Abstract
Poor sleep quality is common in patients receiving maintenance hemodialysis (MHD) and may worsen symptom burden and outcomes. This study evaluated Pittsburgh sleep quality index (PSQI)-defined poor sleep quality and its associated factors in a single-center MHD cohort. This single-center cross-sectional study included 143 patients receiving MHD at the Fifth People’s Hospital of Jinan from June to December 2024. Sleep quality was assessed using the PSQI. Patients were categorized as having good sleep quality (PSQI < 5) or poor sleep quality (PSQI ≥ 5). Demographic, clinical, dialysis-related, and laboratory variables were collected, including dialysis vintage, residual urine volume, weekly dialysis frequency, and dialysis adequacy assessed by standard weekly Kt/V (stdKt/V). An exploratory subgroup analysis was also performed in patients with diabetes mellitus. Of the 143 patients, 76 (53.1%) had poor sleep quality. Significant between-group differences were observed for age, diabetes mellitus, restless legs syndrome (RLS), and dialysis adequacy (all P < .05). In univariable logistic regression, older age (odds ratio [OR] 1.044, 95% CI: 1.013–1.076, P = .005), diabetes mellitus (OR 3.041, 95% CI: 1.518–6.096, P = .002), RLS (OR 3.717, 95% CI: 1.647–8.392, P = .002), and stdKt/V < 2.1 (OR 2.562, 95% CI: 1.218–5.389, P = .013) were associated with poor sleep quality. In multivariable model 1, older age (per 1-year increase; OR 1.041, 95% CI: 1.005–1.079, P = .027), diabetes mellitus (OR 2.625, 95% CI: 1.194–5.772, P = .016), and stdKt/V < 2.1 (OR 2.875, 95% CI: 1.264–6.542, P = .012) remained independently associated with poor sleep quality, whereas female sex was not significant (OR 1.801, 95% CI: 0.775–4.185, P = .171). In model 2, diabetes mellitus (OR 2.609, 95% CI: 1.184–5.749, P = .017), stdKt/V < 2.1 (OR 2.346, 95% CI: 1.039–5.296, P = .040), and RLS (OR 3.002, 95% CI: 1.175–7.668, P = .022) remained significant. In an exploratory subgroup analysis of patients with diabetes mellitus, calcium-phosphorus product > 55 mg2/dL2 was associated with poor sleep quality. Poor sleep quality was common in this MHD cohort and was associated with older age, diabetes mellitus, RLS, and lower dialysis adequacy. In patients with diabetes mellitus, the calcium-phosphorus product finding should be interpreted as exploratory.
Keywords: cross-sectional studies, hemodialysis, risk factors, sleep quality
1. Introduction
Chronic kidney disease (CKD) is a major global public health problem, and many patients with end-stage renal disease require renal replacement therapy to sustain life. Hemodialysis remains one of the most widely used renal replacement modalities worldwide.[1] Sleep-related problems are common in patients receiving dialysis and can negatively affect symptom burden, daily functioning, quality of life, and clinical outcomes.[2-4] These sleep-related problems are heterogeneous and may include insomnia symptoms, sleep-disordered breathing, restless legs syndrome (RLS), and excessive daytime sleepiness.[2] However, the Pittsburgh Sleep Quality Index (PSQI) does not diagnose these disorders individually; rather, it evaluates global subjective sleep quality.
Although sleep problems in hemodialysis populations are well recognized, the factors associated with PSQI-defined poor sleep quality in maintenance hemodialysis patients remain incompletely defined, and findings vary across cohorts. In addition, many studies focus on prevalence rather than clinically useful within-cohort risk stratification. Therefore, this single-center cross-sectional study aimed to evaluate poor sleep quality in patients receiving maintenance hemodialysis (MHD) and to identify associated demographic, clinical, dialysis-related, and laboratory factors. Because diabetes mellitus was common in our cohort, we also retained a predefined exploratory subgroup analysis in patients with diabetes mellitus.
2. Materials and methods
2.1. Research objects
This single-center cross-sectional study included patients receiving MHD at the dialysis center of the Fifth People’s Hospital of Jinan from June 2024 to December 2024. Eligible participants were adults aged ≥18 years who had been diagnosed with end-stage renal disease and had received hemodialysis at our hospital for at least 3 months. Patients with documented severe mental illness, hearing/speech/cognitive dysfunction that precluded questionnaire completion, malignant tumors, severe cardiopulmonary insufficiency, severe infection, or inability to cooperate with the study procedures were excluded. The study was approved by the Ethics Committee of the Fifth People’s Hospital of Jinan (approval number: 25-5-08), and all participants provided informed consent.
