Effects of Spironolactone and Renal Denervation Treatment on Blood Pressure and Its Variability—Different Aspects of Hypertension Treatment

Data from experimental, observational, and interventional studies have clearly shown that lowering blood pressure (BP) reduces cardiovascular (CV) risk. In addition, clinical and population-based cross-sectional and longitudinal studies have demonstrated that circadian BP profiles and BP variability are important hemodynamic markers, both being associated with the development, progression, and severity of organ damage, as well as risk of fatal and nonfatal CV events, independent of and beyond the degree of hypertension.1–3 However, it remains unknown whether an increase in BP variability contributes to or is a complication of hypertensive organ damage and whether or not it should be taken into account when treating hypertension. Accordingly, achieving BP control while reducing variability may serve to maximize CV benefit. Therefore, knowledge about the effects of an antihypertensive therapy on BP and its variability is of clinical importance. Resistant hypertension (RHTN) is a common clinical problem and defined as BP that remains above goal, in spite of use of 3 or more antihypertensive agents of different classes, including a diuretic.4 Patients achieving BP control with 4 or more antihypertensive agents are also considered to have RHTN.4 RHTN has been reported to have a prevalence from 10–15% of patients being treated for hypertension, depending on design of the analysis and the populations studied.5 Patients with confirmed uncontrolled RHTN are at higher risk for CV morbidity and mortality compared with those with controlled RHTN or general hypertension.4 Studies of patients with RHTN indicate that all-cause mortality and CV events are predicted by 24-hour, daytime, and/or nighttime BP levels.6 Data of BP variability in true resistant hypertensive patients compared with values from general hypertensive populations are lacking. In one of the few studies to report such data, the Pressioni Arteriose Monitorate e Loro Associazioni—PAMELA study, the investigators found that systolic BP variability was significantly greater in resistant hypertensive patients as compared to treated controlled or uncontrolled hypertensive patients.7 However, true outcome data guiding optimization of multidrug treatment regimens in patients with RHTN is limited. Experience from recent clinical trials support a standardized 3-drug combination that includes an angiotensin converting enzyme inhibitor or angiotensin II receptor blocker, a calcium channel blocker, a long-acting thiazide or thiazide-like diuretic and, if BP remains uncontrolled, a mineralocorticoid receptor antagonist such as spironolactone as the fourth agent.4,8 While studies have been done to identify the most efficacious combinations of medications for controlling RHTN, a much smaller number of studies have targeted restoration of normal circadian BP profile and reduction of BP variability, with the anticipation that such effects will translate into improved CV outcomes. However, evidence of the effects of the different antihypertensive drug classes on BP variability is limited and inconsistent, and additional data from randomized, controlled trials are needed. This is true especially for spironolactone for which very little has been reported regarding its effects on BP variability. In the Addition of Spironolactone in Patients with Resistant Arterial Hypertension (ASPIRANT) trial, reductions in systolic BP induced by spironolactone compared to placebo were 5mm Hg more pronounced at nighttime than at daytime, suggesting reductions in night-to-day BP ratios indicative of less BP variability, although the results of such an analysis were not reported as part of the original findings.9 The recent publication of the Pathway-2 results provided important confirmation of earlier but generally smaller and uncontrolled studies indicating that spironolactone is effective and safe when used as a fourth agent for treatment of uncontrolled RHTN.8 This large, multicenter, randomized, double-blind, placebo-controlled cross-over study demonstrated unequivocally that spironolactone is superior to a β-blocker or an α-adrenergic receptor antagonist and as an antihypertensive agent when added to a standardized 3-drug regimen of an angiotensin converting enzyme inhibitor or angiotensin II receptor blocker, a calcium channel blocker, and a diuretic. However, while the primary outcome of the Pathway-2 study was change in home systolic BP levels, no data were reported regarding the effects of spironolactone on BP variability. Similarly, the previously published smaller studies documenting the antihypertensive benefit of spironolactone on treatment of RHTN, likewise, did not report effects on BP variability. Patients with uncontrolled RTHN who are not tolerating or responding adequately to pharmacological therapy are being evaluated in numerous ongoing trials of endovascular, catheter-based procedures using radiofrequency or ultrasound ablation of sympathetic afferent and efferent renal nerves, which seemingly lower whole body sympathetic efferent activity, and thereby, reducing BP. Early clinical trials using catheter-based radiofrequency ablation of the renal sympathetic nerves or sympathetic renal denervation (RDN) reported 25–30mm Hg decreases in office BP, leading