Research on cognitive impairment and potential risk factors in peritoneal dialysis patients: An observational study

Abstract

The objective of this study is to investigate the associated risk factors and their effects on cognitive impairment (CI) in patients undergoing peritoneal dialysis. A retrospective analysis was conducted on the basic information of 268 patients who underwent continuous ambulatory peritoneal dialysis (CAPD) at our hospital from January 2020 to September 2023. Cognitive function was assessed using the Montreal Cognitive Assessment Scale during their subsequent dialysis visits. Participants were categorized into a CI group and a cognitively normal group. Blood and other biological samples were collected for relevant biomarker analysis. Subsequently, we analyzed and compared the factors influencing CI between the 2 groups. The prevalence of CI among CAPD patients was 58.2%. Compared to the cognitively normal group, the CI group had a higher prevalence of alcohol consumption, lower levels of education, and reduced serum uric acid levels (P < .05). There was also a higher incidence of autoimmune diseases such as systemic lupus erythematosus in the CI group (P < .05). In terms of dialysis efficacy, the residual kidney Kt/V and residual kidney Ccr were significantly lower in the CI group compared to the cognitively normal group. In blood parameters, the CI group showed elevated total cholesterol levels and lower serum calcium concentrations (P < .05). Logistic regression analysis identified male gender, older age, lower educational attainment, hypercholesterolemia, and elevated high-sensitivity C-reactive protein levels as independent risk factors for CI in CAPD patients (P < .05). Additionally, in this patient cohort, dialysis duration and residual renal function were protective factors against CI (P < .05). CI is prevalent among PD patients. Elevated high-sensitivity C-reactive protein levels, male gender, older age, lower educational attainment, and hypercholesterolemia constitute an independent risk factor for CI in CAPD patients, whereas residual renal function acts as a protective element.

1. Introduction

Peritoneal dialysis (PD) emerges as a vital renal replacement therapy for individuals with end-stage renal disease (ESRD), functioning as a home-based treatment modality that necessitates proficient self-management and operation by the patient.^[1–3] Cognitive impairment (CI), characterized by deficits in memory, computation, spatial-temporal orientation, structural abilities, executive functions, and language comprehension and expression, poses significant challenges.^[4–6] Patients afflicted with ESRD are often burdened with severe anxiety, depression, and CI, conditions that frequently go underdiagnosed and are not given due consideration, thus impacting the patients’ adherence to medical protocols, their independence, and ultimately, their survival prognosis.^[7–9]

Therefore, the evaluation of cognitive function holds critical importance for PD patients. The mainstay kidney replacement therapies for ESRD currently include hemodialysis, PD, and kidney transplantation, with a wealth of research dedicated to examining the cognitive functions of patients undergoing hemodialysis. For the elderly ESRD population suffering from CI, the potential loss of executive function or memory can lead to inaccuracies in managing PD, thereby increasing the risk of PD-associated peritonitis.^[12] The incidence of CI among ESRD patients spans from 28.9% to 67.6%, with CI being identified as an independent predictor of mortality and survival in those undergoing PD treatment.^[4,13]

In light of these considerations, this study aimed to investigate the incidence of CI in patients receiving continuous ambulatory peritoneal dialysis (CAPD) and to explore nontraditional, vascular-related risk factors for CI in this cohort, thereby laying a theoretical foundation for clinical practice.

2. Materials and methods

2.1. General information

The clinical records of 268 patients who underwent CAPD at our institution from January 2020 to September 2023 were retrospectively analyzed. This cohort comprised 121 males and 147 females, with a mean age of 48.2 ± 9.3 years and a median duration of dialysis of 22 months. Within this patient population, 106 were diagnosed with chronic glomerulonephritis, 97 with diabetic nephropathy, 41 with hypertensive nephropathy, and 24 with polycystic kidney disease. During their subsequent visits to our hospital for dialysis, cognitive function was evaluated using the Montreal Cognitive Assessment Scale (MoCA)^[14] and the Mini-Mental State Examination (MMSE). Based on our findings, the prevalence of CI among PD patients was 58.2% (156 cases) as determined by the MoCA score. According to the MMSE score, the prevalence of CI was 30.2% (81 cases), with all CI diagnoses by the MMSE scale showing concordance with MoCA results. In line with previous research, the MMSE exhibits limited diagnostic sensitivity for cognitive function. Consequently, for the purposes of this comparative study, patients were categorized into the CI group (observation group) and the cognitively normal group (control group) based on MoCA score outcomes. This study received ethical approval from the Ethics Committee of our hospital, endorsed by the Institutional Review Board under the approval number BL20200723.

