resistant hypertension: are diuretics harmful?

 i decided to comment extensively on a very recent review article that supports the quite provocative argument that we are treating hypertension incorrectly, especially treatment-resistant hypertension, and that we should avoid the early use of the typical diuretics and instead use spironolactone much earlier in the cascade of meds. this pretty brief article is well-referenced (59 references) to support each of their points and to justify their proposed paradigm-shift in hypertension treatment (see htn resistant diuretics harmful HTN2025 in dropbox, or . DOI: 10.1161/HYPERTENSIONAHA.125.24871):

-- treatment-resistant hypertension (TRH) has historically been defined as the failure to achieve the recommended blood pressure with at least 3 classes of antihypertensive meds, including a diuretic, at target or maximally tolerated doses

    -- in general, the diuretics used are thiazides or chlorthalidone unless there is significant renal failure, when loop diuretics are used

        -- furosemide lowers blood pressure less than thiazides

        -- low-dose torsemide, on the other hand, has similar blood pressure lowering effects to thiazides

            -- torsemide, by the way, seems to be also a better med than furosemide for heart failure: https://gmodestmedblogs.blogspot.com/2017/11/torsemide-seems-to-top-furosemide-for.html

                -- the mechanism of torsemide action has some important differences from furosemide: torsemide inhibits the renin-angiotensin-aldosterone system (and has heart failure benefits even in the setting of high levels of concomitant ACE inhibitor use) along with the associated decrease of sympathetic activation, myocardial fibrosis, and improved ventricular remodeling; torsemide increases diuresis in patients with cirrhosis; and it increases bioavailability with its longer half-life allowing once-a-day dosing: https://gmodestmedblogs.blogspot.com/2023/01/heart-failure-torsemide-vs-furosemide.html

    -- though there has been hesitancy in using thiazides in patients with renal failure, the recent CLICK trial of patients with poorly controlled hypertension assessed by ambulatory blood pressure monitoring (mean 143/75) and stage 4 CKD (mean eGFR 23) on 3.4 antihypertension meds were randomized to chlorthalidone 12.5mg (increasing to max dose of 50mg if needed), leading to an 11 mmHg reduction in the chlorthalidone group and 0.5 mmHg with placebo: https://www.nejm.org/doi/full/10.1056/NEJMoa2110730

    -- the urinary albumin-to-creatinine ratio also improved by 50% with chlorthalidone, though there was more hypokalemia, reversible increased creatinine, hyperglycemia, dizziness, and hyperuricemia

        -- this study led to increased general use of chlorthalidone in those with severe kidney disease

-- many studies have documented quite dramatic blood pressure improvement with the use of spironolactone in those with TRH and on 3 or more classes of meds (and spironolactone has worked 100% of the time in my patients)

        -- there are adverse effects with spironolactone, including hyperkalemia, decreased eGFR (though this is usually transient; see below), or adverse effects on sexual function. about ¼ of people do stop the spironolactone

-- an interesting meta-analysis assessed the cardiovascular benefit of blood pressure reduction in primary and secondary ASCVD prevention, finding similar relative risk reductions by lowering the blood pressure to similar levels, but there was a pretty dramatic residual absolute risk differences in cardiovascular events in those who already had an event (secondary prevention), ie for hypertension treatment "the earlier, the better" prevails over "the lower, the better"

    -- also, independent of the achieved blood pressure, those with TRH have a 2.5-fold higher risk of cardiovascular mortality than those with essential hypertension,  (https://bmcnephrol.biomedcentral.com/articles/10.1186/s12882-019-1315-0)

     -- so, all of this suggests that we should be treating hypertension aggressively from the beginning, prior to a cardiovascular event (there is a higher risk of heart failure, cardiovascular mortality, or progression of CKD to end-stage renal disease despite controlled blood pressure. and mineralocorticoid receptor antagonists (MRAs) should be instituted earlier to maximize the antihypertensive effect

this current article makes the further points:

-- the incidence of primary hyperaldosteronism (PA, as defined by cutpoints above) is common: in 15% of those with essential hypertension and in 20-30% with treatment-resistant hypertension (TRH).

