Inpatient Management of Hyperglycemia Russell Vinik, M.D. Hospitalist, University of Utah
Nov 14, 2014
Inpatient Management of Hyperglycemia
Russell Vinik, M.D.Hospitalist, University of Utah
Overview
Observations related to hyperglycemia and outcomes in hospitalized patientsPotential mechanisms for poor outcomes in hyperglycemic patientsReview interventional studies related to glucose managementStrategies for improving inpatient control
Umpierrez GE et al. J Clin Endocrinol Metab. 2002;87:978-982.
Hyperglycemia Is an Independent Marker of Inpatient Mortality in Patients With Undiagnosed Diabetes
1.73
16
0
2
4
6
8
10
12
14
16
18
In-hospital Mortality Rate
(%)
Newly Discovered
Hyperglycemia
Patients With History of Diabetes
Patients With
Normoglycemia
P < 0.01
P < 0.01
Postoperative Glycemic Control Correlates With Cardiac-Related Mortality
0
5
10
15
<150 150-175 175-200 200-225 225-250 >250
Noncardiac-related mortality Cardiac-related mortality
* (P<.001). Furnary AP et al. J Thorac Cardiovasc Surg. 2003;125:1007-
1021.
Mo
rta
lity
(%)
0.9 1.3
2.34.1
6.0
14.5
Average Postoperative Glucose (mg/dL)
**
*
*
Poor Outcomes Correlate With Hyperglycemia After Acute Ischemic Stroke
Blood Glucose at Admission
OutcomeBG <130 mg/dL
(n=385)BG 130 mg/dL
(n=258) P value
Length of stay
Discharged to home
In-hospital mortality
30-Day mortality
1-Year mortality
6 ± 0.3
79%
5%
5%
11%
7.2 ± 0.4
73%
7%
10%
18%
.015
.07
.15
.018
.009
Williams LS et al. Neurology. 2002;59:67-71.
0%
5%
10%
15%
20%
25%
30%
35%
Infections Infections excluding
UTI's
<220 >220
Rates of Nosocomial Infection in 100 Uninfected Diabetics Undergoing
Elective Surgery
Glucose on Post-Op Day #1
Pomposelli et al. JPEN 1998; 22:2, 77-81
Hyperglycemia and Poor Outcomes Following Myocardial Infarction
0
10
20
30
<110 110-139 140-169 170-199 >200
Arch Intern Med. 2009;169(5):438-446
In H
osp
ital
Mo
rtal
ity
(%)
Average Post-admission Glucose
Basic Science
Hyperglycemia is associated with increased neuron damage following brain ischemiaHyperglycemia leads to increased platelet aggregation and thrombosisInsulin decreases arterial levels of free fatty acidsHyperglycemia leads to increased cytokine levels and inflammationNeutrophil Adherence, chemotaxis, phagocytosis and extravasation are
all inhibited by increased glucose concentrations
Diabetes Care. 2004;27:553-591, Diabetes 1989;38:1031-5, Diabetes Care 2001;24:1634-9
Intervention Studies
Insulin and Sternal Wound Infections
Furnary et al studied 2467 patients undergoing open heart operations
The first 968 patients were treated with a sliding scale to keep glucose near 200
The next 1499 patients received an insulin infusion to keep glucose 150-200
Ann Thorac Surg 1999;67:352-62
Insulin and Sternal Wound Infections
Furnary AP et al. Ann Thorac Surg. 1999;67:352-362.
4
3
2
1
0
DS
WI
(%)
87 88 89 90 91 92 93 94 95 96 97Year
CII
Patients with diabetes
Nondiabetic patients
Insulin and Mortality in CABG patients
0%
2%
4%
6%
8%
10% Non-Diabetics
Diabetics
CII
Year
Furnary AP Endocr Pract. 2004;10(suppl 2):21-33.
Mo
rta
lity
Intensive Insulin in the Critically Ill- (Leuven I)
Van Den Berghe et al enrolled 1548 ventilated patients mostly post cardiac surgery13% of these patients had diabetes Patients were randomized to:
intensive treatment- infusion to maintain glucose between 80-110mg/dl
conventional treatment- targeting a glucose of 180-200mg/dl
Mean glucose in the intensive treatment group was 103 and conventional group was 153Hypoglycemic events (glucose<40) occurred in 5.1% of patients in the intensive treatment group vs. 0.76% of patients in the conventional group
Van den Berghe G et al. NEJM 2001;345: 1359-67
Intensive Insulin in the Critically Ill
Van den Berghe G et al. N Engl J Med. 2001;345:1359-1367.
