Edited by: Haitham Jahrami, Arabian Gulf University, Bahrain
Reviewed by: Yaping Liu, The Chinese University of Hong Kong, China; Khaled Trabelsi, University of Sfax, Tunisia
This article was submitted to Sleep Disorders, a section of the journal Frontiers in Neurology
†These authors share first authorship
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
This study aimed to explore the risk factors and develop a prediction model of sleep disturbance in maintenance hemodialysis (MHD) patients.
In this study, 193 MHD patients were enrolled and sleep quality was assessed by Pittsburgh Sleep Quality Index. Binary logistic regression analysis was used to explore the risk factors for sleep disturbance in MHD patients, including demographic, clinical and laboratory parameters, and that a prediction model was developed on the basis of risk factors by two-way stepwise regression. The final prediction model is displayed by nomogram and verified internally by bootstrap resampling procedure.
The prevalence of sleep disturbance and severe sleep disturbance in MHD patients was 63.73 and 26.42%, respectively. Independent risk factors for sleep disturbance in MHD patients included higher 0.1*age (OR = 1.476, 95%
Older age, lower albumin and calcium levels are higher risk factors of sleep disturbance in MHD, and the prediction model for the assessment of sleep disturbance in MHD patients has excellent discrimination and calibration.
Chronic kidney disease (CKD) is a global public health problem with increasing incidence and prevalence (
Sleep occupies approximately one-third of human life span and is crucial for body health (
Currently, a series of studies have collected the demographic parameters and laboratory indexes of MHD patients to analyze the factors of sleep disturbance, and obtained different results. A study has found that gender, age, education level, diabetes history, blood phosphorus level and depression are associated with the sleep quality of MHD patients (
In addition, few studies on factors influencing severity of sleep disturbance in MHD patients have been reported. However, a previous study has found that the severity of sleep disturbance in hemodialysis patients is positively correlated with the severity of depression and negatively correlated with the quality of life (
The purpose of this study was to explore the risk factors and to construct a risk prediction model of sleep disturbance in MHD patients, in order to provide strategies for the prevention and treatment of sleep disturbance in MHD patients.
This was a single-center and cross-sectional study involving 193 MHD patients who underwent hemodialysis from April 2020 to March 2021 at the Blood Purification Center of the Department of Nephrology, The Third Affiliated Hospital of Soochow University. This study was approved by the Ethics Committee of the Third Affiliated Hospital of Soochow University and registered in the Chinese Clinical Trial Register (clinical trial number: ChiCTR2100042093).
The inclusion criteria were (1) age ≥ 18 years; (2) regular hemodialysis therapy >3 months; (3) ability to complete questionnaires on sleep disturbance, anxiety and depression; (4) signed an informed consent form. The exclusion criteria were (1) the history of sleep disturbance before CKD; (2) the history of dementia, anxiety, depression, Alzheimer's disease or schizophrenia before CKD; (3) A history of trauma, surgery or infection within the past 3 months; (4) complicated with malignant tumor; (5) declined to participate.
All MHD patients enrolled in the study received blood purification 3 times per week (hemodialysis once weekly plus hemodiafiltration twice weekly or hemodialysis twice weekly plus hemodiafiltration once weekly). The low-flux polysulfone membrane dialyzer (B. Braun Diacap LOPS15, Germany) was used for hemodialysis and the high-throughput polysulfone membrane dialyzer (B. Braun Diacap HIPS15, Germany) was used for hemodiafiltration. Each blood purification treatment was ~4 h and low molecular weight heparin was used for anticoagulation. The dialysate was bicarbonate with a flow rate of 500 ml/min, and the average blood flow velocity was 200–280 ml/min. Displacement volume using post-replacement was calculated by ~30% of the ultrafiltration flow rate.
Patients were informed of the study's objectives and instructed to complete questionnaires. The data collection instruments were structured questionnaires, the Hospital Anxiety and Depression Scale (HADS), and the Pittsburgh Sleep Quality Index (PSQI).
Demographic and clinical data were collected from face-to-face interviews conducted by trained nurses using structured questionnaires when MHD patients were awaiting hemodialysis treatment. The following information was recorded: age, duration of dialysis, gender, body mass index (BMI), primary diseases, smoking and alcohol consumption, marital status, educational level, history of hypertension, history of diabetes mellitus, and history of cardiovascular and cerebrovascular diseases. Venous blood samples were collected before the hemodialysis. The following laboratory parameters were measured: hemoglobin (Hb), white blood cell (WBC), red blood cell (RBC), hematocrit (Hct), platelet (PLT), C-reactive protein (CRP), triglyceride (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), albumin (Alb), creatinine (Cr), blood urea nitrogen (BUN), serum phosphorus (P), calcium (Ca2+) and PTH.
