Interventional procedure overview of alcohol-mediated perivascular renal sympathetic denervation for resistant hypertension
Closed for comments This consultation ended on at Request commenting lead permission
Evidence summary
Population and studies description
This interventional procedure overview is based on 480 people from 2 RCTs (Kandzari 2024; Pathak 2023) and 3 single-arm studies (Mahfoud 2020, 2021; Janas 2020; Fischell 2016). Of these 480 people, 271 people had the procedure. This is a rapid review of the literature, and a flow chart of the complete selection process is shown in figure 1. This overview presents 5 studies (6 papers) as the key evidence in table 2 and table 3, and lists other relevant studies in appendix B, table 5.
All the studies were done in Europe and the US, and the follow-up duration ranged from 6 months to 24 months. Only Fischell (2016) was a single-centre study, the other 4 studies were done in 2 or more centres. Both RCTs compared RDN with sham controls (diagnostic renal angiography only).
The mean age ranged from 54 to 60 years across studies. The common morbidities were diabetes (type 2), hyperlipidaemia, and cardiovascular disease. In terms of antihypertensive drugs prescribed, the mean number ranged from 3.4 to 5.1 in 3 studies (Fischell 2016; Mahfoud 2020, 2021; Janas 2020). For the other 2 studies, Kandzari (2024) described 77% of people taking 3 or more medications, and Pathak (2023) reported 45% of people taking 2 or more medications, 30% having 1 medication and 25% of people with no medications. Recognising that escalation of medication burden is a considerable predictor of nonadherence, inconsistency of medication adherence was considered one of the confounding factors.
Table 2 presents study details.
Study no. | First author, date country | Population characteristics | Antihypertensive drugs | Study design | Inclusion criteria | Intervention | Follow up |
|---|---|---|---|---|---|---|---|
Kandzari (2024) 9 countries (99 centres) | n=301 (224 males, 77 females)
Mean 56.1 years | Range 2 to 5 (233 people with 3 or more medications) | RCT (NCT02910414; TARGET BP I; phase 3) | People (18 to 80 years old) with hypertension (office systolic BP of 150 to 180 mmHg and diastolic BP of 90 mmHg or above), despite prescription of 2 to 5 antihypertensive medications. Approximately 1 week before randomisation, people with a mean 24-hour systolic ABPM of 135 to 170 mmHg and confirmed anatomic eligibility. | Alcohol-mediated perivascular RDN: using a novel 3 needle-based delivery device (Peregrine System Infusion Catheter) with 0.6 ml alcohol infused per treated renal artery with a maximum dose of 2.4 ml per person. In addition to the main renal arteries, eligible renal accessory arteries were also treated. Sham: diagnostic renal angiography only. | 6 months (efficacy outcomes limited to 3 months) | |
Pathak (2023) Europe and US (25 centres) | n=106 (78 males, 28 females)
mean 54.1 years | 2 or more:
| RCT (NCT03503773; TARGET BP OFF-MED; phase 2) | People (18 to 80 years old) with hypertension (office systolic BP of 140 to 180 mmHg and diastolic BP of 90 mmHg or above), taking 0 to 2 antihypertensive medications. After a 4-week run-in period, people with 24-hour systolic ABPM of 135 to 170 mmHg and confirmed anatomic eligibility. | Alcohol-mediated perivascular RDN: using a novel 3 needle-based delivery device (Peregrine System Infusion Catheter) with 0.6 ml alcohol infused per treated renal artery with a maximum dose of 2.4 ml. In addition to the main renal arteries, 5 renal accessory arteries were treated. Sham: diagnostic renal angiography only. | 12 months | |
Mahfoud (2020) Europe (9 centres in Poland, Czech Republic, Belgium, and Germany) | n=45 with 94 treated arteries (28 men,17 women; 1 withdrew consent at 6-month follow up) mean 55 years | Mean 5.1 (SD 1.5) | Single-arm study | People with uncontrolled hypertension, despite taking at least 3 antihypertensive medications of different classes for at least 4 consecutive weeks; the renal artery diameter between 4 and 7 mm, with a renal artery length of 5mm or more | Bilateral RDN using the Peregrine Catheter with 0.6 ml alcohol infused per renal artery:
4 people had an accessory artery treated in addition to the 2 main renal arteries. | 6 months | |
