Hypertension: use ARBs over ACE-I

 A recent real-world effectiveness study compared ACE-inhibitors (ACE-I) and angiotensin receptor blockers (ARBs) in the treatment of hypertension, suggesting that ARBs should be the first choice given their better safety profile, and that we should not have an equipoise approach of either being equally appropriate (See htn ARB over ACE htn2021 in dropbox, or DOI: 10.1161/HYPERTENSIONAHA.120.16667). Thanks to Jon Pincus for bringing this to my attention 

 

Details: 

-- retrospective, new-user comparative cohort design study using statistical techniques to minimize confounding:

    -- a few different approaches to propensity scoring to match the groups, involving tens of thousands of measured baseline covariates, including demographics, diagnoses, drug exposures, drug groups, procedures, comorbidities, risk scores

    -- including 76 negative control outcomes (these are outcomes that should not be associated with any of these medications but should help reveal the extent of the placebo effect)

    -- this analysis was extensive and crossed different cultures and medical practices, including all patients initiating monotherapy with ACE-I versus ARBs between 1996 and 2018 across eight databases, from the United States (including commercial insurance claims, Medicare supplemental insurance claims, Medicaid enrollees, Optum Clinformatics databases), Germany (ambulatory care, all ages), and South Korea (National Health Insurance)

-- 2,297,881 patients initiated treatment with ACE-I and 673,938 with ARBs 

-- all patients had at least one year of prior observation in the database before treatment was started 

-- 80% of new ACE-I users received lisinopril (followed by ramapril and enalapril); 45% on ARBs received losartan (followed by valsartan and olmesartan) 

    -- each database, except for South Korea, had a predominant use of ACE-I over ARBs

-- primary outcome: acute myocardial infarction, heart failure, stroke, and a composite of cardiovascular events (these 3 outcomes plus sudden cardiac death) 

-- safety outcome: 51 secondary and safety outcomes including angioedema, cough, syncope, and electrolyte abnormalities 

 

Results: 

-- no statistically significant difference in the primary outcomes, comparing ACE-I versus ARBs: 

    -- myocardial infarction, HR 1.11 (0.95-1.32); 5960 events on ACE-I, 1699 on ARBs 

    -- heart failure, HR 1.03 (0.87-1.24); 8165 on ACE-I, 2468 on ARBs 

    -- stroke, HR 1.07 (0.91-1.27); 6775 on ACE-I, 1991 on ARBs 

    -- composite cardiovascular events, HR 1.06 (0.90-1.25); 18,213 on ACE-I, 5463 on ARBs

        -- essentially all databases showed at least a trend favoring ARBs 

-- safety profile, comparing ACE-I versus ARB: 

    -- angioedema: HR 3.31 (2.55-4.51), p <0.001 

    -- cough: HR 1.32 (1.11-1.59), p <0.001 

    -- pancreatitis: HR 1.32 (1.04-1.70), p=0.002 

    -- G.I. bleeding: HR 1.18 (1.01-1.41), p=0.04 

    -- but, decreased risk of weight gain: HR 0.84 (0.74-0.98), p=0.04 (i.e. favored ACE-I) 

    -- no significant difference in any of the other 49 outcomes, including deterioration of renal function and electrolyte abnormalities 

 

Commentary: 

-- this very large observational analysis with more than 3 million patients from eight databases worldwide found that major clinical cardiovascular outcomes were statistically similar for those initiating therapy with ACE-I versus ARBs, but there was a significantly higher adverse safety outcomes with ACE-I 

    -- of note, the data are less clear on equivalence of these two classes of medications on heart failure therapy (ACE-I may be preferred, from the limited studies)

-- other studies have found ACE-I and ARBs have similar effects on blood pressure control, but the lack of rigorous large-scale RCTs with direct comparisons limits this conclusion. A sensitivity analysis was conducted in the above study from the Optum EHR database (one of the 8 reviewed) finding no significant differences in blood pressure control 

-- at this point the American College of Cardiology/American Heart Association and the European Society of Cardiology/European Society of Hypertension guidelines give both ACE-I and ARBs their strongest recommendation (Class I) as first-line agents for starting antihypertensive therapy with level of evidence A (though other meds were also considered first-line). 

-- there have been only four head-to-head comparisons of ACE-I and ARBs finding conflicting results, likely because of different cohorts of patients (often including patients at high-risk by age or vascular disease) and these were small studies with less than 500 patients and fewer than 10 events. There have been no prior studies on treatment-naïve hypertensive patients not at very high risk

-- there was a network meta-analysis that did find that ACE-I but not ARBs reduce the odds of all-cause death (see htn metanal CKD better with ACE vs ARB  AmJKidDis2016 in dropbox, or doi.org/10.1053/j.ajkd.2015.10.011). this meta-analysis also found that ACE-I seemed to be superior for kidney failure and cardiovascular death (see http://gmodestmedblogs.blogspot.com/2021/09/bp-target-in-those-with-kidney-disease.html)

-- ACE-I are currently far more commonly prescribed for hypertension than ARBs; lisinopril is the most commonly used blood pressure medicine globally  

    -- by the way, there was a really interesting/disturbing editorial in the NY Times addressing generic drugs, including lisinopril, suggesting there is wide variability of the actual potency of the different generics for a given med, with company shenanigans regarding filing false data about med potency testing done: see https://www.nytimes.com/2021/09/18/opinion/drug-market-prescription-generic.html?referringSource=articleShare 

    -- and, there has been concern about a lot of other drugs containing potential carcinogens, including losartan and valsartan, 2 common ARBs

    -- unfortunately, pharmacies tend to vary which generic they stock by price, so saving a couple of pennies leads to changing which generics they purchase, and there may well be better potency consistency by using the same generic. for example, i had some patients who had been stable on their warfarin for years, with very consistent INRs, all of a sudden becoming very erratic in their INR levels, so i switched them to a specific warfarin generic (Jantoven), talked with the pharmacies, and these patients now only get Jantoven with more consistent results....

