Can we use mineralocorticoid receptor blockade in diabetic patients with resistant hypertension? Yes we can! But it may be a double-edged sword

Primary resistant hypertension is defined as failure to reach target blood pressure (BP) in patients treated with an appropriate three-drug regimen, including a diuretic, with all administered at optimal doses [1,2]. Resistant hypertension prevalence ranges from 10 to 30% in surveys or tertiary centres and randomized trials [1,3–5]. It is frequently encountered in certain categories of patients, such as those with type 2 diabetes mellitus or chronic kidney disease in particular [1,6]. It is associated with a high prevalence of end-organ damage [7] and a high incidence of cardiovascular events [4]. The high frequency of resistant hypertension and its severity have led to the development of new treatment strategies based on the use of combinations of available drugs [1,2] or new device-based therapies, such as catheter-based endovascular renal denervation [8] and baroreflex activation therapy [9], the indications of which have yet to be clearly defined [10–12]. The frequent association of resistant hypertension with inappropriate volume retention [13,14] has led to international guidelines recommending principally a reduction in sodium intake and an increase in the intensity of diuretic therapy to treat this condition [1,2]. However, other pathophysiological systems, including the renin–angiotensin–aldosterone system (RAAS), the renal or central sympathetic nervous system and vasomotor systems (endothelin), may also influence BP [15–18]. An understanding of the relative contributions of each of these mechanisms to the BP level in individual patients with resistant hypertension would make it possible to tailor treatment to the individual, by selecting drugs or device-based therapies preferentially interfering with particular systems. However, it remains difficult to determine which patients are most likely to benefit from which treatment, the optimal drug doses to be used, the combinations and optimal order of drug initiation, and the optimal type and timing of device-based therapy. The problem is rendered even more complex by the presence of internal counter-regulation, which may limit the primary biological effect of each of these approaches, partially or fully neutralizing their BP-lowering effects. Finally, in addition to these internal pathophysiological factors, the care of patients with resistant hypertension is also rendered more difficult by problems of non-compliance with treatment, the prevalence of which is regularly reported to be as high as 40–50% in patients with resistant hypertension [19,20]. There is now a body of evidence to suggest that blocking the mineralocorticoid receptor, and hence, the multiple noxious effects of aldosterone in the kidney, vessels and heart [21], is one of the most effective ways of reducing BP in patients with resistant hypertension; indeed, this approach is now recommended by international guidelines [1,2]. There are multiple pathophysiological reasons for this efficacy. The primary mineralocorticoid hormone in humans, aldosterone, plays a major role in the homeostasis of sodium and potassium metabolism and BP control [22]. In addition, aldosterone acts in conjunction with high sodium intake to stimulate inflammatory reactions, cellular hypertrophy, matrix formation and apoptosis in the vessels, heart and kidney [22–24], and is directly involved in causing target organ damage in various cardiovascular and renal disease states [22,25]. A series of observations over the past 10 years have confirmed that relative aldosterone excess is common in patients with hypertension [26–28], particularly in those with resistant hypertension and evidence of intravascular volume expansion [13,14], and in overweight normotensive individuals [29]. It is also associated with insulin resistance [30]. Blockade of the biological effects of aldosterone has mostly been achieved with two mineralocorticoid receptor antagonists, spironolactone and eplerenone [21,31]. Both are used to treat hypertension [21], particularly the forms of hypertension in which aldosterone is thought to play a major role, such as primary aldosteronism [32,33], low-renin hypertension [34], metabolic syndrome [35] and complicated hypertension with left-ventricular hypertrophy [36]. They are also specifically used in patients with resistant hypertension. Indeed, in several studies of patients with resistant hypertension, spironolactone (25–100 mg/day) has been reported to yield additional decreases in BP when used in combination with other antihypertensive drugs [37–39]. However, most of these studies were uncontrolled, non-randomized, open-label, or included poorly characterized patients [40]. The only randomized, double-blind, controlled study published to date, in which a dose of 25 mg of spironolactone was used in patients with resistant hypertension, reported a significant decrease in systolic/diastolic daytime ambulatory BP of 5.4/1.0 mmHg with respect to placebo (P = 0.024/0.358, respectively) [41]. The percentage of patients with resistant hypertension achieving a target systolic office BP of less than 140 mmHg was 54.5% in the spironolactone group and 42.9% in the placebo group. Eplerenone is a short-acting mineralocorticoid receptor antagonist that is less potent than spironolactone [31,33,42]. However, it has the advantage of being more selective than spironolactone for the mineralocorticoid receptor. It does not interfere with progesterone or androgen receptors at the marketed doses (50–100 mg) and, therefore, does not have the sexual side-effects of spironolactone, such as impotence, gynaecomastia, breast tenderness and menstrual irregularities [31]. Eplerenone (50–100 mg daily) was given as an add-on therapy to patients with resistant hypertension in an uncontrolled, non-randomized, open-label trial, in which it was found to control BP in about 40% of patients [43]. However, eplerenone has not been approved for use in the treatment of hypertension in many European countries. In this issue of the Journal of Hypertension, Oxlund et al.