Mahfoud (2021) | n=41 completed the trial (as 4 losses to follow up) age same as above | Same as above | Same as above | Same as above | Same as above | 12 months | |
Janas (2020) Czechia (single centre) and Poland (2 centres) | n=10 (5 males, 5 females) with 20 treated arteries mean 60 years | Mean 4.8 (SD 1.3) | Single-arm study | People with resistant hypertension (office systolic BP >160 mm Hg or >150 mm Hg if people with type 2 diabetes), despite taking at least 3 antihypertensive medications including a diuretic agent for at least 4 weeks; mean 24-hour systolic pressure of 135 mmHg or higher based upon ABPM | Alcohol-mediated RDN, using a novel 3 needle-based delivery device (Peregrine System Infusion Catheter) with 0.3 ml alcohol infused per renal artery:
| 24 months (1 person was lost to follow up at 24 months) | |
Fischell (2016) US (single centre) | n=18 (9 males,9 females) with 37 treated arteries mean 53.5 years | Mean 3.4 (SD 0.7) | Single-arm study | People with resistant hypertension (office systolic BP >160 mm Hg or >150 mm Hg if people with type 2 diabetes), despite taking at least 3 antihypertensive medications, including a diuretic agent | Alcohol-mediated RDN, using a novel 3 needle-based delivery device (Peregrine System Infusion Catheter) with 0.3 ml alcohol infused per renal artery:
| 6 months (1 person died and 1 lost to follow up) | |
The risk of bias (RoB) was assessed using the Cochrane RoB Tool for RCTs and adapted RWE framework tool for single-arm studies. The RoB for RCTs was graded as "low" for low risk, "high" for high risk, and "some concerns". For single-arm studies, each domain of the RoB was rated as "low", "moderate", "serious", or "critical" for RoB. The overall RoB was determined according to the judgment for each domain. For the 2 RCTs (Kandzari 2024; Pathak 2023), some concerns were raised for 1 domain (bias due to missing outcome data) and a low level of concerns was rated for other domains. So, the overall risk of bias was judged as some concerns. For the 3 single-arm studies, Mahfoud (2020, 2021) was rated as low for all RoB domains, so the overall risk of bias was low. For the other 2 single-arm studies, Janas (2020) was rated moderate for 1 domain (bias due to measurement of outcomes) and as low for all other domains whereas Fischell (2016) was judged as moderate for 2 domains (bias due to measurement of outcomes and reporting bias) and as low for all other domains. So, the overall risk of bias was considered moderate. | |||||||
Procedure technique
All the studies described the procedure technique and device used. Under fluoroscopic guidance, a 3-needle-based endovascular delivery device (Peregrine System Infusion Catheter, Ablative Solutions, Inc.,) was used to deliver microdoses of dehydrated alcohol, as a neurolytic agent, locally into the perivascular space of the renal artery to achieve ablation of the afferent and efferent sympathetic nerves. The amount of alcohol infused per renal artery ranged from 0.3 ml (Janas 2020; Fischell 2016) to 0.6 ml (Mahfoud 2020, 2021; Pathak 2024; Kandzari 2024). Most procedures were bilateral RDNs.
When reported, the mean treatment time (from the advancement of the device or infusion catheter insertion to time of retraction) ranged from 7 minutes (Mahfoud 2020) to 10 minutes (Fischell 2016) per artery, and the mean procedure time (from femoral artery access to sheath removal) was between 49 minutes (Mahfoud 2020) and 62 minutes (Pathak 2024). The mean total volume of contrast used was 100 mL (Pathak 2024).
Procedure success (device success with freedom from periprocedural MAE) was reported in 3 studies, ranging from 95% to 100% (Kandzari 2024; Pathak 2023; Fischell 2016).
Efficacy
Reduction in BP
Reduction in BP covered both 24-hour ABPM and office BP. A meta-analysis was done, with the pooled results of 24-hour systolic ABPM and office systolic BP shown in appendix C.
ABPM
24-hour ABPM was assessed and reported in 4 of the 5 studies. After RDN, systolic BP reduced statistically significantly across studies and a MCID of 5 mmHg was met at final follow-up time points in most studies. When comparing RDN with sham procedure, the 2 RCTs showed statistically significant difference in BP reductions at 3 months but not at 12 months when reported in 1 RCT. Considerable BP reductions in the sham controls mitigated the between-group differences.