All patients received conventional bicarbonate hemodialysis using Fresenius 4008S machines with the Bi-bag system. Fresenius FXClass-60 dialyzers (membrane surface area, 1.4 m2) were used. The dialysate flow rate was 500 mL/min, the blood flow rate was 200 to 350 mL/min, and each session lasted 4 hours. Dialysate prescriptions were standardized, with individualized adjustments when clinically necessary. Dialysis water and dialysate quality control followed the ISO 23500 series and relevant Chinese regulations.[5]
2.2. Research methods
2.2.1. General information collection
Demographic and clinical characteristics were collected retrospectively from the medical records, including age, sex, body mass index, dialysis vintage (months), residual urine volume (mL/24 hours), diabetes mellitus, hypertension, pruritus, RLS, vascular access type, and weekly dialysis frequency. Laboratory data from the preceding 3 months were collected, including red blood cell count, hemoglobin, hematocrit, C-reactive protein, ferritin, parathyroid hormone, albumin, prealbumin, blood urea nitrogen, creatinine, uric acid, β2-microglobulin, potassium, sodium, chloride, calcium, phosphorus, magnesium, cholesterol, and triglycerides albumin-corrected calcium and the calcium-phosphorus product were also calculated. The calcium-phosphorus product is one of the core management targets for dialysis patients. About 55 mg2/dL2 has been the most commonly used risk stratification threshold in blood dialysis studies over the past 2 decades.[6] In this study, 55 mg2/dL2 was set as the cutoff value for abnormal elevation of the calcium-phosphorus product. Prescription drugs for CKD-mineral and bone disorder were also collected, including calcium-free phosphate binders (sevelamer), calcium-containing agents, and active vitamin D. Because a minority of patients were treated twice weekly owing to preserved residual renal function, dialysis adequacy was assessed using standard weekly Kt/V (stdKt/V) to allow comparison across dialysis schedules.[7] For dialysis patients who do not follow the 3-times-per-week dialysis regimen, the minimum target for stdKt/V is 2.1.[7] In this study, stdKt/V ≥ 2.1 was considered adequate dialysis. The base stdKt/V was calculated using the Leypoldt–Depner approach; when residual renal function was present, correction was based on the Casino method.[7,8]
2.2.2. Sleep quality assessment
Sleep quality was assessed using the PSQI.[9,10] The scale comprises 7 domains: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleep medication, and daytime dysfunction. Each component is scored from 0 to 3, and the total PSQI score ranges from 0 to 21, with higher scores indicating poorer sleep quality. In this study, patients with a PSQI score <5 were classified as having good sleep quality and those with a PSQI score ≥5 were classified as having poor sleep quality. Importantly, the PSQI evaluates global sleep quality over the previous month rather than establishing specific diagnoses such as insomnia, obstructive sleep apnea, or hypersomnolence disorders.
RLS status was extracted from the clinical record when the diagnosis had been documented according to established clinical criteria and was analyzed as a coexisting clinical condition that could worsen subjective sleep quality rather than as a component of the PSQI-defined outcome.[11,12]
2.3. Statistical analyses
SPSS 23.0 was used for statistical analysis. Normally distributed continuous variables are presented as mean ± standard deviation and were compared with the independent-samples t test. Non-normally distributed variables are presented as median (P25, P75) and were compared with the Mann–Whitney U test. Categorical variables are presented as n (%) and were compared using the chi-square test. Univariable logistic regression was used to identify factors associated with poor sleep quality, and odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. Variables with P < .05 in univariable analyses were entered into multivariable logistic regression. Because age and RLS were considered closely related clinically, 2 complementary multivariable models were presented for interpretability. All tests were 2-sided, and P < .05 was considered statistically significant.
3. Results
3.1. A survey of sleep quality in patients with MHD
A total of 143 stable adult patients receiving MHD were included in this cross-sectional study. The mean age was 61.17 ± 11.94 years; 65.0% were male; 60.1% had diabetes mellitus; and 94.4% had hypertension. Overall, 127 patients (88.8%) received dialysis 3 times/wk and 16 patients (11.2%) received dialysis twice per week. The twice-weekly dialysis group retained meaningful residual renal function, with a 24-hour urine volume of ≥800 mL and Kru > 3 mL/min/1.73 m2, whereas most patients treated 3 times weekly had markedly lower urine output (generally 0–400 mL/24 hours).
The PSQI scores ranged from 0 to 17, with a mean score of 6.58 ± 4.98. Seventy-six patients (53.1%) had poor sleep quality (PSQI ≥ 5), whereas the remaining 67 patients had good sleep quality (PSQI < 5).
3.2. Analysis of single factors affecting sleep quality
Comparison of the general characteristics and laboratory data between the good sleep quality group and the poor sleep quality group showed significant between-group differences in age, diabetes mellitus, RLS, and dialysis adequacy (P < .05; Table 1). Compared with patients with good sleep quality, patients with poor sleep quality were older (63.87 ± 10.73 vs 58.12 ± 12.58 years, P = .004), were more likely to have diabetes mellitus (55 [72.4%] vs 31 [46.3%], P < .001), were more likely to have RLS (30 [39.5%] vs 10 [14.9%], P < .001), and were less likely to achieve adequate dialysis (stdKt/V ≥ 2.1: 32 [48.5%] vs 41 [70.7%], P = .012).