to high hopes for patients and providers that RDN may revolutionize the treatment of RHTN. However, the more recent Symplicity HTN-3 trial, which was the study to utilize a prospective, randomized, double-blinded study design was unable to demonstrate antihypertensive benefit at 6-month follow-up different from sham procedure.10 BP response to RDN was highly variable in actively treated individuals, with identification of a wide range of nonresponders, partial, and ideal responders. Overall, the proportion of ideal responders in the Symplicity-HTN 3 trial was disappointing. Methodological concerns have been raised about the RDN studies in general that may have confounded the observed effects, including procedural aspects; lack of standardization of BP measurements; changes in concomitant antihypertensive therapy; inclusion of heterogeneous study cohorts; inclusion of patients with known secondary causes of hypertension; and use of cohorts with high interindividual and intraindividual BP variation.11 The observed visit-to-visit BP variability in study participants enrolled in the RDN studies has been substantially larger than has been observed in contemporaneous drug trials.12 Separate studies suggest that this increase in BP variability may be related to a variety of factors, including the large number of background antihypertensive medications prescribed, either because of adverse effects, poor medication adherence; or underlying CV abnormalities, such as higher arterial stiffness or chronic kidney disease.11 The prevalence of such CV abnormalities are increased in patients with RHTN compared to patients with more easily controlled hypertension and are associated with greater BP variability and higher CV risk compared to the former.11,12 In the light of uncertainty regarding the true effectiveness of RDN in a setting of a more precise identification of the target population with confirmed RHTN, the DENervation in HyperTensioN (DENERHTN), an open-label randomized controlled trial, was conducted to assess the efficacy, safety, and cost-effectiveness of RDN when added to standardized, stepped-care antihypertensive treatment (SSAHT) compared with the SSAHT alone.13 SSAHT included a diuretic, angiotensin converting enzyme inhibitor or angiotensin II receptor blocker, calcium channel blocker, and after randomization spironolactone, bisoprolol, prazosin, and rilmenidine were sequentially added. At study end, the number and types of medications used were similar in the RDN plus SSAHT and SSAHT alone treatment arms. After 6 months, the investigators reported that RDN plus SSAHT decreased ambulatory daytime systolic BP by 15.8mm Hg which was 5.9mm Hg more than medical therapy alone.13 This observed BP difference of approximately 6mm Hg is clinically meaningful and, if maintained, should result in reduced CV morbidity and mortality.14 In this issue of the American Journal of Hypertension, de la Sierra et al. report the results of a small, but carefully conducted comparative study of 24 patients with RHTN who had participated in the DENERHTN trial.15 In this study, the effects of spironolactone 50mg daily (n = 13) vs. RDN therapy (n = 11) on ambulatory daytime and nighttime systolic and diastolic BP levels, day-to-night BP ratio, BP variability, and heart rate variability were compared between the 2 treatment modalities, using SD and coefficient of variation of 24-hour, daytime, and nighttime BP levels, as well as weighted SD and average real variability.15 The investigators report that spironolactone was more effective than RDN in reducing daytime systolic and daytime diastolic BP (25.6±17.0 vs. 3.4±12.8mm Hg and 10.3±8.4 vs. 1.8±8.3mm Hg, respectively) while the difference in nighttime systolic BP was statistically marginal (23.4±15.6 vs. 7.1±18.2mm Hg). When analyzed for effects on circadian BP profile and BP variability, spironolactone did not alter night-to-day BP ratios. In contrast, RDN reduced 24-hour, daytime, and nighttime diastolic BP variability. The results of this first head-to-head comparison of spironolactone and RDN for treatment of RHTN demonstrate that treatment with spironolactone reduced systolic and diastolic BP more than RDN. However, the greater BP reduction associated with spironolactone was not accompanied by differences in the circadian BP profile, as changes in the night-to-day ratio of BP were minimal and similar between treatment groups. Likewise, spironolactone did not improve BP variability in spite of the larger reduction in BP. In contrast, RDN reduced systolic and diastolic BP variability, although only the former achieved statistical significance, perhaps because of the small sample size. The finding of the larger BP lowering effect with spironolactone adds to recent reports with the Pathway-2 study being the most recent and most important one. It also suggests superiority of spironolactone over RDN regarding BP reduction. If confirmed in larger cohorts, these results provide guidance in suggesting that intensification of diuretic therapy, with a combination of a thiazide diuretic and spironolactone, is superior to RDN for treatment of RHTN. This suggests that RHTN is mostly an issue of excess volume as opposed to excess sympathetic output as recently discussed.16 However, it is possible that patients who fail to control BP with 5 or more antihypertensive agents including a thiazide diuretic and