    -- of particular note, 60-70% of those with TRH have "dysregulated serum aldosterone concentrations" (SAC), consisting of low renin levels but SAC levels that do not increase above the cutpoint for defining their condition as PA (see next paragraph); ie, the SAC is not elevated enough by volume expansion (the natriuretic peptides are elevated reflecting the fluid volume expansion) associated with an increased risk for cardiovascular and renal complications [i assume this is the reason that spironolactone works so well in people with TRH, since it is both a diuretic and an MRA]

        -- obesity-related hypertension is associated with increased aldosterone levels (these track with the BMI level), and this aldosterone production is apparently from adipocytes and associated with the obesity-related cardiovascular complications through the mineralocorticoid receptors: https://pubmed.ncbi.nlm.nih.gov/27357931/

           -- and, even using renin-angiotensin system inhibitors, as is commonly done in those with TRH, CKD or diabetes, there is only a transient (months) effect in lowering SAC levels (these levels return to the normal range over time, which does not happen with spironolactone)

           -- and, both thiazides and furosemide (but not torsemide, and low-dose torsemide does have equivalent blood pressure lowering to thiazide-like meds, unlike furosemide) increase SAC levels and stimulate the sympathetic nervous system activity; chlorthalidone-associated insulin resistance as well as the increased sympathetic activity can be blocked by spironolactone (a high SAC-to-renin ratio or a low renin by itself predicts a good BP response to spironolactone)

        -- bottom line here: the hyperaldosteronism associated with TRH, especially in those with CKD, obesity, or insulin resistance is made worse by thiazide and furosemide!!!!! and these diuretics may well be associated with bad outcomes including poor blood pressure control, CKD, heart failure, and metabolic syndrome/insulin resistance

            -- which reinforces the conclusion that thiazides and furosemide should not be preferentially used in those with TRH in particular (these meds are associated with worse clinical outcomes) and mineralocorticoid antagonists (eg spironolactone) should be used much earlier in the cascade of antihypertensive options)

            -- as an additional point supporting this: the hypokalemia associated with the regular diuretics is harmful, with a serum potassium <4.0 mmol/L being associated with increased cardiovascular risk and greater progression of CKD, including the risk of atrial fibrillation, ventricular arrhythmias and sudden cardiac death. and an observational study found a 73% increase in strokes in patients on diuretics, with a 29% decreased risk in those on spironolactone. similar associations  have been found with cognitive decline

As a reference here for the diagnosis of primary aldosteronism (though, obviously, these are specific cutpoints in a continuous biological spectrum, and differences between being minimally below that cutpoint may not really matter biologically):

-- plasma renin activity (PRA) <= 1 ng/ml/h, or direct renin concentration (DRC) <= 8.2 mU/L

-- aldosterone (immunoassay) >= 10 ng/dL (>277 pmol/L), or (LC-MS/MS) >= 7.5 ng/dL >=208 pmol/L)

 AND

-- aldosterone to renin ratio (ARR):

    -- aldosterone (immunoassay, ng/dL)/PRA (ng/mL/h) >20 ,or aldosterone (immunoassay, pmol/L)/DRC (mU/mL) >70

    -- aldosterone (LC-MS/MS, ng/dL)/PRA  (ng/mL/h) >15, or Aldosterone (LC-MS/MS, pmol/L)/DRC (mU/mL) >52

-- these authors do comment that there are other mineralocorticoid blocking meds that may have a role, though spironolactone is cheap and more available

    -- eplerenone has fewer sexual-related adverse effects but needs twice-a-day intake because of short half-life

    -- finerenone slows progression of CKD and decreases cardiovascular outcomes in patients with heart failure, but in some studies it has less direct effect on blood pressure than spironolactone