Table 1.
Characteristics of the study population.
| Characteristic | Overall (N = 143) | Good sleep quality (N = 67) | Poor sleep quality (N = 76) | P-value |
|---|---|---|---|---|
| Age (yr) | 61.17 ± 11.94 | 58.12 ± 12.58 | 63.87 ± 10.73 | .004 |
| Gender, n (%) | .229 | |||
| Male | 93 (65.0%) | 47 (70.1%) | 46 (60.5%) | |
| Female | 50 (35.0%) | 20 (29.9%) | 30 (39.5%) | |
| BMI (kg/m2) | 22.42 (20.53, 26.22) | 22.20 (20.38, 25.31) | 23.35 (20.61, 26.60) | .177 |
| Dialysis vintage (mo) | 23 (9, 39) | 20 (10, 40) | 23 (9, 39) | .712 |
| Residual urine volume (mL/24 h) | 100 (0, 200) | 100 (0, 250) | 100 (0, 200) | .459 |
| Type of vascular access, n (%) | .804 | |||
| Arterio-venous fistula | 129 (90.2%) | 60 (89.6%) | 69 (90.8%) | |
| Central venous catheter | 14 (9.8%) | 7 (10.4%) | 7 (9.2%) | |
| Frequency of dialysis per week, n (%) | .063 | |||
| 2 times | 16 (11.2%) | 11 (16.4%) | 5 (6.6%) | |
| 3 times | 127 (88.8%) | 56 (83.6%) | 71 (93.4%) | |
| Diabetes, n (%) | <.001 | |||
| Yes | 86 (60.1%) | 31 (46.3%) | 55 (72.4%) | |
| No | 57 (39.9%) | 36 (53.7%) | 21 (27.6%) | |
| Hypertension, n (%) | .101 | |||
| Yes | 135 (94.4%) | 61 (91.0%) | 74 (97.4%) | |
| No | 8 (5.6%) | 6 (9.0%) | 2 (2.6%) | |
| Pruritus, n (%) | .215 | |||
| Yes | 59 (41.3%) | 24 (35.8%) | 35 (46.1%) | |
| No | 84 (58.7%) | 43 (64.2%) | 41 (53.9%) | |
| RLS, n (%) | <.001 | |||
| Yes | 40 (28%) | 10 (14.9%) | 30 (39.5%) | |
| No | 103 (72%) | 57 (85.1%) | 46 (60.5%) | |
| RBC (1012/L) | 4.14 (3.8, 4.41) | 4.09 ± 0.47 | 4.18 ± 0.58 | .313 |
| Hb (g/L) | 120.46 ± 14.05 | 119.00 ± 13.88 | 121.75 ± 14.17 | .244 |
| HCT (%) | 36.60 ± 4.33 | 36.06 ± 4.07 | 37.09 ± 4.53 | .157 |
| PA (mg/L) | 311.5 (258.5, 369) | 319.61 ± 86.73 | 314.82 ± 79.72 | .732 |
| Urea (mmol/L) | 23.05 ± 6.58 | 22.85 ± 7.64 | 23.24 ± 5.53 | .727 |
| Cr (μmol/L) | 803.24 ± 265.63 | 844.20 ± 294.64 | 767.13 ± 233.18 | .083 |
| Ua (μmol/L) | 400 (340, 445) | 401.60 ± 91.90 | 396.11 ± 88.03 | .716 |
| Chol (mmol/L) | 3.78 (3.23, 4.49) | 3.86 ± 0.97 | 3.88 ± 0.90 | .88 |
| CRP (mg/L) | 2.67 (0.89, 6.74) | 2.69 (0.90, 6.56) | 2.67 (0.62, 6.94) | .883 |
| Fer (ng/ml) | 165 (98.05, 300.5) | 163.50 (91.68, 70.25) | 172.00 (98.05, 354.50) | .502 |
| PTH (pg/ml) | 261 (151.25, 436.25) | 242.00 (134.00, 410.00) | 289.00 (156.50, 445.50) | .308 |
| Alb (g/L) | 39.00 (37.00, 42.00) | 40.00 (38.00, 42.00) | 39.00 (37.00, 42.00) | .129 |
| β2-MG (mg/L) | 23.80 (21.19, 25.10) | 24.00 (22.00, 25.50) | 23.40 (20.62, 24.90) | .199 |
| K (mmol/L) | 4.70 (4.40, 5.20) | 4.70 (4.40, 5.30) | 4.70 (4.20, 5.18) | .546 |
| Na (mmol/L) | 138.00 (136.00, 141.00) | 139.00 (137.00, 141.00) | 138.00 (136.00, 140.00) | .061 |
| Cl (mmol/L) | 99.74 ± 3.81 | 100.00 (97.00, 103.00) | 99.00 (97.00, 101.00) | .213 |