[44] provide additional clinical evidence from the use of low-dose spironolactone for resistant hypertension treatment in another population of patients with type 2 diabetes, in which the balance between the benefits and the risk of this approach was needed to be evaluated in the light of the safety concerns reported in the Aliskiren Trial in Type 2 Diabetes Using Cardio-Renal Endpoints trial [45]. Indeed, only 20–40% of the patients with resistant hypertension included to date in clinical trials or observational studies assessing the efficacy of spironolactone had type 2 diabetes [37,38,46]. The findings of this 16-week multicentre, double-blind, randomized, placebo-controlled study can be summarized as follows: 119 overweight or obese type 2 diabetic patients (mean age 63 years) with resistant hypertension defined as mean daytime ambulatory SBP of at least 130 mmHg and DBP of at least 80 mmHg (mean 146/79 mmHg), or either of the above conditions, despite treatment with three or more antihypertensive drugs (range 3–5), including a diuretic and an angiotensin-converting enzyme inhibitor (ACEI) or an angiotensin receptor blocker (ARB), at appropriate doses, were included after a period of unmodified antihypertensive treatment of at least 1 month. A dose of 25 mg spironolactone or placebo was added to the previous treatment. The dose of spironolactone or placebo could be doubled, if a target BP of less than 130/80 mmHg was not achieved after 4 weeks, or halved, if BP was too low. After 16 weeks of treatment, 15% of the patients were taking a daily dose of 12.5 mg spironolactone, 41% were taking 25 mg and 44% were taking 50 mg (mean daily dose 34.9 ± 15.5 mg). The placebo-corrected decrease in mean daytime ambulatory BP was 8.9 (4.7; 13.2)/3.7 (1.5; 5.8) mmHg (P < 0.001) and that of night-time BP was 9.8 (4.0; 15.5)/3.2 (0.4; 6.7) mmHg (P < 0.001). The target daytime ambulatory BP (<130/80 mmHg) was achieved by 30% of the patients in the spironolactone group, versus only 9% in the placebo group (P = 0.019). The significant decrease in BP in the patients on spironolactone was associated with a significant decrease in urinary albumin excretion. Only about half the patients had microalbuminuria/macroalbuminuria at baseline. Small studies have shown that mineralocorticoid receptor blockade added to background ACE inhibition decreases proteinuria in patients with type 2 diabetes [47–49]. However, large clinical trials are required to demonstrate both the safety and the nephroprotective effects of such an approach in the long term [50]. The frequency of adverse events was low in this population of highly selected diabetic patients, from which patients with an estimated glomerular filtration rate (eGFR) below 50 ml/min per 1.73 m2 and previous intolerance to spironolactone were excluded. The patients were exposed to a low dose of spironolactone for only a short period of time (16 weeks). Consequently, no sexual adverse effects were reported. However, despite the dose-dependent nature of the sexual side-effects of spironolactone [51], longer-term exposure to even low doses of spironolactone has been associated with gynaecomastia/breast tenderness, leading to treatment discontinuation in patients with heart failure [52]. As expected, given the combination of a blockade of the renal tubular effects of aldosterone and the background blockade of the renin–angiotensin system (RAS) by ACEI or ARBs, plasma potassium concentration was 0.3 (±0.3) mmol/l higher in the treated patients than in the controls. Mild hyperkalaemia was the most common adverse event, leading to a decrease in spironolactone dose in three patients and treatment cessation in one case, consistent with the findings of other studies (see for review [40]). The risk of hyperkalaemia is of concern and should not be underestimated in patients with resistant hypertension and type 2 diabetes. Indeed, most of these patients are already on RAS inhibitor treatment and may have diabetes-induced hyporeninic hypoaldosteronism or associated conditions, such as diabetic nephropathy, with a low GFR, potentially increasing the risk of hyperkalaemia [53]. However, the risk of hyperkalaemia should not be stressed too strongly either, because it might decrease the rate of prescription of spironolactone in this setting. Careful monitoring of plasma potassium and creatinine concentrations in patients on spironolactone treatment, especially in those with low GFR values, should make it possible to detect hyperkalaemia early, if it occurs, and to adopt appropriate preventive measures [53], including the tapering of spironolactone dose and the avoidance of co-prescription with hyperkalaemic drugs, for example. Are there other alternatives to the administration of low-dose spironolactone in patients with resistant hypertension? New potent dihydropyridine-based third-generation and fourth-generation mineralocorticoid receptor antagonists [54] and aldosterone synthase inhibitors [55] are currently being tested as new tools for antagonizing the effects of aldosterone. However, both these approaches are likely to entail the same risk of adverse events as the first generation of steroidal mineralocorticoid receptor antagonists, including electrolyte disorders, hypotension, renal insufficiency and severe hypoaldosteronism, depending on residual aldosterone effects, initial renal function, dehydration, general anaesthesia, comorbidities (diabetes mellitus) and co-prescriptions (cyclooxygenase inhibitors, RAS blockers, heparin, trimethoprim-sulphamethoxazole, etc.). Another possible way to control BP in patients with resistant hypertension is sequential nephron blockade [46], based on the use of low doses of diuretics operating on different nephron segments to neutralize the effects of intra-renal counter-regulatory mechanisms triggered by the use of diuretics acting at a single site, particularly when used at high doses. In conclusion, this new study by Oxlund et al.[44] adds grist to the mill of those stressing the pathophysiological role of aldosterone in patients with resistant hypertension and an additional condition, type 2 diabetes mellitus. Indeed, diabetes and obesity are known to be independent predictors of the development of resistant hypertension, but the long-term benefit–risk balance of mineralocorticoid receptor blockade in patients with diabetes in ‘real life’ remains to be determined. ACKNOWLEDGEMENTS Conflicts of interest There is no conflict of interest.