Skip Navigation


European Heart Journal Advance Access originally published online on November 6, 2007
European Heart Journal 2008 29(1):31-37; doi:10.1093/eurheartj/ehm503
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
29/1/31    most recent
ehm503v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Related articles in EHJ
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Vijayaraghavan, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vijayaraghavan, R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2007. For permissions please email: journals.permissions@oxfordjournals.org

Local hospital vs. core-laboratory interpretation of the admission electrocardiogram in acute coronary syndromes: increased mortality in patients with unrecognized ST-elevation myocardial infarction

Ram Vijayaraghavan1, Andrew T. Yan2,3, Mary Tan3, David H. Fitchett2,3, Alina A. Georgescu3, Quamrul Hassan3, Anatoly Langer2,3, Shaun G. Goodman for the Canadian Acute Coronary Syndromes Registry Investigators2,3,*,{dagger}

1 Sunnybrook Health Sciences Centre, Toronto, ON, Canada
2 Terrence Donnelly Heart Centre, Division of Cardiology, St. Michael’s Hospital, University of Toronto, 30 Bond St, Rm 6-034Q, Toronto, ON, Canada, M5B 1W8
3 The Canadian Heart Research Centre, Toronto, ON, Canada

Received 13 April 2007; revised 2 October 2007; accepted 4 October 2007; online publish-ahead-of-print 6 November 2007.

* Corresponding author. Tel: +1 416 864 5722, Fax: +1 416 864 5407, Email: goodmans{at}smh.toronto.on.ca

See page 1 for the editorial comment on this article (doi:10.1093/eurheartj/ehm573)


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Limitations
 Conclusions
 Funding
 Acknowledgements
 References
 
Aims: Previous analyses suggest only modest agreement between local site and core-laboratory (core-lab) electrocardiogram (ECG) interpretation in patients with acute coronary syndromes (ACSs); however, this has not been well examined outside of clinical trial populations.

Methods and results: Patients (n = 5277 from 51 hospitals; 4916 with 1 year vital status) participating in the Canadian ACS Registry who were hospitalized with an ACS and had an interpretable initial ECG were included in this study. Core-lab ECG interpretation was blinded to site interpretation and outcomes. There was moderate agreement between site and core-lab regarding the predominant ECG findings ({kappa} = 0.49). Patients with core-lab-defined ST-elevation and cardiac marker elevation (n = 1202) not classified as ST-elevation by the site were less likely to receive acetylsalicylic acid (ASA) (90 vs. 96%, P < 0.0001), heparin (91 vs. 95%, P = 0.04), and reperfusion therapy (14 vs. 76%, P < 0.0001) than patients for whom there was agreement that ST-elevation was present. After adjusting for other validated prognostic factors, site-unrecognized ST-elevation was independently associated with higher mortality (odds ratio = 2.21; 95% CI, 1.46–3.36; P < 0.001).

Conclusions: In patients with ACS, there was only moderate agreement between core-lab and site interpretation of the initial ECG. Site-unrecognized ST-elevation myocardial infarction was associated with underutilization of evidence-based therapies and increased 1-year mortality.

Key Words: Electrocardiogram • Acute coronary syndromes • Prognosis

{dagger} A list of participating Canadian ACS Registry Investigators and Coordinators may be found in the Arch Intern Med 2007;167:1009–1016. Back


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Limitations
 Conclusions
 Funding
 Acknowledgements
 References
 
Performance of an initial electrocardiogram (ECG) in the emergency department is a pivotal step in the evaluation of patients with chest pain. ECG abnormalities correlate with outcome1,2 and guide diagnostic and therapeutic decisions in patients with acute coronary syndromes (ACSs). Accurate interpretation of the initial ECG is therefore crucial in guiding the management of patients with ACS.

