ABO Blood Type Linked to Increased Mortality and Stroke Risk Post–LVAD Implantation

Deborah Brauser

April 28, 2016

WASHINGTON, DC — Patients with a left ventricular assist device (LVAD) who have non-O blood types may be at risk for several outcomes postprocedure, including decreased survival, according to two posters presented here at the International Society for Heart and Lung Transplantation (ISHLT) 2016 Scientific Sessions.

The first, which examined 1150 LVAD patients, showed that those with AB blood type had significantly lower survival at 5 years vs those with A, B, or O blood types (P=0.037)[1]. They also had less freedom from ischemic stroke (P=0.034).

However, only 3% of the study population had the AB blood type (n=31), note the investigators, led by Dr John M Stulak (Mayo Clinic, Rochester, MN). Still, they point out that "knowledge of the differences in bleeding/clotting tendencies according to blood group may assist in tailoring management for patients while on support with an LVAD."

The second study assessed 114 consecutive LVAD patients at a single center[2]. Although there were no significant differences in bleeding events up to 5 years later between the O and non-O groups, a trend was found toward less bleeding in those with O blood type—which was exactly the opposite of the investigators' hypothesis, said lead investigator Dr Robert T Cole (Emory University School of Medicine, Atlanta, GA).

He told heartwire from Medscape that their original prediction was based on patients with O blood type having 25% to 30% lower von Willebrand Factor (vWF) levels vs those with non-O blood type—and that patients with LVADs "are known to develop" vWF deficiency and so more bleeding.

"I don't think right now there's enough evidence to say we should adjust patients' level of anticoagulation according to blood type, based on this case," said Cole. However, he noted that both studies are nice stepping stones to further ABO research.

A Tale of Two Studies

In the Stulak study, 1150 patients (80% men) received continuous-flow LVADs at one of five centers between May 2004 and May 2015 (mean age at implantation 59 years). Of these, 42% had blood type O, 42% had blood type A, 13% had B, and (as mentioned) 3% had AB. All were included in the Mechanical Circulatory Support Research Network (MCSRN) registry.

The survival for the entire group was 83% at 1 year after implant, 60% at 3 years, and 43% at 5 years. Percentages of those who were free from ischemic stroke at 1, 3, and 5 years after LVAD implant were 94%, 89%, and 84%, respectively.

In addition, freedom from gastrointestinal (GI) bleeding was 81%, 74%, and 71%, respectively; and freedom from hemorrhagic stroke was 96%, 92%, and 87%. However, these latter two outcomes were not significantly different between the blood groups.

Dr Robert Cole

In the second study, Cole said the investigators wanted to determine whether patients with LVADs and O blood type were at greater risk for postprocedure bleeding vs non-O blood type patients. They defined bleeding events as any episode that resulted in a visit to the emergency department, hospitalization, or transfusion.

The researchers examined patient records for 114 patients who underwent LVAD implantation at Emory between July 2006 and July 2013. Follow-up continued despite dropouts along the way for up to 1500 days for both groups.

At follow-up, 44% of the participants had 115 total bleeding events (49 in the O blood group vs 66 in the non-O group). The events per patient year were not significantly different (0.58 and 0.77, respectively).

However, the O group had a trend toward less bleeding vs the non-O group, according to univariate analysis (P=0.127). After multiple adjustments, including for older age and sex, the hazard ratio [HR] for bleeding wasn't statistically significant between blood groups, but there was a lower bleeding trend for those with O blood type (HR 0.73, 95% CI 0.41–1.30, P=0.28).

"We did wonder if we had more follow-up over time or if we had had more patients in our single-center study, would the findings have become significant? But based on what we had, the P value wasn't significant," said Cole.

Protective Effect?

As reported by heartwire , a study published last month in Circulation of 1.1 million health blood donors in Denmark and Sweden showed that those with non-O blood type were at significant risk for thromboembolic and vascular events compared with those with O blood type.

Cole said that patients with O blood type tend to have vWF and be at risk for bleeding events, while non-O types tend to have more clotting events.

"Because of what's been shown in huge population studies, we thought, 'Maybe that means that O types with LVADs will have more bleeding events.' But we saw the opposite effect, with the Os in our study having less bleeding," he said.

"In some instances, is a lower von Willebrand factor level protective in some regard? Or is there some other compensating factor?"

He added that he hopes more assessments are now done on a larger scale or with a larger database, such as the Interagency Registry for Mechanically Assisted Circulatory Support (Intermacs) to see whether ABO really does have an impact. "I think there are some interesting questions that deserve to be answered."

Or Is It Epigenetics?

"When it comes to mechanical support, it's important to look at all topics in relation to it and to look for predictors that might help," Dr Rahat Muslem (Erasmus Medical Center, Rotterdam, the Netherlands) commented to heartwire .

Regarding Cole's poster, he noted that although there were no significant differences found between the O and non-O groups, "there are other blood types. And I know some doctors have been talking now about this as a possible predictor of bleeding. I just wish they had provided data on the other specific groups, if they had it."

He reported that in the Netherlands, there's been a lot of talk about treatment with antiplatelet therapy, but "the blood itself is particularly interesting with these patients."

Muslem added that he's not sold that blood type itself is a solid predictor, "because I think there's an underlying epigenetic cause" for bleeding.

For the Stulak poster, "these are very good data, although I don't agree with their conclusion," he said, pointing out the small number of patients in the AB group.

"There might be an association, but I don't think [blood type] is the cause of the survival, especially because you don't know what types of things happened over the 5 years," said Muslem, who was not involved with either study. He noted that the poster did seem to show a particularly strong difference in survival at the 1-year point, but without more details, including demographics and comorbidities, "it's difficult to comment on."

The study authors and Muslem report no relevant financial disclosures.

Follow Deborah Brauser on Twitter: @heartwireDeb. For more from theheart.org, follow us on Twitter and Facebook.


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.