Perioperative covert stroke in patients undergoing coronary artery bypass graft surgery

Abstract

Objectives: Covert stroke is a complication of coronary artery bypass graft surgery that is increasingly recognized as a serious problem. In noncardiac surgery settings, covert stroke is associated with the development of delirium, long-term cognitive decline, and future clinical stroke. Therefore, we sought to determine the feasibility of conducting a large, prospective cohort study of the influence of covert stroke on neurocognitive outcomes in patients undergoing coronary artery bypass graft surgery.

Methods: NeuroVISION Cardiac pilot was a prospective cohort study enrolling patients aged ≥21 years undergoing isolated coronary artery bypass graft surgery to receive diffusion-weighted magnetic resonance imaging of the brain after surgery to identify patients with covert stroke. Patients were screened for postoperative delirium in-hospital and were administered questionnaires of cognitive and global function (once before and twice after surgery). Regional cerebral oxygen saturation was recorded during surgery using near-infrared spectroscopy.

Results: Between March 27, 2017, and February 11, 2018, 50 of 66 patients enrolled (76%) completed the brain magnetic resonance imaging (1 patient per week). Among the 49 patients included in the analysis, 19 (39%; 95% confidence interval, 26%-53%) experienced perioperative covert stroke and 3 (6%) had a clinical stroke within 30 days of surgery. Postoperative delirium occurred in 5 (26%) patients with covert stroke and in 3 (10%) patients who did not experience covert stroke.

Conclusions: The NeuroVISION Cardiac pilot study established the feasibility of conducting a large, prospective cohort study of the determinants and consequences of covert stroke in patients undergoing coronary artery bypass graft surgery.

Keywords: CABG; CABG, coronary artery bypass grafting; DSST, Digit Symbol Substitution Test; DW-MRI, diffusion-weighted magnetic resonance imaging; MRI; MoCA, Montreal Cognitive Assessment; SAGE, Standard Assessment of Global Activities in the Elderly; cognitive; coronary artery bypass graft surgery; covert; delirium; magnetic resonance imaging; oximetry; stroke.

Graphical abstract

Incidence of neurologic complications in patients who experienced or did not experience covert stroke.

Figure 1

Distribution of patients in the pilot study with breakdown of reasons for exclusion. The most common reason for not completing the magnetic resonance imaging (MRI) was patient refusal. Forty-nine patients with evaluable MRI results were included in the final analysis, of whom 19 experienced a covert stroke. Covert strokes were defined based on diffusion-weighted MRI evidence of acute or subacute brain infarcts and the absence of a clinical stroke diagnosis following the brain MRI.

Figure 2

Shown left to right is the incidence of intraoperative cerebral oxygen desaturation events (blue bars), postoperative delirium (red bars), and clinical stroke within 30 days of coronary artery bypass grafting (CABG) surgery (green bars) in patients not experiencing (left) or experiencing (right) a perioperative covert stroke. Cerebral oxygen desaturation events (CODEs) occurred when regional cerebral oxygen saturation (ie, oximetry) values declined ≥20% below baseline in either left, right, or both hemispheres during surgery. Three patients were excluded from the oximetry analysis: 2 patients with incomplete transcripts and 1 patient undergoing off-pump CABG surgery. ∗Covert stroke was associated with higher rates of CODEs during surgery (12 [67%] patients experiencing covert stroke vs 9 [32%] without covert stroke [P = .02]).

Video 1

A video summary of the main results, relevance, and insights gained from the study. Video available at: https://www.jtcvs.org/article/S2666-2736(20)30081-4/fulltext.

Figure 3

Shown left to right is the incidence of intraoperative cerebral oxygen desaturation events (blue bars), postoperative delirium (red bars) and clinical stroke within 30 days of coronary artery bypass grafting (CABG) surgery (green bars) in patients not experiencing (left) or experiencing (right) a perioperative covert stroke. Cerebral oxygen desaturation events (CODEs) occurred when regional cerebral oxygen saturation (ie, oximetry) values declined ≥20% below baseline in either left, right, or both hemispheres during surgery. Cognitive tests were administered before surgery, at hospital discharge and again 30 days after surgery. Acute infarcts (visible on diffusion-weighted magnetic resonance imaging [DW-MRI] sequences) were present only in the brain (schematic diagrams in grey) of patients who experienced a perioperative covert stroke (black dots within the right brain diagram). Brain diagrams (frontal sections only) are not drawn to scale. Three patients were excluded from the oximetry analysis: 2 patients with incomplete transcripts and 1 patient undergoing off-pump CABG surgery. CI, Confidence interval; ∗Covert stroke was associated with higher rates of CODEs during surgery (12 [67%] patients experiencing covert stroke vs 9 [32%] without covert stroke [P = .02]).