IV Insulin

1
IV Insulin

• The Alchemists Dream
• It is Working!!!

2
DIGAMI

• Control Intensive
• Patient numbers 306 314
• Deaths 138 (44) 102(33)
• Absolute reduction mortality 11
• One person saved for every 9 patients treated
• aspirin therapy 80 80
• beta Blockers 70 70
• ACE inhibitors 31 31
• Insulin therapy at discharge 135 (43) 266 (87)
• Insulin therapy 3 months 141 (45) 245 (80)
• Insulin therapy 1 year 141 (49) 220 (72)
• (Malmberg K, BMJ 3141512-1515, 1997)

3
Mortality After MI Reduced by Insulin Therapy
DIGAMI Study
IV Insulin 48 hours, then
4 injections daily
All Subjects
.7
.7
Low-risk and not previously on Insulin
(N 620)
(N 272)
.6
.6
Risk reduction (51)
Risk reduction (28)
.5
.5
P .011
P .0004
.4
.4
.3
.3
.2
.2
.1
.1
0
0
0
1
2
3
4
5
0
1
2
3
4
5
Years of Follow-up
Years of Follow-up
Malmberg, et al. BMJ. 19973141512-1515.
6-11
4
Van Den Berghe

• Conventional Intensive P value
• (N783) (N765)
• Administration of insulin
• No.() 307(39.2) 755(98.7) lt 0.001
• Insulin dose IU/day
• Median 33 71
• Range 17-56 48-100 lt0.001
• Duration of insulin use
• ICU stay
• Median 67 100
• Range 40-100 lt0.001
• Morning blood glucose
• Mgs/dl
• All patients 153/-33 103/-19
  lt0.001
• Patients receiving insulin 173/-33 103/-18
  lt0.001
• (Van Den Berghe NEJM, 34519 1359-1367, 2001)

5
Van Den Berghe

• Conventional Intensive P value no.
  () no.()
• Deaths during ICU
• Total 63/783 (8.0) 35/765 (4.6) lt0.04
• During first 5 days 14 (1.8) 13
  (1.7) 0.9
• ICU gt 5 days 49/243 (20.2) 22/208 (10.6) 0.005
• No history of diabetes 57/680 (8.4) 31/664
  (4.7)
• Causes of death (no.)
• Multiple organ failure
• with sepsis 33 8
• Multiple organ failure
• no sepsis proved 18 14
• Severe brain damage 5 3
• Acute Cardiac 7 10
• In Hospital deaths 85/783 (10.9) 55/765
  (7.2) 0.01
• (Van Den Berghe NEJM, 34519 1359-1367, 2001)

6
Patients in the Study
Van den Berghe, G. et al. N Engl J Med
2006354449-461
7
Van Den Berhge

• Intention to treat Group
  in ICU gt 3 days
• Conventional Intensive Conventional Intensive
• (N605) (N595) (N381) (N386)
• Total deaths ICU
• 162(26.8) 144(24.2)
  145(38.1) 121(31.3)
• P 0.31 P
  0.05
• In Hospital Deaths
• 242(40.0) 222(37.3)
  200(52.5) 166(43.0)
• P 0.33 P 0.09
• No significant differences across groups based on
  apache scores or causes of death.
• In ICU lt 3 days (433 total patients)
• Conventional Intensive
• Total Deaths 42 56

8
Effect of Intensive Insulin Therapy on Morbidity
Van den Berghe, G. et al. N Engl J Med
2006354449-461
9
Goals of Therapy

• ICU patients near normal-glycemia without
  hypoglycemia. The exact target for each group of
  patients is not clear from the literature. The
  original Van Den Berghe trial achieved an average
  glucose of 103. Dr Furnary has in cardiac
  patients utilized a level 90-100.
• Achieving goals recommended for outpatient
  therapy. These goals apply to hospitalized
  patients as well.
• Fasting Glucose lt 110 and gt 70
• 2 hour post meal lt 140 and clearlylt 180
• Modified from AACE Guidelines 2004

10
CAUTIONS

• The Van Den Berghe trials aggressively replaced
  or restored nutrition.
• The accuracy of capillary blood glucose
  measurements can lead to potential pitfalls in
  preventing and recognizing low glucose
  measurements
• An appropriately designed insulin protocol must
  be titrated with relatively stable glucose or
  carbohydrate infusions. If a source of glucose
  is changed ( NPO for a test). The protocol needs
  to be violated to avoid potential risks.

11
IV Insulin at Barnes

• The protocol has been tested in the 84 ICU and
  the results published in The Journal of the
  American College of Surgeons 2005.
• Blood glucoses lt 150 were achieved in 7.2 hours.
  The average blood sugar achieved with this
  protocol was 132. Time to lt 111 was 13.2 hours.
• Hypoglycemia lt 40 was between 1 and 3 . Some of
  the hypoglycemia resulted from failure of the
  nurses to adjust the protocol. The major risk
  for hypoglycemia in this study was similar to Van
  Den Berghes study. Creatinine gt 1.5. 5/ 6
  episodes of hypoglycemia lt 40 was associated with
  titration errors
• The Protocol was modified for patients with renal
  dysfunction.

12
IV Insulin at Barnes

• The literature suggests a benefit of near normal
  glycemia in the critically ill patient. The
  effect of hyperglycemia on markers of
  inflammation can be seen with short term lt 4 hour
  exposure in clamp studies in isolated monocytes.
• The protocol we are using is a modified Portland
  protocol. Can we expect similar outcomes in
  mortality and morbidity?
• The modifications in the Barnes ICU protocol were
  made after bedside study.

13
Diabetes at BJH

• We are currently studying the safety of the
  transition order sets in the 84 ICU. The
  transition order set is based on 60 of the IV
  insulin rate over the last 24 hours. The
  reduction in the dose is based on the patients
  improving condition and to assist in
  transitioning to the floor order sets
• The use of basal Bolus insulin therapy in
  hospitalized patients outside the ICU is also
  being studied. The basis for the dosing being
  0.5 units per Kg. Initial results are suggest
  stable control with a small risk of hypoglycemia

14
Conclusion

• The use of IV insulin in the critically ill
  patient has been shown to have morbidity and
  mortality benefits. Many of these studies have
  been single center studies.
• I will suggest that the use of IV insulin has
  become the standard of care.
• The use of protocols in the studies vary both in
  the use of nutrition replacement as well as the
  rate at which the sugars are normalized.
• The use of protocols requires training and
  supervision. The nursing staff has been
  receptive to the insulin protocol despite an
  increased work-load.