Shunting blood from an iliac artery to a neighboring vein via a percutaneous “coupler” device dramatically reduces blood pressure (BP) in patients with resistant hypertension, according to data presented February 3, 2015, at the International Symposium on Endovascular Therapy in Hollywood, FL.

Findings from the small randomized study were also published online January 23, 2015, ahead of print in the Lancet.

The clinical research—characterized as in the early stages but promising by an accompanying Lancet commentary—comes in the wake of a dramatic setback for renal denervation, another interventional treatment for resistant hypertension.

For the multicenter ROX CONTROL HTN study, investigators led by Melvin D. Lobo, MD, of Queen Mary University of London (London, England), randomized 83 patients with uncontrolled hypertension to implantation of an arteriovenous coupler device in addition to their current antihypertensive regimen (n = 44) or continuation on antihypertensive drugs alone (n = 39). At baseline, all patients had an office systolic BP of at least 140 mm Hg and an average daytime ambulatory pressure of at least 135/85 mm Hg while taking at least 3 antihypertensive medications of different classes, including a diuretic.

The ROX Coupler device (ROX Medical; San Clemente, CA), which is delivered percutaneously, creates a 4-mm anastomosis between the distal external iliac vein and artery, diverting blood flow into the venous circulation. The device holds CE Mark approval in Europe but remains investigational in the United States.

Use of anticoagulation in the coupler arm was at the operator’s discretion. According to the authors, differences in baseline characteristics between the device and control groups were unlikely to be of clinical relevance. There were no differences in the number or type of antihypertensive drugs taken, except for less use of dihydropyridine calcium channel blockers in the coupler patients. Diuretics were used in 94% of the overall cohort. 

In a modified intention-to-treat analysis (42 coupler patients, 35 controls), both mean office systolic pressure and mean 24-hour ambulatory systolic pressure at 6 months (primary endpoint) had dropped markedly in the treatment group but not in controls (table 1).

Diastolic pressures followed the same pattern of decline in the device group but not the control group.

Likewise, in a subset of 17 patients (10 coupler; 7 controls) who had undergone renal denervation within 6 months of enrollment, both systolic and diastolic pressures fell significantly from baseline in device-treated patients but not in those on medical therapy alone.

In terms of antihypertensive medications, 11 patients in the coupler group and 2 in the control group were taking fewer drugs during follow-up (P = .0303), while 4 and 10 patients, respectively, were taking more drugs (P = .0382).

Coupler implantation—76% on the right side—was successful in all but 1 patient. There were 25 procedure- or device-related adverse events, 2 of them (urinary retention and anemia) serious. Late events possibly related to treatment included deep venous thrombosis and lower-limb pain. All events resolved without consequences.

Venous Stenosis Fairly Common

However, 28.6% of patients experienced symptoms of one-sided lower-extremity edema several months after the procedure and were diagnosed with iliac vein stenosis proximal to the anastomosis. The condition was treated with venoplasty and stenting in 11 patients and venoplasty alone in 1 patient.

Hypotension leading to a reduction in drugs was reported in 19% of the arteriovenous coupler group and none in the control group (P = .0056). On the other hand, 8% of control patients were hospitalized for hypertensive crisis vs no coupler patients (P = .0225).

According to the authors, the findings support the notion that incorporation of a vein segment in the central arterial circuit restrains the pulsatile blood flow of the cardiac cycle, resulting in immediate BP reduction through improved arterial compliance and reduced systemic vascular resistance.

Interestingly, Dr. Lobo and colleagues say, BP reductions in the coupler group did not differ depending on prior renal denervation, suggesting that inadequate response to the latter therapy may be partly due to arterial stiffness.

Also, they point out, although arteriovenous anastomosis was associated with late development of venous stenosis, the accompanying symptom of lower-extremity swelling was alleviated by use of self-expanding stents.

The investigators acknowledge that the study lacked a sham procedure in the control arm, even though inclusion of a sham in the SYMPLICITY HTN-3 trial appeared to erase the benefit of renal denervation seen in earlier trials. On the other hand, they note, unlike renal denervation, “technical success with the [arteriovenous] coupler can be documented during the procedure and is associated with an immediate fall in blood pressure,” likely eliminating any placebo effect.

The First Step

“Clearly, arteriovenous anastomosis is not yet ready for clinical implementation,” write Markus Schlaich, MD, and Dagmara Hering, MD, PhD, both of the University of Western Australia (Crawley, Australia), in a commentary accompanying the Lancet paper. In fact, the study is “merely the first step in a series of investigations” aimed at addressing outstanding questions such as the longer-term effects of the procedure and the likely need for antithrombotic therapy after stenting of venous stenosis. In addition, the “absence of adverse consequences of central arteriovenous anastomosis on cardiac output and central hemodynamics will have to be shown,” they observe.

Nevertheless, Drs. Schlaich and Hering assert, the study “could prove to be a milestone in the continuing quest for interventional approaches to treat uncontrolled hypertension.”

In a telephone interview with TCTMD, George Bakris, MD, of the University of Chicago Medical Center (Chicago, IL), did not go that far but said the study’s concept “certainly makes sense from a biophysics standpoint. And interestingly, the drop in BP is immediate—you do the procedure and you see the effect.”

But the treatment is not for everyone with resistant hypertension, he cautioned. Considering the theoretical potential for producing LV hypertrophy and high-output heart failure, patients with diabetes, severe PAD, or heart failure with reduced systolic function are “relatively contraindicated,” he said, noting that such patients were excluded from the trial.

Moreover, the high incidence of venous stenosis suggests that the technique may need refinement or the size of the anastomosis recalibration, Dr. Bakris commented. Meanwhile, the possible need for chronic antiplatelet therapy after stenting of venous occlusions “may be a game changer” for some, he noted. Moreover, it would be interesting to know whether venous stenosis compromises BP reduction, he added.

The coupler procedure “has promise for a small subgroup of people,” Dr. Bakris concluded, and it has an advantage over renal denervation at least in terms of simplicity and transparent efficacy. But a panoply of novel mechanical approaches to resistant hypertension, including baroreflex therapy, have yet to be evaluated, he said, adding that the 1 with the best safety and performance profiles will find a place in routine practice.

Sources:
1. Lobo, MD. Central arteriovenous anastomosis for patients with uncontrolled hypertension: the ROX CONTROL HTN study—a randomized controlled trial. Presented at: 2015 International Symposium on Endovascular Therapy; Hollywood, FL.
2. Lobo MD, Sobotka PA, Stanton A, et al. Central arteriovenous anastomosis for the treatment of patients with uncontrolled hypertension (the ROX CONTROL HTN study): a randomised controlled trial. Lancet. 2015;Epub ahead of print.
3. Schlaich M, Hering D. Central arteriovenous anastomosis in resistant hypertension [comment]? Lancet. 2015;Epub ahead of print.

Related Stories:

• Percutaneous AV Shunt May Do More Harm Than Good in Advanced COPD
• Early Data Show Promise of Endovascular Fistulas as Alternative to Surgical AVFs
• SYMPLICITY HTN-3: Renal Denervation Still Safe, Not Effective at 1 Year