2.2. Inclusion and exclusion criteria

Inclusion criteria: Individuals aged 18 years and above; engagement in regular dialysis for a period exceeding 3 months; clinical stability with comprehensive evaluation and adequate dialysis; willingness to voluntarily participate in the study. Exclusion criteria: presence of severe physical disorders or significant mental illnesses, or current usage of antipsychotic medication; confirmed diagnosis of systemic or localized infection within 4 weeks prior to enrollment; chronic debilitating diseases such as liver cirrhosis, hematological malignancies, tuberculosis, or other forms of cancer; active systemic lupus erythematosus (SLE), systemic vasculitis, or other autoimmune diseases; history of cerebrovascular accident sequelae; illiteracy or severe sensory impairments (hearing or visual), affecting the ability to write, read, or complete assessment scales.

2.3. Methods

2.3.1. Instruments and methods

All participants received their treatment using lactate-based dialysate and dual-lumen tubing systems provided by Baxter Medical Supplies. For PD therapy, PD2 (with 1.77 mmol/L lactate) and PD4 (with 1.24 mmol/L lactate) dialysate formulations were employed.

2.3.2. Cognitive function assessment

The MoCA^[14] was utilized to evaluate cognitive function, with the assessment being conducted within 30 minutes prior to the dialysis treatment. The MoCA encompasses 7 domains: visuospatial/executive function, naming, memory, attention, language, abstraction, and orientation. Conducted by a trained physician in a private and quiet setting, the evaluation adheres to the MoCA scoring scale’s standard guidelines and is completed in under 15 minutes. MoCA maximum achievable score is 30, with an educational adjustment applied: individuals with <12 years of education receive an additional point to their final score, while no adjustment is made for others. A modified score above 26 denotes normal cognitive function; scores below 26 are categorized as CI.^[14] The MMSE assesses orientation, memory, recall abilities, attention, calculation, and language. Like MoCA, MMSE’s total possible score is 30, with a threshold for CI set at <26 points. The MMSE evaluation is performed by a different trained physician in a similar quiet and private environment before the dialysis session.

2.3.3. Blood routine and biochemical index detection

During their subsequent visits to our hospital for dialysis, venous blood samples were collected after an 8-hour fasting period for routine hematological and biochemical analyses. The biochemical parameters assessed included serum albumin, calcium, sodium, triglycerides, total cholesterol, hypersensitive C-reactive protein (hs-CRP), and parathyroid hormone, utilizing the Beckman Coulter AU5800 Biochemical Analyzer for measurements. Records of total urine output and dialysate were compiled 24 hours prior to the clinic visit. Concentrations of urea nitrogen and creatinine were determined in the blood, urine, and dialysate samples. Subsequently, the total urea clearance index, total creatinine clearance, and residual renal function were calculated based on these measurements.

2.4. Observation indicators

Residual renal function is evaluated by averaging the clearance rates of creatinine and urea from 24-hour urine collections gathered before and after dialysis sessions. This methodology is employed to compute various indices of dialysis adequacy, including the total Kt/V (urea clearance index, Kt/V), peritoneal Kt/V, residual renal Kt/V, total creatinine clearance (Creatinine clearance, Ccr), residual renal Ccr, and peritoneal Ccr.

2.5. Statistical analysis

Statistical analyses were conducted using SPSS software, version 23.0. Prior to data comparison, the Kolmogorov–Smirnov test was applied to assess the normality of the distribution. Measurement data conforming to a normal distribution were presented as mean ± standard deviation and compared between groups using the T test. For continuous data not adhering to a normal distribution, values were displayed as the median and interquartile range, with intergroup comparisons conducted using nonparametric tests. Categorical data were expressed as n (%) and analyzed between groups using the χ² test. Logistic regression was utilized to identify risk factors for CI. A P value of <.05 was deemed indicative of statistical significance.

3. Results

3.1. Comparison of general data of patients in CI group (observation group) and cognitively normal group (control group)

In the observation group, the total cholesterol levels were significantly higher than those in the control group (P < .05). Similarly, the education levels and blood uric acid levels in the observation group were significantly lower compared to the control group (P < .05) (Tables 1 and 2). Among the 268 CAPD patients evaluated, the prevalence of CI was 55.56%, with a median MoCA score of 25 (range: 22–27). Across the spectrum of CI domains, delayed recall was diminished in all 156 cases (100%), linguistic capabilities were reduced in 94 cases (60.26%), and visual-spatial and executive functions significantly declined in 87 cases (55.06%), indicating multidimensional impairment. Following adjustments for educational attainment on the MoCA scale, 57.69% of patients with at least a junior high school education exhibited impairments in 3 or more cognitive domains.

