The Clock Drawing Task reveals executive dysfunction in the dialysis population – an underrecognized and underestimated problem

Abstract

Aim of the study

The study aimed to assess the prevalence of executive function impairment among patients with chronic kidney disease (CKD) undergoing dialysis, with no subjective cognitive problems and with normal global cognition on the Mini-Mental State Examination (MMSE). We also investigated the relationship between cardiovascular risk factors and cognitive test results.

Rationale for the study

Patients with CKD, including those undergoing renal replacement therapy, are at a higher risk of developing cognitive impairment (CI) than the general population. Recent research has shown CI to be a growing problem among CKD patients worldwide. Yet, it remains underdiagnosed, even though it may significantly influence the lives of patients.

Materials and methods

In this cross-sectional, prospective study, 58 dialysis patients with no cognitive decline on the MMSE screening were assessed for executive function impairment using the Executive Clock-Drawing Task (CLOX). Moreover, past medical history, demographic data, and laboratory test results were collected.

Results

The mean patient age was 59.47 ± 14.98 years, and the mean duration of dialysis was 45.93 ± 48.49 months. The prevalence of executive function impairment amounted to 8.6%. Moreover, remarkably similar pattern of clock drawing was observed, with numbers written outside the clock face in the CLOX1 test.

Conclusions

Executive dysfunctions in dialysis patients may manifest itself before the onset of global cognitive impairment. There appear to be a deficit in the spatial domain as well. Better education may play a protective role.

Keywords:

• Chronic kidney disease
• dialysis
• executive impairment
• CLOX
• Executive Clock-Drawing Task

Introduction

Patients with chronic kidney disease (CKD), especially those on renal replacement therapy (RRT), are at a higher risk of developing cognitive impairment (CI) than the general population, regardless of age [1]. There are many risk factors, including traditional cardiovascular risk factors such as atrial fibrillation, atherosclerosis, hypertension, diabetes mellitus, hyperlipidemia, tobacco smoking as well as variables related to kidney disease, i.e., inflammation, anemia, endothelial dysfunction, or uremic toxins accumulation, which might lead to cognitive decline [2]. Age is also a strong predictor of cognitive dysfunction. The prevalence of CI in patients on RRT is higher than in CKD patients and is highest in hemodialysis patients (HD) in comparison to peritoneal dialysis (PD) and kidney transplant recipients [3–5]. On the other hand, the prevalence of CI may be underestimated due to the exclusion of patients with severe disabilities such as vision and hearing problems, language and movement disorders, language barrier, as well as patients who refused to participate in the cognitive assessment [6]. Vascular dementia (VaD) may contribute to CI in RRT patients, thought the underlying causes seem to be complex [7–9]. There is a limited number of papers describing patterns of cognitive dysfunction in patients with CKD [3–5,10–12] and verbal fluency, memory, processing speed and attention have been reported to be most affected [3–5,13,14]. We posited that selective impairment of certain cognitive domains may occur in CKD patients with no apparent general cognitive issues. The present study aimed to investigate whether there is executive domain impairment in dialysis patients with normal global cognition on cognitive screening tests. We also examined the relationship between cardiovascular risk factors and cognitive test results.

Materials and methods

In this cross-sectional study, 58 out of 116 dialysis patients (50%), who consented to the study, were recruited in a single Dialysis Unit between 1 July 2023 and 31 August 2023. Among them, there were 13 (22%) and 45 (78%) patients undergoing PD and HD, respectively. Patients were included in the study if they: (i) were diagnosed with end-stage kidney disease (ESKD), (ii) were on ambulatory PD or HD, (iii) were aged 18 years or older, (iv) were Polish-speakers, (v) and the examination occurred after a consultation with their attending physicians’ team. Only clinically stable patients were enrolled in the study, and the following exclusion criteria were applied: (i) no symptoms of an infectious disease in the previous 8 weeks, (ii) no decompensated heart failure or liver failure, (iii) no history of neurodegenerative disorders or delirium. Furthermore, patients on hypnotic, sedative, and antidepressant medication, were also excluded based on clinical records. Executive and visuospatial functions were assessed with the Executive Clock-Drawing Task (CLOX). Cognitive function in HD patients was assessed on the day of the dialysis treatment within its first hours. Additionally, demographic data (including years of education, age, and gender), and past medical history (i.e., primary kidney disease, duration of dialysis, comorbidities, and laboratory blood test results) were obtained from clinical records. The biochemical parameters were selected on the basis of previously published studies showing their possible association with CI [2,15–17]. The Bioethics Committee of the Medical University of Warsaw approved the study protocol (approval number KB/81/2022) and written informed consent was obtained from all of the participants. The study was conducted in the line with the World Medical Association Declaration of Helsinki.

