, or treatment

Wide individual variation Apportioned between basal,

nutritional & correction doses Need decreases as clinical

condition improves

Insulin Requirements in Health & Illness

0

20

40

60

80

100

120

140

Correction

Nutritional

Prandial

Basal

Rel

ativ

e p

rop

orti

on o

f in

suli

n r

equ

irem

ent

(%)*

*Estimations for illustrative purposes: requirements mayvary widely.

Adapted from ADA Technical Review: Management of Diabetes & Hyperglycemia in Hospitals.

Diabetes Care 2004. In press.

Sick/Eating

Healthy

Sick/ NPO

Illness-Related

Correction/supplemental Insulin

Amount of insulin give for unexpected hyperglycemia

a.k.a. “sliding scale”

Sliding Scale InsulinConcerns

Use as only insulin replacement in insulin-deficient patient

Better terminology, i.e., corrective or supplemental insulin

Lack of standardization

SLIDING SCALE INSULIN

171 patients, secondary dx DM, medicine SSI without programmed insulin: 22.8% hypoglycemia (<60 mg/dl) 40.4% hyperglycemia (>300 mg/dl)

SSI alone with 3x risk hyperglycemia

BG

BG

Insulin

Queale & Brancati. 1997. Arch. Int. Med. 157: 545-552

(Normal)

Dextrose

BASALNUTRITIONAL

SUPPLEMENTAL

ILLNESS-RELATED

Terminology:Terminology:Physiologic Insulin Physiologic Insulin

needsneeds

Case #1 22 y/o Female with Acute Leukemia admitted

for neutropenic fever

No prior h/o “diabetes”, but during prior chemo tx regimens, insulin was required.

Diet: medium consistent carb (75g/meal)

New Chemo Rx includes Predisone, 200 mg q A.M.

Example: 22 y/o with Acute Lymphoblastic Leukemia with Neutropenic Fever

0

20

40

60

80

100

120

140

1 4 7 12 16 21

Hospital Day

Uni

ts p

er D

ay

Correction

Insulin Drip

Prandial

Basal

Prednisone

Illness/Stress-Related Insulin

0

50

100

150

200

250

300

1 4 7 16 21

Hospital Day

Avg. Blood Glucose (mg/dl)

Insulin Drip Rate

0

1

2

3

4

5

6

7

8

9

600 800 1000 1200 1400 1600 1800 2000 2200 2400 200 400

Time of Day

Insu

lin

Dri

p R

ate

(un

its/

ho

ur)

Prednisone Dose

Special Circumstances

Perioperative Management

Enteral NutritionParenteral NutritionGlucocorticoid Use

Practical Guidelines:Eating

Programmed Supplemental

basal nutritional

int bid or hs rapid ac rapid ac

or B&D or B,L,D

LA hs or am

insulin drip

Comments: Give rapid insulin 0-15 min ac Glargine usually given as once daily dose at hs Avoid reg & rapid at hs to minimize nocturnal hypoglycemia risk

Rx

Practical Guidelines:Perioperative or peri-procedural

NPO e.g major surgery

Programmed Supplemental/

Basal Nutritional Correction insulin drip n/a or per TPN until resumes po

reg q 4-6 hours enteral guidelines reg q 4-6 hours

rap q 4 hours rap q 4 hours

int, give 1/2 usual

am dose

LA usual daily dose

Comments: If prolonged post-op NPO, insulin drip Rx recommended Periop insulin drip starting dose is 0.2 units/kg/hour

Rx

Enteral Nutrition

Short acting insulin until tolerating well

Continuous enteral regimen Regular insulin q 4-6 hrs during feeding period +/- Basal insulin

Bolus enteral regimen Regular insulin SQ prior to each bolus +/- Basal insulin

Practical Guidelines: Total Parenteral Nutrition

Programmed Supplemental/

basal nutritional correction

reg or rap added reg or rap q 4-6 hours

to TPN bag

Comments: Basal & nutritional needs met with reg or rap insulin added to TPN bag Consider use of separate iv insulin infusion for 24 hours to determine daily

insulin requirement, than add this amount to subsequent bags daily Use subcutaneous insulin with caution, as may lead to erratic BG control

Rx

Initiating Insulin tx Hospitalized Patient

Basal insulin 0.4 units/kg/day (i.e., Glargine)

Prandial &/or nutritional: 0.1 unit/kg/meal (i.e., Novolog or Humalog)

Patients with insulin deficiency always require basal insulin to prevent ketosis

Transition of IV to Subcutaneous InsulinSome Dos & Don’ts

Do overlap SC and IV insulin to minimize “hyperglycemia escape.”

