Chest pain: clinical features and ECG helped in the initial diagnosis but cardiac enzymes did not.

Clinical bottom line (level 1b)

  1. Roughly 40% of patients with chest pain had unstable angina or a myocardial infarction.
  2. In patients with chest pain, unstable angina or MI was more likely if
    • previous history of MI (LR+2.3)
    • pain described as a pressure (LR+1.7)
    • male (LR-1.3)
    • increasing age
    • ECG indicating probable MI (LR+8.7)
    • ECG indicating ischemia or strain not known to be old (LR+3.1)
  3. Unstable angina or MI was less likely if
    • sharp or stabbing pain (LR+0.41)
    • no previous history of MI or angina (LR-0.37)
    • pain pleuritic, positional or reproduced on palpation (LR+0.13)
    (Patients with all three are unlikely to have MI or unstable angina)
  4. Traditional 'cardiac enzymes' taken in the emergency department were not very helpful. AST is most helpful at diagnosing or excluding MI if taken > 12 hours after the onset of chest pain.
Lee et al: Archives of Internal Medicine 1985; 145: 65-69
Expires March 2003

The study

Setting: emergency department, university hospital, USA

596 patients (aged mean 56 years, 52% female) anterior, precordial or left-sided chest pain

Excluded if
  • <25
  • not willing to return in 48-72 hours for repeat ECG and cardiac enzymes
  • local trauma
  • abnormal chest X-ray



Independent blinded reference standard, applied in all patients from a consecutive appropriate spectrum.
Reference standard:
  • MI if any of
    • AST > twice upper limit of normal, which then returned to normal. No intramuscular injection, muscle trauma or hepatic disease
    • CK-MB > 5% total CK
    • LDH1 > LDH2. No haemolytic anaemia or renal infarction
    • ECG - new pathological q waves (> 40 ms duration and > 25% decrease in amplitude of following R wave)
    • scintiscan showing focal uptake in cardiac area; if enzyme peak occurred before hospital admission and patient had no previous MI or valve calcification
    unstable angina if any of
    • senior clinician's diagnosis not contraindicated by follow-up
    • abnormal exercise tolerance test, abnormal angiogram or follow-up history
    • known angina pain which worsened
Diagnostic test:
  • history and physical
  • CK, AST, LDH

The evidence

pre-test probability of MI: 17%, (95% CI: 14% to 21%)
pre-test probability of unstable angina: 24%, (95% CI: 21% to 27%)

diagnostic test MI or unstable angina no MI or unstable angina LR
(95% CI)		 post-test probability
aged 80+ 25 14 3.53
(1.94 to 6.42) 		71%
aged 70-79 43 28 2.17
(1.39 to 3.39) 		61%
aged 60-69 75 60 1.77
(1.31 to 2.38) 		56%
aged 50-59 56 70 1.13
(0.83 to 1.54) 		44%
aged 40-49 31 88 0.50
(0.34 to 0.72) 		26%
aged 30-39 5 61 0.12
(0.047 to 0.28) 		8%
aged 25-29 2 28 0.10
(0.024 to 0.42) 		7%
total 247 349


diagnostic test MI or unstable angina no MI or unstable angina LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
male 135 151 1.3
(1.1 to 1.5) 		47% 0.80
(0.68 to 0.94) 		36%
pressure 137 116 1.7
(1.4 to 2.0) 		54% 0.67
(0.57 to 0.78) 		32%
aching 25 54 0.65
(0.42 to 1.0) 		32% 1.1
(1.0 to 1.1) 		43%
burning/ indigestion 19 24 1.1
(0.63 to 2.0) 		44% 0.99
(0.95 to 1.0) 		41%
sharp or stabbing 35 122 0.41
(0.29 to 0.57) 		22% 1.3
(1.2 to 1.5) 		48%
other 31 33 1.3
(0.84 to 2.1) 		48% 0.97
(0.91 to 1.0) 		41%
previous history of MI or angina 184 115 2.3
(1.9 to 2.7) 		62% 0.38
(0.30 to 0.48) 		21%
pain pleuritic, positional or reproduced by palpation 13 138 0.13
(0.077 to 0.23) 		9% 1.6
(1.4 to 1.7) 		53%
total 247 349


diagnostic test MI or unstable angina no MI or unstable angina LR
(95% CI)		 post-test probability
pleuritic pain 0 36 0.0
(0.0 to 0.12) 		0%
partly pleuritic pain 13 83 0.22
(0.13 to 0.39) 		14%
pain not pleuritic 234 230 1.4
(1.3 to 1.6) 		50%
total 247 349


diagnostic test MI or unstable angina no MI or unstable angina LR
(95% CI)		 post-test probability
positional pain 2 22 0.13
(0.03 to 0.54) 		8%
pain partly positional 20 92 0.31
(0.19 to 0.48) 		18%
pain not positional 225 235 1.4
(1.3 to 1.5) 		49%
total 247 349