The HADS includes 14 items assessing depression and anxiety in general hospital patients. It was divided into depression subscale (HADS-D) and anxiety subscale (HADS-A). The two subscales each contain 7 items, with scores ranging from 0 to 3 points (
The PSQI was used to assess the sleep quality of the subjects over the past month (
In the analysis of factors influencing, SPSS 25.0 (IBM Corp., Armonk, NY, USA) was used for statistical analyses and GraphPad Prism 8 (GraphPad Software Inc., San Diego, CA, USA) was used for plotting. The normality of data distribution was assessed using the Shapiro-Wilk test. Continuous variables with normal distribution were shown as mean ± standard deviation, and data with a skewed distribution were shown as medians (25–75% interquartile ranges); categorical variables were expressed as frequency (percentage). The differences between the two groups were evaluated by the Student's
The prediction model was developed using R software (version 4.1.2). The initial prediction model was established by using the variables with
Of the 236 MHD patients were enrolled, 43 were excluded from this study: 11 patients with hemodialysis therapy <3 months, 13 patients failed to complete PSQI and HADS, 6 patients with a history of sleep disturbance, 3 patients with a history of dementia, 2 patients with a history of depression, 1 patient had malignant tumor, 5 patients declined to participate, and 2 patients withdrew from the study (
Clinical characteristics of MHD patients with or without sleep disturbance.
Age (years) | 53.09 ± 11.68 | 54.80 ± 11.50 | 50.09 ± 11.46 | 2.744 | 0.007a |
Male [ |
127 (65.8) | 77 (62.6) | 50 (71.4) | 1.545 | 0.214c |
BMI (kg/m2) | 21.93 ± 3.26 | 21.98 ± 3.13 | 21.86 ± 3.51 | 0.242 | 0.809a |
Primary diseases | 2.984 | 0.394c | |||
Chronic glomerulo-nephritis [ |
40 (20.7) | 24 (19.5) | 16 (22.9) | ||
Diabetic nephropathy [ |
26 (13.5) | 18 (14.6) | 8 (11.4) | ||
Hypertensive nephropathy [ |
39 (20.2) | 21 (17.1) | 18 (25.7) | ||
Others [ |
88 (45.6) | 60 (48.8) | 28 (40.0) | ||
Duration of dialysis | −1.161 | 0.246b | |||
<3 years [ |
40 (20.7) | 22 (17.9) | 18 (25.7) | ||
3–5 years [ |
20 (10.4) | 13 (10.6) | 7 (10.0) | ||
>5 years [ |
133 (68.9) | 88 (71.5) | 45 (64.3) | ||
Smoking [ |
52 (26.9) | 33 (26.8) | 19 (27.1) | 0.002 | 0.962c |
Drinking [ |
19 (9.8) | 9 (7.3) | 10 (14.3) | 2.441 | 0.118c |
Married [ |
188 (97.4) | 120 (97.6) | 68 (97.1) | 0.031 | 0.860c |
High school or above [ |
103 (53.4) | 68 (55.3) | 35 (50.0) | 0.501 | 0.479c |
Hypertension [ |
163 (84.5) | 104 (84.6) | 59 (84.3) | 0.002 | 0.961c |
Diabetes mellitus [ |
35 (18.1) | 25 (20.3) | 10 (14.3) | 1.096 | 0.295c |
Cardiovascular and cerebrovascular diseases [ |
33 (17.1) | 24 (19.5) | 9 (12.9) | 1.394 | 0.238c |
Hb (g/L) | 108.73 ± 18.47 | 108.75 ± 19.36 | 108.69 ± 16.92 | 0.022 | 0.982a |
WBC (109/L) | 5.75 (4.85–6.92) | 5.86 (4.68–6.88) | 5.63 (5.04–7.01) | −0.176 | 0.861b |
RBC (1012/L) | 3.68 ± 0.66 | 3.71 ± 0.69 | 3.63 ± 0.62 | 0.762 | 0.447a |
Hct (L/L) | 0.33 ± 0.06 | 0.33 ± 0.06 | 0.33 ± 0.05 | 0.158 | 0.875a |
PLT (109/L) | 177.86 ± 58.66 | 175.54 ± 62.01 | 181.93 ± 52.44 | −0.726 | 0.469a |