100
96
92
88
84
80
0
0 20 40 60 80 100 120 140 160
Intensive treatment
Conventional treatment
Intensive treatment
Conventional treatment
Su
rviv
al i
n I
CU
(%
)
100
96
92
88
84
80
0
0 50 100 150 200 250
In-H
osp
ital
Su
rviv
al (
%)
Days After Admission Days After Admission
42.5% reduction in mortality with intensive treatment; P<.04
34% reduction in mortality with intensive treatment; P<.01
Benefits of IV Insulin Treatment in Critically Ill Hospitalized Patients
-60
-50
-40
-30
-20
-10
0Total Mortality Blood Infection
Acute RenalFailure Transfusions Polyneuropathy
Van den Berghe G et al. N Engl J Med. 2001;345:1359-1367.
Red
uct
ion
(%
)
34%
46%41%
50%
44%
Krinsley JS. Mayo Clin Proc. 2004;79:992–1000.
The Stamford Project
Mixed medical/surgical/cardiac ICU
“Before-and-after” design Developed an insulin protocol and followed
800 consecutive patientsCompared the outcomes in these patients to
a control group of 800 consecutive patients immediately prior to protocol institution
Targeted a blood glucose of less than 140
Stamford Project: Improvement in Mortality
0
10
20
30
40
50
60
70
All Cardiac Respiratory Septic Shock Neurologic GeneralSurgical
*
*P<.01 compared with control group.†P<.05 compared with control group.
†
*
†
Krinsley JS. Mayo Clin Proc. 2004;79:992-1000.
Dec
reas
e in
Mo
rtal
ity
Wit
h T
reat
men
t (%
)
Leuven II- Intensive Insulin in MICU Patients with expected LOS >3 days
Van den Berghe G et al. N Engl J Med. 2006;354;5:449-61.
* p<.05
0%
10%
20%
30%
40%
50%
60%
Conventional Intensive treatment
*
*
*
Intensive Insulin Meta-analysis of 29 Trials
8432 Patients
JAMA 2008;300(8):933-44
0%
5%
10%
15%
20%
25%
30%
Total Mortality (p=.7)
Mortality in Surgical pts
Mortality in Medical pts
Septicemia
Hypoglycemia (glucose <40)
Conventional Intensive treatment
NICE-SUGARRandomized trial of 6104 patients 42 hospitals in Australia, New Zealand, and Canada Enrolled patients with an expected LOS of 3 days and had an arterial
line Intervention discontinued when patient was eating or discharged from
ICU
Reason for ICU admission: 37% Operative 63% Non-Operative
Patients were randomized to: intensive treatment- infusion to maintain glucose between 81-108
mg/dl conventional treatment- targeting a glucose of <180mg/dl
Mean glucose in the intensive treatment group was 115mg/dl and conventional group was 144mg/dl
NEJM 2009;360:1283-97
NICE-SUGAR Results
Critical Care Med 2008;36:12 1-8
0%5%
10%15%20%25%30%35%40%45%
Conventional Intensive treatment
How do we use this data to care for our patients?
American Diabetes Association 2009* “Standards of Medical Care”
Diabetes Care 2009 32;Supp 1:S14-61 Circulation 2008;117;1610-19 Critical Care Med 2008 35:296-327
International Guidelines for Management of Severe Sepsis and Septic Shock: 2008
Critical <150 mg/dL 2C
Patients Fasting Non-Fasting Evidence levelCritical surgical 110 mg/dL 110 mg/dL ACritical non-surgical <140 <140 CNoncritical <126 mg/dL <180-200mg/dL E (Expert Consensus)
American Heart Association: Hyperglycemia and Acute Coronary Syndrome
Critical 90-140 mg/dL CNoncritical <180 mg/dL C
The Endocrine Society- Position Statement March 2009
Critical <144-180 mg/dL
* ADA/AACE statement March 09 promised new guidelines and recommended targets similar to the “conventional” arm of NICE-SUGAR
Barriers to Inpatient Glucose Control
Infection, fever, stress, glucocorticoids, surgery all exacerbate hyperglycemia
Patients may eat less or have meals held
Timing of insulin administration and meals are often disrupted
Oral medications are often held
Limitations of Oral Agents for Managing In-Hospital Hyperglycemia
Sulfonylureas No rapid dose adjustment Risk of hypoglycemia in patients not eating normally
Metformin No rapid dose adjustment Mostly contraindicated due to increased risk of lactic acidosis in
hospitalized patients (ie, intravenous contrast, renal failure, congestive heart failure)
Thiazolidinediones No rapid dose adjustment Mostly contraindicated in heart failure, hepatic dysfunction
Clement S et al. Diabetes Care. 2004;27:553-591.
