MI & Hyperglycaemia: Intensive insulin therapy
reduces mortality!
|
In patients with a blood glucose > 11 mmol/L
following a heart attack, for every 9 patients given an insulin-glucose
infusion for ³ 24 hrs then qid
subcutaneous insulin for ³ 3
months, compared to standard therapy, one less death occurs over the next
3.5 years. (95% CI 5 to 34)
Level 1+ evidence
|
Citation/s:
Prospective randomised study of intensive insulin treatment on long term
survival after acute MI in patients with diabetes mellitus. BMJ 1997; 314(7093):
1512-1515.
Lead author's name and fax: Klas Malmberg, DIGAMI study group, Dept of
Cardiology, Karolinska Hospital, S-171 76 Stockholm, Sweden
Three-part Clinical Question:
Among hyperglycaemic patients with acute myocardial infarction, does more
intensive insulin therapy reduce mortality?
Search Terms: hyperglycaemia AND myocardial infarction
The Study: Non-blinded
concealed randomised controlled trial with intention-to-treat.
The Study Patients: All patients admitted to CCU with an acute MI within
the previous 24 hours (with or without a previous diagnosis of diabetes
mellitus) whom had a blood glucose >11 mmol/L.
Control group (N = 314; 314 analysed): Standard CCU care. These patients
did not receive any insulin unless clinically indicated. The
clinical indications were not clearly stated either in this paper or in
the previous one J Am Col Cardiology 1995; 26: 57-65. If there were no
contraindications, thrombolysis, B blockers & aspirin was started.
Experimental group (N = 306; 306 analysed): Standard CCU care plus an
insulin-glucose infusion for ³ 24 hours
and then qid subcutaneous insulin for ³
3 months. If there were no contraindications, thrombolysis, B blockers & aspirin
was started.
The Evidence:
|
Outcome |
Time to
Outcome |
CER |
EER |
RRR |
ARR |
NNT |
|
Mortality
at 1 year |
1 year
|
0.261 |
0.190 |
27% |
0.071 |
14 |
|
95% Confidence
Intervals: |
2% to 52% |
0.005 to
0.137 |
7 to 182 |
|
Mortality
at end of follow up (Median 3.5yrs) |
End of
follow up 3.5 years |
0.439 |
0.333 |
24% |
0.106 |
9 |
|
95% Confidence
Intervals: |
7% to 41% |
0.030 to
0.182 |
5 to 34 |
Comments:
-
Benefit greatest in low risk
patients (<70 years old, no prior MI, no CHF, no digoxin therapy) not
previously on insulin.
-
50% thrombolysed; by discharge 80%
given aspirin, 70% given beta-blockers, and 31% given ACE-inhibitors. No
difference between groups.
-
72 % in the infusion group & 49% in
the control group were taking insulin at 1 year.
-
The infusion group required 2 extra
hospital days.
Appraised by: Kevin Rooney,
Department of Anaesthesia & Intensive Care Medicine, Royal Alexandra Hospital,
Paisley; 30 December 2002 Email:
kd.rooney@virgin.net
Kill or Update By: December
2004
Printer friendly format
|