Left Bundle Branch Block ECG Characteristics in the Evaluation of Acute Coronary Syndrome

Left Bundle Branch Block Clinical Trial

Official title:

Correlation of Clinical Outcomes With ECG Findings in Patients With Left Bundle Branch Block Being Evaluated for Acute Coronary Syndrome - a Prospective Cohort Study

Clinical Trial Details — Status: Terminated

Administrative data

• NCT number: NCT02283619
• Other study ID #: Pro00051429
• Status: Terminated
• Phase: N/A
• First received: October 31, 2014
• Last updated: September 24, 2015
• Start date: November 2014
• Est. completion date: August 2015

Study information

• Verified date: August 2015
• Source: Duke University
• Contact: n/a
• Is FDA regulated: No
• Health authority: United States: Institutional Review Board
• Study type: Observational

Clinical Trial Summary

Purpose of the study:

The goal of this study is to evaluate the relationship of electrocardiogram (ECG) findings with clinical outcomes in a cohort of patients with left bundle branch block (LBBB) who are being evaluated for acute coronary syndrome (ACS) in the Emergency Department (ED).

Background and significance:

The significance of specific ECG findings in patients with LBBB being evaluated for ACS has been inadequately studied, and this gap in knowledge is a barrier to optimal management of this population. Due to the speed, availability, low cost, and non-invasive nature of the ECG, it would be ideal to identify ECG characteristics that help to risk stratify these patients in order to inform clinical decision-making, reduce unnecessary invasive testing, and conserve resources.

Methods:

In this prospective observational study the investigators will identify a consecutive series of adult patients with LBBB presenting to the ED with suspicion of ACS. The investigators will collect data including demographics, cardiac risk factors, initial ECG measurements, lab and radiographic results, procedure results, and clinical outcomes such as 30-day death or myocardial infarction (MI). The investigators will analyze the data using a cohort study design to calculate odds ratios between ECG characteristics and the outcomes of interest.

Description:

Purpose of the study:

The goal of this study is to identify initial electrocardiogram (ECG) findings (including ST/S ratio and T-wave morphology) that predict clinical outcomes (myocardial infarction, positive stress test, revascularization and death), in a cohort of Emergency Department (ED) patients with left bundle branch block (LBBB) who are being evaluated for acute coronary syndrome (ACS).

Design & procedures:

The investigators will prospectively identify all adult (18yrs and older) patients presenting to their institution's ED who are being evaluated for ACS and have LBBB on ECG. The investigators will collect data on demographics (including age, but not date of birth), cardiac risk factors, date and time of presentation to the ED, ECG measurements, lab and radiographic results (cardiac biomarker data including date and time of the sample, stress test results, coronary imaging results, coronary catheterization results) and clinical outcomes such as 30-day death or MI. They will collect copies of ECGs and measure the following: ST/S ratio (ratio of the amplitude of ST segment deviation to the S wave amplitude); T-wave morphology (entirely discordant, entirely concordant, or biphasic to the PR interval); and standard ST segment deviation. A logistic regression analysis will be used to identify which ECG features are associated with 30-day death, MI, and revascularization. In the sub-cohort of subjects with negative "modified Sgarbossa criteria" a subanalysis will identify the outcome rates for the following specific sub-populations: those less than 2 consecutive partially-concordant T-wave leads and those with 2 or more consecutive partially-concordant T-wave leads.

Selection of Subjects:

The population will be patients age 18 or older who present to the ED with LBBB and suspected ACS. The investigators expect the mean age of the study population to be approximately 60 years old, with most patients between 40 and 90. They anticipate that these patients will on average have multiple medical comorbidities given that those with LBBB and ACS are reported to be a sicker, older population with more cardiac risk factors than the usual ACS patient without LBBB. Patients under age 18 will be excluded because they extremely rarely have ACS and/or LBBB. The investigators anticipate 300 subjects.

Inclusion criteria: age 18 and older, LBBB on ECG, presenting to the ED and being evaluated for ACS.

Exclusion criteria: prisoners, institutionalized individuals, patients who are incapable of giving legally effective consent, patients who cannot understand spoken English.