Using Insulin in the Hospital
First, Determine Source/Route of Nutrition
Second, Estimate a Starting Dose of Scheduled Insulin
Third, Know the Kinetics of the insulin you are using and make a plan
Prandial insulin
Source of Nutrition- Effects on Insulin Secretion
B L D B L D
Basal insulinBasal insulin Basal insulin
Prandial insulin
The Eating Patient Pt. Receiving Continuous Feeds
Estimating a Starting Dose
Use patient’s home regimen Adjust as clinically indicated
Make a weight based estimate* Start 0.4units/kg for glucose 140-200 Start 0.5 units/kg for glucose 201-400 Consider lower starting dose with significant renal or hepatic
impairment
Estimate basal insulin and carb count Difficult to achieve in the hospital If attempting, estimate basal insulin (.2-.25 units/kg/day)
Type I: Give 1 unit per 15g carbohydrates Type II: Give 1 unit per 10g carbohydrates
Diabetes Care 30:2181-2186, 2007
Kinetics of Insulins
Regular
NPH
0 126 18 24
aspart/glulisine/lispro
glargine
Mimicking Nature With Insulin Basal/Bolus ConceptPhysiologic Insulin Secretion
Insu
lin
(µU
/mL
)
Glu
cose
(mg
/dL
)
9
B L D
150
100
50
07 8 91011121 2 3 4 5 6 7 8
AM PMTime of Day
Basal glucose
50
25
0
24-hr profile
Basal insulin
Adapted from Bergenstal RM et al. In: DeGroot LJ, Jameson JL, eds. Endocrinology. 4th ed. Philadelphia, Pa: WB Saunders Co.; 2001:821
Insulin aspart/glulisine/lispro
Insulin glargine
Basal-Bolus Insulin Therapy: Insulin Glargine at HS
and Mealtime Lispro or Aspart
B DL HS
Insu
lin E
ffe
ct
Adapted from Leahy J. In: Leahy J, Cefalu W, eds. Insulin Therapy. New York, NY:
Marcel Dekker Inc.; 2002:87
Example: Patient’s Total Daily Insulin Estimate=60
Units
10 units aspart
glulisine lispro
10 unitsaspart
glulisine lispro
10 unitsaspart
glulisine lispro
30 unitsglargine
Insu
lin E
ffe
ct
Adapted from Leahy J. In: Leahy J, Cefalu W, eds. Insulin Therapy. New York, NY: Marcel Dekker Inc.; 2002:87
Twice-Daily Split-Mixed Regimens
Regular
NPH
B DL HS B
Endogenous insulin
Dawn phenomenon
Hyperglycemia
13 units NPH
Example: Patient’s Total Daily Insulin Estimate=60 Units
27 units NPH
40 units of insulin in the a.m.
20 units of insulin in the p.m.
+13 units regular +7 units regular
Insulin Requirement During Continuous Dextrose, TPN or Enteral Feedings,or Negligible Carbohydrate Exposure
8 12 6 10
6 12 6 12
NPH
6 12 6 12
Regular at least 50 % - - - -
Glargine not more than 50 %
Regular ~ 33 % (hold if low) - - -
NPH ~ 67 % —
q 6 – 8 h
Regimens for patient while NPO, on IV’s,
or receiving continuous enteral
feedings.
The Insulin Infusion
Many Protocols ExistDIGAMI (IV insulin glucose infusion followed by outpatient multidose subcutaneous insulin regimen)van den Berghe (IV insulin therapy to maintain blood glucose between 80 and 110 mg/dL)Portland protocol (perioperative use of IV insulin)Markovitz (IV insulin therapy to maintain blood glucose between 120 and 199 mg/dL)Yale Protocol (IV insulin therapy to maintain blood glucose between 100 and 139 mg/dL)Stamford Protocol (IV/SQ insulin only given if glucose is greater than 140)Duke Protocol (IV insulin to maintain blood glucose 101-150)
Malmberg K. BMJ. 1997;314(7093):1512–1515.van den Berghe G, et al. N Engl J Med. 2001;345:1359–1367.Markovitz LJ, et al. Endocr Pract. 2002;8:10–18.Goldberg PA, et al. Diabetes Care. 2004;27:461–467.Krinsley JS. Mayo Clin Proc. 2004;79:992-1000.Lien LF., et al. Endocr Pract. 2005;11: 240-53.
43
8 12 6 10
aspart, glulisine, lispro
Glargine
Peri-Procedural Management
8 12 6 10RegularNPH
Hold short acting insulin and give ½ regular dose of NPH
8 12 6 10
Insulin Requirement High a.m. dose corticosteroids
Solution B.
8 am 12 6 10
R - - -
NPH —
8 12 6 10
Lispro or aspart
Glargineduringcortico-steroids( ~ 30 % )
Solution A.
Thank You