Download this article's Factoid in PDF (& PPT for Gold Subscribers)

Among patients with advanced chronic obstructive pulmonary disease (COPD), an arteriovenous (AV) shunt improves physiological parameters such as cardiac output and oxygen delivery but does not have a positive impact on exercise capacity or quality of life. Moreover, it is associated with serious adverse events, according to a pilot study published online February 7, 2012, ahead of print in Circulation: Cardiovascular Interventions.

In 15 patients with advanced chronic COPD, researchers at the CardioVascular Center Frankfurt (Frankfurt, Germany), led by Horst Sievert, MD, percutaneously created an AV shunt in the femoral/iliac region. Using a femoral approach, they tracked a guidewire to a point where artery and vein were adjacent, then pushed a crossing needle through the venous wall into the adjacent artery. A self-expanding nitinol coupler (Rox Medical, San Clemente, CA) was deployed to maintain the connection. Balloon inflation was used to achieve a fistula with a lumen of about 5 mm.

No Clinical Improvement—and Suggestion of Decline

The procedure was technically successful in all patients. At 3 months, a trend was seen toward a decrease from baseline in 6-minute walk distance (the primary endpoint), with a more pronounced decline among the 8 patients who did not respond to oxygen supplementation at baseline. There was no change in the St. George Respiratory Questionnaire score, a measure of overall health, daily life, and perceived well-being (table 1).

Table 1. Changes in Quality of Life from Baseline to 3 Months

 
On the other hand, several physiological parameters improved (table 2).

Table 2. Physiological Changes from Baseline to 3 Months

However, tests of lung function such as total lung volume did not change, nor did measures of dyspnea or the BODE index, a predictor of COPD progression. NYHA class, though, did improve (from a median 3 to a median 2; P < 0.01). In addition, oxygen saturation was similar at baseline and 3-month follow-up regardless of whether or not patients were on oxygen or had it measured before or after the 6-minute walk test.

Serious Safety Concerns

All but 1 patient reported at least 1 adverse event considered to be related to some extent to the procedure or the device, including:

• Right heart failure (n =4)
• Edema (n = 10)
• Venous stenosis (n = 7)
• Venous thrombosis (n = 4)
• Hematoma (n = 4)
• Pseudoaneurysm (n = 1)
• Gingival/nose bleeding (n = 2)

One patient diagnosed with both right heart failure and deep venous thrombosis likely related to the shunt died at 9 months. With the exception of the hematomas and pseudoaneurysm, the average interval between the procedure and the first adverse event was 110 days (range 2 to 337 days).

Importantly, in 8 patients, adverse events—ipsilateral external iliac vein stenosis in 1 and/or ipsilateral lower extremity edema in 7—were severe enough to warrant fistula closure with a covered stent graft. In the remaining patients, the events were mild and/or managed medically.

Still a Viable Concept?

According to Dr. Sievert and colleagues, there is good reason to believe that creating an AV shunt might improve functional capacity in COPD patients: Increasing oxygen saturation in shunted blood may increase systemic arterial oxygen content. Together with increased cardiac output, this could improve oxygen delivery to tissues. In fact, an earlier small trial of AV shunting (Faul JL, et al. Chest. 2010;138:52-58) did yield improvements in the 6-minute walk test.

The authors say the divergent results of the 2 studies should prompt reevaluation of the rationale for AV shunting. Crucially, they note, “it is not clear whether the increase in cardiac output and total oxygen delivery seen in our study translate into better tissue perfusion or oxygenation.” Moreover, after AV shunt creation in the current study, no improvement was seen in the second key component of tissue oxygenation, namely arterial oxygen content. This value declined in the earlier study.

As for safety, the authors admit a legitimate concern exists that a supraphysiologic increase in cardiac output may exacerbate pulmonary hypertension and accelerate pulmonary vascular disease, although no such tendency has been observed thus far. Another potential adverse effect is heart failure. Despite a reduction in systemic vascular resistance, an increase in left ventricular stroke work after AV shunt formation has been shown, and 4 patients in the current study developed right heart failure. Indeed, “compromise in cardiac performance may have contributed to the lack of clinical benefit,” the authors speculate.

Dr. Sievert concludes, “Given the present lack of evidence pointing to a clear benefit and the number of concerning adverse events, careful further investigations of the chronic effect of [AV] fistula creation on tissue oxygenation in an animal model are warranted before further pursuit of the concept of AV shunt creation in patients with COPD.”

In a telephone interview with TCTMD, Louis A. Cannon, MD, of Northern Michigan Regional Hospital (Petoskey, MI), said the rationale for the strategy is understandable. And AV fistulas are hardly novel. “We create them all the time—accidentally when we do a femoral puncture, and [deliberately] in the arm when we do hemodialysis,” he noted.

Risk Destined to Outweigh Benefit

 But Dr. Cannon took issue with the study design.

“The risks have to be less than the potential benefit, and in this study we’re talking about a group of patients that are not critically ill. So you ended up with risks…that may have resulted in death in patients who didn’t have severe pulmonary hypertension.” If fact, he pointed out, those with a pulmonary arterial pressure higher than 35 mm Hg were excluded. Yet these patients are “the ones that are really suffering” and would be most likely to benefit, he suggested.

According to Dr. Cannon, the situation is analogous to TAVR because in order for the treatment to be extended to asymptomatic aortic stenosis patients who are good surgical candidates, the risk from TAVR would have to be quite low, he noted. But here the potential for adverse effects in noncritical patients from a high-flow shunt was considerable, and in fact the procedure made them worse. “In my mind, this was not the correct group to focus on,” he said.

Dr. Cannon specifically criticized the fact that the 1 patient who died did not undergo an autopsy. “If the patient died of a cerebrovascular accident or paradoxical embolus, that would be critical to know,” he said. “It’s really unfortunate that that those data were lost.”

A Restricted Future

On the other hand, Dr. Cannon allowed, for patients who have no other options, the creation of a temporary AV shunt might help. Given the current data, he said, “the avenues [for future research] have been significantly narrowed, but I don’t think they are completely closed.”

One alternative approach, he suggested, might be to look at renal failure patients who have pulmonary arterial hypertension and are on home oxygen, and receive artificial AV [fistulas] to accommodate dialysis. “These patients could simply be followed to see if their oxygenation gets better and their hypercardia and pulmonary arterial hypertension go away.”

Study Details

A minimum of 75 mg of aspirin daily for at least 12 months and 75 mg of clopidogrel daily for 12 weeks was recommended.