Consent process:

After the initial screening process, if the potential subject meets the inclusion/exclusion criteria and does not object to learning more about the study during initial discussion with their primary provider, then one of the key personnel trained and authorized to consent will approach the potential subject to discuss the study by explaining the following: the purpose of the study and the future benefits it may provide in subsequent patients, the prospective and observational nature of the study, the study procedures, the risks of loss of confidentiality, and the safeguards against that risk. All information will be clearly printed in plain language on the consent form. The potential subject will then be given the choice to enroll in the study.

Only the potential subject will be allowed to provide consent. The potential subject will have until discharge or transfer from the Emergency Department to decide whether or not to participate in the study. The minimal risks and minimal subject involvement justify this practical and logical time requirement. The consent process will occur in the potential subject's room in the Emergency Department.

Data Analysis & Statistical Considerations:

The data will be analyzed with a cohort study design, calculating odds ratios between ECG characteristics and outcomes. The two primary ECG characteristics investigated will be 1) presence of at least 2 consecutive leads with at least a partially concordant T-wave (biphasic or entirely concordant), and 2) the ST/S ratio (the ratio of the ST-segment deviation to the S or R wave amplitude, whichever greater). Secondary ECG features measured will include heart rate, rhythm, intervals, bundle branch blocks, PR and ST segment deviations and concordance/discordance, P/Q/R/S/T amplitudes, QRS durations, QT intervals, T wave inversions, segment contours and concavity/convexity. The primary outcomes of interest will be 1) positive cardiac troponin T biomarker levels on index visit, and 2) positive stress test results on index visit. Secondary outcomes will include the rates of 30-day myocardial infarction, 30-day death, and 30-day revascularization.

The investigators will present summary statistics on subject demographics, ECG findings, lab/imaging/procedure results, and outcomes. They will determine association between dichotomous outcomes of cardiac biomarker (cTNT) levels, stress test, MI, death, revascularization, and normally distributed continuous clinical and demographic characteristics using t-tests for normally distributed continuous data or Wilcoxon tests for non-normally distributed data, and with categorical characteristics using Fisher's exact or Chi-squared tests. Associations between continuous variables will be assessed using Spearman or Pearson correlations. Significance of the tests will be assessed at alpha = 0.05. Analysis will be conducted using SAS 9.3 (SAS Institute, Inc., Cary, NC).

Recruitment information / eligibility

• Status: Terminated
• Enrollment: 8
• Est. completion date: August 2015
• Est. primary completion date: August 2015
• Accepts healthy volunteers: No
• Gender: Both
• Age group: 18 Years and older

Eligibility

Inclusion Criteria:

• Age 18 and older

• LBBB on ECG

• Presenting to the ED and being evaluated for ACS

Exclusion Criteria:

• Prisoners

• Institutionalized individuals

• Patients who are incapable of giving legally effective consent

• Patients who cannot understand spoken English

Study Design

Observational Model: Cohort, Time Perspective: Prospective

Related Conditions & MeSH terms

• Acute Coronary Syndrome
• Bundle-Branch Block
• Heart Block
• Left Bundle Branch Block

Intervention

Other:
• None - no intervention, observational study
No intervention, observational study

Locations

Country	Name	City	State
n/a

Sponsors (1)