Cognitive function evaluation

All dialysis patients underwent baseline general cognitive screening using the MMSE test. The MMSE is a 30-item cognitive test battery, which assesses several cognitive domains, including orientation, attention, language, counting, and memory. It is broadly used in the detection of cognitive decline among the elderly, and it takes 7–10 min to complete. The cutoff of 26 points was used as an indication of cognitive impairment [18]. The results were presented either as a raw overall test score or as an adjusted outcome using the Mungas correction for age and education. Applying this correction contributed to higher specificity and sensitivity in the diagnosis of CI [19].

The CLOX test is divided into two parts to examine executive (CLOX1) and visuospatial (CLOX2) functions, respectively. The whole examination lasts less than 5 min and requires only a pen and a piece of paper. CLOX1 is an unprompted task, while CLOX2 is a copied version of the same instruction. First of all, the patient is asked to draw a clock on a blank sheet of paper with all the numbers placed on its face and the clock hands pointing to quarter to two. The instructions can be repeated as needed until the patients fully understand them. Afterwards, the patient’s performance is rated as CLOX1 with the quantitative scoring system of 0 to 15 points, according to the evaluation criteria. Subsequently, the patient watches the researcher drawing the clock for them with the same instruction, placing 12, 3, 6, and 9 first. Then the patient is asked to duplicate the examiner’s clock (CLOX2), which is assessed again against the same criteria [20]. The cutoffs of 10 and 12 points for CLOX1 and CLOX2 respectively, have been shown to be a sensitive and specific value for cognitive impairment [21].

Statistical analysis

The analysis was conducted using IBM SPSS Statistics 29. The Kolmogorov–Smirnov test was used for normal distribution analysis. We used both parametric tests, such as the Student’s t-test (for a normal distribution), and non-parametric tests, such as the Mann–Whitney U test (for a non-normal distribution), to explore variations in different parameters. The chi-square test of independence was used to compare groups. We employed the Spearman rank correlation test to examine correlations between variables. No multiple comparisons were performed in the analyses. A significance level of p < 0.05 was applied.

Results

The study included 58 out of 116 dialysis patients, 37.9% of whom were women. Two patients refused to participate in the study. Additionally, three patients were excluded from the study due to the language barrier, another nine due to physical disabilities, and 44 due to either expressed subjective cognitive concerns or cognitive assessment with the Mini-Mental State Examination (MMSE). The demographic and clinical characteristics of the study group are summarized in the Table 1. The mean patient age was 59.47 ± 14.98 years, and the mean number of the years of education amounted 14.84 ± 3.26. The primary kidney disease was unknown in the majority of patients (29.3%), and the other most common causes of ESKD were polycystic kidney disease (15.5%), diabetic nephropathy (8.6%), and chronic glomerulonephritis (8.6%), respectively (data not shown). Hemodialysis was a more frequent form of dialysis (78%), and the mean dialysis vintage was 45.93 ± 48.49 months. The results of the cognitive screening are presented in the Figure 1. The mean CLOX scores were 12.19 ± 2.07 and 14.60 ± 0.59 points for CLOX1 and CLOX2, respectively. All patients had normal CLOX2 scores (above 12 points), while 8.6% of patients had impaired CLOX1 scores. The cognitive functions were not significantly impacted by creatinine levels and clearance, uremic toxins concentration, parathyroid hormone (PTH), serum phosphate, fasting glucose, serum albumin, and C-reactive protein (CRP) levels. Furthermore, a better CLOX1 test score correlated statistically with the number of years of education (p = 0.028) and with female gender (p = 0.034). The correlation coefficient between cognitive screening scores and years of education (shown in Table 2) revealed a significant relationship between MMSE and CLOX2 scores, Mungas correction of MMSE and both CLOX1 and CLOX2 scores, and a statistical trend association between CLOX1 and CLOX2 scores. Finally, we observed that 13 patients exhibit a remarkably similar pattern of executive dysfunction with numbers written outside the clock face in the CLOX1 test (Figure 2). No association between comorbidities such as hypertension, diabetes mellitus, ischemic heart disease, or stroke, was found.