In an RCT of 301 people who had RDN (n=148) or a sham procedure (n=153), Kandzari (2024) reported that the mean change at 3 months from baseline in 24-hour systolic ABPM was greater in the RDN group compared with the sham group (-10.0 mmHg compared with -6.8 mmHg), with a statistically significant between-group difference, favouring RDN (-3.2 mmHg; 95% CI -6.3 to 0.0; p=0.0487). But there was no statistically significant difference in 24-hour diastolic ABPM reduction between RDN and sham control (-5.4 mmHg compared with -4.1 mmHg; treatment difference, -1.4 mmHg; 95% CI -3.1 to 0.3; p=0.1146).
In an RCT of 106 people who had RDN (n=50) or a sham procedure (n=56), Pathak (2023) found statistically significant reductions in 24-hour systolic ABPM at 6 months and 12 months after the procedure in both groups (RDN: -13.9 mmHg and -10.6 mmHg at 6 and 12 months, respectively; all p<0.0001; sham: -13.4 mmHg and -15.9 mmHg at 6 and 12 months, respectively; all p<0.0001). But the authors did not see a statistically significant difference in BP reduction between groups at both time points (-0.55 mmHg and 5.3 mmHg at 6 and 12 months, respectively; all p>0.05). For 24-hour diastolic ABPM, the authors observed statistically significant reductions at 6 and 12 months in both groups, and reported a statistically significant difference in BP reduction between groups at 12 months, favouring sham procedure (2.5 mmHg, p=0.0341), but not at 6-month follow up.
Mahfoud (2020, 2021) reported that, of the 45 people who had RDN, 24-hour systolic and diastolic ABPM statistically significantly reduced at 6 months (mean change in systolic BP, -11 mmHg; mean change in diastolic BP, -7 mmHg; both p<0.001), and at 12 months from baseline (mean change in systolic BP, -10 mmHg; mean change in diastolic BP, -7 mmHg; both p<0.001).
Janas (2020) found that, of the 10 people who had RDN, 24-hour systolic ABPM statistically significantly reduced at 12 months from baseline (mean change, -6 mmHg; p<0.05) but not at 6 months (mean change, -3 mmHg) and 24 months (mean change, -1 mmHg). For 24-hour diastolic ABPM, the mean change was 2 mmHg at 6 months, -3 mmHg at 12 months and -1 mmHg at 24 months.
Office BP
Office BP was evaluated in all 5 studies. After RDN, both statistically and clinically (a MCID of 10 mmHg) significant reductions in systolic BP were shown across all studies. When comparing RDN with the sham procedure, the 2 RCTs reported statistically significant difference in BP reductions at 3 months but not at 12 months when reported in 1 RCT. Notable BP reductions in the sham controls lessened the between-group differences.
Kandzari (2024) reported that the mean reduction in office systolic BP at 3 months was -12.7 mmHg (SD 18.3) for the RDN group compared with -9.7 mmHg (SD 17.3) for the sham control group. The difference between groups was not statistically significant (-3.0 mmHg; 95% CI -7.0 to 1.0; p=0.173). Similarly, there was no statistically significant difference observed in office diastolic BP reduction between groups.
Pathak (2024) found that the mean changes in office systolic BP at 6 and 12 months were -12.9 mmHg (SD 15.6; p<0.0001) and -11.0 mmHg (SD 15.3; p<0.0001) in the RDN group and -14.7 mmHg (SD 15.7; p<0.0001) and -13.2 mmHg (SD 16.6; p<0.0001) in the sham group. Between-group comparison did not show any statistically significant differences in mean changes at both time points (6 months, 1.8 mmHg [95% CI -4.5 to 8.2], p=0.724; 12 months, 2.2 mmHg [95% CI -4.5 to 8.9], p=0.6823). For office diastolic BP, the mean changes at 6 and 12 months were -10.0 mmHg (SD 9.0; p<0.0001) and -9.4 mmHg (SD 9.4; p<0.0001) in the RDN group and -8.4 mmHg (SD 9.5; p<0.0001) and -9.6 mmHg (SD 11.0; p<0.0001) in the sham group. No statistically significant difference in mean changes between groups at both follow-up durations (6 months, -2.5 mmHg [95% CI -6.1 to 1.2], p=0.3575; 12 months, -1.6 mmHg [95% CI -5.4 to 2.1], p=0.6375).
Mahfoud (2020, 2021) reported statistically significant reductions in office systolic and diastolic BP at 6 months (mean change in systolic BP -18 mmHg; mean change in diastolic BP -10 mmHg; both p<0.001), and at 12 months from baseline (mean change in systolic BP -20 mmHg; mean change in diastolic BP -10 mmHg; both p<0.001).
Janas (2020) found that office systolic BP statistically significantly reduced at 6, 12 and 24 months from baseline (mean change, -28 mmHg, -21 mmHg and -25 mmHg, respectively; all p<0.001). For office diastolic BP, the authors reported that the reduction was 7 mmHg, 2 mmHg and 6 mmHg at 6, 12 and 24 months, respectively.
Fischell (2016) described that, of the 18 people who had RDN, office systolic and diastolic BP reduced at 6 months (mean change in systolic BP, -24 mmHg; mean change in diastolic BP, -12 mmHg). The authors also reported similar BP outcomes in the 12 people with accurate medication data (mean change in systolic BP, -24 mmHg; mean change in diastolic BP, -12 mmHg).
Use of antihypertensive medications
Data on antihypertensive medications was described in all 5 studies. But medication adherence was evaluated and reported in 2 studies only, with the rate of full adherence between 50% and 60% over time (Kandzari 2024; Mahfoud 2020).
Kandzari (2024) described that there were no statistically significant differences between groups in prescribed medication changes from baseline to 3 months, and no notable differences between groups in dose titration score, defined daily does, or medication index at 3 months. Adherence testing revealed that 43% and 41% of people in the RDN and sham groups, respectively, were fully adherent to their prescribed medications at baseline (p=0.712). At 3 months, the adherence rates increased to 51% and 49%, respectively (p=0.765). The rates of partial and complete non-adherence were similar between groups at all time points and did not statistically significantly vary.
Pathak (2024) reported that the medication burden was statistically significantly lower in the RDN group than the sham group at 12 months (mean daily defined dose: 1.5 compared with 2.3; p=0.017). The authors also described that following primary endpoint collection at 8 weeks, antihypertensive medication was uptitrated to achieve a target office systolic BP of 140 mmHg or less, and that the medication burden increased from 8 weeks in both groups (RDN, 0.08; sham, 0.12).
Mahfoud (2020) described that urine toxicological analyses revealed that adherence to the antihypertensive regimen remained relatively consistent over time (75%, 82%, and 78% after 1, 3, and 6 months of follow up). This was also reflected in the proportion of people who were fully adherent with their antihypertensive regimens, with 52%, 61%, and 59% at 1, 3, and 6 months of follow up, respectively. At 12 months, Mahfoud (2021) reported that the number of antihypertensive medications reduced from baseline (-0.1; 95% CI -0.5 to 0.4).
Janas (2020) found that antihypertensive medications increased in 2 people, reduced in 1 person and remained the same in 6 people at 24 months after RDN.
Fischell (2016) reported that the number of antihypertensive medications reduced to 2 at 6 months from 3.4 at baseline in the 12 people with accurate medication data.
Renal function
Renal function was reported in all 5 studies. There were no statistically significant changes in renal function after RDN except for Mahfoud (2021) who found a statistically significant reduction in eGFR at 12 months.
Kandzari (2024) described that renal function remained unchanged from baseline through 3 and 6 months in both the RDN and sham groups (mean change in eGFR through 6 months: -1.2 and -0.9 mL/min/1.73m2 for the RDN and sham control groups, respectively; p=0.728)
Pathak (2023) reported that eGFR remained stable in the RDN group (mean change, -2.1 ml/min/1.73m2) but decreased in the sham group up to 12-month follow up (mean change, -6.4 ml/min/1.73m2). The difference between groups was statistically significant (p=0.0224) although baseline values were comparable between groups.
Mahfoud (2020, 2021) found that eGFR statistically significantly reduced at 12 months from baseline (mean change, -3.9±10.3 ml/min/1.73m2; p=0.02) but not at 6 months (mean change, -2.7±12.1 ml/min/1.73m2; p=0.15). The authors did not observe any statistically significant changes in serum creatinine, cystatin C and spot urine albumin at both 6 and 12 months.
Janas (2020) reported that there was no deterioration in renal function (including blood urea nitrogen, serum creatinine and eGFR) during the 24-month follow up and no significant divergences between the follow-up time points. The authors noted that 25% or more reduction in eGFR was observed in 1 person at 12 months.
Fischell (2016) described that eGFR increased from 66 ml/min/1.73m2 at baseline to 75 ml/min/1.73m2 at 6 months, but this change was not statistically significant (p=0.15). The authors also found that serum creatinine, blood urea nitrogen, and electrolytes remained stable over the study period.
Safety
Major or serious adverse events
Major or serious adverse events were reported in all 5 studies and the observation period ranged from 1 to 6 months. The rate of MAEs was up to 7% in 3 studies (Kandzari 2024; Pathak 2024; Mahfoud 2020) and the rate of serious adverse events was 20% in Janas (2020). Although death was reported in the Fischell (2016) study, it was unrelated to the device or procedure.
Kandzari (2024) found that at 30 days, the proportion of people with MAEs was 5% (n=7) for the RDN group and zero for the sham control group, with a statistically significant difference between groups (p=0.007). In the RDN group, MAEs included 1 major vascular complication, 1 hypertensive emergency and 6 hypotension needing intervention or medication change. By 6 months, cumulative occurrence of MAEs was similar between groups (RDN, 5% [n=11]; sham, 4% [n=6], p=0.224). In the RDN group, the 11 events included 1 death (unrelated to the procedure, device or drug), 1 myocardial infarction, 1 major vascular complication, 2 hypertensive emergencies, and 7 hypotension needing intervention or medication change. In the sham group, the 6 events consisted of 1 myocardial infarction, 2 hypertensive emergencies, and 3 hypotension needing intervention or medication change.
Pathak (2024) reported that the rate of MAEs up to 30 days after treatment was 2% in the RDN group and less than 2% in the sham group. The authors described that 1 person experienced a hypertensive crisis up to 30 days after RDN and 1 person had a vascular complication after the sham procedure (the person developed a small subcutaneous haematoma, subsequently diagnosed as aneurysm spurium).
Mahfoud (2020) reported that the proportion of people with primary safety endpoint events (any periprocedural major vascular complications or major bleeding, acute kidney injury, or death within 1 month of the procedure) within 1 month after RDN was 4% (n=2), including periprocedural major vascular complications/vascular access pseudoaneurysm (n=2) and major bleeding (TIMI classification, n=1). Within 6 months postprocedure, the proportion of people with MAEs was 7% (n=3), including major vascular complications (n=2) and severe hypotension or syncope (n=1).
Janas (2020) found that there were 3 serious adverse events in 2 people, including inflammation of the duodenum mucosae membrane at 6 months (n=1), upper respiratory tract infections (n=1) and diabetes intensification (n=1). The last 2 events happened to the same person.
Fischell (2016) reported death in 1 person at 9 weeks after RDN but this was unrelated to the device or procedure.
Anecdotal and theoretical adverse events
Expert advice was sought from consultants who have been nominated or ratified by their professional society or royal college. They were asked if they knew of any other adverse events for this procedure that they had heard about (anecdotal), which were not reported in the literature. They were also asked if they thought there were other adverse events that might possibly occur, even if they had never happened (theoretical).
They listed the following anecdotal and theoretical adverse events: angiography related complications (such as rupture of arteries and cholesterol emboli) and specific complications (such as loss of a kidney).
Eight professional expert questionnaires for this procedure were submitted. Find full details of what the professional experts said about the procedure in the specialist advice questionnaires for this procedure.
Validity and generalisability
The key evidence includes 2 RCTs and 3 single-arm studies. Most studies had a follow up of 6 to 12 months and only Janas (2020) reported the data for 24 months. All studies were done in Europe and the US.
In addition to the RoB assessment detailed in table 2, 1 RCT (Kandzari 2024) was adequately powered for 24-hour systolic ABPM (a between-group difference of 6 mmHg). But the other RCT (Pathak 2023) was a phase 2 trial and underpowered for statistical comparisons of efficacy or safety events. The sample sizes for the 3 single-arm studies were small, so there was no statistically powered endpoint in these studies. Also, all 5 studies were sponsored by Ablative Solutions, Inc., and had more than 1 author with conflicts of interest reported.
