Unstable angina: LMWH was better than heparin.

Clinical bottom line (level 1b)

  1. LMWH reduced the number of revascularisation procedures (NNT = 19 at 14 days) , and number of recurrent angina, MI and death after one month (NNT = 29 at 14 days) .
  2. The side effect profiles were similar, with an increase in minor bleeds (NNH = 21 at 14 days) for LMWH.
Cohen et al: New England Journal of Medicine 1997; 337 (7): 447-452
Expires July 2003

The study

Double-blinded concealed randomised trial with intention-to-treat
Setting: 176 centres, USA, Canada, South America, Europe, 1994 to 1996

3171 patients (aged mean 64 years, 66% male) recent onset of angina at rest lasting more than 10 minutes occurring <24 hours before randomisation and one of:
  • new ST depression, ST elevation or T-wave inversion in at least two contiguous leads
  • documented prior MI or revascularisation procedure
  • results of invasive/noninvasive testing of coronary artery disease


Excluded if
  • <18 years old
  • pregnancy
  • left bundle branch block or pacemaker
  • persistent ST-T elevation
  • angina with established cause (anaemia/CHF)
  • contraindications to anticoagulation
  • creatinine clearance of less than 30 ml/min
  • refusal to give consent


    • Control Group: (n = 1564, 1564 analysed): 5000 U bolus unfractionated heparin and dose adjusted to keep aPTT 55-85 secs for 2 to 8 days and placebo LMWH
    Experimental Group: (n = 1607, 1607 analysed): enoxaparin 1 mg/kg subcutaneously twice daily and placebo heparin for 2 to 8 days
    All patients received aspirin 100 mg to 325 mg po once daily.
    100% followed for 30 days

    The evidence

    Outcome Time to outcome CEREERRRR
    (95% CI)    		ARR
    (95% CI)    		NNT
    (95% CI)
    death 30 days 56
    (3.60%)    		 47
    (2.90%)    		 20%
    (-17% to 45%)    		 0.66%
    (-0.58% to 1.89%)
    (NNT = 190 to infinity;
    NNH = 51 to infinity)
    MI 30 days 81
    (5.20%)    		 63
    (3.90%)    		 26%
    (-3% to 46%)    		 1.26%
    (-0.19% to 2.71%)    		 76
    (NNT = 800 to infinity;
    NNH = 36 to infinity)
    recurrent angina 30 days 282
    (18.0%)    		 252
    (15.7%)    		 13%
    (-2% to 25%)    		 2.35%
    (-0.26% to 4.96%)    		 43
    (-390 to 2-)
    death, MI or recurrent angina 30 days 364
    (23.3%)    		 318
    (19.8%)    		 15%
    (3% to 26%)    		 3.49%
    (0.63% to 6.35%)    		 29
    (16 to 160)
    revascularisation 30 days 503
    (32.3%)    		 434
    (27.0%)    		 16%
    (6% to 25%)    		 5.15%
    (1.98% to 8.33%)    		 19
    (12 to 50)
    major bleed 30 days 106
    (6.78%)    		 101
    (6.29%)    		 7%
    (-21% to 29%)    		 0.49%
    (-1.23% to 2.21%)    		 200
    (NNT = 81 to infinity;
    NNH = 45 to infinity)
    minor bleed 30 days 109
    (6.97%)    		 188
    (11.7%)    		 -68%
    (-110% to -34%)    		 -4.73%
    (-6.74% to -2.71%)    		 -21
    (-37 to -15)
    thrombocytopaenia (>50%) 30 days 56
    (3.58%)    		 39
    (2.43%)    		 32%
    (-1% to 55%)    		 1.15%
    (-0.04% to 2.34%)    		 87
    (NNT = 2800 to infinity;
    NNH = 43 to infinity)

  • No statement on LVEF and multi-vessel disease was documented.
  • No details on other antianginal treatments received which influence prognosis (eg. beta-blockers).
  • Treatment prematurely discontinued in 367 patients (11.6%; 207 on unfractionated heparin, 160 on LMWH).

    • Citation

    1. Cohen M, Demers C, Gurfinkel EP, et al: A comparison of low-molecular-weight heparin with unfractionated heparin for unstable coronary artery disease. New England Journal of Medicine 1997; 337 (7): 447-452
    Contributor: Nick Shenker and Chris Ball, July 2000
    Reviewer:

    Clinical Question.
    Patient unstable angina
    Intervention or Exposure LMWH
    Comparison unfractionated heparin
    Outcome death, pain, revascularisation