Table 1.

Comparison of general data of patients with CAPD between observation group and control group.

Indicator	Observation group (n = 156)	Control group (n = 112)	t/X 2	P
Male [n(%)]	36 (40.00)	36 (50.00)	0.54	.62
Age (yr)	50 (44,59)	42 (33,51)	−1.655	.51
Dialysis time (mo)	24 (12,36)	19 (10,48)	−0.479	.29
CDK disease course (mo)	55.8 (29.61,87.24)	49.21 (31.43,89.13)	0.131	.242
Body mass index (kg/m2)	22.9 (21.0,25.3)	22.7 (20.7,24.7)	−0.661	.73
Educational level [n (%)]
Junior high school and below	72	21	15.434	<.001
High school	15	27
Graduate and above	3	24
Primary disease [n (%)]
Glomerulonephritis	71 (45.5)	51 (45.5)	0.104	.991
Diabetic nephropathy	34 (21.8)	24 (21.4)
Hypertensive nephropathy	32 (20.5)	22 (19.6)
Others	19 (12.2)	15 (13.4)
Previous history
Hypertension	125 (80.1)	86 (76.8)	0.435	.510
Diabetes	13 (8.3)	7 (6.3)	0.410	.522
Anaemia	133 (85.3)	97 (86.6)	0.098	.755
Smoking	36 (23.1)	29 (25.9)	0.281	.600
Drinking	51 (32.7)	22 (19.6)	5.601	.018
Rheumatoid arthritis	21 (13.5)	9 (8.0)	1.931	.165
Systemic lupus erythematosus	19 (12.2)	3 (2.7)	7.810	.005
Dialysis evaluation index (wkly) M (P25, P75)
Peritoneum Kt/V	1.53 (1.15,1.77)	1.58 (1.02,1.73)	0.366	.233
Residual kidney Kt/V	0.08 (0.00,0.43)	0.31 (0.00,0.51)	0.239	.011
Total Kt/V	1.82 (1.36,2.13)	1.84 (1.34,2.31)	1.032	.753
Peritoneum Ccr	38.78 (31.49,45.93)	35.41 (30.82,41.94)	1.711	.155
Residual kidney Ccr	0.02 (0.00,16.47)	2.33 (0.00,20.00)	−0.136	.024
Total Ccr	46.73 (40.13,55.14)	51.33 (43.37,64.39)	1.941	.069
MoCA	20.56 ± 2.76	27.34 ± 1.08	−24.676	<.001
MMSE	25.22 ± 3.33	27.87 ± 1.62	−7.78	<.001

Table 2.

Comparison of physical indicators of patients with CAPD between observation group and control group.

Indicator	Observation group (n = 156)	Control group (n = 112)	t/χ2	P
Hemoglobin (g/L)	115.1 ± 17.83	113.38 ± 18.29	−0.349	.215
Parathyroid hormone (pg/mL)	389.4 (152.9622.0)	352.1 (259.868.6)	−0.47	.513
Serum calcium (mmol/L)	2.27 ± 0.19	2.33 ± 0.21	−2.44	.015
Serum sodium (mmol/L)	138.13 ± 3.13	137.36 ± 7.61	1.14	.256
Serum albumin (g/L)	36.70 ± 4.69	36.03 ± 4.16	−0.55	.718
High-sensitivity C-reactive protein (mg/L)	2.90 (0.89,5.56)	1.34 (0.57,3.99)	−0.142	.293
Uric acid (µmol/L)	357.77 ± 64.02	391.5 ± 56.96	2.024	.016
Triglyceride (mmol/L)	1.92 ± 1.06	1.86 ± 1.18	−0.187	.195
Total cholesterol (mmol/L)	4.90 ± 1.13	4.32 ± 0.76	−2.221	.029
Urine volume (mL)	600 (300,950)	600 (100,1100)	−0.227	.817
Wkly Kt/V up to standard [n(%)]	54 (60.00)	42 (58.30)	0.046	.83

3.2. Analysis of risk factors for cognitive function

PD patients were categorized into an observation group (n = 156) and a control group (n = 112), with statistically significant differences observed between the 2 groups in terms of education level, CRP (C-reactive protein), total cholesterol, and residual kidney function (P < .05). No significant differences were found in sex, age, hemoglobin levels, parathyroid hormone levels, serum albumin, uric acid, and weekly Kt/V (P > .05). Regarding residual renal function, a higher Residual kidney Ccr/Kt/V ratio significantly reduces the incidence of CI. Surprisingly, SLE was found to affect cognitive function in dialysis patients (Table 3).