Clinical trials of therapies in ACS often rely largely on specific ECG abnormalities for entry criteria. Despite this, it has been shown that agreement between the site and blinded core-laboratory (core-lab) in the interpretation of the entry ECG varies widely from fair to good.3,4 Furthermore, one study demonstrated a worse outcome among patients with core-lab-defined ST-segment elevation myocardial infarction (MI) that was not recognized by the site.4

As important as these findings are with respect to the validity of clinical trials, the implications are potentially greater for clinical practice. There are conflicting data from relatively small studies describing the accuracy of ECG interpretation in ACS outside of clinical trials, and no examination of its association with clinically relevant outcomes.5–7 Thus, as part of the Canadian ACS Registry, we undertook a prospective, multi-centred, observational study evaluating the agreement between site and core-lab interpretation of the admission ECG. We describe the impact of differences in ECG interpretation on utilization of therapies and 1 year mortality with a particular focus on patients with ST-elevation MI, in whom the decision to administer reperfusion therapy is strongly influenced by ECG findings.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Limitations
 Conclusions
 Funding
 Acknowledgements
 References
 
Study population and design
Details of the ACS Registry rationale and methods have been published previously.8 Patients were recruited from 51 hospitals in nine provinces in Canada if: (i) they were ≥18 years old; (ii) they were admitted to hospital with a suspected ACS (defined by symptoms consistent with acute cardiac ischaemia within 24 h of onset); and (iii) the qualifying ACS was not deemed to be precipitated by a significant co-morbidity such as trauma or gastrointestinal bleeding. There were no other specific exclusion criteria, and all sites were encouraged to enrol consecutive patients. At each site, the designated physician or study co-ordinator recorded demographic and clinical data, relevant laboratory results, in-hospital treatment and outcomes, discharge medications, and site interpretation of the admission ECG on standardized case report forms (CRFs). The site ECG interpreters were practicing clinicians, including emergency physicians, internists, and cardiologists and did not include house officers. The CRF, along with a copy of the presenting ECG, was forwarded to the co-ordinating centre (Canadian Heart Research Centre, Toronto, Canada) and scanned directly into an electronic database (TeleformTM, Version 7.0, Cardiff, San Diego, CA, USA). To ensure accuracy, central data checks were performed and queries were sent for correction. The study complies with the Declaration of Helsinki and was approved by the local research Ethics Committee of each participating hospital, and informed consent was obtained from all patients after hospital discharge.

From September 1999 to June 2001, 5312 patients with suspected ACS were enrolled in the Canadian ACS registry. Of these patients, 5277 (99%) had interpretable admission ECG data and were included in the current analysis. One-year outcome data were available for 4916 patients (93%). Baseline characteristics and degree of agreement between the site and core-lab were similar in those with and without (n = 361) 1 year follow-up. A final diagnosis of ACS was ultimately given to 4598 patients by their respective treating physicians. Although our overall agreement and {kappa} analyses (see below) were performed on the larger cohort of 4916 patients with suspected ACS, repeat analysis of the final ACS diagnosis cohort gave similar results.

Electrocardiogram analysis
Core-lab interpretation of the admission ECG was performed by one of three non-cardiologist physicians blinded to CRF designation and patient outcomes. The co-ordinating centre core-lab has previous experience in the systematic evaluation of ECGs; inter- and intra-observer agreements have been previously demonstrated to be 93–999,10 and 100%,10 respectively. Multiple findings from the same ECG were recorded (e.g. ST-depression and T-wave inversion). If there were multiple ECG findings on a single CRF or according to the core-lab evaluation, the primary ECG diagnosis was assigned in the following mutually exclusive hierarchy: ventricular/paced rhythm, left bundle branch block (LBBB), ST-segment elevation (≥0.1 mV in ≥2 contiguous leads), ST-segment depression (≥0.1 mV in ≥2 contiguous leads), T-wave-inversion (≥0.1 mV in ≥2 contiguous leads) or other (including non-specific changes). These criteria for ST elevation, ST depression, and T-wave inversion were listed on the CRF.

Primary outcome
The primary outcome was cumulative all-cause mortality at 1 year. In-hospital deaths were recorded on the CRFs. Vital status at 1 year follow-up was ascertained by standardized telephone interview for hospital survivors.

ST-elevation myocardial infarction subgroup
For the purposes of a pre-specified subgroup analysis, ST-elevation MI at entry into the registry was defined by the presence of both core-lab ECG categorization as ST-elevation and cardiac biomarker [creatine kinase (CK)/creatine kinase MB fraction (CK-MB) or troponin] elevation. A total of 1310 patients with a final diagnosis of ACS met these criteria and were included in the subgroup analysis; 1202 patients had available 1 year outcome data.