Figure 1. Qualitative differences in CLOX performance of patient without general cognitive decline.

Figure 2. CLOX1, CLOX2 and pentagon results from a patient’s MMSE with no overall cognitive decline.

Table 1. Demographic characteristics of the study group.

	All patients (n = 58)	CLOX1 ≥ 10* (n = 53)	CLOX1 < 10* (n = 5)	p-Value*
General characteristics
Age, years, mean ± SD;(Q1; Q3)	59.47 ± 14.98 (49; 70.25)	59.30 ± 14.62 (49; 70.5)	61.20 ± 20.32 (43; 78)	0.79^b
Female, n (%)	22 (37.9)	20 (37.7)	2 (40.0)	0.034^c^,**
Years of education, mean ± SD	14.84 ± 3.26	15.13 ± 3.10	11.80 ± 3.77	0.028^a^,**
Dialysis vintage, months ± SD (Q1; Q3)	45.93 ± 48.49 (17.75; 55.75)	45.26 ± 48.92 (18; 50.5)	53.00 ± 48.18 (16.5; 105.5)	0.75^b
Dialysis modality Hemodialysis, n (%) Peritoneal dialysis, n (%)	45 (78) 13 (22)	40 (75.5) 13 (24.5)	5 (100.0) 0 (0.0)	0.21^c0.21^c
Biochemical data (mean ± SD)
Creatinine level (mg/dl)	9.30 ± 2.44	9.30 ± 2.51	9.29 ± 1.55	0.99^a
Creatinine clearance (ml/min/1.73m2) (Q1; Q3)	5.40 ± 1.79 (4; 6)	5.45 ± 1.86 (4; 6)	4.80 ± 0.45 (4.5; 5)	0.44^b
Urea level (mg/dl)	111.22 ± 42.34	112.17 ± 43.27	101.20 ± 32.55	0.58^a
PTH level (pg/ml)	500.95 ± 351.30	519.19 ± 359.29	307.60 ± 171.42	0.20^a
Serum phosphate level (mmol/l) (Q1; Q3)	1.93 ± 0.68 (1.52; 2.21)	1.92 ± 0.71 (1.49; 2.22)	1.98 ± 0.23 (1.77; 2.19)	0.85^b
Glucose level fasting (mg/dl) (Q1; Q3)	113.12 ± 45.79 (83.25; 124)	111.17 ± 45.21 (82.5; 122.5)	133.80 ± 52.17 (83.5; 179.5)	0.30^b
CRP (mg/l) (Q1; Q3)	9.02 ± 14.37 (1.40; 11.45)	9.60 ± 14.89 (1.45; 12.6)	2.88 ± 2.63 (0.80; 5.35)	0.13^b
Serum albumin (g/dl) (Q1; Q3)	3.91 ± 0.42 (3.7; 4.1)	3.91 ± 0.43 (3.7; 4.1)	3.88 ± 0.30 (3.65; 4.15)	0.40^b
Comorbidities, n (%)
KTx before	8 (13.8)	7 (13.2)	1 (20.0)	0.67^c
Smoking	8 (13.8)	6 (11.3)	2 (40.0)	0.14^c
Hypertension	50 (86.2)	46 (86.8)	4 (80.0)	0.67^c
Diabetes mellitus	14 (24.1)	12 (22.6)	2 (40.0)	0.39^c
Ischemic heart disease	18 (31.0)	17 (32.1)	1 (20.0)	0.58^c
Stroke/TIA before	2 (3.4)	2 (3.8)	0 (0.0)	0.66^c

n: number of patients; SD: standard deviation; Q1: Quartile 25.; Q3: Quartile 75.; PTH: parathormone; CRP: C-reactive protein; KTx: kidney transplantation; TIA: transient ischemic attack.