| Mg (mmol/L) | 1.07 (1.00, 1.19) | 1.07 (1.01, 1.19) | 1.08 (0.99, 1.20) | .659 |
| TG (mmol/L) | 1.52 (0.98, 2.23) | 1.43 (0.96, 2.06) | 1.55 (1.06, 2.51) | .132 |
| Ca (mmol/L) | 2.13 ± 0.19 | 2.13 ± 0.21 | 2.13 ± 0.17 | .955 |
| Albumin-corrected calcium (mmol/L) | 2.29 ± 0.25 | 2.29 ± 0.17 | 2.29 ± 0.21 | .999 |
| P (mmol/L) | 1.81 (1.46, 2.18) | 1.81 (1.38, 2.09) | 1.82 (1.48, 2.23) | .393 |
| Calcium-Phosphorus product, n (%) | .281 | |||
| ≤55 (mg2/dL2) | 96 (67.1%) | 48 (71.6%) | 48 (63.2%) | |
| >55 (mg2/dL2) | 47 (32.9%) | 19 (28.4%) | 28 (36.8%) | |
| stdKt/V, n (%) | (N = 124) | (N = 58) | (N = 66) | .012 |
| ≥2.1 | 73 (58.9%) | 41 (70.7%) | 32 (48.5%) | |
| <2.1 | 51 (41.1%) | 17 (29.3%) | 34 (51.5%) | |
| Calcium-containing agents, n (%) | 107 (74.8%) | 47 (70.1%) | 60 (78.9%) | .226 |
| Calcium-free phosphate binders | ||||
| Sevelamer, n (%) | 44 (30.8%) | 16 (23.9%) | 28 (36.8%) | .094 |
| Active vitamin D, n (%) | 36 (25.2%) | 12 (17.9%) | 24 (31.6%) | .060 |
Bold values indicate P < .05 after statistical testing.
Ca (mmol/L) in the table refers to serum total calcium before albumin correction.
Albumin-corrected calcium (mmol/L) = total Ca (mmol/L) + 0.02 × [40 − Alb (g/L)].
Calcium-phosphorus product (mg2/dL2) ≈ 12.4 × [albumin-corrected Ca (mmol/L) × P (mmol/L)].
Alb = albumin, BMI = body mass index, Chol = cholesterol, Cr = creatinine, CRP = C-reactive protein, Fer = ferritin, Hb = hemoglobin, HCT = hematocrit, PA = prealbumin, PTH = parathyroid hormone, RBC = red blood cell count, RLS = restless legs syndrome, stdKt/V = standard weekly Kt/V, TG = triglycerides, Ua = uric acid, Urea = blood urea nitrogen, β2-MG = beta-2 microglobulin.
3.3. Multifactorial logistic regression analysis of factors affecting sleep quality
In univariable logistic regression, older age (OR 1.044, 95% CI: 1.013–1.076, P = .005), diabetes mellitus (OR 3.041, 95% CI: 1.518–6.096, P = .002), RLS (OR 3.717, 95% CI: 1.647–8.392, P = .002), and stdKt/V < 2.1 (OR 2.562, 95% CI: 1.218–5.389, P = .013) were associated with poor sleep quality.
Because age and RLS were closely related clinically, we presented 2 complementary multivariable models while retaining sex, diabetes mellitus, and stdKt/V in both models. In model 1, older age (per 1-year increase; OR 1.041, 95% CI: 1.005–1.079, P = .027), diabetes mellitus (OR 2.625, 95% CI: 1.194–5.772, P = .016), and stdKt/V < 2.1 (OR 2.875, 95% CI: 1.264–6.542, P = .012) remained independently associated with poor sleep quality, whereas female sex was not significant (OR 1.801, 95% CI: 0.775–4.185, P = .171). In model 2, diabetes mellitus (OR 2.609, 95% CI: 1.184–5.749, P = .017), stdKt/V < 2.1 (OR 2.346, 95% CI: 1.039–5.296, P = .040), and RLS (OR 3.002, 95% CI: 1.175–7.668, P = .022) remained independently associated with poor sleep quality (Table 2).
Table 2.