Statistical analysis
Continuous variables are reported as mean values unless otherwise specified. Likelihood ratio {chi}2 statistics or Fisher’s exact tests were used for differences in categorical variables, including outcomes, among patient groups. Wilcoxon rank-sum tests were used to assess the differences in continuous variables. Kappa values were derived to express degree of concordance between site and core-lab ECG interpretation. A two-sided P-value of <0.05 was considered statistically significant. To adjust for differences in baseline risk and determine the independent prognostic significance of unrecognized ST-elevation by site investigators, we first categorized patients in the ST-elevation MI subgroup by tertiles of the GRACE risk score.11 The GRACE risk score has previously been validated as a predictor of all-cause mortality in the Canadian ACS Registry12 and includes the following clinical variables: age, systolic blood pressure, Killip class, heart rate, serum creatinine level, cardiac arrest during presentation, ST deviation, and positive initial cardiac enzymes; the latter two variables were, by definition, present in all patients for the analysis. We then performed multivariable logistic regression using GRACE risk score tertile, ECG interpretation agreement, and administration of reperfusion therapy. We also performed a sensitivity analysis by adjusting for other differences in baseline characteristics (in addition to the GRACE risk score) in the multivariable model. Data processing and statistical analyses were performed by the Canadian Heart Research Centre, using SAS software package, version 8.2 (SAS Institute, Cary, NC, USA).


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Limitations
 Conclusions
 Funding
 Acknowledgements
 References
 
Admission electrocardiogram
The core-lab characterized the primary abnormality of the admission ECG as ST-elevation in 32.8%, ST-depression in 13.3%, T-wave-inversion in 13.7%, and ‘other’ (including normal or non-specific abnormalities) in 33.7%. Left bundle branch block (5%) and ventricular/paced (1.4%) were identified in a minority of patients.

Table 1 compares the ECG classification between the core-lab and the sites. Overall concordance between core-lab and site interpretation of the admission ECG was 62%. Thus, there was disagreement of admission ECG classification in 38% of all patients. The calculated kappa value was 0.49 (95% CI, 0.47–0.50), indicating modest agreement. Concordance between core-lab and site evaluation was highest for LBBB and ‘other’ ECG findings, and approached nearly 80%. However, of the ECGs classified by the core-lab as having ST-elevation, ST-depression, and T-wave-inversion, the site’s classification was the same in only 59, 50, and 40% of patients, respectively. The remainder tended to be classified by the site as having ‘other’ ECG changes.


View this table:
[in this window]
[in a new window]

  		Table 1 Comparison of core-lab and site classification of admission electrocardiograms

 
Kappa values were similar in academic ({kappa} = 0.49; 95% CI, 0.47–0.52) and non-academic ({kappa} = 0.48; 95% CI, 0.46–0.50) hospitals, and in hospitals with ({kappa} = 0.49; 95% CI, 0.46–0.52) and without ({kappa} = 0.49; 95% CI, 0.47–0.51) on-site cardiac catheterization facilities.

Baseline characteristics
Table 2 compares baseline characteristics of patients with and without agreement in ECG interpretation between site and core-lab. Compared with patients with core-lab and site agreement in ECG interpretation, those in the group without agreement had a higher prevalence of prior MI, angina, congestive heart failure, and coronary bypass surgery (P < 0.001 for all comparisons). Patients in whom there was disagreement in ECG interpretation were also more likely to be male (70 vs. 65%, P = 0.001).


View this table:
[in this window]
[in a new window]

  		Table 2 Overall baseline characteristics stratified by presenting ECG interpretation agreement between site and core-lab

 
Outcome
Of the 4916 patients with available 1 year outcome data, mortality was 9.4% (n = 464). As compared with patients in whom the core-lab and site agreed on the ECG classification, those in whom the core-lab and site disagreed had a significantly higher mortality at 1 year (10.8 vs. 8.6%, P = 0.01) (Figure 1).


Figure 1
View larger version (13K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1 One year mortality stratified by agreement in ECG interpretation between core-lab and site. *P < 0.05 for agreement vs. disagreement; {dagger}core-lab-defined ST-elevation and positive CK, CK-MB, and/or troponin{ddagger}

 
ST-elevation myocardial infarction subgroup
A total of 1310 patients with a final ACS diagnosis met the criteria for ST-elevation MI based on core-lab ECG designation and positive cardiac biomarkers. Baseline characteristics of these patients are listed in Table 3. The site did not identify ST-elevation in 29% (n = 383) of core-lab-defined ST-elevation MI patients. Patients with disagreement in ECG interpretation were older and more likely to have diabetes or a prior history of angina, MI, congestive heart failure, or coronary artery bypass surgery, but less likely to be smokers. Gender was similar between the two groups.