--the purported mechanism for ACE-I causing the adverse effects of cough and angioedema is that the ACE-I prevent degradation of bradykinin, which can be associated with these presumably bradykinin-induced events (though there have been reports of increased incidence of cough with losartan, and ARBS do not lead to bradykinin increases). Pancreatitis might be related to the accumulation of bradykinin causing localized edema of the pancreatic duct and pancreas, and this has been found in one case-controlled study. Unclear why there is an increase in GI bleeding 

--so, why do we tend to prescribe ACE-I over ARBs??

    -- a lot of the reason is likely inertia and "medical culture". i suspect all of us have seen patients develop cough on ACE-I, leading to extra clinician visits, perhaps chest xrays/radiation exposure, then transition/titration to ARBs and more clinic visits to assess/adjust dose. and, i suspect many of us have seen patients with really bad outcomes: anaphylaxis and angioedema (i have seen several). yet we continue with ACE-I first.... Perhaps just because we always did. Perhaps because that's what everyone else around us does, including our mentors and medical leaders.

    --there is a similar situation with prescribing hydrochlorothiazide as a solo agent for hypertension. several (but not all) studies have found this to be a significantly inferior antihypertensive in terms of clinical outcomes, including one study (MRFIT) dating back to the 1970s (chlorthalidone was a better agent). still it is very often the first agent prescribed in the US. see http://gmodestmedblogs.blogspot.com/2018/05/hctz-and-melanoma-risk-and-not-such.html which summarizes much of the more current data

    -- the US made the unwise choice of advancing rosiglitazone as the TZD of choice for diabetics (vs Europe, which chose pioglitazone, which also improves lipids). unfortunately, rosiglitazone causes lots of heart disease mortality and pioglitazone is cardioprotective. yet rosiglitazone continued to be prescribed even after the pretty clear studies showed that rosiglitazone was a bad actor (the FDA did step in, tepidly). again, likely related to some inertia (clinicians and patients were using it for a long time), in this case aided and abetted by fierce drug company pressure: see http://gmodestmedblogs.blogspot.com/2014/03/rosiglitazone-and-fda.html

 

Limitations: 

-- the lack of large-scale randomized controlled trials makes it difficult to draw firm conclusions, though the huge numbers of patients in the databases and the quite rigorous statistical techniques employed do buttress their conclusions 

    -- and, given that these medications are generic, it is hard to imagine there will be very large and undoubtedly expensive RCTs in the future to really answer the question

    -- but, despite their rigorous statistical analysis, there still is the potential for residual confounding (ie unsuspected/unmeasured patient characteristics which could explain the outcomes)

    -- and, there are advantages to real-world evidence, as in the above study, since this is not conducted in the context of a prescribed research study (with patients who are willing to participate, the study's inherent biases in how patients are recruited, the narrower demographics for the patients, involvement of study personnel to help assure adherence and answer questions, etc) and does not include the inclusion/exclusion criteria of the RCTs. i suspect that most of us actually practice in the context of the "real-world" 

-- the data also did not break down specific drugs within the class of ACE-I and ARBs, and is making the assumption that these results above are class effects [ie, not all ACE-I's are necessarily the same...] 

-- the study involved first-line treatment only and we cannot necessarily generalize these results to patients on combination therapies or switching from one medication to the other 

 

So, it seems to make sense for us to preferentially choose ARBs over ACE-I when initiating therapy in hypertensive patients, despite the equipoise of current guidelines. 

-- and, i would argue that the above conclusion also likely would apply to add-on therapy, preferentially choosing an ARB versus an ACE-I, since the adverse effects of ACE-I continue as second-line meds

-- though, 2 indications for preferring ACE-I might be in those with chronic kidney disease or heart failure, as noted above

by this article i do not mean that ARBs (or ACE-I) are the first line drugs for hypertension, just that for the case a clinician is deciding between ARBs and ACE-I as their med-of-choice.  there is a reasonable argument that amlodipine, for example, may be a somewhat better choice since it has a longer 24-hour blood pressure control (and, several studies have found higher stroke rates with ACE-I than amlodipine, perhaps from waning effect of the former in the wee hours of the AM) and is associated with less blood pressure variability (studies have shown that this variability seems to confer higher cardiovascular risk independent of blood pressure control): see http://gmodestmedblogs.blogspot.com/2018/05/hctz-and-melanoma-risk-and-not-such.html for more comments on this

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

For access to the dropbox, go to link: https://www.dropbox.com/sh/0bmvtita8mzms11/XDTwHySFFg

Then go to "clinic", then to "clinical stuff" for articles, or go to https://www.dropbox.com/sh/nyle22q1fn6lkpk/AAB9B2hBj5Kw4gtrJAkI-UF8a?dl=0 for the powerpoint presentations

 

or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.

 

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list