Transcript Intensive Insulin Therapy: The Critical Facts

University of Minnesota – School
of Nursing Spring Research Day
Glycemic Control of Critically Ill Patients
Lynn Jensen, RN; Jessica Swearingen, BCPS, PharmD; Peggy
Hoeft, RN; Pam Richardson, RN; Robert Miner, MD
Abbott Northwestern Hospital
Objectives



Understand benefits of intensive (“tight”)
glycemic control in critically ill patients
Describe the Intensive Insulin ICU protocol
implementation experience at Abbott
Northwestern (ANW) Hospital
Share patient outcome data associated with
ANW Intensive Insulin ICU protocol
utilization
Hyperglycemia in the Critically
Ill Patient Population





Hyperglycemia occurs in >50% in ICU patients
Multiple etiologies (e.g., counterregulatory
hormone release, medications)
Historically, hyperglycemia treated only at very
high blood glucose levels
Hyperglycemia-related adverse effects (e.g.,
osmotic diuresis, impaired immune function) well
established
More recent evidence suggests close correlation
between hyperglycemia & clinical outcome
Hyperglycemia Clinical Trials in
Critically Ill Patients

Open heart surgery patients with history of DM &
mean BG >206 mg/dL post-op had increased risk for:
– leg & chest wound infections
– pneumonia
– urinary tract infections

AMI patients with history of DM or hyperglycemia on
hospital admission randomized to tight control (BG
126-200 mg/dL) for 3 months or usual care
– mortality at 1 yr & 3.4 yrs  by 7.5% & 11%, respectively
– reinfarction & new cases of CHF decreased
Golden et al. Diabetes Care 1999;22(9):1408-14;
Malmberg et al. J Am Coll Cardiol 1995;26(1):57-65
Hyperglycemia Clinical Trials in
Critically Ill Patients

Mechanically ventilated, surgical ICU patients
– majority of patients had no history of DM
– randomized to tight control or standard care
– after transfer from ICU both groups received standard care

Results
– mortality  by 3.4% for tight control group
– mortality in patients with ICU stay >5 days  by 9.6%
– significant  in deaths due to sepsis & MODS
– tight control  blood transfusions (28.6% vs. 31%); dialysis
(4.8% vs. 8.2%); mechanical ventilation >14 days (7.5% vs.
11.9%); or ICU stay >14 days (11.4% vs. 15.7%)
Van den Berghe et al. NEJM 2001;345(19):1359-67.
Hyperglycemia Clinical Trials in
Critically Ill Patients


Observational trial in Med/Surg/Neuro/Cardiac ICU
Before & after design
– historical controls vs. consecutive protocol patients
– protocol group received insulin infusion after 2 successive BG
levels >200 mg/dL
– BG goal <140 mg/dL

Results
– mean BG  from 152.3 mg/dL to 130.7 mg/dL
– protocol significantly  mortality from 20.9% to 14.8%
– most striking  in mortality for septic shock, neurologic &
surgical patients
– BG>200 mg/dL  from 16.7% to 7.1%
– hypoglycemia did not increase (0.35% vs. 0.34%)
Krinsley et al. Mayo Clin Proc 2004;79(8)992-1000
ANW Intensive Insulin Protocol
Implementation Experience






Multidisciplinary team of physicians,
pharmacists & nurses from each ICU
Revision of existing Med/Surg/Neuro ICU
protocol
Desktop computer protocol developed
New protocol implemented in all ICUs May 2004
Nurses in all ICUs educated
Additional resources available during first 5
days of protocol implementation
ANW Intensive Insulin Protocol
Implementation Experience


ANW blood glucose goal range:
mg/dL
All protocol patients received:
90-120
– insulin infusion
– hourly blood glucose checks until within goal range, then
every two hours

Data collected on:
–
–
–
–
mean blood glucose
efficacy attaining goal range
episodes of hypoglycemia
patient outcomes
ANW Intensive Insulin Protocol
Implementation Experience


Continued to make changes to protocol &
provide feedback
Challenges during implementation
–
–
–
–
–
physician (surgeon) acceptance
limited glucometer availability
multiple patient sticks/blood draws
nursing acceptance due to  workload
computer dosing based on last 2 BG values
ANW Intensive Insulin Protocol
Implementation Experience
Protocol Example
R eturn t
What is the rationale for using this intensive regular insulin infusion protocol?
Research in critically ill patients has demonstrated decreased morbidity (sepsis, ventilator
days, ICU LOS, dialysis, etc.) and mortality when glucose concentrations are kept below 110
mg/dL.
van den Berghe G, Wouters P, Weekers F, et al. Intensive Insulin Therapy in Critically Ill Patients. N Engl J Med 2001;
345(19):1359-67
How does the intensive regular insulin infusion protocol improve morbidity and mortality?
Intense control of glucose concentration may improve immune function since white blood cell
function is more effective when the glucose concentration is normal rather than when it is
R eturn t
greater than 200 mg/dL.
Can the intensive regular insulin infusion protocol be used in all critically ill patients?
No. The protocol may be beneficial in most critically ill patients with acute hyperglycemia,
even those with no prior diagnosis of diabetes. However, the protocol IS NOT to be used in
R eturn t
patients with diabetic ketoacidosis or in women who are pregnant.
What are some side effects of an insulin infusion?
Hypokalemia: insulin and glucose cause potassium to shift out of the blood and into cells. The
end result may be to excessively lower the concentration of potassium in the blood. To avoid
this, monitor potassium concentrations and implement the potassiu
Hypoglycemia: Because the glucose target range is narrow and lower than that of the past,
the risk for hypoglycemia a concern. The insulin infusion rate may be too high for a specific
patient and excessively lower his/her blood glucose concentration. T
R eturn t
Neurologic events: Severe hypoglycemia may cause seizures or obtundation. Again, close
monitoring of glucose concentrations is indicated to minimize the risk of these occurring.
ANW Intensive Insulin
Protocol Data
CVICU
CCU
Pre-protocol
(81 patients;
5227 BG
values)
Post-protocol
(139 patients;
14192 BG
values)
Pre-protocol
(25 patients;
1639 BG
values)
Post-protocol
(65 patients;
8141 BG
values)
BG in goal
range (90–
120 mg/dL)
22%
36%
20%
33%
Mean BG
(mg/dL)
158
133
162
138
BG ≥ 200
mg/dL
18%
7%
23%
10%
BG < 60
mg/dL
0.6%
0.6%
1.1%
0.6%
ANW Intensive Insulin
Protocol Data
Patient Demographics
Pre-Protocol
(n = 50)
Post-Protocol
(n = 50)
66.8
65.6
Sex (% male)
59
67
Ventilated (%)
76
80
History of Diabetes (%)
84
74
High Risk for Hyperglycemia (%)
70
61
-Cardiovascular
69
75
-Renal
10
5
-Pulmonary
6
10
-Other
15
10
Mean Age (years)
Admit Diagnosis (%)
ANW Intensive Insulin
Protocol Data
Outcomes
Pre-Protocol
(n = 50)
Post-Protocol
(n = 50)
Mean Blood Glucose (mg/dL)
168
133
Hypoglycemic Events (%)
0.22
0.23
Hospital Mortality (%)
14
11
New Onset Renal Dysfunction (%)
44
31
Mean Hospital Length of Stay (days)
17
13
Blood Product Administration (%)
61
52
Conclusions



Tight glycemic control can significantly
improve morbidity & mortality in critically
ill surgical patients
Barriers to implementation can be
overcome
Nurses can significantly impact mortality
& patient outcome by managing blood
glucose more tightly
Any Questions?