Table 3.

Univariate analysis of the relevant influencing factors of CI.

Independent variable	OR (95% confidence interval)	P
Gender (male vs female)	1.3 (0.233–5.532)	.285
Age (middle-aged vs young)	2.164 (0.343–7.136)	.013
Dialysis time (mo)	0.839 (0.781–1.116)	.328
Educational level (higher middle school vs lower middle school)	0.113 (0.021–0.381)	.016
Hemoglobin (g/L)	1.013 (0.885–1.029)	.667
Parathyroid hormone (pg/mL)	0.989 (0.988–1.000)	.872
Serum albumin (g/L)	1.025 (0.814–1.174)	.421
CRP (increased vs normal)	4.53 (1.352–13.912)	.003
Total cholesterol (mmol/L)	1.854 (1.113–3.429)	.017
Uric acid (mmol/L)	1.303 (1.131–1.879)	.133
Residual kidney Kt/V	0.681 (0.306–1.548)	.021
Residual kidney Ccr	0.786 (0.147–1.691)	.001
Systemic lupus erythematosus	1.012 (0.983–1.772)	.042

Multivariate logistic regression analysis was performed on variables that exhibited differences in the univariate analysis. The multivariate logistic regression results indicated that advanced age, lower educational attainment, hypercholesterolemia, and elevated levels of hypersensitive C-reactive protein (hs-CRP) are independent risk factors for CI in CAPD patients. Conversely, residual renal function emerged as a protective factor against cognitive dysfunction, as detailed in Table 4. In this study, Residual kidney Ccr/Kt/V is used as a measure of residual renal function, and due to the strong correlation between these 2 indicators, only one of them was included in the multivariate regression analysis. It was listed in the table as the Residual kidney function item.

Table 4.

Multivariate analysis of the relevant influencing factors of CI.

Independent variable	OR (95% confidence interval)	P
Gender (male vs female)	1.070 (0.205–5.582)	.912
Age (middle-aged vs young)	5.831 (1.942–36.077)	.013
Dialysis time (mo)	0.970 (0.940–1.002)	.109
Educational level (higher middle school vs lower middle school)	0.100 (0.026–0.387)	.002
CRP (increased vs normal)	4.671 (1.181–18.463)	<.001
Total cholesterol (mmol/L)	2.009 (1.043–3.869)	.008
Residual kidney function (yes vs no))	0.084 (0.010–0.733)	.017
Systemic lupus erythematosus	1.010 (0.971–1.673)	.052

4. Discussion

The findings of this research reveal that the prevalence of CI among CAPD patients is 58.2%, aligning with the outcomes of other domestic studies.^[15] Identified independent risk factors for CI in CAPD patients include advanced age, a lower level of education, hypercholesterolemia, and elevated hypersensitive C-reactive protein (hs-CRP) levels, while preserved residual renal function acts as a protective factor against CI.

This study observed lower uric acid levels in the observation group compared to the control group. Previous research indicates that blood uric acid may have neuroprotective effects, with lower levels associated with cognitive decline and depression. This protective mechanism is attributed to uric acid’s antioxidant properties, which can neutralize oxygen free radicals, thereby mitigating oxidative stress and potentially delaying neuropathology, reducing CI incidence.^[16] Conversely, some studies have noted higher uric acid levels in patients with mild CI related to cerebrovascular diseases than in healthy controls, with serum uric acid inversely correlated with MoCA total scores, executive function, spatial discrimination ability, and general cognition.^[17]

Significant declines in delayed recall, visuospatial and executive abilities, and language skills were observed among CAPD patients across various CI domains. After adjusting for education level, it was found that 40.7% of participants with at least junior high school education exhibited impairments in 3 or more cognitive domains. Palmer et al^[18] reported that 79% of hemodialysis patients had deficits in one or more cognitive aspects, including learning and memory, complex attention, executive function, language, and perceptual motor skills; 42% showed impairments in 3 to 5 areas. Further research^[19] suggests that cognitive decline in PD patients predominantly manifests as impairments in memory, attention, and executive function.

