Hearts and minds : an assessment of cognitive impairment after cardiac surgery

Abstract

Impaired cognition following cardiac surgery is a common complication. Estimates of the incidence of post-operative cognitive decline/dysfunction (POCD) vary widely, ranging from 20-70% of patients within the first week after cardiac surgery and declining to 10-40% by 6 weeks. While there is a need to better identify the patient risk factors and surgical techniques that contribute to POCD, it is difficult to reach meaningful conclusions from the present data, due to a lack of concordance in experimental design and data analysis. The overall aim of this thesis was to investigate the incidence ofPOCD following various forms of cardiac surgery, using a study design that addressed methodological issues which have limited other studies in the field: i) additional control groups were included; ii) a mood state measure for depression, anxiety and stress was included; iii) a more robust statistical definition of POCD was used to analyse the data; iv) a battery of cognitive tests, including a novel computer-based screening tool of cognition, was used; and v) pre-operative assessment was performed for all neuropsychological measures. Three separate studies were conducted. The first study (Chapter 2) examined community dwelling elders using a novel computer-based task, the Subtle Cognitive Impairment Test (SCIT), and 'gold-standard' neuropsychological tests, with the addition of an emotional status measure assessing depression, anxiety and stress. The second and third studies (Chapter 3 and 4) utilized the revised methodology to assess POCD after coronary artery bypass graft surgery (Chapter 3), and compared POCD after conventional valve surgery and robotically-assisted valve surgery (Chapter 4). Chapter 2 revealed that the SCIT is useful as a brief screening tool of global cognition in the elderly. The SCIT correlated with every gold standard test used, distinguished between good and poor performance on these other cognitive tests, and was associated with performance in multiple cognitive domains. Factor analysis revealed that the SCIT is sensitive to an underlying factor that may reflect signal processing speed (response time) and reduced efficacy of signal processing and decision-making (error rate). Chapter 3 and 4 demonstrated that, regardless of the type of surgery, for most patients POCD is temporary or does not occur at all. This finding differed from the conclusions of most papers in the field, which have generally reported that POCD is permanent in a substantial proportion of patients. This finding indicates that research design contributes significantly to the apparent cognitive outcomes following cardiac surgery and that greater consistency may be achieved if studies utilise appropriate control groups. It was also evident that individual patient performance varies considerably, and that results from a small proportion of discrepant patients can strongly influence the overall results at the group level. Demonstration of this heterogeneity was important because it illustrates that some individuals are more vulnerable than others. In the future it might be possible to use this type of analysis to identify cardiac patients who are more likely to experience greater degrees ofPOCD, so that the factors involved can be managed and the risk reduced.