View this table:
[in this window]
[in a new window]

  		Table 3 Selected clinical characteristics of the st-elevation myocardial infarction subgroup stratified by agreement in ECG interpretation between site and core-lab (n = 1310)

 
There was significantly less use of conventional therapies in patients in whom ST-elevation was not identified by the site (Figure 2). For agreement and disagreement in ECG interpretation, respectively, the use within the first 24 h of hospital presentation of ASA was 96 vs. 90% (P < 0.0001), heparin 95 vs. 91% (P = 0.04), and reperfusion therapy (fibrinolysis or primary PCI) 76 vs. 14% (P < 0.0001). Fibrinolysis was the reperfusion treatment in the vast majority of cases (96%), and there was no difference between the groups regarding the use of primary percutaneous coronary intervention (PCI; 2.7% for agreement vs. 1.7% for disagreement, P = 0.27). 14.6% of patients with core-lab-defined ST-elevation MI identified by the site had contraindications to fibrinolytic administration, compared with 24.2% of patients in whom ST-elevation was not recognized. These contraindications included presentation beyond 12 h of symptom onset, significant blood loss from any site that would be considered a relative contraindication, and major surgery or trauma within 2 weeks of ACS presentation; however, other potential contraindications were not specifically addressed by the CRF.


Figure 2
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2 Selected therapies in the ST-elevation myocardial infarction subgroup* stratified by agreement in ECG interpretation between core-lab and site. *Core-lab-defined ST-elevation and positive CK, CK-MB, and/or troponin; {dagger}P < 0.05 for agreement vs. disagreement; {ddagger}ticlopidine or clopidogrel; §fibrinolytic therapy or primary percutaneous coronary intervention

 
A total of 1202 patients in the ST-elevation MI subgroup (92%) had available 1 year outcome data. Of these patients, 29% (n = 352) not identified by the site experienced a significantly higher 1 year mortality of 14 vs. 9% (P = 0.01) in those patients where the core-lab and site agreed (Figure 1).

Of all the patients with core-lab-defined ST-elevation MI who were not identified by the site, 19% had ST-elevation of at least 2 mm in two contiguous precordial leads (V1–V6).

Multivariable analysis
The GRACE risk score demonstrated very good discrimination for 1 year mortality in the ST-elevation MI subgroup (c-statistic = 0.77; 95% CI, 0.72–0.81; P < 0.001). Patients were categorized into tertiles of the GRACE risk score in order to characterize baseline risk of mortality. The group in which ECG interpretation agreed had a significantly higher risk score (P < 0.001); however, adjustment for GRACE risk tertiles revealed that disagreement in ECG interpretation between site and core-lab was independently associated with increased mortality [adjusted odds ratio (OR), 2.21; 95% CI, 1.46–3.36; P < 0.001]. A similar association (adjusted OR, 1.93; 95% CI, 1.26–2.96; P = 0.003) was noted when analysis was performed using the GRACE score as a continuous variable. This relationship was consistently observed irrespective of hospital type (academic vs. non-academic, those with vs. without on-site cardiac catheterization facilities). Further adjustment of other differences in baseline characteristics did not attenuate the results (adjusted OR, 2.25; 95% CI, 1.39–3.66; P = 0.001) in the sensitivity analysis. However, after adjustment for use of reperfusion therapy and in-hospital revascularization, the independent association of disagreement in ECG interpretation with mortality became less strong and was no longer statistically significant (OR, 1.41; 95% CI, 0.86–2.31; P = 0.17).

Comments
In this study of ‘real-world’ ACS patients, comparison of ECG interpretation between site and core-lab demonstrated only modest agreement. Lack of agreement in ECG interpretation was associated with increased mortality at 1 year. In the subgroup of patients with core-lab-defined ST-elevation MI that was not identified by the site, there was underutilization of proven therapies and higher 1 year mortality. Furthermore, a significant proportion of patients with ST-elevation MI who were apparently not identified by the site had profound ECG abnormalities.

