Was it optimal to drop a diuretic as a first-line choice of drug treatment in the 2020 International Society of Hypertension Guidelines?

In this issue of Blood Pressure, the lead authors of the 2020 International Society of Hypertension (ISH) Guidelines comment [1] on our recent editorial [2] regarding their first choice of antihypertensive therapy [3]. Our editorial questioned the recommendation by ISH to leave out diuretics from steps 1 and 2 in initiating antihypertensive therapy [2,3]. Our editorial is supported by the 2017 American [4] and the 2018 European Hypertension Guidelines [5], which both recommended diuretic treatment as a choice for both the first and the second steps in the medical treatment of hypertension. In contrast, the 2020 ISH Guidelines recommend angiotensin converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARB) plus calcium channel blockers (CCB) only in both steps of treatment initiation [1,3].

Modern hypertension guidelines are evidence-based in that they recommend medical treatments that have been proven effective in lowering blood pressure and preventing cardiovascular disease outcomes and death in randomised controlled clinical trials and/or in well-designed observational studies with large patient populations. The decision by ISH [3] to move thiazide or thiazide-like diuretics such as chlorthalidone or indapamide to step 3 in the treatment scheme was based in part on results of the avoiding cardiovascular events through combination therapy in patients living with systolic hypertension (ACCOMPLISH) trial, in which the ACE inhibitor benazepril in combination with the CCB amlodipine prevented cardiovascular endpoints more effectively than the ACE inhibitor in combination with the diuretic hydrochlorothiazide [6]. The CCB plus the ACE inhibitor, if needed, was also more effective than the beta-adrenergic blocker plus the diuretic, if needed, in the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT) [7].

We participated in both ACCOMPLISH and ASCOT, and recognise them as two of the very few outcome trials with sufficient statistical power to show significant differences in the ability of different classes of antihypertensive drugs to prevent cardiovascular disease events [6,7]. However, we find it peculiar that the 2020 ISH change in policy, which was directed at a worldwide audience, was based on outcomes of these trials, which were performed in the Nordic countries (approximately 13,500 patients), United Kingdom (approximately 9000 patients) and the United States (approximately 8000 patients). The data that provided the evidence base for the most important ISH recommendation, choice of first and second line treatments for hypertensive patients throughout the world, were derived from trials that enrolled mostly Caucasian patients, few blacks and almost no Hispanics, Asians or others.

In ASCOT, the diuretic, which was given as the second drug, if needed, in the atenolol + thiazide arm, was severely under-dosed early in the trial, resulting in less effective blood pressure lowering in the beta-adrenergic blocker + diuretic arm compared to the CCB + ACE-inhibitor arm. As a consequence, the study secretary, first author of the current ISH letter, contacted the Nordic leaders and requested higher dosing of bendroflumethiazide in an attempt to minimise the difference in blood pressure between the two arms. This attempt was only partly successful as lower blood pressure in the CCB + ACE-inhibitor arm sustained throughout the length of the ASCOT study [7]. Study leadership did not conclude that the diuretic component of the treatment was inferior [7], but rather that the beta-blocker atenolol was inferior to the CCB amlodipine, leaving ACCOMPLISH [6] as the only trial ever to show a diuretic to be inferior to a CCB when administered in combination with an ACE inhibitor. ACCOMPLISH included a rigid protocol with forced up-titration of blinded study drugs to achieve blood pressure targets [6]. It reached its blood pressure targets in more than 80% of study participants, but required a protocol that differed greatly from how treatment of hypertension is carried out by most health care providers around the world. The ACCOMPLISH findings [6] need to be confirmed in other similarly designed and properly powered outcome trial(s) before concluding that a diuretic is inferior to other drug classes as first or second line treatment. Besides, black patients in sub-Saharan Africa had similar blood pressure responses to amlodipine plus hydrochlorothiazide as they had to amlodipine plus perindopril [8].

As we pointed out in our editorial [2], numerous outcome trials in hypertension have shown the benefit of thiazide or thiazide-type diuretics in preventing cardiovascular disease outcomes, including stroke, heart failure, myocardial infarction, left ventricular hypertrophy and aortic aneurysm, compared to placebo controls. Importantly, several common hypertension-related comorbidities, including aging, obesity, diabetes and renal function impairment, are associated with salt sensitivity, which favours diuretic treatment. However, insufficient diuretic treatment is one of the most frequent reasons for failure to achieve blood pressure targets, and there is increasing evidence that all diuretics are not equal in terms of efficacy and tolerability [9]. The increasing use of lower doses of diuretics in combination with ACE inhibitors or ARBs further supports diuretic use, as adverse effects of diuretics are reduced by both dose reduction and by the effects of the accompanying ACE inhibitor or ARB, which counteract the potential adverse effects of diuretics on circulating potassium, calcium, glucose, lipids and uric acid levels.

There are numerous arguments for choosing various first and second-line antihypertensive drugs. In Table 1, we have elaborated on various possibilities regarding CCB vs. diuretic as an add-on treatment to ACE inhibitor or ARB. While ACE inhibitor or ARB + CCB is an excellent combination, as shown in ACCOMPLISH, in light of the strong track record of diuretics in controlling blood pressure and preventing cardiovascular disease, and all the clinical conditions in Table 1 which may favour the diuretic choice, we argue for maintaining diuretics as a choice in first and second line treatments of hypertension.

Table 1. Potential choices of add-on to angiotensin-converting enzyme inhibitor or angiotensin receptor blocker treatment in first-line combination treatment for most hypertensive patients: Dihydropyridine calcium channel blocker (CCB) or thiazide-type of diuretic for initial combination therapy? Clinical conditions and preferences discussed in 2017 American [4] and 2018 European Hypertension Guidelines [5].

Favours dihydropyridine CCB	Favours thiazide or thiazide-like diuretic such as chlorthalidone or indapamide
Metabolic syndrome	Fluid overload
Gout	Ankle edema (but not caused by CCB)
Diuretic intolerance	CCB intolerance
Patient preference	Patient preference
Left ventricular hypertrophy	High salt intake
Angina pectoris*	Unhealthy diet (high ‘junk food’ intake)
Type 2 diabetes (ACCOMPLISH data) [10]	Type 2 diabetes with salt sensitivity
Stable coronary disease	Overweight with salt sensitivity
Renal failure (ACCOMPLISH data) [11]	Renal failure with sodium retention
Chronic obstructive pulmonary disease	Hyperkalaemia
Atherosclerosis in lower limbs with	Symptomatic treatment in imminent or
claudicatio intermittens*	suspected heart failure

*Anti-ischemic properties.

In conclusion, the 2020 ISH hypertension guidelines [3] diverge from the 2017 American [4] and 2018 European hypertension guidelines [5] regarding choice of initial drug treatment. We are concerned that this could be a step in the wrong direction because a thiazide or thiazide-type diuretic, e.g. chlorthalidone or indapamide, is at the top of the list of evidence-based first-line antihypertensive drugs because of unsurpassed efficacy in lowering blood pressure and preventing cardiovascular disease when given in combination with an ACE inhibitor, ARB and/or a CCB.

Disclosure statement

SEK, KN, MB and SO are editors of blood pressure and report no relevant conflicts of interest to disclose related to this editorial.