SAN FRANCISCO, CA — The SAFE-PCI study testing the safety and effectiveness of radial-access PCI in women failed to show that the radial approach reduced the risk of bleeding or vascular complications.

However, investigators did observe a reduction in the risk of bleeding or vascular complications with radial access in the entire randomized cohort of women, which included women undergoing PCI and those who underwent a diagnostic cardiac catheterization. Interpreting the results of the trial, however, remain challenging, given that the study was stopped early because of lower-than-expected event rates.

"The question of whether this is a positive or negative trial is really a clinical-trials question," lead investigator Dr Sunil Rao (Duke University Medical Center, Durham, NC) told heartwire . "I think that from a clinical-trials standpoint, you can't say either way, because it didn't reach its planned sample size. But from a clinical standpoint, you have to consider what's worse—having a major bleeding event or having to bail out to the traditional way of doing the procedure. I think most interventionalists would say, probably, that having the bleeding event is worse."

Presenting the results of the study at TCT 2013 , Rao said the results suggest radial-access PCI is a reasonable option in women, although there should be an expectation that some women will require conversion to femoral access. The lack of reduction in bleeding and vascular complications in the PCI subgroup is likely an error resulting from the trial being stopped early.

Dr Sunil Rao

In total, 1787 women from 60 US sites were randomized, including 893 women to radial-access PCI or diagnostic catheterization and 894 women to the same procedures via the femoral artery. Of these, 290 women underwent PCI via the radial artery and 292 women underwent conventional PCI.

As part of their periodic reviews, the data safety and monitoring board recommended the trial be stopped because of significantly lower-than-expected rates of Bleeding Academic Research Consortium (BARC)-defined (2, 3, or 5) events or vascular complications in the patients treated via the femoral artery. Given that the trial was stopped early and significantly underpowered, Rao said the investigators did not expect to see a difference in the primary efficacy end point. The statistically significant reduction in bleeding or vascular complications in the entire cohort, the study's secondary end point, was surprising, he added. Only enough patients were enrolled in the trial to complete the quality-of-life analysis.

No Significant Difference in Primary End Point

Regarding the primary end point, the rate of BARC-defined bleeding or vascular complications occurred in 1.2% of patients undergoing PCI via the radial artery and 2.9% of those who underwent conventional femoral-access PCI, a nonsignificant difference. In the PCI cohort, crossover to femoral access occurred in 6.1% of patients.

In the entire cohort of women, including those undergoing PCI and diagnostic catheterizations via the radial artery, the rate of bleeding/vascular complications was significantly reduced compared with conventional PCI. Again, crossover to femoral access occurred in more than 6% of radial-access treated patients.

Primary Efficacy and Feasibility in PCI Cohort Alone

End point Radial, n=345 (%) Femoral, n=346 (%) p
BARC 2, 3, or 5 bleeding or vascular complications 1.2 2.9 0.12
Access-site crossover 6.1 1.7 0.006

Efficacy and Feasibility in PCI and Diagnostic Catheterization Cohort

End point Radial, n=89 (%) Femoral, n=894 (%) p
BARC 2, 3, or 5 bleeding or vascular complications 0.6 1.7 0.03
Access-site crossover 6.7 1.9 <0.001

To heartwire , Rao said the researchers looked for an interaction between patients treated for PCI and those who underwent diagnostic catheterization and did not observe any heterogeneity, a finding that suggests the difference between the groups is the result of a type-2 error. "We just didn't have enough patients enrolled," said Rao. "We enrolled less than half of the PCI patients we wanted. We wanted 1800 patients, but we only got 690 patients."

High Bivalirudin Use in the Study

Rao said that his center has been performing radial-access PCI since 2005 and despite their expertise, they still prep the femoral artery in case patients need to be converted. "We also counsel patients when we see them in the preprocedure area that our primary approach is to go by the wrist because we think it's safer, but there is a possibility that we might have to go in via the leg," he said.

Dr Alexandra Lansky

The lower-than-expected event rate in the femoral-access arm might be tempered by the high rate of bivalirudin use (60%) in the trial. In addition, Rao said that operators who adopt the radial approach might be "so clued into bleeding complications that they just might be better femoral operators." At this stage, though, the reasons are only hypotheses.

Dr Alice Jacobs

During the press conference announcing the results, Dr Alexandra Lansky (Yale University School of Medicine, New Haven, CT) said bleeding remains an important issue for women undergoing PCI and is a major contributor to mortality. On the whole, Lansky said the results were positive, while Dr Alice Jacobs (Boston University School of Medicine, MA) said it was simply impossible to make conclusions about the results, given that the trial was stopped early.

Dr Roxana Mehran

Dr Roxana Mehran (Mount Sinai School of Medicine, New York), the scheduled discussant during the late-breaking clinical-trials session, said she was "thrilled" to see a study focusing entirely on women, a subgroup of patients consistently overlooked in clinical trials despite having higher rates of bleeding compared with men.

In the trial, the projected bleeding rate in the femoral-access PCI arm was 8%. So, said Mehran, a bleeding/complication rate of 2.9% in the PCI subgroup and 1.7% in the PCI and diagnostic catheterization subgroup suggests US operators are getting better at reducing the risks of the procedure.

"So how do we look at this?" asked Mehran. "Is the glass half full or is it half empty?"

For the purists, the trial can tell researchers very little given the small sample size, the early stoppage, and the nonsignificant difference between the two treatment arms. However, like Rao, Mehran said clinicians will see the results as half full, with the data suggesting the procedure reduces the risk of bleeding or vascular complications in women. However, physicians should keep in mind that approximately three patients will cross over to femoral access in order to prevent one bleeding event, said Mehran.

Of note, SAFE-PCI was a registry-based trial and cost just $5 million, a relatively small amount for such a large clinical trial. The randomization was conducted within the CathPCI Registry of the National Cardiovascular Data Registry.

Rao consults for the Medicines Company and AstraZeneca.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.