* Compared sub-populations; **p < 0.05; ^aStudent’s t test; ^bU Mann–Whitney test; ^cχ2: Chi-squared.

Table 2. Correlation coefficient between test scores and years of education (Spearman’s rho).

	MMSE	MMSE Mungas correction	CLOX1	CLOX2	Years of education
MMSE		0.63**	0.34**	0.27*	0.09
MMSE Mungas correction	0.63**		0.27*	0.27*	−0.38**
CLOX1	0.34**	0.27*		0.48**	−0.13
CLOX2	0.27*	0.27*	0.48**		0.02
Years of education	0.09	−0.37**	−0.13	0.02

* p < 0.05; **p < 0.1.

Discussion

Summary of main results

In the present study, we showed that 8.6% of dialysis patients, both PD and HD, with no cognitive problems on the cognitive screening test (MMSE) had executive function impairment. Although, the cause of CI in patients with CKD is multifactorial, cerebrovascular disease is probably one of the main contributory factors [22]. It is well known that patients with CKD are more prone to developing multiple lacunes or confluent white matter hyperintensities, brain atrophy, which can lead to VaD [23]. Executive function is a cognitive domain that is typically impaired in VaD, caused by small-vessel disease [24]. Patients with executive dysfunction may have difficulty initiating, organizing, planning, or completing tasks. They may be unable to multitask, have problems with short-term memory or problem solving, and have trouble controlling emotions. In the present study patients with executive dysfunction had problems with the unprompted clock drawing task (CLOX1), whereas the copying tasks, both clock drawing (CLOX2) and MMSE pentagons were performed correctly (Figure 2). These findings suggested that only executive function was impaired, while visuospatial abilities remained unaffected. However, we also observed that some patients with normal overall CLOX1 and CLOX2 scores demonstrated analogous patterns of clock error with numbers written outside the clock face layout. This finding may suggest impairment of the non-dominant right hemisphere involving the right parietal lobe and reflects early spatial and planning deficits [25]. Finally, patients with a higher level of education and women showed statistically significantly better scores on the executive function test (CLOX1), which may suggest that education serves a protective role against cognitive decline. There was no significant impact of cardiovascular disease, diabetes mellitus, and smoking on the occurrence of executive dysfunction, but due to the small group size, these results should be treated with caution.

Comparison between this and other studies

Our results are in line with other studies indicating that not all cognitive domains are impaired to equal level of severity, as executive function was found to be often impacted in patients with CKD undergoing RRT. The study conducted by Sarnak et al. evaluated cognitive function in 314 patients on HD. The authors showed that executive function impairment was prevalent in HD patients despite normal MMSE scores [26]. According to Tamura et al. patients with ESKD demonstrated problems in tests assessing executive function in the first months of RRT, while there were no difficulties in general cognitive skills [27]. Furthermore, a few studies, assessing cognitive abilities in CKD patients, including our previously published study, found verbal fluency to be the most affected domain, which reflects executive dysfunction [3,4,14].

Limitations

There are some limitations to this study. The study sample was relatively small, and non-homogenous (for instance, different dialysis types and causes of renal failure). Moreover, the examination took place on the day of the dialysis treatment. The dialysis unit consisted of several treatment areas, which may lead to distractions during the cognitive assessment.

Conclusions

In dialysis patients, executive dysfunctions may occur before the onset of global cognitive impairment. There appears to be a deficit in the spatial domain as well. Better education may play a protective role.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.