TG (mmol/L) | 1.60 (1.17–2.50) | 1.62 (1.25–2.40) | 1.60 (1.11–2.84) | −0.405 | 0.686b |
TC (mmol/L) | 4.11 ± 0.97 | 4.05 ± 0.97 | 4.21 ± 0.96 | −1.098 | 0.274a |
HDL-C (mmol/L) | 0.90 (0.76–1.15) | 0.87 (0.74–1.11) | 0.98 (0.80–1.16) | −1.540 | 0.124b |
LDL-C (mmol/L) | 2.30 ± 0.74 | 2.25 ± 0.73 | 2.38 ± 0.76 | −1.168 | 0.244a |
Alb (g/L) | 38.30 (36.30–40.15) | 38.00 (36.20–39.60) | 39.55 (36.78–41.43) | −3.303 | 0.001b |
P (mmol/L) | 1.97 (1.61–2.34) | 2.11 (1.62–2.43) | 1.89 (1.59–2.26) | −1.576 | 0.115b |
Ca2+ (mmol/L) | 2.28 ± 0.18 | 2.25 ± 0.18 | 2.34 ± 01.7 | −3.497 | 0.001a |
CRP (mg/L) | 2.70 (2.20–3.40) | 2.80 (2.20–3.60) | 2.60 (2.10–3.10) | −1.628 | 0.103b |
PTH (ng/L) | 370.60 (197.35–609.25) | 388.60 (199.50–614.20) | 341.25 (185.08–591.45) | −0.525 | 0.599b |
Cr (umol/L) | 896.08 ± 186.95 | 891.84 ± 188.19 | 903.52 ± 185.87 | −0.416 | 0.678a |
BUN (mmol/L) | 22.08 (18.11–26.30) | 22.61 (18.10–26.30) | 22.02 (18.51–25.43) | −0.549 | 0.583b |
Anxiety n (%) | 30 (15.5) | 25 (20.3) | 5 (7.1) | 5.905 | 0.015c |
Depression n (%) | 34 (17.6) | 27 (22.0) | 7 (10.0) | 4.390 | 0.036c |
PSQI score of MHD patients.
Subjective evaluation of sleep quality (points) | 1.19 ± 0.82 | 1.55 ± 0.75 | 0.56 ± 0.50 |
>1 point | 49 (25.39) | ||
Sleep latency (points) | 1.58 ± 1.09 | 2.07 ± 0.91 | 0.71 ± 0.82 |
>1 point | 110 (56.99) | ||
Total sleep duration (points) | 1.17 ± 1.07 | 1.59 ± 1.01 | 0.44 ± 0.75 |
>1 point | 82 (42.49) | ||
Habitual sleep efficiency (points) | 0.98 ± 1.15 | 1.45 ± 1.18 | 0.17 ± 0.42 |
>1 point | 56 (29.02) | ||
Sleep disturbance (points) | 1.06 ± 0.54 | 1.23 ± 0.49 | 0.77 ± 0.49 |
>1 point | 32 (16.58) | ||
Use of sleep drugs (points) | 0.67 ± 1.19 | 1.03 ± 1.34 | 0.04 ± 0.36 |
>1 point | 45 (23.32) | ||
Daytime dysfunction (points) | 1.33 ± 0.97 | 1.71 ± 0.93 | 0.66 ± 0.63 |
>1 point | 75 (38.86) | ||
Total PSQI score (points) | 7.98 ± 4.74 | 10.62 ± 3.86 | 3.36 ± 1.44 |
>5 points | 123 (63.73) | ||
>10 points | 51 (26.42) |
The multivariate binary logistic regression analysis between sleep disturbance of MHD patients and clinical characteristics (
The multivariate binary logistic regression analysis between severe sleep disturbance of MHD patients and clinical characteristics (
Age, duration of dialysis, CRP, TG, HDL-C, Alb, Ca2+, PTH, anxiety and depression were screened out to establish initial predictive model based on univariate binary logistic regression analyses between sleep disturbance or severe sleep disturbance and clinical characteristics, combined with variables considered to be meaningful in previous studies. Furthermore, age, CRP, Alb, Ca2+, PTH and anxiety were screened by two-way stepwise regression. A collinearity test showed that no collinearity existed among these variables. Multivariate binary logistic regression analysis was conducted with the selected age, CRP, Alb, Ca2+, PTH and anxiety as variables (
Binary logistic regression analysis between sleep disturbance of MHD patients and predictors. AIC, Akaike information criterion; Alb, albumin; BIC, Bayesian information criterion; Ca2+, calcium; CRP, C-reactive protein; PTH, parathyroid hormone; Ref., reference.
