Chest pain: LBBB: ECG changes and cardiac enzymes helped diagnose a myocardial infarction

Clinical bottom line (level 2b)

  1. One in seven patients with suspected myocardial ischaemia and LBBB had a myocardial infarction.
  2. The following tests helped diagnose a myocardial infarction
    • elevated initial CK-MB/RI or ECG changes with concordant ST-elevation or depression (LR + 44)
    • elevated initial CK-MB/RI (LR + 29)
    • clinical impression of high risk (LR + 13)
    • ECG changes (LR + 6.6)
  3. The following tests made a myocardial infarction less likely
    • normal initial CK-MB/RI and no concordant ST-elevation or depression on ECG (LR - 0.38)
    • normal initial myoglobin (LR - 0.39)
    • old LBBB on ECG (LR - 0.41)
Kontos et al: Ann Emerg Med 2001; 37 : 431-438
Expires March 2004

The study

Setting: emergency department, unversity hospital, USA

182 patients (aged mean 66, 70% male) presenting with possible myocardial ischaemia and a left bundle branch block (LBBB) on ECG

Independent blinded reference standard, applied in all patients from a consecutive inappropriate spectrum.
Reference standard:
  • serial sampling of CK and CK-MB; a relative CK index was calculated using (CK-MB x 100)/CK. Myocardial infarction diagnosed if CK-MB 8.0 ng/ml or higher and CK relative index 4.0, and there was a classic rise and fall in enzymes.
Diagnostic test: ECG (read by 2 cardiologists) and initial biochemical markers. ECG changes assessed were
  • ST-elevation of 1 mm concordant with the QRS complex
  • ST-depression of 1 mm in leads V1, V2 or V3
  • ST-elevation of 5 mm discordant with QRS complex

The evidence

pre-test probability of myocardial infarction: 13%, (95% CI: 8.3% to 18%)

diagnostic test myocardial infarction no myocardial infarction LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
ECG changes 11 11 6.6
(3.2 to 13) 		50% 0.58
(0.40 to 0.84) 		8%
new or indeterminate LBBB 20 93 1.4
(1.1 to 1.8) 		18% 0.41
(0.16 to 1.0) 		6%
new LBBB 10 56 1.2
(0.70 to 2.0) 		15% 0.90
(0.63 to 1.3) 		12%
clinical impression (high risk) 6 3 13
(3.5 to 49) 		67% 0.76
(0.61 to 0.96) 		10%
total 24 158


diagnostic test myocardial infarction no myocardial infarction LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
initial myoglobin elevated 12 20 4.4
(2.6 to 7.3) 		38% 0.39
(0.20 to 0.76) 		5%
total 18 131


diagnostic test myocardial infarction no myocardial infarction LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
elevated initial CK-MB/RI 10 2 29
(6.8 to 130) 		83% 0.59
(0.42 to 0.83) 		9%
elevated initial CK-MB/RI or ECG with concordant ST-elevation or depression 15 2 44
(11 to 180) 		88% 0.38
(0.23 to 0.64) 		6%
elevated initial myoglobin, or elevated initial CK-MB/RI or ECG with concordant ST-elevation or depression 15 20 4.4
(2.6 to 7.3) 		43% 0.44
(0.26 to 0.74) 		7%
total 24 140

Comments

  1. Emergency physicians are less accurate at reading ECGs than cardiologists, so the diagnostic tests are likely to be less useful in reality.
  2. The diagnostic test results that include CK-MB should be treated with caution, since these are not independent of the reference standard, and therefore are likely to be overestimated. Furthermore these results come only from patients admitted to hospital.

Citation

  1. Kontos MC, McQueen RH, Jesse RL, et al: can myocardial infarction be rapidly identified in emergency department patients who have left bundle-branch block?. Ann Emerg Med 2001; 37 : 431-438
Search Terms: from ACP Journal Club
Contributor: Chris Ball, March 2002
Reviewer:

Clinical Question.
Patient suspected myocardial ischaemia and LBBB
Intervention or Exposure ECG and cardiac enzymes
Outcome myocardial infarction