-- the increased potassium from spironolactone, if too high, can be controlled with SGLT-2 inhibitors, and these also reduce proteinuria and add to blood pressure reduction; other meds (eg patiromer) can further reduce potassium if needed, allowing the continued use of spironolactone

-- the worsening of eGFR associated with spironolactone Is typically transient and in fact is renoprotective, similar to the effect of SLGT-2s and ACE inhibitors; a 5.9 year study found that MRAs are associated with a 28% lower risk of renal replacement therapy initiation in a dose-dependent fashion: https://www.ahajournals.org/doi/10.1161/HYPERTENSIONAHA.121.18360

-- so, their conclusion is that spironolactone should be the initial diuretic in patients being put on diuretics for hypertension, and not the 4th-line one….

    -- as an aside, HCTZ is just a bad choice as a sole agent for blood pressure control, since it has a very short clinical half-life; the measured blood pressure decrease during daytime office visits typically seems good, belying the fact that for half the day control is poor : https://gmodestmedblogs.blogspot.com/2013/02/hypertension-abpm-not-use-hctz.html, and see the chart below from Messerli et al.

 

my synthesis of a reasonable approach to hypertension treatment:

-- it is important for hypertension control (and multiple other medical benefits) to continually engage patients in the importance of the nonpharmacologic benefits of diet, exercise, and achieving a normal BMI

-- it should be remembered that ambulatory blood pressure monitoring (ABPM) is the gold standard measurement of hypertension in terms of actual clinical outcomes associated with hypertension, and not the in-clinic blood pressure: https://gmodestmedblogs.blogspot.com/2015/10/uspstf-guidelines-on-blood-pressure.html

-- home-based blood pressure (HBBP) monitoring is also a very useful approach, which empowers patients to take control of their hypertensive condition:  https://gmodestmedblogs.blogspot.com/2024/01/home-based-blood-pressure-variability.html

    -- there are only a few rigorous studies confirming the benefit of HBBP, and the studies done often have used different methodology for the measurements (though the self-measured BP does correlate well with ABPM in several studies, and not with the clinic blood pressure measurements; HBBP decreases finding both the white-coat and masked hypertension readings): https://jamanetwork.com/journals/jama/fullarticle/2782296

    -- and, it makes sense to me that we should ask patients doing home-based monitoring to sit in a quiet room for a couple minutes, take deep breaths in and out, think of pleasant thoughts, and then measure their blood pressure (and highlighting the 2nd and third value taken a bit later over the first one). This preferentially should be at different times during the day and night

-- and, as a general hypertension treatment approach, studies have found that for those with inadequate med response, it is far more effective to add another med vs doubling the dose of the initial med: https://gmodestmedblogs.blogspot.com/2018/03/adding-yet-another-bp-med-helps-alot.html and

so, i think this article is a game-changer in hypertension treatment:

-- it makes sense to check all hypertensive patients for primary hyperaldosteronism, PA, (as per the current clinical recommendations by the 2025 Endocrine Society Clinical Practice Guideline, more on this in the next blog), given PA's high prevalence and the documented finding that these patients do better with surgery if appropriate than with medical therapy (more on this in the next blog).

    -- those who qualify as having primary aldosteronism might benefit from further workup to see if they are potentially surgical candidates (assuming surgery would be something the patient was interested in)

-- those who do not make criteria for primary aldosteronism by having decreased renin levels and increased aldosterone levels but not to the level of PA, referred to as dysregulated aldosteronism in the current study (much more common than PA), should be treated primarily with spironolactone as the diuretic, perhaps initially or in addition to other blood pressure meds (i could not find any information in the medical literature supporting this claim, with only one study finding blood pressure benefit for spironolaction as initial therapy in people with stage 1  hypertension: https://pmc.ncbi.nlm.nih.gov/articles/PMC5777944/

    -- studies have found that just a decreased renin level is associated with good response to spironolactone

-- and we should be wary of using the traditional diuretics in treating hypertension

geoff

-----------------------------------

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

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them