Stroke After PCI, CABG Signals Higher Long-term Mortality in Modern Era

The rate at which periprocedural stroke occurs after PCI and CABG in contemporary practice is “not negligible” and is associated with a higher long-term mortality risk, according to findings from the Japanese CREDO-Kyoto PCI/CABG registry Cohort-3.

Stroke occurred within 30 days of the procedure in 0.96% of the PCI group and 2.13% of the CABG group (P < 0.001), with most of these patients leaving the hospital with at least mild disability, researchers led by Ko Yamamoto, MD (Kyoto University, Japan), report in a study published online last week in the American Journal of Cardiology.

Periprocedural stroke was associated with a higher risk of mortality regardless of the mode of revascularization through a median follow-up of 5.7 years.

“When we consider invasive coronary revascularization on top of optimal medical therapy in patients with coronary artery disease, these results would be important information for physicians, patients, and their families,” senior author Takeshi Kimura, MD (Kyoto University), told TCTMD in an email.

Periprocedural Stroke in the New-Generation DES Era

Prior studies have presented data on the incidence of periprocedural stroke after coronary revascularization, but there is limited information from more-contemporary cohorts of patients from “rapidly aging societies,” Kimura said.

To help fill that gap, the investigators turned to the CREDO-Kyoto PCI/CABG registry Cohort-3, which included patients enrolled at 22 Japanese centers. The current analysis covers 14,867 patients who underwent first-time coronary revascularization (89% PCI and 11% CABG) between 2011 and 2013.

A unique aspect of the study is that the researchers assessed the severity of the periprocedural strokes, most of which were ischemic. The majority of strokes (68% in the PCI group and 77% in the CABG group) were considered major, defined with a modified Rankin Scale score of 2 or higher, indicating at least mild disability at hospital discharge.

Independent risk factors for periprocedural stroke in PCI-treated patients included ACS, carotid artery disease, age 75 and older, heart failure, and end-stage renal disease. For CABG-treated patients, they were ACS, carotid artery disease, A-fib, chronic obstructive pulmonary disease, malignancy, and severe frailty.

Among patients who survived at least 30 days after revascularization, periprocedural stroke was associated with an elevated long-term mortality risk, with a stronger relationship seen after CABG (51.1% vs 14.9%; HR 4.55; 95% CI 2.79-7.43) than after PCI (39.2% vs 15.2%; HR 1.71; 95% CI 1.25-2.33). Those increased risks were driven by CV death.

Rates ‘Higher Than Commonly Assumed’

Commenting for TCTMD, Davide Capodanno, MD, PhD (University of Catania, Italy), pointed out via email that the periprocedural stroke rates “were slightly higher than commonly assumed for both PCI and CABG, which may be partly due to the old and relatively sick population enrolled in the registry. Of note, most strokes carried some degree of disability. These data highlight the nonnegligible incidence and impact of this complication, persisting in the contemporary era.”

Capodanno pointed to the fact that periprocedural stroke in this study was defined as events occurring within 30 days according to criteria from the Neurologic Academic Research Consortium (Neuro ARC)—a point also highlighted by the investigators. Prior studies had focused on in-hospital strokes.

Looking over the longer span “allows capturing the onset and impact of later events due, for example, to new episodes of atrial fibrillation,” Capodanno said.

Kimura said the higher rate of periprocedural stroke seen after CABG is consistent with prior studies, adding that the greater severity of the events in the CABG patients might explain why the relationship with long-term mortality risk was stronger in that group than in the PCI-treated patients.

“Therefore, PCI rather than CABG might be a preferable coronary revascularization strategy in patients with high risk for periprocedural stroke,” Kimura said.

Capodanno, however, would not draw a conclusion about the choice between revascularization strategies on the basis of these data.

“I do not see the registry as a comparison of PCI and CABG for this endpoint (we have randomized data and meta-analyses for that purpose), but it is a valuable source of data, with large numbers, to define predictors of stroke in either PCI or CABG patients,” he said. “Interestingly and perhaps not surprisingly, some of these predictors overlapped in the two data sets (eg, ACS, carotid disease).”

For Kimura, “the most important thing is to discuss the risk of periprocedural stroke of individual patients in a multidisciplinary team preferably including a neurologist when we plan not only CABG, but also PCI.”