Background: Increasing evidence links postoperative cognitive dysfunction (POCD) to surgery and anesthesia. POCD is recognized as an important neuropsychological adverse outcome in surgical patients, particularly the elderly. This prospective cohort study aimed to investigate whether POCD is associated with impaired intraoperative cerebral autoregulation and oxygenation, and increased levels of biomarkers of brain injury.
Methods: Study subjects were patients ≥65 years of age scheduled for major noncardiac surgery. Cognitive function was assessed before and 1 week after surgery. POCD was diagnosed if a decline of >1 standard deviation of z-scores was present in ≥2 variables of the test battery. The incidence of POCD 1 week after surgery was modeled as a multivariable function of the index of autoregulation (MxA) and tissue oxygenation index (TOI), adjusting for baseline neuropsychological assessment battery (Consortium to Establish a Registry for Alzheimer's Disease-Neuropsychological Assessment Battery [CERAD-NAB]) total score and the maximum C-reactive protein (CRP) concentration. The biomarkers of brain injury neuron-specific enolase and S100β protein, age, and level of education were included in secondary multivariable logistic regression analyses.
Results: Of the 82 patients who completed the study, 38 (46%) presented with POCD 1 week after surgery. In the multivariable regression analysis, higher intraoperative MxA (odds ratio [OR; 95% confidence interval (CI)], 1.39 [1.01-1.90] for an increase of 0.1 units, P = .08 after Bonferroni adjustment), signifying less effective autoregulation, was not associated with higher odds of POCD. The univariable logistic regression model for MxA yielded an association with POCD (OR [95% CI], 1.44 [1.06-1.95], P = .020). Tissue oxygenation index (1.12 [0.41-3.01] for an increase of 10%, P = 1.0 after Bonferroni adjustment) and baseline CERAD-NAB total score (0.80 [0.45-1.42] for an increase of 10 points, P = .45) did not affect the odds of POCD. POCD was associated with elevated CRP on postoperative day 2 (median [interquartile range]; 175 [81-294] vs 112 [62-142] mg/L, P = .033); however, the maximum CRP value (OR [95% CI], 1.35 [0.97-1.87] for a 2-fold increase, P = .07) had no distinct effect on POCD.
Conclusions: Impairment of intraoperative cerebral blood flow autoregulation is not predictive of early POCD in elderly patients, although secondary analyses indicate that an association probably exists.