Registry Study: Embolic Protection Does Not Appear to Improve SVG PCI Outcomes

Use of embolic protection devices during saphenous vein graft (SVG) interventions is associated with an increase in procedural complications without corresponding improvements in clinical outcomes for up to 3 years afterward, according to a study published in the March issue of Circulation: Cardiovascular Interventions.

The findings “challenge both the safety and effectiveness of distal [embolic protection devices] and question the relevance of existing randomized data in contemporary practice,” write J. Matthew Brennan, MD, MPH, of the Duke Clinical Research Institute (Durham, NC), and colleagues.

But there might still be a role for the devices in high-risk graft interventions, they observe, adding that the new data warrant further randomized evaluation of embolic protection in contemporary practice.

The investigators linked data from the National Cardiovascular Data Registry CathPCI Registry with Medicare inpatient fee-for-service claims information on 49,325 patients aged 65 years and older who underwent SVG PCI at 1,001 hospitals from 2005 to 2009. Median patient age was 75 years, and median graft age was 12.2 years.

Embolic protection devices—including SpideRX and SpiderFX (eV3) and FilterWire EX and FilterWire EZ (Boston Scientific)—were used in 21.2% of interventions. Use was greater in certain prespecified high-risk subgroups: patients with ACS vs no ACS, with de novo vs restenotic lesions, and with graft body lesions vs aortic and distal anastomotic lesions.

Procedural success was high (almost 96%) regardless of the use of embolic protection devices, although procedural complications were more frequent when the devices were employed. The differences remained after risk adjustment (table 1).

Embolic protection was not associated with improvement in either short- or long-term outcomes. The in-hospital mortality rate was 1.1% and the 30-day rate of MACE (death, MI, or repeat revascularization) was 5.5%, with no between-group differences.

By 3 years, 25% of patients had died, 15% had experienced an MI, and 30% had undergone repeat revascularization. In propensity-matched analysis, embolic protection was not tied to lower risks of death, MI, or repeat revascularization. The findings were generally consistent in the high-risk subgroups.

The researchers then compared outcomes between the 5.6% of hospitals that used embolic protection devices routinely (in at least 50% of SVG PCIs) and the 31.5% of centers that did not use them at all during the study period. After adjustment for case mix, hospitals with routine use provided a lower risk of death (HR 0.84; 95% CI 0.75-0.94) but not lower rates of MACE, MI, or repeat revascularization.

However, “no clinical benefit was apparent in the acute setting (as it was in the SAFER trial), thereby making this reduction in mortality more likely to be the effect of differences in downstream management (eg, discharge medications) than procedural outcomes,” the authors write. Centers with routine use of embolic protection also were more likely to use certain evidence-based medications at discharge, including beta-blockers, ACE inhibitors, and statins.

Low Use Despite Strong Guideline Recommendation

Embolic protection devices were introduced to counter the high risk of periprocedural major adverse events early in the experience with SVG stenting. In the SAFER trial, the only study with adequate power to compare clinical events following SVG PCI with or without embolic protection, use of the devices improved outcomes due to reductions in periprocedural MIs and no reflow. Based on those results, embolic protection received a class I (level of evidence B) recommendation in the 2011 PCI guidelines.

However, use of embolic protection during these procedures has remained low for several reasons. The study authors cite potential explanations including the time and complexity added by using the devices, the possible risks of distal vessel dissection, and trends toward lower rates of no reflow and periprocedural MI in SVG PCI, stemming from aggressive preprocedural platelet inhibition and advances in techniques and devices.

These changes in care have made “distal embolic protection unnecessary in routine clinical practice,” Dr. Brennan and colleagues argue. They add, however, that the study cannot address the value of embolic protection in certain high-risk groups.

“Others have demonstrated the association of high graft degeneration scores, long lesion length, and visible thrombus to an increased risk of no reflow and periprocedural MI,” they write. “Furthermore, anecdotal experience has repeatedly demonstrated the use of distal filters for capturing intravascular debris in these cases. Given the limitations of the CathPCI Registry database, identification of a subset of cases involving these high-risk graft interventions was not possible.”

The researchers also note additional study limitations, including the nonrandomized design and uncertain applicability to younger cohorts.

“Together, these limitations emphasize both the hypothesis-generating nature of this observational analysis and the need for a contemporary randomized trial of embolic protection in vein-graft PCI,” they write. “Given the enrollment difficulties observed in the TRAP trial, this may be an ideal setting for a pragmatic, registry-based clinical trial.”

No Definitive Conclusions

In an accompanying editorial, Ron Waksman, MD, and Edward Koifman, MD, of MedStar Washington Hospital Center (Washington, DC), assert that the study’s drawbacks make it impossible to draw firm conclusions about the efficacy and safety of embolic protection devices. They point to selection bias, shortcomings of the data available in the CathPCI Registry, and the lack of information on plaque morphology and volume, vein graft degenerative score, and procedural techniques.

“There is no doubt that [embolic protection] can prevent embolization of plaque debris,” Drs. Waksman and Koifman write. “Its benefits, however, rely on appropriate patient selection and proper device use—both of which require familiarity and experience with the specific device used to prevent potential complications, such as distal embolization because of misuse of the device, device entrapment, or dissection or perforation of the graft or native coronary.”

Decisions to use embolic protection, they say, should be “based on the degree of risk of distal embolization and facilitation of the complexity of the anatomy to use [such a device] safely.” They suggest using 3 categories of risk in decision making:

• High: For patients with ACS and thrombus-containing lesions, a degenerative graft, or high lipid-containing plaque

• Intermediate: For a focal lesion, calcified and tortuous anatomy, or borderline landing zone

• Low: For proximal and distal anastomotic lesions, fibrotic lesions, or in-stent restenosis.

“Intracoronary imaging modalities, such as optical coherence tomography and near-infrared spectroscopy, may be helpful to stratify high-risk morphology with high lipid-containing plaque and can be used as guidance for the need of [embolic protection] for SVG intervention,” Drs. Waksman and Koifman write. “Future study design should take these factors and the heterogeneous nature of the disease into consideration.”

Interventional programs and continuing education should help operators use the devices appropriately, they add.

They liken embolic protection to use of seat belt: “Until definitive data proves otherwise, it is too early and too risky to take the seat belt off when performing interventions to SVG lesions; therefore, we need to consider routine use of [embolic protection] during SVG PCI when appropriate.”

Sources:

1. Brennan JM, Al-Hejily W, Dai D, et al. Three-year outcomes associated with embolic protection in saphenous vein graft intervention: results in 49 325 senior patients in the Medicare-linked National Cardiovascular Data Registry CathPCI Registry. Circ Cardiovasc Interv. 2015;Epub ahead of print.

2. Waksman R, Koifman E. Embolic protection device for saphenous vein graft intervention: too early to take off the seat belt [editorial]. Circ Cardiovasc Interv. 2015;Epub ahead of print.

Related Stories:

• Interceptor Plus Noninferior to Other Embolic Protection Devices for SVG Lesions
• Embolic Protection for SVG: Is It Needed?
• Even in Trial Setting, Embolic Protection Underused with SVG Stenting