Don’t switch to oral agents alone from IV insulin.

Arrange for follow up of patients placed on temporary insulin.

Ensure adequate food intake when switching patients with to SC insulin.

Bedside Glucose Monitoring Strong quality control program essential

Some systems can give falsely elevated readings

Specific situations render capillary tests inaccurate Shock, hypoxia, dehydration Extremes in hematocrit Elevated bilirubin, TG’s Drugs

Prevention/Tx of Hypoglycemia

Proactive Approach Missed meal, tube feeding D/C’d Schedule procedures in the AM

Establish a nurse-driven protocol for starting dextrose and test hourly glucose testing if hypoglycemia anticipated.

Education:Core knowledge for physicians

Impact of BG on hospital outcomes Hospital targets for BG Terminology basal/nutritional/correction Insulins Hypoglycemia prevention & treatment Avoid SSI Special circumstances

Education: Core competency for Nurses

Bedside glucose monitoring technique Critical and target BG values Insulin administration technique Optimum timing of SQ insulin shots Hypoglycemia prevention & treatment BG & insulin dose documentation When to call the MD

Patient Education:Content areas

What is diabetes? Symptoms & signs of high and low BG Hypoglycemia Rx Medications (specifics of discharge regimen) Self-glucose monitoring (keep a log) When to call the doctor Education resources

Adapted from American Association of Diabetes Educators

Survival Skills Education Guidelines.

GU Hospital Initiatives2004 - 2006

In-service all M.D.’s and nursing units on proper basal/bolus insulin therapy

Laminated cards

I.V. Drip changes and any SC insulin injection requires second nurse check dose and sign

Implement IV insulin protocol outside of ICU

GU Hospital Initiatives(cont) 2004 - 2006

Piloted standard order form and MAR for s.q. insulin administration

Eliminated Regular insulin except for enteral feeding and insulin drips

Roll out of revised order form for entire hospital (July ’05.)

Components of the standardized subcutaneous insulin protocol

BG monitoring frequency Target BG range Programmed insulin orders: Suggested lag times for prandial insulin Correction dose algorithm Call parameters for high & low BG Hypoglycemia Rx guidelines or reference

to hypo protocol

Insulin-Glucose Flow Sheet

GU Hospital ResultsInsulin Error Rate

0

5

10

15

20

25

30

35

% Errors

2005 2006

Future GU Hospital Initiatives

Within 8 hours of admission DM patients will have lab glucose and Bedside BG’s started

Any patient with a lab value > 200 mg/dl will be checked to see if bedside BG orders are written and A1C ordered

Future Initiatives (cont.)

Patients with two or more bedside or lab BG values > 300 or < 60 receives automatic consult by diabetes NP

Outcomes: Mean BG and range of BG levels for all patients Mortality, LOS

TEAM APPROACH TO THE TREATMENT TEAM APPROACH TO THE TREATMENT OF THE HOSPITALIZED DIABETIC PATIENTOF THE HOSPITALIZED DIABETIC PATIENT

PhysicianPhysician Nurse EducatorNurse Educator

DietitianDietitian

Endocrinologist Pharmacist

Blood Glucose

0

50

100

150

200

=

Glucose Control Matters

Collaborators

Penny Smith, CNP Susan Braithwaite, M.D. Michelle Magee, M.D. Andrew Ahmann, M.D. Rebecca Schaffer, R.D. Irl Hirsch, M.D. American Diabetes Assoc. AACE