Clinical trials analyses suggest that agreement between site and independent core-lab interpretation of the presenting ECG in patients with ACS varies widely.3,4 In a substudy of the Thrombin Inhibition in Myocardial Ischemia (TRIM) trial, local investigator ECG interpretation was compared with blinded analysis by core-lab experts in 516 patients presenting with non-ST-elevation ACS.3 There was poor agreement on the presence and extent of ST-segment elevation ({kappa} = 0.05) and fair agreement on ST-segment depression ({kappa} = 0.38). The authors hypothesized that this discrepancy may have been a consequence of bias towards including patients in the trial who would have normally been excluded, i.e. those with ST-elevation. An analysis of the GUSTO-IIb trial in 12 142 patients with ACS demonstrated a higher 1 year mortality in patients with site-defined non-ST-elevation ACS who actually had core-lab-defined ST-elevation, compared with those in whom the designation of non-ST-elevation was concordant (8.8 vs. 6.8%, P = 0.0093).4 This occurred despite better overall agreement in ECG interpretation ({kappa} = 0.67) than in the TRIM substudy.3

Thus, the clinical trial data have supported the twin hypotheses that there are significant discrepancies between local investigator and core-lab interpretation of the presenting ECG in patients with ACS, and that these differences probably impact adversely on outcomes, especially in those patients with unrecognized ST-elevation who are potential candidates for reperfusion. However, the factors affecting ECG interpretation may be different in the clinical trial setting than those in less selected populations with ACS. Recruitment expectations, economic incentives, and possible gain for patients may, in part, influence characterization of the presenting ECG in patients undergoing assessment for inclusion in clinical trials.3

The limited registry data available demonstrate conflicting results. A recent single-centre Swiss study of 692 ACS patients found that emergency physicians failed to recognize 5% of cases of transient ST-elevation compared with expert interpreters7; overall concordance was 74% ({kappa} = 0.51). Another single-centre emergency room study showed that only 1% of ST-elevation MI was missed.6 Both these smaller studies postulated that the clinical effects of inaccurate ECG diagnoses would be minimal. However, an older but much larger multi-centre study of 2320 patients presenting to the emergency room with possible ACS found that treating physicians failed to recognize 41% of ST-segment abnormalities compared with ECG specialists, and there was an association with suboptimal triage for coronary care unit admission.5

Our study is the first to offer additional insights on the impact of failure to recognize ECG changes and outcomes in a less-selected population of patients with ACS. In contrast to previous study populations, the Canadian ACS registry had a broader selection of patients with both ST-elevation and non-ST-elevation ACS. Furthermore, this registry population is likely less fraught with selection bias than those of clinical trials. Our finding of overall modest agreement of 62% ({kappa} = 0.49) is, therefore, more likely to be generalizable than previous estimates. This level of concordance is consistent with that in clinical trial settings3,4 as well as the largest registry experiences.5,7 One-year mortality was higher by 1.8% (absolute risk difference) in the group of patients in whom ECG classification was discordant. This suggests that the magnitude of ECG misdiagnoses may be sufficient to impact adversely on patient outcome.

Patients in whom ECG interpretation disagreed between site and core-lab comprised a high-risk group with higher rates of previous congestive heart failure, previous MI, coronary artery bypass surgery, angina, and male gender. This likely contributed to their higher overall mortality compared with patients in whom ECG interpretation agreed. However, it can be argued that such high-risk patients are precisely those who stand to gain the greatest absolute benefit of targeted medical and interventional therapies and, therefore, in whom accurate ECG classification is most important.

ST-elevation myocardial infarction subgroup
We pre-defined a subgroup of confirmed ACS patients with ST-elevation by core-lab criteria and positive cardiac biomarkers in order to capture a high-risk population that would be particularly sensitive to differences in ECG interpretation. Given that ECG classification is pivotal in decision-making regarding treatment, it is not surprising that we found underutilization of reperfusion therapies in patients with ST-elevation not identified on the CRF by the site (see Figure 2). We have likely overestimated this difference in the use of reperfusion therapy for two reasons. First, 14% (n = 53) of patients who were ‘missed’ by the site (n = 383) were, in fact, treated with reperfusion therapy within the first 24 h. Thus, we may presume that in these cases ST-elevation on the presenting ECG was, in fact, recognized by the treating physician despite the CRF indicating another ECG finding. Secondly, a proportion (~24%) of the patients with ST-elevation unrecognized by the site had contraindications to fibrinolytic therapy, including presentation more than 12 h after onset of symptom (22%), active bleeding (1.9%), or recent surgery or trauma (0.3%). However, this still leaves the majority of patients (290/383) with unclear reasons as to why reperfusion therapy was not initiated.

