|
|
|||||
|
|
The CARE-EF1 Study
|
| a multinational e-mail/web-based study | |
| Copyright ©2000-2002, CARE-EF1 | |
| Summary | Study Question | Background | Objectives | Methods |
|
The blood pressure response to the Valsalva maneuver can be easily observed at the bedside. This response is abnormal when the left ventricular ejection fraction (LVEF) is low. It would be useful to know when patients have significant cardiac dysfunction since therapeutic choices may change and since echocardiograms are sometimes unavailable.
This study of congestive heart failure patients will investigate the association of the LVEF obtained by echocardiography to the Phase IV blood pressure response to the Valsalva maneuver.
|
What is the accuracy of the Phase IV blood pressure response to the Valsalva manoever for the estimation of Left Ventricular Ejection Fraction (LVEF)?
|
The Valsalva maneuver is defined as expiration against a closed glottis that increases the intrathoracic pressure by at least 30-40 mm Hg and lasts for at least 10 seconds. The normal arterial blood pressure response to the Valsalva maneuver has been divided into four phases. During straining, there is an initial transient increase (Phase I) followed by a gradual decrease (Phase II) in the blood pressure. Once the Valsalva maneuver is released, there is a brief further drop in the blood pressure (Phase III) followed by a rapid rise to and above the baseline blood pressure (Phase IV). The fourth phase is also referred to as the "overshoot" phase.
The Valsalva maneuver has been used in clinical situations since 1704 when it was described by Valsalva as a technique for expelling pus from the middle ear by expiration against a closed nose and mouth. Later, in 1851, Weber noted an "imperceptible pulse" which occurs when a patient performs the Valsalva maneuver. Currently, the Valsalva maneuver is taught as an adjunct to listening to murmurs and as a therapeutic tool for converting arrhythmias.
Another use of the Valsalva maneuver has been described since 1944 where an abnormal response of the blood pressure occurred with congestive heart failure (CHF). These responses included an increase (or no change)of the phase II blood pressure, and an absence of the phase IV overshoot. A number of papers have advocated using these responses as a means of detecting heart failure at the bedside, however it is not a commonly taught or learned technique. Zema et al. proposed that the reason for the maneuver’s absence from most textbooks was a lack of studies demonstrating its sensitivity and specificity. Thus, he demonstrated in two studies a strong correlation between the overshoot blood pressure response and the LVEF. The authour demonstrated a sensitivity of 69% and a specificity of 90% which give a positive likelihood ratio (LR) of 6.9 and a negative LR of 0.34 Although referenced in more recent papers on the diagnosis of CHF, the blood pressure response to the Valsalva maneuver has still not become regularly implemented.
Diagnosing CHF and its severity has become increasingly important as therapeutic options have shown significant benefits on mortality and rehospitalization. However, the diagnosis is sometimes uncertain, and echocardiograms are often not readily accessible to all physicians. Assessing the severity of left ventricular dysfunction at the bedside would be useful for earlier diagnosis and for simpler means of monitoring.
|
This study will analyze the association between the LVEF and the Phase IV blood pressure response to the Valsalva maneuver and assess its validity as a diagnostic test.
|
This project will be a blinded cohort study and will compare a bedside diagnostic test with its gold standard, trans-thoracic echocardiography. Ejection fraction will be reported as a relative grade using ranges of <20%, 20-40%, 40-60%, and >60%.
Consecutive patients seen and examined by a general medicine service or referred to an echo lab with known or suspected left ventricular heart failure who meet eligibility criteria will be approached for consent to participate in the study.
A sample size of 96 patients will be required based on the original data recorded by Zema. This has been estimated based on prior data that an abnormal blood pressure response will be found in 5% and 28% of patients with normal and abnormal LVEF's respectively. This was calculated using "Epiinfo" software, a confidence interval of 95%, and power of 80%. It is expected that the prevalence of 41% found in the original study will be higher than that in the current study due to generous inclusion criteria, therefore a larger sample size will inevitably be required. A doubling to a sample of 200 patients should account for a change in prevalence and consequent changes in frequencies of positive and negative tests.
General medicine services or echo laboratory at your institution.
All patients admitted to the general medicine CTU or referred to an echo lab with known or suspected left ventricular heart failure based on history and clinical examination will be considered for entrance into the study. Suspicion of left ventricular failure is based on the initial examiners criteria for diagnosing LVEF, and is purposely left vague to include a wide variety of patients. This will also include patients with conditions previously demonstrated to have abnormal blood pressure responses to the Valsalva maneuver (such as aortic stenosis, constrictive pericarditis, coarctation of the aorta, or ASD).
Blinding will be reinforced by excluding patients if the examiner has knowledge of their current ejection fraction or the echocardiographer has knowledge of the result from the bedside test. Patients unable to follow the instructions for doing the diagnostic test will also be excluded. Patients with previous echocardiograms or cardiac catheter investigations will be included if the above conditions can be met.
After consent has been obtained, standard examination protocol sheets and data collection sheets will be provided to the examiner. These sheets are included in the Appendix.
Echocardiography will be performed in the same manner with each patient using methods recommended by cardiology.
The data collected will be analyzed using non-parametric statistical tests to ascertain discriminant function. This will normally be done by using the Chi-Square test. If marginal expected cell size is less than 5, then the Fisher exact test will be used.
In addition the sensitivity and specificity of the test will be calculated along with the relative positive and negative LR's for each cut point of overshoot response.
Proposed by Neil Gibson of Edmonton, Alberta
* Any comments or concerns about this protocol, please email carestudy@rogers.com.
|
|
|||||
