FP686A STUDY OF THE EFFECT OF HAEMODIALYSIS ON COGNITIVE FUNCTIONS WITH ONE- WEEK FOLLOW UP: A CALL FOR CARE AND CONCERNS

Abstract

Introduction and Aims: Cognitive functions have been shown to decline in patients on long term haemodialysis (HD), with 70% of HD patients aged 55 years and older having moderate to severe chronic cognitive impairment. No evidence is currently available to identify if cognition returns back to baseline beyond 24 hours of a HD session or if physiological factors are associated with cognition during a session. The main aim of this study was to monitor cognitive function over three time periods in order to examine whether cognitive function remains stable between HD sessions. The secondary aim was to measure associations of cognitive performance with physiological parameters during HD. Methods: A prospective observational study was completed in three satellite HD units, involving 100 patients aged over 50 without a known diagnosis of dementia. Cognitive assessments using the Montreal Cognitive Assessment (MOCA), Geriatric Depression Scale (GDS) and Confusion Assessment Method (CAM) were performed up to one hour into a HD session (T1) and in the last hour before the end of the same HD session (T2). They were then repeated, one hour into a HD session, on average seven days later (T3). During each assessment period, the blood pressure, heart rate and oxygen saturation of each participant were obtained. Total MOCA scores were converted to percentages to standardize comparisons and a value of P <0.05 was considered significant. Results: The median time of T1 and T3 was 4 minutes and 19.5 minutes into the start of HD respectively and for T2 was 52.5 minutes before the end of HD. The total median (IQR) for MOCA at T1 was 70 [(60-80), n=100], at T2 was 63 [(56-73), n=100] and T3 was 69 [(60-83, n=78].From T1 to T2, the attention (p=0.02), language (p=0.000), abstract (p=0.005) and delayed recall (p=0.000) domains of MOCA statistically significantly declined, with a statistically significant decline in total MOCA (p<0.001) and an effect size of 0.36. From T2 to T3, the language (p=0.001) and delayed recall (p=0.001) domains of MOCA statistically significantly increased, with a statistically significant increase in total MOCA (p=0.007) and an effect size of 0.2. The difference in MOCA between T1 and T3 was not statistically significant (p=0.925). 57% of participants had scores indicative of mild or severe depression at T1, with 45% at T2 and 53% at T3. There was a statistically significant decline in GDS score from T1 to T2 (p= 0.016).There were no significant associations between cognitive changes and physiological parameters. Conclusions: Cognition significantly declines during a single HD session, in particular the attention, language, abstract and delayed recall domains of cognition and does not correlate to changes in intradialytic physiological parameters. This is important because tasks that require cognition to make important decisions following HD, such as driving, should not be undertaken. However, on average seven days later, cognition returns to just below baseline. In the long term, this repeated pattern may lead to significant cognitive impairment such as dementia. There is a need to repeat this study over a longer period of time, whilst also developing strategies to deal with long term cognitive decline in HD patients.Our study also showed that the incidence of both depression and cognitive impairment is high amongst HD patients and remains under diagnosed. It is important that clinicians carefully screen for both psychiatric disorders to improve the health care of patients on HD.