Domain-specific cognitive dysfunction after cardiac surgery. A secondary analysis of a randomized trial.

BACKGROUND: Brain injury and cognitive dysfunction are serious complications after cardiac surgery. In the perfusion pressure cerebral infarcts (PPCI) trial, we allocated cardiac surgery patients to a mean arterial pressure of either 70-80 mm Hg (high-target) or 40-50 mm Hg (low-target) during cardiopulmonary bypass. In this secondary analysis, we aimed to assess potential differences in domain-specific patterns of cognitive deterioration between allocation groups and to investigate any associations of postoperative cognitive dysfunction (POCD) with diffusion-weighted magnetic resonance imaging (DWI)-detected brain lesions.

METHODS: Of the 197 patients randomized in the PPCI trial, 89 in the low-target group and 80 in the high-target group had complete DWI datasets, and 92 and 80 patients had complete data for an evaluation of cognitive function at discharge respectively. Cognitive function was assessed prior to surgery, at discharge and at 3 months. DWI was obtained at baseline and on postoperative days 3 to 6.

RESULTS: We found no statistically significant differences between the two groups when comparing the proportion of patients with a domain-specific deterioration over the pre-defined critical level in seven individual test variables at discharge. Significant deterioration was most common in tests thought to assess cognitive flexibility and interference susceptibility and least common in the memory test. POCD at discharge was more frequent in patients with DWI-positive brain lesions (OR adjusted for age and group allocation: 2.24 [95% CI 1.48-3.00], P = 0.036).

CONCLUSIONS: Domain-specific patterns of POCD were comparable between groups. A significant association was seen between DWI-positive brain lesions and POCD.