diagnostic test MI or unstable angina no MI or unstable angina LR
(95% CI)		 post-test probability
pain reproduced by chest wall palpation 9 115 0.11
(0.057 to 0.21) 		7%
pain partially reproduced by chest wall palpation 8 26 0.43
(0.20 to 0.94) 		24%
pain not reproduced by chest wall palpation 230 208 1.6
(1.4 to 1.7) 		53%
total 247 349


diagnostic test MI or unstable angina no MI or unstable angina LR
(95% CI)		 post-test probability
probable MI 74 12 8.7
(4.8 to 16) 		86%
ischaemia or strain not known to be old 78 36 3.1
(2.1 to 4.4) 		68%
ischaemia or strain or infarction but changes known to be old 33 27 1.7
(1.1 to 2.8) 		55%
abnormal but not diagnostic of ischaemia 18 54 0.47
(0.28 to 0.78) 		25%
non-specific ST or T wave changes 38 112 0.48
(0.34 to 0.67) 		25%
normal 6 108 0.078
(0.035 to 0.18) 		5%
total 247 349


diagnostic test MI no MI LR
(95% CI)		 post-test probability
probable MI 63 23 13
(8.5 to 20) 		73%
ischaemia or strain not known to be old 29 85 1.6
(1.1 to 2.3) 		25%
ischaemia or strain or infarction but changes known to be old 4 56 0.34
(0.13 to 0.91) 		7%
abnormal but not diagnostic of ischaemia 3 69 0.21
(0.066 to 0.64) 		4%
non-specific ST or T wave changes 4 146 0.13
(0.049 to 0.34) 		3%
normal 1 113 0.042
(0.0059 to 0.30) 		1%
total 104 492


diagnostic test MI or unstable angina no MI or unstable angina LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
sharp or stabbing pain; no prior MI or angina 6 92 0.092
(0.041 to 0.21) 		6% 1.3
(1.2 to 1.4) 		48%
sharp or stabbing pain; pain pleuritic, positional or reproduced by palpation 2 64 0.044
(0.011 to 0.18) 		3% 1.2
(1.2 to 1.3) 		46%
sharp or stabbing pain; no prior MI or angina; no prior ; MI or angina pain pleuritic, positional or reproduced by palpation 0 48 0.0
(0.0 to 0.088) 		0% 1.2
(1.1 to 1.2) 		45%
total 247 349


diagnostic test MI no MI LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
CK >180 (in emergency department) 41 70 1.8
(1.3 to 2.5) 		37% 0.77
(0.65 to 0.91) 		20%
total 102 320


diagnostic test MI no MI LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
AST >60 (in emergency department) 24 15 6.1
(3.5 to 10) 		62% 0.43
(0.29 to 0.64) 		10%
total 39 148


diagnostic test MI no MI LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
AST >47 (in emergency department) 25 28 3.4
(2.3 to 5.1) 		47% 0.44
(0.29 to 0.68) 		10%
total 39 148


diagnostic test MI no MI LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
LDH >200 0.59 0.30 2.0
( to ) 		35% 0.59
( to ) 		14%
total


diagnostic test MI no MI LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
CK or AST abnormal 57 103 1.7
(1.4 to 2.2) 		36% 0.65
(0.51 to 0.82) 		17%
total 101 314


diagnostic test MI no MI LR
(95% CI)		 post-test probability
AST level in patients with chest pain onset >12 hours ago: 100 0.30 0.01 30
( to ) 		89%
80 0.09 0.04 2.3
( to ) 		39%
60 0.28 0.05 5.6
( to ) 		60%
50 0.02 0.05 0.40
( to ) 		10%
40 0.13 0.22 0.59
( to ) 		14%
30 0.12 0.37 0.32
( to ) 		8%
<30 0.02 0.26 0.078
( to ) 		2%
total


diagnostic test MI no MI LR
(95% CI)		 post-test probability
AST level in patients with chest pain onset <12 hours ago: 100 0.07 0.03 2.3
( to ) 		39%
80 0.05 0.03 1.7
( to ) 		32%
60 0.11 0.03 3.7
( to ) 		50%
50 0.13 0.04 3.3
( to ) 		47%
40 0.16 0.26 0.62
( to ) 		14%
30 0.29 0.39 0.74
( to ) 		17%
<30 0.19 0.22 0.86
( to ) 		19%
total

  • Data for the final two tables was obtained from ROC curves.

Comments

  1. CK was found to be unhelpful in diagnosing MI if patients attended >12 hours after the onset of chest pain.
  2. The study was performed before CK-MB was introduced- this is a helpful test at <12 hours.

Citation

  1. Lee TH, Cook F, Weisberg M, et al: Acute chest pain in the emergency room: identification and examination of low-risk patients. Archives of Internal Medicine 1985; 145: 65-69
Search Terms: bibliographic reference from 'Evidence-based Cardiology'
Contributor: Chris Ball and Clare Wotton, March 2000
Reviewer:

Clinical Question.
    Patient chest pain
    Intervention or Exposure clinical features and ECG
    Outcome diagnosis of MI or unstable angina