Adverse cognitive outcome is well recognized after coronary artery bypass grafting (CABG) while little is known about the extent and duration of decline after cardiac valve surgery. We investigated changes in cognitive function following conventional cardiac valve surgery over up to 4 years.
Among 36 patients (65.2 ± 9.2 years, 36% women) who received valve surgery, we assessed serial cognitive function with a battery of 11 standardized tests across 3–4 years. Cognitive function was analysed to identify: (1) cognitive decline (i.e. within-patient changes in test scores) and (2) cognitive deficit (i.e. drop of score ≥1 SD in ≥3 tests). Diffusion-weighted magnetic resonance imaging (DW-MRI) was applied pre- and post-procedure to detect ischaemic brain injury. Data were compared to a historical cohort of 39 patients undergoing CABG.
After both valve surgery and CABG, a significant decline at discharge was detected in 7 of 11 cognitive tests. The rate of patients with a cognitive deficit after valve surgery vs CABG was 39% vs 56% at discharge, 14% vs 23% at 3 months, and 16% vs 26% at 3–4 years (not significant, [n.s.]). After valve surgery, DW-MRI identified 19 (53%) patients with evidence of 50 new focal ischaemic lesions (CABG: 20 [51%] patients with 42 lesions, n.s.). Cumulative cerebral ischaemic load per patient was not significantly different between the valve surgery group and CABG group (503 ± 485 mm3 vs 415 ± 234 mm3). After correction for multiple potential risk factors in both groups, reduced verbal memory at discharge could be identified as a predictor of long-term cognitive impairment in CABG patients only (P = 0.04). For both the valve surgery and CABG group, no association between cognitive impairment and new ischaemic cerebral lesions was found.
The course of cognitive performance after valve surgery and CABG was similar with early postoperative decline followed by subsequent recovery. Although silent small brain infarcts were present in about half of all patients, they did not impact cognitive performance neither at early nor during long-term follow-up.