1

Screen failure theory

How is it that a randomized, placebo (sham)-controlled trial following strict inclusion/exclusion criteria with the majority of enrollment coming from hypertensive centers of excellence possibly enroll patients with pseudoresistant hypertension? The answer may be likened to finding a “needle in a haystack” in that the prevalence of patients with HTN who are truly resistant is quite small, probably around 10% , such that for purposes of enrolling patients, theoretically only 1 out of 10 screened would be included. Acknowledging the obvious selection bias for choosing which hypertensive patients are considered candidates for RDN, this ratio is likely reduced to some degree. But while secondary causes of HTN are rare compared with essential, the prevalence is drastically increased for resistant HTN. For example, obstructive sleep apnea (OSA) is a treatable form of secondary HTN with a three-fold greater prevalence amongst resistant patients .

In designing Symplicity HTN-3, the proposed population power analysis called for screening 1,060 patients to find 530 suitable for randomization (i.e., truly resistant). While a screen failure rate of 50% (1 out of every 2 screened) seems quite exclusive on the surface, the true prevalence was likely underestimated and thus even with adherence to screening guidelines, it is possible a statistically meaningful number of patients who were not truly resistant were entered the trial.

While the use of ambulatory blood pressure monitoring (ABPM) was implemented to further exclude pseudo-resistant patients, an important question to consider is whether the final blood pressure screening inclusion criteria (24-hour ABPM average ≥ 135/80) equate to an average office blood pressure > 160/90, and thus resistant HTN. As research suggests, an office reading of 140/90 most closely correlates with an ABPM of 131/74, it seems unlikely that the proposed ABPM inclusion criteria represent resistant HTN. Concerns over the magnitude of blood pressure reduction previously seen in renal denervation studies lacking a randomized, placebo-control design now seem well founded .

The prevalence of secondary HTN in this population and the information that can be obtained with ABPM were probably underappreciated and underutilized, respectively. ABPM provides more than just confirmation of resistant HTN, it also offers insight into medication compliance, pseudo-resistance, and a screening method for secondary HTN, such as OSA by assessing nocturnal blood pressure fluctuations. Future research will likely focus on a more sensitive blood pressure inclusion requirement using ABPM as the gold standard, with a threshold yet to be determined.

2

Threshold theory

While the evidence is compelling for the contribution of renal sympathetic nerve stimulation with subsequent plasma norepinephrine spillover as a primary role in the pathogenesis of resistant HTN, it appears only 65% of essential HTN is related to excessive sympathetic tone . The degree of sympathetic hyper-responsiveness is likely to correlate with the success of RDN, which undoubtedly varies among hypertensive etiologies. Furthermore, a graded response in the degree of blood pressure reduction seen after RDN is likely related to the sympathetic activity present at baseline . In other words, not all resistant HTN will be receptive to the mechanism of RDN, particularly in those patients where the pathophysiology is not entirely dependant on the excessive sympathetic overdrive theory.

Take for example the most common form of secondary HTN, obesity related with obstructive sleep apnea. In this population, there is minimal involvement of sympathetic outflow to the heart and cardiac norepinephrine is reduced despite renal sympathetic activation . Another limitation with the “one size fits all” concept is that in patients aged > 60 years, cardiac and renal norepinephrine spillover is often normal. In such cases, the basis of RDN therapy may not hold up. Furthermore, whether norepinephrine spillover is an appropriate surrogate for blood pressure response to RDN is debatable given the mixed results of previous studies, the heterogeneity of studied populations, populations (including the jump from translational to human) with variability in response, and the unproven knowledge as to what constitutes a significant reduction in sympathetic activity. What will help prove or disprove this theory will be to correlate blood pressure reduction with plasma norepinephrine in both patient groups in Symplicity HTN-3. Unfortunately, this important correlation may not occur as it was not described as being routinely measured in the methods or as a study end point.

To ensure a “clean” population of resistant HTN, patients with suspected obesity-induced HTN and OSA probably should not have been included in this trial. Multiple studies have shown a reduction in both daytime and nocturnal blood pressures after initiating continuous positive airway pressure, thus providing effective therapy for this resistant form of HTN . Because a high prevalence of OSA exists within the resistant hypertensive population, and compliance with therapy varies greatly and cannot be controlled for, if such etiologies are included, you will undoubtedly dilute your response to RDN. For instance, if control arm patients have their secondary HTN treated prior to the 6-month primary end point analysis, compared with the treatment arm, an improvement with RDN will either not be seen, or the response will appear blunted. On the other hand, if such patients are enrolled in the treatment arm, a further benefit from RDN is less likely.

Unlike percutaneous coronary intervention for obstructive coronary artery disease, HTN treatment is a moving target, with multiple etiologies and pathophysiological mechanisms often at play, and it would be quite naive for us to believe we found the “silver bullet” with RDN. Anticipating post-hoc analyses showing a benefit or at least a trend towards such in certain populations, we will again need to proceed with caution given the small sample size of the overall study. Unlike coronary studies with tens of thousands of patients, subgroup analysis in this situation will be fairly limited, statistically speaking.

Besides Medtronic’s Symplicity, four other catheter systems have received CE Mark approval for use in Europe, including St Jude Medical’s EnligHTN (EnligHTN-1 study), Boston Scientific’s Vessix V2 (REDUCE-HTN study), Covidien’s OneShot (RHAS study), and ReCor’s Medical Paradise (REALISE study). The haste at which each company with skin in the game pulled up stakes after Medtronic announced their negative results was impressive, as if skepticism had been somewhat present all along. Whether in fact skepticism, or possibly awareness and reconsideration of an appropriate study design with revised inclusion criteria, or simply a watch and wait strategy will be determined in the months to come. While the granular details of Symplicity HTN-3 are not yet available, we anticipate that we still have a lot to learn. The formal presentation and publication of Symplicity HTN-3 will most likely be hypothesis generating rather than the final nail in the coffin for renal denervation.