spironolactone may be more likely to have a neurogenic component and may favorably respond to RDN.17 Additionally, de la Sierra et al. demonstrate that treatment with spironolactone treatment did not benefit BP variability while RDN had a positive effect. Establishing the effects of spironolactone and RDN on the composite of CV hemodynamic parameters such as BP, BP pattern, and variability is of considerable interest and 2 aspects related to the current findings deserve discussion. Firstly, the divergence between the degree of BP reduction and improvement in variability with 2 different treatment modalities, would suggest, at least in part, differences in underlying mechanisms, and, if identified, might allow for better elucidation of mediators of BP control vs. stability. Secondly, there are some aspects of the role of an intermediate endpoint such as BP variability in treatment of RHTN that need to be considered. From a mechanistic standpoint factors likely influencing the magnitude of BP variability include vascular, humoral, and neural factors. Vascular factors include an increase in the stiffness of the conduit arteries secondary to aging, hypertension, and atherosclerosis; a decrease in the compliance of the resistant arteries by hypertensive remodeling; and an increase in the vascular tone of the resistance arteries by humoral factors; as well as impairment of baroreflex sensitivity and heightened sympathetic output.2,18 Vascular factors may be the largest mediator of BP variability in patients with hypertension, especially in elderly patients, patients with a long history of hypertension, and patients with advanced atherosclerosis. Experimental evidence is emerging that suggests that humoral factors, e.g., at the levels of the heart, are affected by increases in BP variability which promotes hypertensive cardiac remodeling by inducing inflammation via stimulation of the cardiac renin–angiotensin system and cardiac mineralocorticoid receptors.19 Furthermore, although organ damage induced by high BP variability seems to be similar to that induced by hypertension, experimental comparative studies in sinoaortic-denervated and spontaneously hypertensive rats revealed that, e.g., aortic hypertrophy is a sensitive index of BP variability while left ventricular hypertrophy is a sensitive index of high BP levels.19 If confirmed, class effects of different antihypertensive agents on the renin–angiotensin–aldosterone system may be relevant in terms of differentially targeting high BP levels vs. BP variability. Although speculative at this point, optimization of antihypertensive treatment regimens to both control BP and improve BP variability may be an important treatment strategy and translational and clinical studies are needed to provide further evidence. Reversing the effects of neural factors such as impairment of baroreflex sensitivity and increased sympathetic activity with therapies that modulate baroreflex function and/or reducing sympathetic outflow may accomplish improvements in BP control while reducing BP variability. In terms of the latter effect, post hoc analyses of RDN trials have often, but not always shown, reductions in systolic and diastolic BP variability.11 Attempting to identity factors associated with reductions in BP as well as improvement in BP variability, demonstrating that these effects are sustained with accompanying benefit on CV outcomes will be an important goal of future studies. In that regard, the study of de la Sierra et al. is the first to directly compare the differential effects of intensified diuretic therapy with use of spironolactone with presumed sympatholytic therapy, i.e., RDN. The findings suggest that intensified diuretic therapy better controls BP, while sympatholytic therapy better improves BP variability. Secondly, as recently outlined by the European Society of Hypertension Working Group on BP monitoring and CV variability, there are several unknowns regarding BP variability.20 Future research needs to clarify which BP variability index has the best performance as a risk factor; which component of BP variability (short-term, mid-term, long-term) is the best parameter for risk stratification; to what extent BP variability influences vascular damage and/or vascular damage exacerbates BP variability; which BP variability threshold best relates to increased CV risk and can be targeted with treatment; and whether treatment-induced reductions in BP variability improve outcome. Overall, it is clear that many questions regarding BP variability as a risk factor and as a potential goal of therapy remain unanswered. In that regard, more data from randomized controlled intervention trials like the current findings from de la Sierra et al. are needed elucidating the effects of different antihypertensive drug classes on BP variability. It will also be important to determine if reducing BP variability is associated with less target organ damage and if a reduced BP variability translates into improved CV outcomes. If the predictive value of BP variability and thresholds are established, then it would support recommendations to index BP variability with the same BP monitors/devices that are used to quantify BP levels. Studies that directly compare different treatment modalities, as done by de la Sierra et al. in the current study, provide important opportunities to address some of these deficiencies.