October 10th, 2013
How Well Do Standard ECG Criteria Detect RV Hypertrophy?
CardioExchange’s John Ryan interviews Isaac Whitman and Steven Kawut about their study of the sensitivity and specificity of electrocardiographic criteria in screening for MRI-detected right ventricular hypertrophy in patients without clinical cardiovascular disease. The study is published in JACC.
Researchers studied data from 4062 Multi-Ethnic Study of Atherosclerosis participants without clinical cardiovascular disease who had cardiac MRIs that were interpretable for right ventricular morphology. Electrocardiographic (ECG) criteria for right ventricular hypertrophy (RVH) came from ECG recommendations published by the AHA in 2009. Of the 3719 participants with normal LV function, 6% had RVH, most often mild.
Traditional ECG criteria had high specificity (often >95%) but low sensitivity (often <10%) for identifying MRI-detected RVH. Positive predictive values (similar to the underlying prevalence of RVH) and negative predictive values (equal to the prevalence of RVH-free subjects) were not high enough to be clinically useful. Findings were consistent across demographic subgroups and whether participants with abnormal LV mass or function were included. No combination of ECG variables, nor changing cutoff points, improved how well the ECG criteria predicted RVH.
Ryan: Most patients in this study had mild RVH. How clinically relevant is it to detect mild RVH?
Whitman and Kawut: RVH reflects an increase in afterload (e.g., from diseases affecting the left side of the heart, lungs, and breathing during sleep). Even mild RVH is an independent predictor of clinical heart failure and cardiovascular death. It is possible that identifying RVH at a subclinical state could provide an opportunity for earlier treatment and a reduction in the untoward sequelae of RVH. However, this approach is untested and, for now, unproven.
Ryan: As you know, most of our patients with pulmonary arterial hypertension have moderate-to-severe RVH. Are the ECG criteria valid for detecting moderate-to-severe RVH?
Whitman and Kawut: The ECG criteria for RVH were first developed from very small cadaveric dissection studies in patients with known advanced cyanotic congenital heart disease, cor pulmonale, or mitral stenosis; the applicability to modern-day adults without symptoms or a known cardiopulmonary diagnosis is unclear. We therefore do not recommend basing clinical decisions solely on the presence or absence of these findings in adults who do not have clinical cardiovascular disease. The availability and acceptability of transthoracic echocardiography make this the preferred approach to initial testing if pulmonary hypertension or RV abnormalities are suspected on the basis of clinical risk factors, other testing, or physical examination findings.
Ryan: Your study did not identify a superior method of characterizing RVH on ECG. What do you now tell your house staff and trainees when they say, “no RVH on ECG”?
Whitman and Kawut: Given our results, we believe that in adults without known clinical cardiovascular disease, the absence (or presence) of the AHA’s ECG criteria for RVH may not be helpful in ruling out (or suggesting) RVH. We do not know the implications of the presence or absence of these criteria, whether diagnostic or prognostic, in a patient with known or suspected clinical cardiopulmonary disease.
JOIN THE DISCUSSION
Will this study change your approach to patients without clinical cardiovascular disease for whom an ECG does not detect RVH?