In the dialysis population, 30% to 50% of patients exhibit a state of microinflammation. hs-CRP, a sensitive marker of inflammation, is expressed at levels 8 to 10 times higher in the dialysis population than in the general public.^[20] Duan et al^[21] identified elevated hs-CRP as an independent risk factor for overall cognitive decline. Additional studies^[22–24] have linked plasma CRP levels with CI in older adults, highlighting the relationship between inflammation and cognitive dysfunction. Our study found that hs-CRP levels were higher in the control group than in the observation group, though the difference was not statistically significant. Dividing hs-CRP levels into normal and elevated groups, binary multivariate logistic regression analysis confirmed elevated hs-CRP as an independent risk factor for CI in CAPD patients. This may be due to pro-inflammatory factors stimulating microglia cell proliferation and activation in the central nervous system, leading to neuronal degeneration and necrosis via neurotoxic substance production, culminating in CI.

Residual renal function aids CAPD patients in maintaining stable volume status, reducing inflammatory responses, and clearing molecular substances, thus enhancing quality of life. Several studies^[25–27] suggest that uremic toxin accumulation can trigger oxidative stress, chronic inflammation, and blood-brain barrier abnormalities, leading to cerebrovascular and neurological degeneration and cognitive decline. Residual renal function efficiently eliminates plasma macromolecular toxins, thereby delaying neuropathy. Consistent with this study’s findings, Tian et al^[28–30] indicated that PD patients with better residual renal function demonstrated superior cognitive performance, underscoring residual renal function as a critical protective factor for cognitive health.

Previous studies have extensively investigated the risk factors associated with CI in dialysis patients, identifying age and educational level as key cognitive risk factors in this population, findings largely consistent with those of our study.^[31,32] This consistency suggests that our sample is representative. Additionally, we often observe autoimmune diseases in clinical settings among dialysis patients, which correlate with CI, thereby prompting us to include autoimmune disease factors in our analysis based on previous research. Interestingly, the history of SLE was only significant in univariate analyses, but not in multivariate analysis, potentially due to the effects of C-reactive protein (CRP) or total cholesterol, or possibly due to the limitations of our single-center sample. However, the role of autoimmune diseases in CI among dialysis patients is emerging as a potentially interesting area for future research.

However, this study is a single-center retrospective study with a small sample size. Missing data may introduce biases that compromise the representativeness of the sample in specific subgroups, potentially diminishing the statistical power to detect subtle changes in certain indicators. These issues can be addressed by expanding the scope of the study, which is a goal for our future research, including the conduct of multicenter studies.

5. Conclusions

The results of this study showed that the incidence of CI was consistent with domestic and foreign studies. However, early CI is easily ignored by clinicians. The results of this study suggest that more attention should be paid to patients with advanced age, low level of education, hypercholesterolemia elevated hs-CRP, and loss of residual renal function, and early screening of CI in the population to identify intervention groups. Meanwhile, paying attention to the prevention and treatment of chronic inflammatory response in CAPD patients and protecting residual renal function may be effective measures to delay the progression of CI and improve patients’ treatment compliance.

Author contributions

Conceptualization: Chunxia Shi, Feng Shao, Yanan Shi, Zhongxin Li.

Data curation: Chunxia Shi, Xiaoqi Wang, Conghui Liu.

Funding acquisition: Chunxia Shi, Xiaoqi Wang, Zhongxin Li.

Investigation: Chunxia Shi, Shujing Jia, Conghui Liu, Feng Shao.

Project administration: Chunxia Shi, Shujing Jia, Conghui Liu, Feng Shao, Zhongxin Li.

Resources: Chunxia Shi, Shujing Jia, Xiaoqi Wang, Yanan Shi, Zhongxin Li.

Software: Chunxia Shi, Shujing Jia, Conghui Liu, Feng Shao.

Validation: Chunxia Shi.

Writing – original draft: Chunxia Shi.

Formal analysis: Shujing Jia, Feng Shao, Yanan Shi.

Supervision: Shujing Jia, Xiaoqi Wang, Conghui Liu, Feng Shao, Zhongxin Li.

Visualization: Shujing Jia, Yanan Shi.

Methodology: Xiaoqi Wang, Yanan Shi, Zhongxin Li.

Writing – review & editing: Zhongxin Li.