Lead Sponsor	Collaborator
Duke University

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Sensitivity of non-discordant T waves for prediction of positive cardiac biomarkers on index visit	Diagnostic properties (sensitivity, specificity, likelihood ratios) of item 1 for item 2 below: Presence of at least two ECG leads with non-discordant T waves (meaning either entirely concordant or biphasic, relative to the PR interval); Positive (above the assay's normal reference range) cardiac biomarkers (cardiac troponin I, T, or CK-MB) within 48 hours of presentation to the ED.	The ECG will be the ECG on presentation to the ED. The maximum cardiac troponin within 48 hours of index visit will be used as the outcome.	No
Primary	Specificity of non-discordant T waves for prediction of positive cardiac biomarkers on index visit	Diagnostic properties (sensitivity, specificity, likelihood ratios) of item 1 for item 2 below: Presence of at least two ECG leads with non-discordant T waves (meaning either entirely concordant or biphasic, relative to the PR interval); Positive (above the assay's normal reference range) cardiac biomarkers (cardiac troponin I, T, or CK-MB) within 48 hours of presentation to the ED.	The ECG will be the ECG on presentation to the ED. The maximum cardiac troponin within 48 hours of index visit will be used as the outcome.	No
Primary	Positive likelihood ratio of non-discordant T waves for prediction of positive cardiac biomarkers on index visit	Diagnostic properties (sensitivity, specificity, likelihood ratios) of item 1 for item 2 below: Presence of at least two ECG leads with non-discordant T waves (meaning either entirely concordant or biphasic, relative to the PR interval); Positive (above the assay's normal reference range) cardiac biomarkers (cardiac troponin I, T, or CK-MB) within 48 hours of presentation to the ED.	The ECG will be the ECG on presentation to the ED. The maximum cardiac troponin within 48 hours of index visit will be used as the outcome.	No
Primary	Negative likelihood ratio of non-discordant T waves for prediction of positive cardiac biomarkers on index visit	Diagnostic properties (sensitivity, specificity, likelihood ratios) of item 1 for item 2 below: Presence of at least two ECG leads with non-discordant T waves (meaning either entirely concordant or biphasic, relative to the PR interval); Positive (above the assay's normal reference range) cardiac biomarkers (cardiac troponin I, T, or CK-MB) within 48 hours of presentation to the ED.	The ECG will be the ECG on presentation to the ED. The maximum cardiac troponin within 48 hours of index visit will be used as the outcome.	No
Secondary	Sensitivity of various exploratory ECG findings for positive cardiac biomarkers	Logistic regression will be used to test multiple ECG measurements (other than the primary outcome, listed above) for diagnostic utility to predict positive cardiac biomarkers. Potentially relevant ECG findings will be explored via an ROC curve to find optimal cutpoints for diagnostic accuracy, and presented by means of diagnostic properties including sensitivity, specificity, and likelihood ratios. These findings will be explicitly listed as post-hoc, exploratory findings meant only to direct further research.	The ECG measurements at presentation to the ED will be used, to predict maximum cardiac troponin within 48 hours of presentation to the ED.	No
Secondary	Specificity of various exploratory ECG findings for positive cardiac biomarkers	Logistic regression will be used to test multiple ECG measurements (other than the primary outcome, listed above) for diagnostic utility to predict positive cardiac biomarkers. Potentially relevant ECG findings will be explored via an ROC curve to find optimal cutpoints for diagnostic accuracy, and presented by means of diagnostic properties including sensitivity, specificity, and likelihood ratios. These findings will be explicitly listed as post-hoc, exploratory findings meant only to direct further research.	The ECG measurements at presentation to the ED will be used, to predict maximum cardiac troponin within 48 hours of presentation to the ED.	No
Secondary	Positive likelihood ratio of various exploratory ECG findings for positive cardiac biomarkers	Logistic regression will be used to test multiple ECG measurements (other than the primary outcome, listed above) for diagnostic utility to predict positive cardiac biomarkers. Potentially relevant ECG findings will be explored via an ROC curve to find optimal cutpoints for diagnostic accuracy, and presented by means of diagnostic properties including sensitivity, specificity, and likelihood ratios. These findings will be explicitly listed as post-hoc, exploratory findings meant only to direct further research.	The ECG measurements at presentation to the ED will be used, to predict maximum cardiac troponin within 48 hours of presentation to the ED.	No
Secondary	Negative likelihood ratio of various exploratory ECG findings for positive cardiac biomarkers	Logistic regression will be used to test multiple ECG measurements (other than the primary outcome, listed above) for diagnostic utility to predict positive cardiac biomarkers. Potentially relevant ECG findings will be explored via an ROC curve to find optimal cutpoints for diagnostic accuracy, and presented by means of diagnostic properties including sensitivity, specificity, and likelihood ratios. These findings will be explicitly listed as post-hoc, exploratory findings meant only to direct further research.	The ECG measurements at presentation to the ED will be used, to predict maximum cardiac troponin within 48 hours of presentation to the ED.	No