Insulin Therapy in the ICU: Hyperglycemic Protocols

Bradley J. Phillips, M.D.

Critical Care Medicine

Boston Medical Center

Boston University School of Medicine

TRAUMA-ICU NURSING EDUCATIONAL SERIES

Insulin in the ICU…Hypergylcemia associated with insulin resistance

is common in ICU patients, even those who have not previously had diabetes.

• Reports of pronounced-hyperglycemia leading to multiple complications– a lack of clinical trials to support

• High serum levels of insulin-like growth factor-binding protein 1 increases the risk of death– reflects an impaired response of the hepatocyte to insulin

NEJM 2001

Landmark Paper

• Van Den Berghe et al. Intensive Insulin Therapy in Critically Ill Patients. NEJM 2001;345 (19): 1359-67.

– Prospective, Randomized, Controlled study

– 1,548 Adults admitted to a SURGICAL-ICU receiving Mechanical Ventilation

– 2 Groups Assigned• Intensive-Insulin: Blood Glucose 80 – 110• Conventional: Insulin therapy only if Blood Glucose > 215 with a

maintenance between 180 – 200

NEJM 2001: Hypothesis

Hyperglycemia or relative insulin deficiency

(or both) during critical illness may

directly or indirectly confer a predisposition

to complications, such as

severe infections,

polyneuropathy,

multiple-organ failure, and death.

NEJM 2001: Purpose

To determine

whether normalization of blood glucose levels

with intensive insulin therapy

reduces mortality and morbidity

among critically ill patients.

Some of the Logistics (1)

• Conventional Group– IV Insulin was started if the Blood Glucose exceeded 215

– Infusion was adjusted to maintain level between 180-200

• Intensive-Insulin Group– Started if Blood Glucose exceeded 110

– Infusion was adjusted to maintain level between 80 – 110

– Maximal rate of insulin was set at 50 IU per hr.

– Dose adjustment was via strict algorithm followed by ICU-nurses and assisted by a single study-physician that was NOT involved in the clinical mgmt of the patient

Some of the Logistics (2)

• On admission, all patients were fed continuously with IV Glucose (200 – 300 g/24 hrs).

• The next day, TPN, Combined Enteral-Parenteral, or Total Enteral Feeding was instituted according to a standardized schedule– 20-30 nonprotein kilocalories/kg/24 hrs– AND a balanced formula

• 0.13-0.26 g/N2/kg/24 hrs• 20-40 % of nonprotein calories via lipid solution

• Total Enteral Feeding was attempted as early as possible

Some of the Logistics (3)

• Original Plan was to enroll 2,500 patients in order to detect an absolute difference in mortality of 5%

• Interim analysis (conducted every 3 months) of overall mortality required the study be terminated early

• Sponsors were not involved in the study design, data collection, analysis, interpretation of the data, or preparation of the manuscript…

Demographics

• ½ of the pts were CT Surgery

•Note:

• the AGE

• the Hx of Cancer

• Hx of Diabetes

• % of pts above 200…

Method – Serious Study

• All patients admitted to the SICU from February 2, 2000 through January 18, 2001 were considered for enrollment – after consent was obtained

• Only 14 pts were excluded– 5 because of participation in other studies– 9 pts were moribund or DNR

A Few Points (1)

• 98% of the pts in the Intensive-Insulin Group required therapy– Mean Morning Blood Glucose Level: 103 +/- 19 mg/dl

• 39% of the pts in the Conventional Group required therapy– Treated group: Mean Morning Blood Glucose Level:

173 +/- 33 mg/dl– Untreated group: Mean Morning Blood Glucose Level:

140 +/- 25 mg/dl.

Results (1)

Results (2)

Mortality in Perspective (1)

• 35 pts in the Intensive Group Died (4.6 %)• 63 pts in the Conventional Group Died (8.0 %)

– Apparent Risk Reduction of 42 %– Unbiased Risk Reduction of 32 %

• Due to having to adjust for repeated interim analysis

• Intensive therapy also reduced the in-hospital mortality – mostly in those pts with multiple-organ failure secondary to a septic focus, regardless if there was a history of diabetes or hyperglycemia.

Results were similar in patients who had undergone CT Surgery versus other types of surgery

Results (4)

Mortality in Perspective (2)

• Since the introduction of Mechanical Ventilation, few direct interventions have actually improved ICU Survival.

Treatment of sepsis with Activated Protein C

results in a 20 % relative reduction

in mortality at 28 days…

glycemic control reduces R.R. of mortality by 42 %.

A Few Points (2)

• Hypoglycemia (Blood Glucose < 40 mg/dl)

– 39 pts in the Intensive Group • 2 of the 39 pts had associated sweating and agitation

– 6 pts in the Conventional Group

There were no instances of hemodynamic deterioration or convulsions !