Patients with core-lab-defined STEMI not identified by the site experienced a statistically, significantly higher 1 year mortality rate, representing an excess absolute mortality of 5% and a relative risk increase of >50%. Multivariable analysis was performed to adjust for differences in baseline risk as captured by the GRACE risk score; this confirmed the independent association of disagreement in ECG interpretation with mortality. Interestingly, the association was no longer statistically significant when we included the administration of reperfusion therapy in the model. This suggests that the effect of ECG interpretation on outcome is, as one would expect, mediated by the decision to administer reperfusion therapy. An overview of nine trials of fibrinolytic therapy for acute ST-elevation MI has shown an 18% relative risk reduction in 35-day mortality,13 which persists over the long term.14,15 This is concordant with the hypothesis that a significant portion of the mortality increase in patients with ST-elevation MI not identified by the site is attributable to the lack of administration of reperfusion therapy.

Finally, to emphasize the sometimes non-trivial nature of core-lab-defined ST-segment elevation not identified by the site, 19% of those cases apparently not recognized by the site had, in fact, at least 2 mm of ST-elevation in at least two contiguous precordial leads (V1–V6).


   Limitations
Top
 Abstract
 Introduction
 Methods
 Results
 Limitations
Conclusions
 Funding
 Acknowledgements
 References
 
There are a number of important limitations of this study to consider. First, abstraction of site classification of the presenting ECG from the patient chart may have led to misrepresentation of the treating physician’s ECG diagnosis. This is evidenced by the fact that 14% of patients with core-lab-defined ST-elevation unrecognized by the site were, as noted above, treated with reperfusion therapy within the first 24 h. In addition, the treating physician may not have characterized patients as having an ST-elevation ACS when the ST-elevation was known to be ‘old’. We sought to minimize this problem by including only patients with a final ACS diagnosis in the ST-elevation MI subgroup analysis. Both these ECG classification ‘errors’ would have led to an overestimation of the number of reperfusion-eligible patients among those in whom ST-elevation was apparently not identified by the site. Thus, using the core-lab interpretation as the ‘gold standard’ also means that some patients might have inappropriately received (e.g. fibrinolytic) therapy if indeed the site interpretation was correct.

Secondly, we did not examine all possible contraindications to fibrinolytic therapy. However, we would not expect these to comprise a substantial number. Thirdly, this registry included a non-random sample of Canadian hospitals between 1999 and 2001. This may limit generalizability to other patient populations and treating facilities. However, our estimate of ECG interpretation agreement appears compatible with those reported previously, involving non-Canadian centres; we can only speculate that the impact on outcomes in other jurisdictions is likely similar. Finally, the use of aggregate risk stratification with the GRACE risk score in the multivariable analysis raises the possibility that we have failed to control for all differences in baseline risk that could influence outcome. However, inclusion of other baseline characteristics did not improve the discriminatory performance of the GRACE risk score,11 which has also been validated in the ACS Registry population.12 Furthermore, our sensitivity analysis, which adjusted for other differences in baseline characteristics in addition to GRACE risk score, confirmed the robustness of our results.


   Conclusions
Top
 Abstract
 Introduction
 Methods
 Results
 Limitations
 Conclusions
Funding
 Acknowledgements
 References
 
In this study of ‘real-world’ ACS patients, the comparison of ECG interpretation between site and core-lab demonstrated only modest agreement. In the subgroup of patients with core-lab-defined ST-elevation MI, ST-elevation unrecognized by the site was independently associated with increased mortality, which appeared to be mediated by underutilization of reperfusion therapy. Although contemporary registry data have shown that up to 30% of eligible patients do not receive reperfusion therapy,16,17 the results of this analysis support the existence of a significant additional population of patients who are deprived of reperfusion therapy and its attendant benefits, based on misdiagnosis of their presenting ECG. Our results suggest that treatment gaps may be even larger than previously recognized, and there remain important opportunities to improve the care of ACS patients.


