Cardiac conduction system disorders on the electrocardiogram (ECG) are common. Left bundle branch block (LBBB) is associated with adverse outcomes, and variants previously considerednormal, suchas rightbundlebranchblock (RBBB),1 left anterior fascicularblock,2orPR intervalprolongation,3may also portend morbidity andmortality. At present, no paradigm to treat ECG conduction abnormalities inpatientswithout cardiomyopathyorheart blockexists. Such conduction abnormalities are associated with hypertension, left ventricular hypertrophy (LVH), and atrial fibrillation, anda linkwith inflammation and fibrosis has been suggested.1-3 Although recentlydescribedgeneticdeterminants for conduction abnormalities4 in some patients argue against treatment, the involvement of modifiable or preventable risk factors in others would make prevention of ECG conduction defects an attractive strategy. In this issue of JAMA Internal Medicine, Dewland et al5 present an important study in this area. The authors hypothesized that the antifibrotic and anti-inflammatory properties of angiotensin-converting enzyme inhibitors or statins may reduce incident conduction abnormalities. To test this hypothesis, they studied the effect of randomized administration of lisinopril vs other antihypertensives and pravastatin vs standard care on the development of ECG conduction abnormalities in high-risk patients with hypertension and moderate hypercholesterolemia in the Antihypertensive Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT).6 They assessed 21 004 individuals 55 years or older with hypertension and at least 1 other cardiac risk factor from 33 357 in the parent trial. They excluded 10 691 patients without serial ECGs or with baseline conduction disease or paced rhythm. They assessed 12-lead ECGs every 2 years for incident ECG conduction abnormalities. During the 5-year follow-up, LBBB was detected in 389 individuals; RBBB, in 570; and any conduction defect, in 1114. Lisinopril was associated with a statistically significant 19% reduction in incident ECG conduction abnormalities compared with chlorthalidone. The reduction in incident LBBB was of borderline statistical significance (adjusted hazard ratio, 0.77; 95% CI, 0.60-1.00; P = .05), and the reduction in incident RBBB was not statistically significant. No differences were found for amlodipine compared with chlorthalidone or for pravastatin compared with usual hyperlipidemia treatment. Left ventricular hypertrophy, male sex, age, smoking, coronary disease, diabetes, and body mass index were associated with incident ECG abnormalities. To our knowledge this is the first report that incident conduction system disease may be prevented by drug therapy, in this case lisinopril. This important observation is consistent with established knowledge that angiotensinconverting enzyme inhibition has salutary effects on inflammation, hypertrophy, and fibrosis. If confirmed in other populations, the ability to prevent conduction system disease could have substantial clinical and research implications. First, the major goal of therapy would likely be to prevent incident LBBB. More specific definition of individuals who may benefit, beyond associations with age and diabetes, may be needed to facilitate such preventive strategies. Second, the authors suggest studies to determine whether pharmacologic treatment can affect existing conduction abnormalities, including the need for pacemaker implantation. It would have been intriguing to study whether patients with existing conduction abnormalities in ALLHAT had a slower progression of disease depending on the randomization limb, if a sufficient number of such patients were present at baseline. Third, the association of lisinopril with reduced incident conduction disease5 was not attributable to blood pressure control. Indeed, lisinopril was significantly less effective than chlorthalidone in reducing blood pressure and preventing major cardiovascular outcomes in ALLHAT.6 One hypothesis that could reconcile this paradox is that lisinopril may cause greater regression in left ventricular mass over time than chlorthalidone or amlodipine, independently of blood pressure, particularly in white individuals.6 This finding suggests that interventions that prevent or treat LVH may prevent conduction system disease. Some limitations of this study should be recognized. Echocardiographic indices were reported in a minority of patients,6 and ECG indices of LVH are less reliable (particularly once conduction abnormalities develop). This limitation should be considered when interpreting the lack of interaction between baseline LVH and antihypertensive treatment randomization. Another limitation is that a large number of patients (n = 10691) were excluded owing to lack of ECGs. Remaining patients were more likely to be white men without diabetes or baseline LVH than the parent ALLHAT cohort. Although this difference does not alter the association of incident ECG conduction abnormalities with comorbidities, it may have introduced differences in baseline comorbidities in subgroups randomized to each antihypertensive. Finally, atrial fibrillation, which is linked with atrial conduction slowing,7 was not reported, and only 2 to 3 ECGs recorded during the 5 years precluded detection of transient conduction abnormalities (such as those caused by medications). However, these effects would likely be minor. In conclusion, Dewland et al5 should be congratulated for this important study from ALLHAT showing that treatment with the angiotensin-converting enzyme inhibitor Author Audio Interview at jamainternalmedicine.com
Read full abstract
Mar 27, 2017
Min-Jae Kang + 5
Open Access