Morbidity (1)

Intensive therapy reduced the duration of ICU stay

but not overall-hospital stay

• Intensive therapy reduced episodes of septicemia by 46 %

• Fewer pts in the Intensive Group required prolonged ventilatory support and renal replacement therapy – yet the number of patients that required inotropic or vasopressor support were the same between groups

Morbidity (2)

Variable Conventional Intensive p Val.Cr > 2.5 12.3 % 9.0 % 0.04Plasma Urea N2 > 54 11.2 % 7.7 % 0.02Dialysis or CVVH 8.2 % 4.8 % 0.007

Bilirubin > 2 26.7 % 22.4 % 0.04

Septicemia 7.8 % 4.2 % 0.003Tx with Abx > 10 days 17.1% 11.2% < 0.001

EMG-Polyneuropathy 51.9 % 28.7 % < 0.001

# Transfusions per Pt 2 1 < 0.001

Some Critique

• European Study (Belgium)• Not Blinded

– Team of ICU Nurses and a Specific Study Physician following Pre-designed Protocol

• Nutritional Protocol is not described or reported

• Insulin Protocol is not described or reported

– Independent of Clinical Decision-making Process

• SICU-specific patient population• Are the results “too good”… ?

NEJM 2001: Conclusions

the use of exogenous insulin

to maintain

blood glucose at a level

less than 110 mg/dl

reduces morbidity and morality

among critically ill patients in the Surgical ICU,

regardless of whether there is a

history of diabetes or hyperglycemia.

So, where are we going ?

“we need to re-adjust our thinking…”

“there is a set-point (similar to a thermostat)

that we must adjust clinically in order to apply this

information at the bedside…”

“no longer can we accept Blood Sugars

outside of the normal physiologic range”

Blood Sugars: Insulin Management in the ICU

Tisha K Fujii, DO, Bradley J. Phillips, MD

• Traditional Thinking: Blood Sugar less than 200 is adequate…after all, the kidney dumps sugar above 180.

• 2002 Thinking: The human system is designed to function with a Glucose between 80 and 120. It is a matter of will that we, as healthcare workers, force it to do otherwise.

The following is a suggested protocol to allow appropriate “blood sugar control” in the intensive care unit. We have employed its

use successfully in a variety of units (i.e. trauma, surgical, medical) and

believe that focusing specific attention at undue hyperglycemia is well-worth the

effort required.

ISPUB.COM

Blood Sugars in the ICU (in-press)

• If Glucose is 121 - 150: Give 2 unit bolus injection and start drip at 1 u/hr.• If Glucose is 151 - 175: Give 3 unit bolus injection and start drip at 1 u/hr.• If Glucose is 176 - 200: Give 4 unit bolus injection and start drip at 2 u/hr.• If Glucose is 201 - 250: Give 6 unit bolus injection and start drip at 2 u/hr.• If Glucose is 251 - 300: Give 8 unit bolus injection and start drip at 3 u/hr.• If Glucose is 301 - 350: Give 10 unit bolus injection and start drip at 3 u/hr.• If Glucose is 351 - 400: Give 12 unit bolus injection and start drip at 4 u/hr.• If Glucose is above 401: Give 15 unit bolus injection and start drip at 4 u/hr.

• Accuchecks q 1 hr. until Glucose is “steady-state” between 80 - 150, then q 2hrs ATC. Adjust Drip Rate as Necessary to fit Target Parameters.

• Remember, the real goal is 80 - 120, but for practical reasons we accept the range of 80 - 150.

* Hourly adjustments are usually in increments of 1-2 units (most patients seem to reach a “steady-state” in the range of 3-5 units/hr.). We have had multiple patients intermittently require rates of 8-12 units per hour.

Blood Sugars in the ICU (in-press)

A Tight Sliding Scale is also a component of Therapy:

Accucheck Treatment70 or below Give 1/3 amp D50. Recheck in 1 hr.71 - 80 Recheck in 1 hr.81 - 120 No direct treatment121 - 150 2 units and recheck in 1 hr.151 - 175 3 units and recheck in 1 hr.176 - 200 4 units and recheck in 1 hr.201 - 250 6 units and recheck in 1 hr.251 - 300 8 units and recheck in 1 hr.301 - 350 10 units, recheck in 1 hr..? Insulin Drip351 - 400 12 units, recheck in 1 hr..? Insulin Drip401 or greater 15 units, recheck in 1 hr., & notify MD.

ISPUB.COM

BMC Version: Insulin Protocol

Currently in development

• Critical Care Medicine

• ICU Staff

• Pharm. D.’s

• Committee and more committees…

WHY ??

NEJM 2001: Hypothesis

Hyperglycemia or relative insulin deficiency

(or both) during critical illness may

directly or indirectly confer a predisposition

to complications, such as

severe infections,

polyneuropathy,

multiple-organ failure, and death.

Questions & Comments

Thank you….

Insulin Therapy in the ICU:

Hyperglycemic Protocols

Bradley J. Phillips, M.D.

Critical Care Medicine

Boston Medical Center

Boston University School of Medicine