Univariable and multivariable analyses of factors associated with poor sleep quality in MHD patients.
| Factors | Univariable analyses | Multivariable analyses | |||
|---|---|---|---|---|---|
| Odds ratio (95% CI) | P-value | Odds ratio (95% CI)* | P-value | ||
| Age | 1.044 (1.013, 1.076) | .005 | Model 1 | ||
| Female | 1.533 (0.764, 3.076) | .230 | Age | 1.041 (1.005, 1.079) | .027 |
| BMI (kg/m2) | 1.066 (0.981, 1.159) | .132 | Female | 1.801 (0.775, 4.185) | .171 |
| Vintage (mo) | 1.005 (0.994, 1.017) | .373 | Diabetes | 2.625 (1.194, 5.772) | .016 |
| Residual urine volume (mL/24 h) | 0.999 (0.997, 1.000) | .056 | stdKt/V < 2.1 | 2.875 (1.264, 6.542) | .012 |
| Type of vascular access (Central vein) | 0.870 (0.289, 2.621) | .804 | Model 2 | ||
| Frequency of dialysis per week (3 times) | 2.789 (0.916, 8.495) | .071 | Female | 1.700 (0.733, 3.941) | .216 |
| Diabetes | 3.041 (1.518, 6.096) | .002 | Diabetes | 2.609 (1.184, 5.749) | .017 |
| Hypertension | 3.639 (0.709, 18.683) | .122 | stdKt/V < 2.1 | 2.346 (1.039, 5.296) | .040 |
| Pruritus | 1.529 (0.780, 2.998) | .216 | RLS | 3.002 (1.175, 7.668) | .022 |
| RLS | 3.717 (1.647, 8.392) | .002 | |||
| RBC (1012/L) | 1.385 (0.737, 2.606) | .312 | |||
| Hb (g/L) | 1.014 (0.990, 1.039) | .244 | |||
| HCT (%) | 1.058 (0.979, 1.143) | .157 | |||
| PA (mg/L) | 0.999 (0.995, 1.003) | .730 | |||
| Urea (mmol/L) | 1.009 (0.960, 1.061) | .724 | |||
| Cr (μmol/L) | 0.999 (0.998, 1.000) | .086 | |||
| Ua (μmol/L) | 0.999 (0.996, 1.003) | .714 | |||
| Chol (mmol/L) | 1.028 (0.718, 1.473) | .879 | |||
| CRP (mg/L) | 1.013 (0.974, 1.053) | .511 | |||
| Fer (ng/mL) | 1.001 (0.999, 1.003) | .260 | |||
| PTH (pg/mL) | 1.000 (0.999, 1.002) | .705 | |||
| Alb (g/L) | 0.970 (0.893, 1.054) | .472 | |||
| β2-MG (mg/L) | 0.963 (0.894, 1.037) | .320 | |||
| K (mmol/L) | 0.853 (0.566, 1.287) | .450 | |||
| Na (mmol/L) | 0.912 (0.822, 1.013) | .085 | |||
| Cl (mmol/L) | 0.929 (0.850, 1.015) | .103 | |||
| Mg (mmol/L) | 0.432 (0.044, 4.193) | .469 | |||
| TG (mmol/L) | 1.315 (0.956, 1.810) | .093 | |||
| Ca (mmol/L) | 1.052 (0.181, 6.105) | .955 | |||
| P (mmol/L) | 1.304 (0.748, 2.275) | .349 | |||
| Calcium-phosphorus product >55 (mg2/dL2) | 1.474 (0.727, 2.988) | .282 | |||
| stdKt/V < 2.1 | 2.562 (1.218, 5.389) | .013 | |||
| Calcium-containing agents | 0.627 (0.293, 1.340) | .228 | |||
| Calcium-free phosphate binders | |||||
| Sevelamer | 1.859 (0.896, 3.858) | .096 | |||
| Active vitamin D | 2.115 (0.960, 4.660) | .063 | |||
Bold values indicate P < .05 after statistical testing.
Age OR is expressed per 1-year increase. Poor sleep quality was defined as PSQI ≥ 5.
Alb = albumin, BMI = body mass index, Ca = calcium, CI = confidence interval, Chol = cholesterol, Cr = creatinine, CRP = C-reactive protein, Fer = ferritin, Hb = hemoglobin, HCT = hematocrit, Mg = magnesium, OR = odds ratio, P = phosphorus, PA = prealbumin, PTH = parathyroid hormone, PSQI = Pittsburgh sleep quality index, RBC = red blood cell count, RLS = restless legs syndrome, stdKt/V = standard weekly Kt/V, TG = triglycerides, Ua = uric acid, Urea = blood urea nitrogen, β2-MG = beta-2 microglobulin.
Model 1 included age, sex, diabetes mellitus, and stdKt/V. Model 2 included sex, diabetes mellitus, stdKt/V, and RLS.
3.4. Exploratory subgroup analysis in patients with diabetes mellitus combined with MHD
Among the 86 patients with MHD and diabetes mellitus, 55 (64.0%) had poor sleep quality. In this exploratory subgroup, age and calcium-phosphorus product differed significantly between the good and poor sleep quality groups (Table 3).
Table 3.