   Funding
Top
 Abstract
 Introduction
 Methods
 Results
 Limitations
 Conclusions
 Funding
Acknowledgements
 References
 
The Canadian ACS Registry was sponsored by the Canadian Heart Research Centre (a federally incorporated not-for-profit academic research organization) and Key Pharmaceuticals, Division of Schering Canada Inc. The industrial sponsor had no involvement in the study conception or design; collection, analysis, and interpretation of data; in the writing, review, or approval of the manuscript; and in the decision to submit the manuscript for publication.


   Acknowledgements
Top
 Abstract
 Introduction
 Methods
 Results
 Limitations
 Conclusions
 Funding
 Acknowledgements
References
 
We thank Sue Francis for her secretarial assistance, and are indebted to the physicians, study co-ordinators, and patients who participated in the registry.

Conflict of interest: D.H.F., A.L., and S.G.G. have received speaker and consulting honoraria, and research grant support from Key Schering.


   Footnotes
 
{dagger} A list of participating Canadian ACS Registry Investigators and Coordinators may be found in the Arch Intern Med 2007;167:1009–1016. Back


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Limitations
 Conclusions
 Funding
 Acknowledgements
 References
 

  1. Savonitto S, Ardissino D, Granger CB, Morando G, Prando MD, Mafrici A, Cavallini C, Melandri G, Thompson TD, Vahanian A, Ohman EM, Califf RM, Van de Werf F, Topol EJ, for the Global Use of Strategies to Open Coronary Arteries Investigators (GUSTO). Prognostic value of the admission electrocardiogram in acute coronary syndromes. JAMA (1999) 281:707–713.[Abstract/Free Full Text]
  2. Cannon CP, McCabe CH, Stone PH, Rogers WJ, Schactman M, Thompson BW, Pearce DJ, Diver DJ, Kells C, Feldman T, Williams M, Gibson RS, Kronenberg MW, Ganz LI, Anderson HV, Braunwald E. The electrocardiogram predicts one-year outcome of patients with unstable angina and non-Q-wave myocardial infarction: Results of the TIMI III registry ECG ancillary study. J Am Coll Cardiol (1997) 30:133–140.[Abstract]
  3. Holmvang L, Hasbak P, Clemmensen P, Wagner G, Grande P. Differences between local investigator and core-laboratory interpretation of the admission ECG in patients with unstable angina pectoris or non-Q-wave myocardial infarction (A thrombin inhibition in myocardial ischemia (TRIM) substudy). Am J Cardiol (1998) 82:54–60.[CrossRef][Web of Science][Medline]
  4. Goodman SG, Fu Y, Langer A, Barr A, Tan M, Wagner GS, Barbagelata A, Sgarbossa EB, Birnbaum Y, Granger CB, Califf RM, Van de Werf F, Topol EJ, Armstrong PW, for the GUSTO-IIb Investigators. The prognostic value of the admission and pre-discharge electrocardiogram in acute coronary syndromes: the GUSTO-IIb ECG Core Laboratory experience. Am Heart J (2006) 152:277–284.[CrossRef][Web of Science][Medline]
  5. Jayes RL Jr, Larsen GC, Beshansky JR, D’Agostino RB, Selker HP. Physician electrocardiogram reading in the emergency department–accuracy and effect on triage decisions: findings from a multicenter study. J Gen Intern Med (1992) 7:387–392. Jul-Aug.[Web of Science][Medline]
  6. Brady WJ, Perron A, Ullman E. Errors in emergency physician interpretation of ST-segment elevation in emergency department chest pain patients. Acad Emerg Med (2000) 7:1256–1260.[Web of Science][Medline]
  7. Berger A, Meier JM, Stauffer JC, Stauffer JC, Eckert P, Schlaepfer J, Gillis D, Cornuz J, Yersin B, Schaller MD, Kappenberger L, Wasserfallen JBl. ECG interpretation during the acute phase of coronary syndromes: in need of improvement? Swiss Med Wkly (2004) 134:695–699.[Medline]
  8. Yan AT, Tan M, Fitchett D, Chow CM, Fowlis RA, McAvinue TG, Roe MT, Peterson ED, Tu JV, Langer A, Goodman SG, for the Canadian Acute Coronary Syndromes Registry Investigators. One-year outcome of patients after acute coronary syndromes (from the Canadian Acute Coronary Syndromes Registry). Am J Cardiol (2004) 94:25–29.[CrossRef][Web of Science][Medline]
