Unstable angina or MI: hirudin prevented more infarcts but caused more bleeding.

Clinical bottom line (level 1b)

  1. Hirudin prevented more MI in patients with chest pain and ST-T changes on ECG than heparin (NNT = 110 at 30 days) , but had no effect on mortality.
  2. However, more patients had non-life threatening bleeds (NNT = 92 at 30 days) , and possibly more life-threatening ones.
GUSTO-IIb Investigators : New England Journal of Medicine 1996; 335 (11): 775-782
Expires July 2003

The study

Double-blinded concealed randomised trial with intention-to-treat
Setting: 373 hospitals in 13 countries, North America, Europe, Australasia

12142 patients (aged mean 65 years, 70% male) chest discomfort in the last 12 hours associated with transient or persistent ECG changes (ST-T depression/elevation or T-wave changes)

Excluded if
  • taking warfarin
  • active bleeding
  • history of stroke
  • contraindication to heparin therapy
  • renal insufficiency (Cr>177 mmol/dl)
  • systolic blood pressure >200 mmHg or diastolic blood pressure >110 mmHg
  • women of childbearing age


Control Group: (n = 6073, 6073 analysed): heparin (bolus of 5000 units iv, followed by 1000 units per hour, adjusted so aPTT 60-85 secs) and placebo for 3 to 5 days
Experimental Group: (n = 6069, 6069 analysed): hirudin (bolus of 0.1 mg/kg iv, followed by 0.1 mg/kg/hr adjusted so aPTT 60-85 secs) and placebo for 3-5 days
All patients had aspirin, nitrates, beta-blockers, calcium channel blockers and ACE inhibitors as required. In patients with ST elevation, thrombolysis was given (streptokinase or tPA) if clinician decided.
100% followed for 30 days
Outcome notes:

  • MI : CK-MB elevated above two times limit of normal with appropriate signs, symptoms and ECG changes
  • severe bleed : intracranial haemorrhage or major bleed requiring transfusion

The evidence

Outcome Time to outcome CEREERRRR
(95% CI)		ARR
(95% CI)		NNT
(95% CI)
death 30 days 285
(4.69%)		 273
(4.50%)		 4%
(-13% to 18%)		 0.19%
(-0.55% to 0.94%)		 510
(NNT = 110 to infinity;
NNH = 180 to infinity)
MI 30 days 382
(6.29%)		 328
(5.40%)		 14%
(1% to 26%)		 0.89%
(0.05% to 1.72%)		 110
(58 to 2000)
severe bleed 30 days 69
(1.14%)		 73
(1.20%)		 -9%
(-52% to 22%)		 -0.10%
(-0.48% to 0.28%)		 -1000
(NNT = 360 to infinity;
NNH = 210 to infinity)
moderate bleed 30 days 468
(7.71%)		 534
(8.80%)		 -14%
(-29% to -1%)		 -1.09%
(-2.07% to -0.11%)		 -92
(-880 to -48)
intracranial haemorrhage 30 days 12
(0.20%)		 18
(0.30%)		 -50%
(-211% to 28%)		 -0.10%
(-0.28% to 0.08%)		 -1000
(NNT = 1300 to infinity;
NNH = 360 to infinity)

Comments

  1. Raw data recalculated from percentages given in paper.
  2. Study not large enough to show differences in benefit for patients with ST elevation compared to those without ST elevation.
  3. The study was somewhat biased against heparin since heparin was not administered in a weight-based fashion as is now standard practice.
  4. Hirudin costs more than heparin without substantial advantages. LMWH has been shown to be significantly better and safer than heparin. Direct comparisons of LMWH and hirudin in large clinical trials are currently lacking

Citation

  1. GUSTO-IIb Investigators , : A comparison of recombinant hirudin with heparin for the treatment of acute coronary syndromes. New England Journal of Medicine 1996; 335 (11): 775-782
Contributor: Nick Shenker and Clare Wotton, July 2000
Reviewer: Arnold Baas

Clinical Question.
Patient chest pain
Intervention or Exposure heparin
Comparison hirudin
Outcome death or MI