Background: Recent sham-controlled randomized clinical trials have confirmed the safety and efficacy of catheter-based renal denervation (RDN). Long-term safety and efficacy data beyond 3 years are scarce. Here, we report on outcomes after RDN in a cohort of patients with resistant hypertension with an average of ≈9-year follow-up (FU). Methods: We recruited patients with resistant hypertension who were previously enrolled in various RDN trials applying radiofrequency energy for blood pressure (BP) lowering. All participants had baseline assessments before RDN and repeat assessment at long-term FU including medical history, automated office and ambulatory BP measurement, and routine blood and urine tests. We analyzed changes between baseline and long-term FU. Results: A total of 66 participants (mean±SD, 70.0±10.3 years; 76.3% men) completed long-term FU investigations with a mean of 8.8±1.2 years post-procedure. Compared with baseline, ambulatory systolic BP was reduced by −12.1±21.6 (from 145.2 to 133.1) mm Hg ( P <0.0001) and diastolic BP by −8.8±12.8 (from 81.2 to 72.7) mm Hg ( P <0.0001). Mean heart rate remained unchanged. At long-term FU, participants were on one less antihypertensive medication compared with baseline ( P =0.0052). Renal function assessed by estimated glomerular filtration rate fell within the expected age-associated rate of decline from 71.1 to 61.2 mL/min per 1.73 m 2 . Time above target was reduced significantly from 75.0±25.9% at baseline to 47.3±30.3% at long-term FU ( P <0.0001). Conclusions: RDN results in a significant and robust reduction in both office and ambulatory systolic and diastolic BP at ≈9-year FU after catheter-based RDN on less medication and without evidence of adverse consequences on renal function.
BACKGROUND: Renal denervation is a recognized adjunct therapy for hypertension with clinically significant blood pressure (BP)-lowering effects. Long-term follow-up data are critical to ascertain durability of the effect and safety. Aside from the 36-month follow-up data available from randomized control trials, recent cohort analyses extended follow-up out to 10 years. We sought to analyze study-level data and quantify the ambulatory BP reduction of renal denervation across contemporary randomized sham-controlled trials and available long-term follow-up data up to 10 years from observational studies. METHODS: A systematic review was performed with data from 4 observational studies with follow-up out to 10 years and 2 randomized controlled trials meeting search and inclusion criteria with follow-up data out to 36 months. Study-level data were extracted and compared statistically. RESULTS: In 2 contemporary randomized controlled trials with 36-month follow-up, an average sham-adjusted ambulatory systolic BP reduction of −12.7±4.5 mm Hg from baseline was observed ( P =0.05). Likewise, a −14.8±3.4 mm Hg ambulatory systolic BP reduction was found across observational studies with a mean long-term follow-up of 7.7±2.8 years (range, 3.5–9.4 years; P =0.0051). The observed reduction in estimated glomerular filtration rate across the long-term follow-up was in line with the predicted age-related decline. Antihypertensive drug burden was similar at baseline and follow-up. CONCLUSIONS: Renal denervation is associated with a significant and clinically meaningful reduction in ambulatory systolic BP in both contemporary randomized sham-controlled trials up to 36 months and observational cohort studies up to 10 years without adverse consequences on renal function.
Muscle sympathetic nerve activity (MSNA) controls the diameter of arterioles in skeletalmuscle, contributing importantly to the beat-to-beat regulation of blood pressure (BP). Although brain imaging studies have shown that bursts of MSNA originate in the rostral ventrolateral medulla, other subcortical and cortical structures—including the dorsolateral prefrontal cortex (dlPFC)—contribute. We tested the hypothesis that MSNA and BP could be modulated by stimulating the dlPFC. dlPFC. In 22 individuals MSNA was recorded via microelectrodes inserted into the common peroneal nerve, together with continuous BP, electrocardiographic, and respiration.Stimulation of the right (n=22) or left dlPFC (n=10) was achieved using transcranial alternating current (tcACS; +2 to −2mA, 0.08 Hz,100 cycles), applied between the nasion and electrodes over the F3 or F4 EEG sites on the scalp. Sinusoidal stimulation of either dlPFC caused cyclicmodulation of MSNA, BP and heart rate, and a significant increase in BP. We have shown, for the first time, that tcACS of the dlPFC in awake humans causes partial entrainment of MSNA, heart rate and BP, arguing for an important role of this higher-level cortical area in the control of cardiovascular function.
