Friday, August 31, 2012

Angiography can have bleeding complications - be more sure of your STEMI diagnosis in high risk patients

An over 90 yo patient complained of 1 hour of chest pain.  There was a h/o CAD with CABG.  The prehospital BP was 200 systolic.  The paramedics recorded an ECG which could not be obtained but which was identical to the ED ECG recorded here:
There is sinus rhythm.  There is some STE in aVL, with reciprocal ST depression in II, III, aVF, and STD in V6.  There is also a QR-wave in aVL,  consistent with old MI.  Although one might be worried about acute MI here, it is likely that the ST elevation, without large T-wave and with Q-wave, is due to old MI (LV aneurysm morphology).

The paramedics activated the cath lab.  The CP resolved with NTG and ASA, and the ED BP was 170/90.  She was given heparin and abciximab and taken to the cath lab.  There were several severe chronic calcified lesions, no culprit, and no PCI was performed.

However, there were severe bleeding complications and she died of a retroperitoneal hemorrhage.

There are several lessons one can be reminded of by this case:

1. Not all ST elevation is acute STEMI
2. Acute STEMI usually has large T-waves and, unless anterior, does not develop Q-waves rapidly.  When you see inferior or lateral QR-waves, think of old MI with persistent ST elevation.
3. Hypertension can cause ischemia with chest pain
4. If the pain goes away and the ECG is equivocal, then slow down
5. PCI has risks, mostly bleeding risks, and these are much higher in the elderly, especially women

Summary of this case:

ECG differentiation of old vs. new MI is frequently difficult, so assess risks and uncertainties:




Known previous MI with CABG
Chest pain resolved.
BP 200 systolic

Bleeding Risk Factors  in PCI

GRACE registry 24000 patients; Eur Ht J 24:1815; 2003

Major bleeding definition: life threatening:
  • Transfusion of ≥2 units PRBC or
  • Decrease in hematocrit of ≥10% or
  • Intracranial hemorrhage or
  • Death
The overall incidence of Major Bleeding in PCI:
  • 4.8% in patients with STEMI
  • 4.7% in patients with Non-STEMI
Risk Factors for Major Bleeding in PCI:
  • Advanced age
  • Female sex
  • History of bleeding
  • Renal insufficiency
  • (Diabetes and stroke in other studies)
  • After adjustment, major bleeding is:
    • Significantly associated with increased risk of hospital death (OR 1.64, 95% CI: 1.18, 2.28).
    • Also: Transfusion is associated with worse ischemic outcomes!
How to Lower Risk in High Risk Patients 
Assess risk benefit

Be more certain of diagnosis

High risk vs. lower risk STEMI
  • Anterior vs. other locations
  • Poor LV function on echo
  • Hemodynamics (shock, blood pressure, pulse)
  • Pulmonary edema
  • ST score (how many leads involved? How widespread?)
  • Ongoing symptoms (vs. resolved)
  • Resolution of ECG findings associated with much lower risk
Appropriate dosing of Heparin
60 U/kg Loading dose: IBW + 0.4(TBW-IBW), maintenance 12 U/kg/hr: IBW + 0.4(TBW-IBW)

Use of bivalirudin
  • (always with clopidogrel) vs. heparin + abciximab
    • HORIZONS-AMI, 3 yr outcome, 2011
      • Lower mortality (5.9 % vs. 7.7%)
      • Lower cardiac mortality (2.9% vs. 5.1%
      • Lower major bleeding (6.9% vs. 10.5 %)
    • ISAR-REACT 2011
      • No difference in ischemic outcomes or death
      • Major Bleeding 4.6 % vs. 2.6%
Radial artery access
Clopidogrel 300 mg, vs. 600 mg, vs. Prasugrel, vs. Ticagrelor (the latter have less ischemia, more bleeding)


  1. I agree with the importance of getting the diagnosis right, but I think a fair amount of blame should be placed on the heparin and glycoprotein inhibitor in this case. Shouldn't we be questioning why two medicines that provide no patient-important outcome benefits but have clear bleeding harms continue to be prescribed, when benefits are pretty limited to things like "improved TIMI 3 flow" and subjective endpoints like "need for urgent revascularization?"

    1. Thanks, Graham:

      I don't want to "blame" anyone. And certainly not the paramedics.

      The physicians in the ED could have de-activated the cath lab, but it's hard to stop the momentum once it is going, and you have to really be good at reading the EKG.

      The interventionalist could have, too. But it's late at night and his whole team got up to come in. And he is also not sure of the diagnosis.

      If you're going to do PCI, you have to have heparin or bivalirudin; it can't be done without. And you have to have the antiplatelet agents, too, for the best outcome. The medications alone, of course, did not cause the bleeding: it was the combination of getting a large groin needle stick AND the meds.

      Bottom line is that there will always be false positive cath lab activations and some of them will have bleeding complications. The ones at highest risk are most likely to have the complications and these are the patients we must be most careful about not having a false positive.

      Steve Smith

  2. Hey Steve -- "blame" wasn't the right word, you're right.

    Why do you have to have heparin or bivalirudin? Perhaps during the procedure, but do they need to be maintained on a heparin drip after?

    My point was really that we shouldn't be loading people up multiple medications that increase bleeding (and have clear, well-documented harms) when there aren't clear, well-documented, patient-important benefits as well, or when the risks are clearly greater than the benefits.

    1. They are not necessary for the angiogram, but without heparin, or bivalirudin + clopidogrel, the wires used will dangerously clot.

  3. what if chest pain resolved but ECG still showing STE with high certainty? do you give TPA?

    1. Not in this case, because the ECG suggests old MI, not acute STEMI. In other cases: if the STE is unequivocally diagnostic of acute STEMI, not a look-alike, not a "maybe", and remains so with serial ECGs, then I would give tPA even if symptoms had resolved. But I would be much more skeptical of the ECG findings and scrutinize them much more.

  4. Hey doctor smith! I know this is an old case, but do you think the pain resolving with nitro as well as some of the ST changes could've been exacerbated by the severe HTN? It may be her baseline due to old MI, but I feel it's possible it could've normalized further as well as her pain been eliminated with adequate blood pressure control.

    1. Cortland,
      Yes, that is possible.
      I don't have subsequent ECGs any longer with which to answer definitively.

  5. I thought the small q waves in aVL are septal q waves. Overall tough ECG if its not acute. If not a high risk patient then activating cath lab for such an ECG justified in a symptomatic patient?

    1. I do not think this looks like acute OMI. I would not have activated based on this ECG, especially after pain resolved with NTG.

    2. So what the ECG changes suggestive of?
      Given the 90yr old age , i would have gotten an echo before shifting to cath esp when chest pain has resolved as i read it as acute MI; tough for me since its not.

    3. Suggestive of old MI with persistent STE.


DEAR READER: We welcome your Comments! Unfortunately — due to a recent marked increase in SPAM — we have had to restrict commenting to Users with a GOOGLE Account. If you do not yet have a Google account — it should not take long to register. Comments give US feedback on how well Dr. Smith’s ECG Blog is addressing your needs — and they help to clarify concepts of interest to all readers. THANK YOU for your continued support!

Recommended Resources