Across studies, there was variation in the population groups and procedure techniques; this might affect the efficacy and safety outcomes. For the population groups, 3 studies (Mahfoud 2020, 2021; Janas 2020; Fischell 2016) included people with resistant hypertension, 1 study (Kandzari 2024) recruited people with moderate uncontrolled hypertension and resistant hypertension, and 1 study (Pathak 2023) selected people with mild to moderate uncontrolled hypertension.
In terms of the procedure techniques, there were different doses of alcohol infused per renal artery. A low dose of 0.3 ml was used in 2 studies (Janas 2020; Fischell 2016) but a high dose of 0.6 ml was applied in 3 studies (Kandzari 2024; Pathak 2023; Mahfoud 2020, 2021). In addition, it is acknowledged that complete renal artery treatment is important for BP reduction; in particular treating accessory arteries, which has been shown to contribute to the sympathetic innervation of the renal parenchyma. But there was no measure of effective or complete ablation of renal sympathetic nerves (Kandzari 2024; Pathak 2023).
Taken together, the evidence suggested statistically and clinically significant reductions in 24-hour systolic ABPM and office systolic BP after RDN from baseline. But this pre- and post-RDN effect should be interpretated with caution and might be vulnerable to Hawthorne effect such as improvement in adherence to antihypertensive medications and change in lifestyles.
When comparing RDN with the sham procedure, the effect of RDN on BP lowering was only found at 3 months (not beyond this time point). But its magnitude was small, given unexpected large BP reductions observed in the sham controls. This indicated a potential placebo effect on BP lowering after sham procedures. Also, Pathak (2023) argued that most people in their trial were recruited during the COVID-19 pandemic, which might contribute to the increase in systolic BP and potentially introduce additional confounding factors. To support this point, the authors found larger and clinically meaningful BP changes in people who were enrolled before the start of the COVID-19 pandemic. Pathak (2023) also claimed that there was a possibility that the confounding effect of the COVID-19 pandemic was not evenly distributed between people and treatment groups (RDN and sham).
Data on antihypertensive medications was reported across studies, and the changes were generally small. Pathak (2023) found that antihypertensive medication use was lower in the RDN group at 3, 6, and 12 months postprocedure. But antihypertensive medication utilisation increased from 8 weeks to 12 months in both groups. This was because people stopped their antihypertensive medications 4 weeks before randomisation and, after primary endpoint collection at 8 weeks, antihypertensive medication was uptitrated to achieve a target office systolic BP. Notably, across all the studies, medication adherence was only measured in 2 studies and the rate of full adherence was between 50% and 60%. This highlighted the potential for confounding factors relating to inconsistency of medication adherence and indicated medication adherence being an ongoing challenge in hypertension management. With regard to renal function, evidence generally suggested that it remained stable after RDN.
For the safety outcomes, the rates of MAEs were up to 7% in 3 studies and the rate of serious adverse event was 20% (2/10) in the Janas (2020) study. Device- or procedure-related, nonserious adverse events included transient microleaks of alcohol, minimal pain, minor vessel dissection and device deficiency; these events did not raise significant safety concerns.
In summary, the evidence suggested that absolute reductions in both 24-hour systolic ABPM and office systolic BP achieved within the RDN groups were statistically significant and clinically relevant. But the relative differences between RDN and sham controls were of uncertain clinical significance and in particular the duration of effect was unclear. The effect of RDN on BP reductions could be affected by various factors. So, adequately powered RCTs and other well-designed studies with larger samples and longer follow ups are warranted to conclusively determine the BP-lowering effect of alcohol-mediated RDN in the management of hypertension and to provide greater assurance of procedural safety.
It is noted that the Kandzari (2024) study (NCT02910414), included in the key evidence, is still ongoing to continue examining whether the theoretical advantages of alcohol-mediated denervation with the Peregrine System Kit translate into clinical benefits. At the time of preparing this overview, the estimated completion date is December 2025. No other ongoing trials have been identified.
How are you taking part in this consultation?
You will not be able to change how you comment later.
You must be signed in to answer questions