Cognitive dysfunction after cardiac surgery

Overall Aim: The overall aim of the current research was to evaluate the methodological efficiency and effectiveness of current neuropsychological research in the cardiac surgery context. Background: Conducting a reliable assessment of cognitive impairment after cardiac surgery is crucial, particularly when treatment selection to remediate deficits depends on the conclusion drawn from the assessment. Studies examining possible neuropsychological decline following cardiac surgery have produced a variety of outcomes. Possible reasons for this include methodological differences across studies and inadequate or inconsistent definitions of what constitutes a significant change in functioning. The 1995 consensus statement on the assessment of post cardiac surgery neurobehavioural deficits attempted to rectify the methodological differences by proposing guidelines for conducting research and a core test battery for inclusion in all research within the field (Murkin, Newman, Stump, & Blumenthal, 1995). The current research consists of a series of studies designed to examine post consensus research in the field and to further elucidate methodological problems in current research strategies. Method and Results: The current effort commenced with a review of outcomes published in the literature since the 1995 consensus statement's release. That review sought to examine current research methodology in the field and to highlight how various methodological practices influence conclusions drawn from the research. The review concluded that a large range of assessment and decision methodologies currently exist in the literature and that, as a result, consistency in outcomes and conclusions has not been obtained. The review was followed by a series of four studies designed to further demonstrate the methodological inadequacies in the field. Study One consisted of a replication, within cardiac surgery related publications, of the power and effect size research conducted by Bezeau and Greaves (2001). Aim: The aim of that analysis was to establish the overall effect size and power of cardiac surgery clinical neuropsychological research. Analysis'. The analytical method described by Zakzanis (2001) was used for between group comparisons. To avoid potential negative impact by higher correlations between pre and postoperative scores, the method used to derive power and effect size for within group studies was that described by Dunlap, Cortina, Vaslow, and Burke (1996). Results: The current analyses indicate that the statistical power and population effect sizes are underreported in cardiac surgery research publications. Analyses also revealed that very small effect sizes exist for both independent groups and repeated measures designs, and that the power of research findings was on average poor. Conclusion: It was concluded from those results that neuropsychological research into the effects of cardiac surgery requires substantial improvement to ensure that conclusions being drawn from investigations are reliably robust. Study Two was an analysis of the structure of four test batteries used in previously published research studies for which raw data was made accessible by the studies' authors. Aim: The aim of this study was to examine the structure of those test batteries in order to establish if the originally hypothesised design was correct. Analysis: Facet theory was used to analyse the intercorrelations between test scores across each individual test battery. Results: The results of the facet analyses indicate that none of the batteries consistently supported their proposed methodological structures. Conclusion: It is concluded from those results that test batteries must be carefully considered and validated on the specific cohort before being used in research or recommended for use in clinical contexts. Study Three involved an analysis of the effect sizes for the core test battery proposed in the 1995 consensus statement. Suitable published studies that were included in study two were used as the data source for this analysis. Aim: The aim of this study was two fold. Firstly, it sought to identify cohort specific effect sizes for each core test. Secondly, it sought to provide estimates of effect sizes that could be used by future researchers to optimise their research strategies, to ensure suitably powerful results are achieved. Analysis: The analytical methods used were the same as those used in study one. Results: The analyses indicate that in recent studies using the core neuropsychological test battery tests, the population effect sizes detected ranged up to -.7, which equates to distribution overlaps of at least 57%. Additionally, it was found that the effects being detected differed across assessment intervals. Conclusion: In accordance with the aim of the study, the results identified population effect sizes currently in operation in the cardiac research field for the core test battery tests. The estimates established provide valuable information for researchers planning research endeavours in the field using the core neuropsychological test battery. Study Four was a preliminary investigation of the structural integrity of the consensus statement core test battery on a small sample of cardiac surgery patients. Aim: The aim of this study was to examine the structure of the core test battery to clarify the relationship between tests and to identify whether the structure of the battery proposed by the consensus group is replicable. Analysis: Facet Theory was used to examine the relationships (intercorrelations) between tests in the core test battery. Results: The results of the facet analysis indicate that the relationships between tests does not conform to the structure proposed by the consensus group, and that the structure of the battery varied across assessment intervals. Additionally, when both assessment intervals were considered contemporaneously, only two general facets of cognitive functioning could be consistently observed. Conclusion-. The conclusion drawn from those results is that the core test battery requires careful consideration and validation to ensure that the resulting conclusions are appropriate to what is actually being examined. Overall Conclusions: Reviewing the relevant literature reveals that deleterious effects due to cardiac surgery are continuously present in the cognitive domains of attention, psychomotor speed, memory, visuoperception, executive functioning, and general cognitive functioning. However, current investigations demonstrated that research methodologies in the field are insufficient for the purposes for which they are being used. While that finding does not altogether exclude the previous conclusions drawn in the literature, it does indicate that substantial methodological improvement is required to ensure the robustness of the conclusions being drawn. In an attempt to assist in the improvement of research methods in the field, statistical power and population effect sizes in research on cardiac surgery samples were identified. By identifying those values, foundation knowledge has been provided to assist future researchers to enhance their research design process.