Low-frequency sinusoidal galvanic vestibular stimulation (sGVS) can induce perceptions of sway and nausea through entraining vestibular afferent firing to the sinusoidal stimulus. As recently shown, concurrent dorsolateral prefrontal cortex (dlPFC) stimulation via transcranial alternating current stimulation (tACS) greatly attenuates these vestibular perceptions. Given that both vestibular afferents and dlPFC efferents project to the insular cortex, it was reasoned that the insula is the most likely area for the top-down inhibitory interaction to take place. To identify the sites of this interaction, blood-oxygen-level-dependent (BOLD) functional magnetic resonance imaging (fMRI) was collected whilst simultaneously delivering sinusoidal electrical stimulation (±2 mA, 0.2 Hz, 60 cycles) to 20 participants. These stimuli were randomly applied as follows: (i) bilateral sGVS alone via the mastoid processes; (ii) tACS of the dlPFC alone at electroencephalogram site F4; and (iii) sGVS and tACS together. Altered BOLD signal-intensity patterns were identified in the parieto-insular vestibular cortex and thalamus when comparing both sGVS and dlPFC stimulation to concurrent stimulation. Within the brainstem, signal-intensity increased in the inferior olivary nucleus and decreased in the nucleus of the solitary tract during concurrent stimulation, when analysed relative to single stimuli. Because concurrent stimulation elicited different activation patterns in each of these regions compared to the single stimuli, they were considered to be key for the interaction. Given the role each plays in dlPFC and vestibular pathways, the inhibitory function exerted by the dlPFC on vestibular processing likely involves ongoing modulation of one or several of these cortical or subcortical areas.
Transcranial magnetic stimulation (TMS) is applied both in research settings and clinically, notably in treating depression through the dorsolateral prefrontal cortex (dlPFC). We have recently shown that transcranial alternating current stimulation of the dlPFC partially entrains muscle sympathetic nerve activity (MSNA) to the stimulus. We, therefore, aimed to further explore the sympathetic properties of the dlPFC, hypothesizing that single-pulse TMS could generate de novo MSNA bursts. Microneurography was performed on the right common peroneal nerve in 12 participants. TMS pulses were then delivered to the ipsilateral dlPFC at resting motor threshold (MT) of the first dorsal interosseous muscle and at powers 20 below, 10 below, 10% above, and 20% above MT. The MT and 10% above MT intensities were also used to stimulate the right motor cortex and shoulder. Comparisons between stimulus intensities at the same site and between sites at the same intensities revealed no differences in MSNA burst frequency, burst incidence, or single MSNA spikes. Most stimulus trains, however, showed reduced burst frequency and incidence from baseline, regardless of site. This suggests that the TMS itself was evoking arousal-based sympathoinhibition, independent of dlPFC influences. It seems the dlPFC is capable of modulating MSNA but cannot directly generate bursts.
The cortical and subcortical circuitry underpinning motion sickness is poorly understood. However, certain regions – including the nucleus of the solitary tract (NTS) and the parieto-insular vestibular cortex (PIVC) – are known to be involved in the generation of nausea. The PIVC, in particular, is believed to serve an important role in the processing of vestibular inputs. We have previously used low-frequency sinusoidal galvanic vestibular stimulation (sGVS) to induce perceptions of side-to-side motion, which also induces nausea in ~ 40% of people. While sGVS has been previously used with brain imaging to investigate vestibular pathways, those studies used shorter bouts of stimulation (~ 30 s) which did not induce nausea. Here, we have used functional magnetic resonance imaging (fMRI) to record blood-oxygen-level-dependent (BOLD) signal-intensity changes at baseline (5 min), during low-frequency sGVS for a sustained period of time (± 2 mA, 0.08 Hz, 100 cycles; ~21 min) and during a recovery period (9 min) in 17 participants. Surprisingly, however, only four participants reported incipient or overt nausea during stimulation. Nevertheless, two regions displayed BOLD signal-intensity changes associated with the long-form (as opposed to previously used short-form) stimulation: the nucleus accumbens and locus coeruleus. Though sensations of nausea were not as apparent as expected (perhaps due to the necessity of participants adopting a supine position with their heads fixed in place), these regions would appear to be more associated with a longer period of vestibular stimulation, perhaps required to evoke sensations of nausea.