By Dr. Geoffrey Modest
A secondary analysis of the ALLHAT hypertension study (see doi:10.1001/jamainternmed.2016.2502 ) found that patients in the lisinopril arm had decreased development of cardiac conduction system disease, which the authors suggested was due to lisinopril’s anti-inflammatory and antifibrotic properties.
Details:
- ALLHAT was a community-focused hypertension study in 623 North American sites with 3 medication arms: lisinopril, amlodipine, and chlorthalidone. There was also a lipid component, and patients with fasting LDL 120-189 mg/dl (or 100-129 if known atherosclerosis) were randomized to pravastatin vs placebo. Follow-up 5 years
- 21,004 people: 56% men, mean age 66.5, 92% white/6% black, baseline BP=174/98, BMI 28, Framingham risk score 22%, 16% smokers
Results:
- 1114 developed conduction system disease: LBBB in 389, RBBB in 570, intraventricular conduction delay in 155
- lisinopril vs chlorthalidone: 19% reduction in developing conduction system disease [HR 0.81 (0.69-0.95, p=0.01] with lisinopril
- Amlodipine was non-significantly different from lisinopril
Commentary:
- It was interesting that in this study no one developed lesser conduction system abnormalities such as 1st degree AV block, LAFB, or incomplete RBBB, which suggests that there might be a different pathophysiology than simply a progression of conduction system dysfunction (e.g. progressive fibrosis of Lev). The data on progression of 1stdegree AV block, for example, are mixed, with some long term studies finding this to be a benign condition, but some studies (e.g. Framingham Study) did find that an increase in atrial fibrillation (HR 2.1), likelihood of progression to require a pacemaker (HR 2.9) and higher all-cause mortality (HR 1.4): see Cheng S. JAMA 2009; 301(24): 2571.
- LBBB and RBBB are clearly associated with increased cardiac mortality, as well as the potential progression to complete heart block
- Lisinopril was better than chlorthalidone in this secondary analysis, controlling for an array of clinical variables, including demographics, BMI, smoking, aspirin use, diabetes, CAD, LVH, lipids. And this wasdespite the fact that the achieved blood pressure reduction was inferior with lisinopril.
- Concerns about the study:
- This is a secondary analysis, and there was some selection bias (those who received serial EKGs were more likely to be men and white, not have diabetes, and to be on aspirin)
- >40% of all participants were on at least one additional step 2 or step 3 drug, which brings up 2 issues:
- The choice of the second and third agents was prescribed (either atenolol, clonidine or reserpine as step 2, hydralazine as step 3) and these were not standard community practice even at that time, at least in the Boston area (e.g.: adding atenolol to lisinopril was not done much then, since that combo was felt to be less synergistic, both being renin-active agents. The other agents were not used much at all)
- There were no granular data presented separating out patients just on lisinopril as a single agent vs those on lisinopril in combination. This is problematic if lisinopril plus one of the second line meds did have some synergistic effect in combination for preventing conduction system abnormalities. So, for example, maybe the lisinopril did nothing alone, but did so only in combination with another med. in that case the conclusion that lisinopril is good for protecting the conduction system would be erroneous.
- If lisinopril were protective, why would that be the case?
- I think positing the anti-inflammatory effect is a bit of a stretch, since those on pravastatin (a pretty potent anti-inflammatory) had no protection and in fact a trend to worse conduction system disease outcomes
- Clearly the issue is not lowering the blood pressure per se, since the lisinopril group had less BP improvement
- My guess is that the issue was decreasing LVH, since LVH was the strongest single predictor of incident conduction system disease in multivariate analysis. (The association with lisinopril above did control for LVH, but only LVH at baseline. And only by the relatively poorly sensitive EKG). And, ACE inhibitors/ARBs are the best agents for reversing LVH. Also, LVH itself does have significant mortality associated with it, which is decreased when an antihypertensive agent decreases EKG-determined LVH (see the LIFE trial, Dahlof B. Lancet 2002; 359: 995, which showed that the ARB losartan was better than atenolol in decreasing cardiac events, but that with LVH regression by either drug, there were fewer cardiac events – i.e., it was the LVH regression that mattered, not the drug. unfortunately, they did not report on conduction system dysfunction). The data on reversal of LVH is almost as good for calcium blockers (e.g. amlodipine) as with ACE inhibitors (e.g. lisinopril). Diuretics (like chlorthalidone) are significantly less likely to reverse LVH.
- So, my conclusions:
- This study reinforces my use of dihydropyridine calcium channel blockers as my first drug for hypertensive patients. As mentioned in several prior blogs, I am concerned with using hydrochlorothiazide as the initial agent (the most commonly used one), since its durability over 24 hours is quite limited (see blog below), and this ALLHAT secondary analysis reinforces not using a diuretic. Amlodipine has a much longer duration of action and has much less blood pressure variability (perhaps an additional clinical benefit). This was pretty much the conclusion of NICE, in their 2011 analysis, suggesting that all Afro-Caribbean hypertensive individuals and all white people >55yo have a calcium channel blocker as the first agent. They do suggest an ACE inhibitor if white and <55yo (mostly because of the higher likelihood of high renin hypertension in this group), though they also supported using chlorthalidone (but not hydrochlorothiazide)
- And, as in the LIFE study, the current study does support preferentially using an ACE inhibitor or ARB for those with LVH by EKG (and, I would extend this to echo LVH). Though, again, amlodipine (or other dihydropyridine calcium channel blockers) are also reasonable. (i.e., the 2011 NICE guidelines are basically upheld by this study….)
See https://stg-blogs.bmj.com/bmjebmspotlight/2016/05/04/primary-care-corner-with-geoffrey-modest-md-chlorthalidone-is-better-than-hctz-for-hypertension/ which showed that HCTZ has poor 24-hr duration, with chlorthalidone being much better
(Visited 6 times, 1 visits today)