Exploratory subgroup comparison in MHD patients with diabetes mellitus.
| Characteristic | Overall (N = 86) | Good sleep quality group (N = 31) | Poor sleep quality group (N = 55) | P-value |
|---|---|---|---|---|
| Age | 62 (56.75, 69) | 59 (55, 62) | 65 (59, 71) | .002 |
| Calcium-phosphorus product, n (%) | .003 | |||
| ≤55 mg2/dL2 | 61 (70.9%) | 28 (90.3%) | 33 (60%) | |
| >55 mg2/dL2 | 25 (29.1%) | 3 (9.7%) | 22 (40%) | |
Bold values indicate P < .05 after statistical testing.
This was an exploratory subgroup analysis among patients with MHD and diabetes mellitus.
Calcium-phosphorus product was calculated using albumin-corrected calcium.
MHD = maintenance hemodialysis.
In multivariable analysis adjusted for sex, calcium-phosphorus product >55 mg2/dL2 was associated with poor sleep quality (OR 9.894, 95% CI: 2.235–43.799, P = .003; Table 4). Because calcium-phosphorus product was not associated with poor sleep quality in the overall cohort, this subgroup finding should be interpreted as exploratory and hypothesis-generating rather than definitive.
Table 4.
Exploratory multivariable analysis of poor sleep quality in diabetic MHD patients.
| Factors | Odds ratio (95% CI) | P-value |
|---|---|---|
| Age | 1.093 (1.028, 1.162) | .004 |
| Female | 1.212 (0.407, 3.611) | .731 |
| Calcium-phosphorus product >55 mg2/dL2 | 9.894 (2.235, 43.799) | .003 |
Bold values indicate P < .05 after statistical testing.
Exploratory multivariable model adjusted for sex. Age OR is expressed per 1-year increase.
Calcium-phosphorus product was calculated using albumin-corrected calcium.
MHD = maintenance hemodialysis, OR = odds ratio.
4. Discussion
Poor sleep quality was common in this single-center MHD cohort, affecting 53.1% of patients. According to previous studies, the prevalence of sleep disorders in hemodialysis patients around the world is 40% to 85%.[4] Poor sleep quality has a serious impact on the quality of life of patients and increase the risk of cardiovascular disease and mortality.[2,13] The main contribution of this study is not the description of prevalence alone, but the identification of demographic, clinical, and dialysis-related factors associated with PSQI-defined poor sleep quality in a contemporary Chinese MHD cohort.
Older age was independently associated with poor sleep quality in the present cohort. A previous study also reached the same conclusion.[14,15] This finding is biologically plausible because aging is accompanied by greater sleep fragmentation, reduced deep sleep, and a higher burden of comorbidity. The main reason may be that as age increases, the body’s physiological functions and immunity decline, leading to an increase in related complications. Melatonin levels also decrease, and since melatonin regulates the body’s sleep-wake cycle and its rhythmic changes, a decrease in melatonin weakens its regulatory role, making sleep problems more likely.[16]
The results of this study showed that the prevalence of poor sleep quality in patients with MHD and diabetes mellitus was 64%. Diabetes mellitus was also consistently associated with poor sleep quality in both multivariable models. In this study, diabetes was analyzed as a comorbidity rather than as a surrogate for the primary cause of kidney failure. Diabetes may worsen sleep through autonomic dysfunction, metabolic dysregulation, and the broader symptom burden associated with chronic disease.[17-19]
The assessment of dialysis adequacy, which evaluates the efficiency of dialysis in eliminating waste products and excess fluid from the body, is critical for optimizing dialysis regimens and improving patient prognosis.[20,21] Our research indicates that dialysis adequacy may also contribute to poor sleep quality. Because some patients in our cohort were treated twice weekly on the basis of preserved residual renal function, we used stdKt/V rather than a single-session adequacy measure alone to compare dialysis adequacy across schedules.[7] Patients with stdKt/V < 2.1 were more likely to have poor sleep quality in both multivariable models. Previous studies have reported inconsistent associations between dialysis adequacy and sleep outcomes, but our findings suggest that inadequate solute clearance may be 1 potentially modifiable correlate of poor sleep quality.[22-24]
RLS, also known as Willis-Ekbom diseases, is a neurological sensory-motor disorder, the core manifestation of which is intense discomfort in the lower limbs at rest (especially at night or before bedtime), forcing patients to move their limbs to alleviate the symptoms, seriously affecting sleep and quality of life.[11,25,26] The prevalence of RLS is higher in patients with CKD compared to the general population. Most studies of dialysis patients report a prevalence of RLS in the range of 15% to 30%.[27] Some studies have shown that dialysis patients with comorbid RLS are more likely to have poor sleep quality,[27] which is consistent with the results of this study. RLS was independently associated with poor sleep quality in model 2. This does not mean that RLS was treated as equivalent to the study outcome. Rather, the outcome was global subjective sleep quality as measured by the PSQI, whereas RLS was analyzed as a coexisting clinical condition that can worsen sleep latency, nocturnal discomfort, and sleep fragmentation. This conceptual distinction is important because the PSQI does not classify specific sleep disorder phenotypes.
This study additionally found that a calcium-phosphorus product >55mg2/dL2 increased the risk of sleep disorders in patients with MHD combined with diabetes. However, calcium-phosphorus product was not significant in the overall cohort, and contemporary CKD-mineral and bone disorder guidance emphasizes serial assessment of calcium, phosphorus, and parathyroid hormone rather than reliance on a single Ca × P target.[28,29] Accordingly, this subgroup association should be regarded as exploratory.
Several limitations of this study should be acknowledged. First, this was a single-center cross-sectional study with a limited sample size, which precludes causal inference. Second, because of the retrospective design, the primary kidney disease could not be fully ascertained in some patients. HbA1c data were incomplete in the retrospective records and therefore could not be included in the current analysis. Third, although multiple sociodemographic and clinical variables were included, other potentially relevant factors, such as socioeconomic status, psychological condition, and lifestyle behaviors, were not fully accounted for. Fourth, sleep was assessed using the PSQI rather than objective or diagnostic tools, and specific sleep disorders could not be distinguished. Finally, no structured psychiatric assessment was performed, limiting control for psychiatric confounding and further exploration of the interplay between sleep disturbance and psychological problems.
Despite these limitations, the present findings may help clinicians identify patients at higher risk of poor sleep quality and support more targeted screening and intervention strategies in routine dialysis care.
5. Conclusion
Poor sleep quality was common in this MHD cohort and was associated with older age, diabetes mellitus, RLS, and lower dialysis adequacy. Among patients with diabetes mellitus, the observed association between calcium-phosphorus product and poor sleep quality should be regarded as exploratory. Clinicians should pay closer attention to sleep complaints in MHD patients and consider routine screening, especially in patients who are older, diabetic, have RLS, or have lower dialysis adequacy.
Author contributions
Conceptualization: Liyan Liu.
Data curation: Chen Qu, Yun Liu.
Investigation: Na Tang.
Methodology: Min Qi.
Resources: Jin Wang.
Writing – original draft: Xuemei Liu.
Abbreviations:
- CI
- confidence interval
- CKD
- chronic kidney disease
- MHD
- maintenance hemodialysis
- OR
- odds ratio
- PSQI
- Pittsburgh sleep quality index
- RLS
- restless legs syndrome
- stdKt/V
- standard weekly Kt/V
This work was supported by the Shandong Provincial Medical and Health Science and Technology Development Plan Project (202003050422 to Liyan Liu), and Science and Technology Development Plan Project of Jinan Municipal Health Commission (2025301014 to Liyan Liu).
This study was approved by the Ethics Committee of the Fifth People’s Hospital of Jinan (approval number: 25-5-08). All participants provided informed consent. All procedures were performed in accordance with the ethical standards of the institutional review board and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
The authors have no conflicts of interest to disclose.
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
How to cite this article: Liu X, Qu C, Tang N, Qi M, Wang J, Liu Y, Liu L. Risk factors associated with poor sleep quality in maintenance hemodialysis patients: A single-center cross-sectional study. Medicine 2026;105:19(e48573).
Contributor Information
Xuemei Liu, Email: Pretty423@163.com.
Chen Qu, Email: 340021705@qq.com.
Na Tang, Email: 564920083@qq.com.
Min Qi, Email: 635309976@qq.com.
Jin Wang, Email: 987902361@qq.com.
Yun Liu, Email: Pretty423@163.com.
References
- [1].Htay H, Bello AK, Levin A, et al. hemodialysis use and practice patterns: an international survey study. Am J Kidney Dis. 2021;77:326–35.e1. [DOI] [PubMed] [Google Scholar]
- [2].Lyons OD. Sleep disorders in chronic kidney disease. Nat Rev Nephrol. 2024;20:690–700. [DOI] [PubMed] [Google Scholar]
- [3].Benetou S, Alikari V, Vasilopoulos G, et al. Factors associated with insomnia in patients undergoing hemodialysis. Cureus. 2022;14:e22197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Cukor D, Unruh M, McCurry SM, Mehrotra R. The challenge of insomnia for patients on haemodialysis. Nat Rev Nephrol. 2021;17:147–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Kawanishi H, Moriishi M, Takahashi N, Tsuchiya S. Preparation and quality management of fluids for hemodialysis. Contrib Nephrol. 2017;189:153–9. [DOI] [PubMed] [Google Scholar]
- [6].National Kidney Foundation. K/DOQI clinical practice guidelines for bone metabolism and disease in chronic kidney disease. Am J Kidney Dis. 2003;42(4 Suppl 3):S1–201. [PubMed] [Google Scholar]
- [7].National Kidney Foundation. KDOQI clinical practice guideline for hemodialysis adequacy: 2015 update. Am J Kidney Dis. 2015;66:884–930. [DOI] [PubMed] [Google Scholar]
- [8].Casino FG, Roblero MFS, González-Sanchidrian S, et al. Prescribing the dialysis dose and treatment frequency in home haemodialysis. Nephrol Dial Transplant. 2024;39:445–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Buysse DJ, Reynolds CF, 3rd, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28:193–213. [DOI] [PubMed] [Google Scholar]
- [10].Carpi M. The Pittsburgh Sleep Quality Index: a brief review. Occup Med (Lond). 2025;75:14–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Winkelman JW, Wipper B. Restless legs syndrome: a review. JAMA. 2026;335:703–14. [DOI] [PubMed] [Google Scholar]
- [12].Gossard TR, Trotti LM, Videnovic A, Louis EK, St. Restless legs syndrome: contemporary diagnosis and treatment. Neurotherapeutics. 2021;18:140–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Yoshida N, Tanaka T, Suzuki Y, et al. Sleep quality and its association with quality of life and mortality in hemodialysis patients. J Clin Med. 2025;14:8729. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Yue Z, Zhang Y, Cheng X, Zhang J. Sleep quality among the elderly in 21st century shandong province, china: a ten-year comparative study. Int J Environ Res Public Health. 2022;19:14296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Chen Y, Zhang B. Latent classes of sleep quality and related predictors in older adults: a person-centered approach. Arch Gerontol Geriatr. 2022;102:104736. [DOI] [PubMed] [Google Scholar]
- [16].Hasannia E, Derakhshanpour F, Vakili MA. Effects of melatonin on salivary levels of cortisol and sleep quality of hemodialysis patients: a randomized clinical trial. Iran J Psychiatry. 2021;16:305–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Rutters F, Nefs G. Sleep and circadian rhythm disturbances in diabetes: a narrative review. Diabetes Metab Syndr Obes. 2022;15:3627–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Cheng W, Chen H, Tian L, Ma Z, Cui X. Heart rate variability in different sleep stages is associated with metabolic function and glycemic control in type 2 diabetes mellitus. Front Physiol. 2023;14:1157270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Mao Y. Sleep architecture changes in diabetes. J Clin Med. 2024;13:6851. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Ai S, Xu Q, Chen G, Zheng K, Qin Y, Li X. Effects of hemodialysis adequacy on chronic kidney disease complications using latent class trajectory modeling: a real-world study based on long-term observation of Kt/V. Front Med (Lausanne). 2024;11:1449919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Jeon J, Kim GO, Kim BY, et al. Effects of Kt/V(urea) on outcomes according to age in patients on maintenance hemodialysis. Clin Kidney J. 2024;17:sfae116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Weerakoon D, Perera R, Ambillapitiya R, et al. Determinants of poor sleep quality in hemodialysis patients: a multicenter cross sectional study in Sri Lanka. Frontiers in Medicine. 2026;13:1780694. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Afaghi E, Beigmohammadi H, Rajai N, Pishgooie AH. Effectiveness of Otago exercise on sleep quality and dialysis adequacy in hemodialysis patients. BMC Nephrol. 2026;27:154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Jamelo JP, Valdepenas AN, Fadriquela DL, Gueco I. #235 Quality of sleep and the quality of life in chronic kidney disease patients undergoing hemodialysis in a tertiary hospital in the Philippines. Nephrol Dial Transplant. 2024;39(Supplement_1):235. [Google Scholar]
- [25].Broström A, Alimoradi Z, Odzakovic E, et al. Quality of life among patients with restless legs syndrome: a systematic review and meta-analysis. J Clin Neurosci. 2024;122:80–91. [DOI] [PubMed] [Google Scholar]
- [26].Castillo-Álvarez F, Marzo-Sola ME. Restless legs syndrome. Pathophysiology, diagnosis and treatment. Med Clin (Barc). 2025;164:84–90. [DOI] [PubMed] [Google Scholar]
- [27].Safarpour Y, Vaziri ND, Jabbari B. Restless legs syndrome in chronic kidney disease- a systematic review. Tremor Other Hyperkinet Mov (N Y). 2023;13:10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Ketteler M, Block GA, Evenepoel P, et al. Executive summary of the 2017 KDIGO Chronic Kidney Disease-Mineral and Bone Disorder (CKD-MBD) Guideline Update: what’s changed and why it matters. Kidney Int. 2017;92:26–36. [DOI] [PubMed] [Google Scholar]
- [29].Ketteler M, Evenepoel P, Holden RM, et al. ; Conference Participants. Chronic kidney disease-mineral and bone disorder: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int. 2025;107:405–23. [DOI] [PubMed] [Google Scholar]