  9. Fu Y, Goodman S, Chang W-C, Van de Werf F, Granger CB, Armstrong PW, for the ASSENT-2 Investigators. Time to treatment influences the impact of ST-segment resolution on one-year prognosis. Circulation (2001) 104:2653–2659.[Abstract/Free Full Text]
  10. Goodman SG, Barr A, Langer A, Wagner GS, Fitchett D, Armstrong PW, Naylor CD. Development and prognosis of non-Q-wave myocardial infarction in the thrombolytic era. Am Heart J (2002) 144:243–250.[Web of Science][Medline]
  11. Fox KAA, Goldberg RJ, Pieper K, Cannon CP, Van de Werf F, Goodman SG, Dabbous O, Granger CB, for the Global Registry of Acute Coronary Events Registry Investigators (GRACE). Predictors of in-hospital mortality in the Global Registry of Acute Coronary Events. Arch Intern Med (2003) 163:2345–2353.[Abstract/Free Full Text]
  12. Yan AT, Jong P, Yan RT, Tan M, Fitchett D, Chow CM, Roe MT, Pieper KS, Langer A, Goodman SG, for the Canadian Acute Coronary Syndromes (ACS) Registry Investigators. Clinical trial-derived risk model may not generalize to real-world patients with acute coronary syndrome. Am Heart J (2004) 148:1020–1027.[CrossRef][Web of Science][Medline]
  13. Fibrinolytic Therapy Trialists’ (FTT) Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet (1994) 343:311–322.[CrossRef][Web of Science][Medline]
  14. Baigent C, Collins R, Appleby P, Parish S, Sleight P, Peto R, for the ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. ISIS-2: 10-year survival among patients with suspected acute myocardial infarction in randomised comparison of intravenous streptokinase, oral aspirin, both, or neither. BMJ (1998) 316:1337–1343.[Abstract/Free Full Text]
  15. Franzosi MG, Santoro E, De Vita C, Geraci E, Lotto A, Maggioni AP, Mauri F, Rovelli F, Santoro L, Tavazzi L, Tognoni G, for the GISSI Investigators. Ten-year follow-up of the first mega-trial testing thrombolytic therapy in patients with acute myocardial infarction: results of the Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto-1 study. Circulation (1998) 98:2659–2665.[Abstract/Free Full Text]
  16. Eagle KA, Goodman SG, Avezum A, Budaj A, Sullivan CM, Lopez-Sendon J, for the GRACE investigators. Practice variation and missed opportunities for reperfusion in ST-segment-elevation myocardial infarction: findings from the Global Registry of Acute Coronary Events (GRACE). Lancet (2002) 359:373–377.[CrossRef][Web of Science][Medline]
  17. Rogers WJ, Canto JG, Lambrew CT, Tiefenbrunn AJ, Kinkaid B, Shoultz DA, Frederick PD, Every N. Temporal trends in the treatment of over 1.5 million patients with myocardial infarction in the U.S. from 1990 through 1999. J Am Coll Cardiol (2000) 36:2056–2063.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?

Related articles in EHJ:

Interpretation of the electrocardiogram: clinical correlation suggested
William J. Brady and Robert E. O'Connor
EHJ 2008 29: 1-3. [Extract] [FREE Full Text]  



This article has been cited by other articles:


Home page
 JAMAHome page
A. T. Yan, R. T. Yan, and S. G. Goodman
Misinterpretation of Electrocardiograms and Cardiac Catheterization Laboratory Activations
JAMA, April 23, 2008; 299(16): 1897 - 1897.
[Full Text] [PDF]

Home page
 Eur Heart JHome page
W. J. Brady and R. E. O'Connor
Interpretation of the electrocardiogram: clinical correlation suggested
Eur. Heart J., January 1, 2008; 29(1): 1 - 3.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
29/1/31    most recent
ehm503v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Related articles in EHJ
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Vijayaraghavan, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vijayaraghavan, R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?