Nomograms predict the risk of sleep disturbance in patients with MHD. Alb, albumin; Ca2+, calcium; CRP, C-reactive protein; PTH, parathyroid hormone.
Taking the original data of the prediction model as the data set, the ROC curve for predicting sleep disturbance in MHD patients was plotted, and the area under the curve was 0.764 (
Internal validation of the prediction model.
According to statistical analyses, CKD patients account for ~9.1% of the global population, and the prevalence of ESKD is as high as 550 people per million (
The incidence of sleep disturbance in 193 MHD patients included in this study was 63.73%, which is similar to previous study results (
The association between poor sleep quality and old age has long been recognized among the general population and has also been shown in MHD patients (
Alb in MHD patients with sleep disturbance was significantly lower than that in without sleep disturbance, and the incidence of severe sleep disturbance was significantly reduced with the increase of Alb. Ling et al. (
The relationship between blood lipid and sleep quality in MHD patients is still unclear. A clinical study suggested that increased TG is associated with increased risk of sleep disturbance in MHD patients (
Depression, anxiety and sleep disturbance are usually associated with a lower quality of life in MHD patients. Studies have found that depression is widespread in MHD patients and is a risk factor for sleep disturbance in MHD patients (
Although a series of previous studies have focused on prediction models related to sleep quality, no study has discussed the risk prediction models for sleep disturbance in MHD patients. Li et al. (
In the present study, age, CRP, Alb, Ca2+, PTH and anxiety were used as predictors to develop a prediction model for sleep disturbance in MHD patients. As previously mentioned, age, Alb, Ca2+ and anxiety play important roles in MHD patients with sleep disturbance. The relationship between CRP and sleep quality has been demonstrated in multiple studies, with higher CRP levels indicating poorer sleep quality (
There are some limitations in our study. First, this study had a single-center and cross-sectional design, which was not representative of the population and did not allow confirming causal relationship between sleep disturbance and risk factors. Further multicenter prospective cohort studies are needed to clarify the clinical value of risk factors in sleep disturbance in MHD patients. Second, our study developed a risk prediction model for sleep disturbance in MHD patients, providing a clinical guidance tool for improving sleep quality. However, this model has not been verified externally. In the future, external validation is needed to further improve the prediction efficiency of the model. Third, fewer patients with depression and anxiety were included in this study, which failed to adequately describe the relationship between depression, anxiety and sleep disturbance in MHD patients. Future studies should expand the sample size to further verify the interaction between depression, anxiety and sleep disturbance. Finally, our study relied on the scale to evaluate sleep quality, and we expect to use sleep monitoring devices such as polysomnography in the future, so as to provide better objective indicators for sleep monitoring of MHD patients.
In conclusion, sleep disturbance in MHD patients may be associated with age, Alb, Ca2+ and anxiety. In addition, we developed a risk prediction model for sleep disturbance in MHD patients, and early intervention with reversible predictors, such as CRP, Alb, Ca2+ and PTH, will contribute to the prevention and treatment of sleep disturbance in MHD patients. Subsequent studies should elucidate the exact pathogenesis of sleep disturbance in MHD patients and develop effective therapeutic strategies.
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
The studies involving human participants were reviewed and approved by the Ethics Committee of the Third Affiliated Hospital of Soochow University. The patients/participants provided their written informed consent to participate in this study.
RX, LM, BZ, and RJ conceived the study. RX and LM performed the study and data analyses and drafted the manuscript. JN, YD, YS, and CY conducted the patient enrolment and collected the data. BZ and RJ revised the manuscript. All authors contributed to the article and approved the submitted version.
This work was supported by the National Natural Science Foundation of China (81974171 and 81703482), Innovative and Entrepreneurial Team of Jiangsu Province (JSSCTD202144), the Program of Major Science and Technology Project of Changzhou Health Commission (No: ZD202101), and Changzhou Sci&Tech (Program Grant No. CJ20210090).
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
The authors thank all the patients for their cooperation in this study.
The Supplementary Material for this article can be found online at: