Three Part Question
In [patients with chest pain suggestive of myocardial ischaemia] does [the presence of chest wall tenderness] [help to rule out myocardial ischaemia]?
During a busy shift in the Emergency Department, you see a 50 year-old man with dull central chest pain and feel that, although he is clinically stable and the initial ECG is normal, myocardial ischaemia ought to be ruled out. He is very keen to get back to work, doesn't like hospitals and doesn't want to spend the day awaiting blood tests. On examination you elicit chest wall tenderness. You wonder if this sign is sufficiently reliable to allow the exclusion of an acute coronary syndrome.
OVID Medline 1966 – May 2005 using the OVID interface
EMBASE 1974 – May 2005 using Dialog datastar
CINAHL 1982 - May 2005 using Dialog datastar
Cochrane Library 2005 Issue 2
(exp Angina Pectoris/ OR exp Coronary Disease/ OR exp Myocardial Ischemia/ OR (myocard$ ADJ ischaem$).mp. OR (myocard$ ADJ ischem$).mp. OR (myocard$ ADJ infarct$).mp. OR exp Myocardial Infarction/ OR heart attack.mp. OR exp Angina, Unstable/ Or exp Coronary Thrombosis/ OR (acute ADJ coronary ADJ syndrome).mp. OR MI.mp. OR AMI.mp.) AND (tender$.af. OR palpat.$.af. OR exp Palpation/)
[myocardial ADJ ischaemia OR Heart-Muscle-Ischemia#.DE. OR Coronary-Artery-Disease#.DE OR Ischemic-Heart-Disease#.DE. OR Angina-Pectoris#.DE. OR Unstable-Angina-Pectoris#.DE. OR acute ADJ coronary ADJ syndrome OR Heart-Muscle-Ischemia#.DE. OR Acute-Heart-Infarction#.DE.] AND [palpat$.af. OR palpation#.W..DE. OR tender$.af.]
(Myocardial-Infarction#.DE. OR Myocardial-Ischemia#.DE. OR (myocard$ ADJ infarct$).mp. OR (myocard$ ADJ ischem$).mp. OR (myocard$ ADJ ischaem$).mp. OR (heart ADJ attack).mp. OR (acute ADJ coronary ADJ syndrome).mp.) AND (tender$.mp OR palpat$.mp. OR Palpation#.W..DE.)
([MeSH descriptor Myocardial Ischemia] OR [MeSH descriptor Myocardial Infarction] OR [MeSH descriptor Angina, Unstable] OR [MeSH descriptor Angina Pectoris]) AND (palpat* OR tender* OR [MeSH descriptor Palpation])
Altogether 121 papers were identified using OVID, 207 papers using EMBASE and 61 papers using CINAHL. The same three relevant papers were identified using OVID and EMBASE, one of which was also identified using CINAHL. 8 papers were identified using the reported Cochrane search, none of which were relevant.
|Author, date and country
||Study type (level of evidence)
|Panju AA et al|
|Patients from three studies that enrolled patients presenting to Emergency Departments with symptoms consistent with acute coronary syndromes (total number of patients not stated)||Systematic review and meta-analysis||Acute myocardial infarction (diagnostic standard taken as cardiac enzyme rises and ECG changes)||Likelihood ratio 0.2-0.4||Study designed to evaluate the predictive values of many other clinical features in the diagnosis of acute coronary syndromes. As such, the design is suboptimal for the purposes of this three-part question
No attempt to address publication bias|
|Goodacre et al|
|893 consecutive patients admitted to the Chest Pain Observation Unit (CPOU) of the Emergency Department at a large urban teaching hospital.
25% of all patients attending with chest pain were admitted to the unit and included in the study. Patients with ST elevation or depression, deep T wave inversion, new left bundle branch block, co-morbidity (such as arrhythmia or heart failure), serious alternative pathology necessitating admission or definite unstable angina with prolonged or recurrent pain were not admitted to CPOU and not included.||Prospective observational cohort||Acute myocardial infarction (WHO criteria) at presentation||Presence of chest wall tenderness: Multivariate analysis: Odds ratio 0.2 (95% confidence interval 0.05-0.97, p=0.045).||Study designed to evaluate the predictive values of many other clinical features in the diagnosis of acute coronary syndromes.
No sample size calculation
Inter- and intra-observer variability not evaluated
Exact sites and methods for eliciting chest wall tenderness not described|
|Absence of chest wall tenderness:Sensitivity 91.7% (74.2-97.7%), Specificity 27.8% (24.6-31.2%); positive predictive value 4.2% (2.8-6.3); negative predictive value 99.0% (96.3-99.7); Positive likelihood ratio 1.27 (1.12-1.44), negative likelihood ratio 0.30 (0.08-1.14). Multivariate analysis: Odds ratio 0.6 (95% confidence interval 0.3-1.2, p=0.18)|
|Chun AA; McGee SR|
United States Of America
|442 patients with symptoms consistent with acute coronary syndromes enrolled into three studies||Systematic review and meta-analysis||Acute myocardial infarction (diagnostic standard taken as elevated cardiac isoenzyme levels, diagnostic ECG changes or both)||If chest wall tenderness present: Likelihood ratio 0.3 (95% confidence interval 0.2-0.4); If chest wall tenderness absent: Likelihood ratio 1.3 (1.1-1.4); Sensitivity 3-15%; specificity 64-83%||Study designed to evaluate the predictive values of many other clinical features in the diagnosis of acute coronary syndromes.
No attempt to address publication bias
Heterogeneity of effect not described or taken into account|
Ruling out acute coronary syndromes in clinically stable patients with acute chest pain who have a non-diagnostic ECG presents a significant challenge to the emergency physician. It is often impossible to confidently exclude myocardial ischaemia on the basis of history alone. Chest wall tenderness is a clinical sign that often persuades physicians to make a diagnosis of musculoskeletal pain.
The three studies identified strongly suggest that chest wall tenderness predicts a reduced likelihood of acute myocardial infarction and acute coronary syndrome. However, given that the pre-test probability of myocardial infarction was 12.5 - 17.4% in the studies utilised by the two meta-analyses and taking a likelihood ratio of 0.3, the post-test probability of myocardial infarction is 4.3 - 6.3%. Given the potential implications of missing this diagnosis, implementing chest wall tenderness as an independent rule out strategy in this patient group would lead to an unacceptably high rate of false negative diagnoses.
Clinical Bottom Line
In patients with acute chest pain, chest wall tenderness suggests that acute coronary syndrome is less likely but it does not effectively rule the diagnosis out.
Level of Evidence
Level 1 - Recent well-done systematic review was considered or a study of high quality is available.
- Panju AA, Hemmelgarn BR, Guyatt GH, Simel DL Is This Patient Having A Myocardial Infarction? 1998; 280(14): 1256-1263
- Goodacre S, Locker T, Morris F, Campbell S How Useful Are Clinical Features In The Diagnosis Of Acute, Undifferentiated Chest Pain? Academic Emergency Medicine 2002; 9: 203-208
- Chun AA; McGee SR Bedside Diagnosis of Coronary Artery Disease: A Systematic Review American Journal Of Medicine 2004; 117: 334-343