OBJECTIVE

There is a high risk of asymptomatic hypoglycemia associated with hemodialysis (HD) using glucose-free dialysate; therefore, the inclusion of glucose in the dialysate is believed to prevent intradialytic hypoglycemia. However, the exact glycemic fluctuation profiles and frequency of asymptomatic hypoglycemia using dialysates containing >100 mg/dL glucose have not been determined.

RESEARCH DESIGN AND METHODS

We evaluated the glycemic profiles of 98 patients, 68 of whom were men, with type 2 diabetes undergoing HD (HbA1c 6.4 ± 1.2%; glycated albumin 20.8 ± 6.8%) with a dialysate containing 100, 125, or 150 mg/dL glucose using continuous glucose monitoring.

RESULTS

Sensor glucose level (SGL) showed a sustained decrease during HD, irrespective of the dialysate glucose concentration, and reached a nadir that was lower than the dialysate glucose concentration in 49 participants (50%). Twenty-one participants (21%) presented with HD-related hypoglycemia, defined by an SGL <70 mg/dL during HD and/or between the end of HD and their next meal. All these hypoglycemic episodes were asymptomatic. Measures of glycemic variability calculated using the SGL data (SD, coefficient of variation, and range of SGL) were higher and time below range (<70 mg/dL) was lower in participants who experienced HD-related hypoglycemia than in those who did not, whereas time in range between 70 and 180 mg/dL, time above range (>180 mg/dL), HbA1c, and glycated albumin of the two groups were similar.

CONCLUSIONS

Despite the use of dialysate containing 100–150 mg/dL glucose, patients with diabetes undergoing HD experienced HD-related hypoglycemia unawareness frequently. SGL may fall well below the dialysate glucose concentration toward the end of HD.

Diabetes is a major cause of end-stage kidney disease and cardiovascular disease (1,2). The prognosis of patients with diabetes undergoing maintenance hemodialysis (HD) is worse than that of patients without diabetes undergoing HD (3), but it remains unclear whether differences in glycemic profile affect the high mortality and morbidity of these patients (4,5). Appropriate glycemic control slows the progression of micro- and macrovascular complications (6). However, in addition to high hemoglobin A1c (HbA1c) levels (4,7), low levels are also associated with higher mortality and comorbidity in patients with diabetes undergoing HD (4,811). Furthermore, strict glycemic control increases the incidence of hypoglycemia, of which patients are often unaware, increases glycemic variability, and sometimes induces fatal complications in those with diabetes undergoing HD. High glycemic variability and hypoglycemia are associated with higher incidences of cardiovascular events and other fatal complications (12), and HD-related hypoglycemia develops more frequently in patients with diabetes than in those without (1315). Although the glycemic profile of each patient should be monitored and properly controlled, conventional glycemic markers, such as HbA1c and glycated albumin (GA), do not provide sufficient information with which to predict large glycemic fluctuations or hypoglycemia in those with type 2 diabetes undergoing HD (16).

Asymptomatic hypoglycemia is commonly associated with HD when glucose-free or glucose-poor dialysate solution is used, and glucose-supplemented dialysate is believed to prevent these episodes (14,15,17,18). However, most previous studies included a small number of patients and monitored blood glucose concentrations intermittently. In contrast, continuous glucose monitoring (CGM) has become established as an ideal method of obtaining well-defined glucose profiles (19). Recent studies using CGM have revealed the frequency of all-cause hypoglycemia in patients with diabetes undergoing HD to be 16.7–23.5% (2022) and shown that patients have distinct glycemic profiles on consecutive days on and off HD (2124). The use of dialysate containing 100 mg/dL glucose has been reported to prevent hypoglycemia during HD, on the basis of CGM recordings, although the glycemic patterns obtained are unpredictable, often including sporadic episodes of asymptomatic hypoglycemia during HD (20,25,26). However, the exact glucose fluctuation pattern and the incidence of hypoglycemia related to HD have not been well characterized using CGM (27).

To evaluate HD-related hypoglycemia and HD-induced glucose variability, we assessed the glucose profiles of patients with type 2 diabetes undergoing HD using CGM on consecutive days on and off HD. Close assessment of the CGM data prompted us to focus on the glucose profiles during and immediately after the completion of HD performed using dialysates containing glucose concentrations ≥100 mg/dL.

Study Design and Population

We enrolled 98 consecutive patients with type 2 diabetes who were undergoing maintenance HD at either Kitasato University Hospital or Sohbudai Nieren Clinic, up to June 2017. This study was approved by the ethics committees of Kitasato University Hospital (B17–147) and Sohbudai Nieren Clinic (20170905). All study methods were performed in accordance with the relevant guidelines and regulations of Kitasato University Hospital as well as the Ethical Guidelines for Medical and Health Research Involving Human Subjects in Japan and under the Code of Ethics of the Helsinki Declaration. Type 2 diabetes was defined using the American Diabetes Association criteria. Patients with malignancy, liver cirrhosis, hematologic disease, thyroid dysfunction, pregnancy, or hemoglobinopathy were excluded from the study. Participants were confirmed to have stable glycemic control during at least the preceding 3 months, while taking the same dose of erythropoietin-stimulating agent, and not to have undergone blood transfusion during this period.

We performed CGM using CGMS System Gold (Medtronic Minimed, Northridge, CA) in 40 patients and Medtronic iPro2 CGM (Medtronic Minimed) in 58 for 2 consecutive days, both when on and off HD. The complete sensor glucose level (SGL) data were analyzed in CGM records obtained between 0000 and 2400 h of the 2-day period. Hypoglycemia was defined as SGL <70 mg/dL (3.9 mmol/L) for >20 min (28,29). Hypoglycemia during HD and post-HD hypoglycemia were defined as hypoglycemia that occurred during HD therapy and during the period between the completion of HD and the meal following, respectively. We defined HD-related hypoglycemia as hypoglycemia during HD and/or post-HD hypoglycemia. The following markers of glycemic variability were calculated using SGL data for the entire 48-h recording period: range (maximum to minimum SGL), SD, and coefficient of variation (CV). Furthermore, the glucose management indicator (GMI) was calculated from the 48-h mean SGL, and the percentages of time of SGL between 70 and 180 mg/dL (time in range; TIR), <70 mg/dL (time below range; TBR70), <54 mg/dL (TBR54), and >180 mg/dL (time above range; TAR) for 2 consecutive days, an HD day and non-HD day, were calculated (29).

Biochemical Measurements

The following tests were performed to assess the eligibility of potential participants for the study: HbA1c, GA, plasma glucose, C-peptide immunoreactivity, complete blood count, serum albumin, and serum creatinine. GA was measured by an enzymatic synthesis method using a glycated albumin-L assay kit (Lucica; Asahi Kasei Pharma, Tokyo, Japan) (CV <0.3%).

Statistical Analysis

Statistical analyses were performed using GraphPad Prism (version 5.02; GraphPad Software Inc., San Diego, CA) and JMP (version 14; SAS Institute, Cary, NC). Data are presented as means ± SDs, unless otherwise indicated. Unpaired Student t or Mann-Whitney tests were used to compare data from two groups, and ANOVA was used for the evaluation of the glucose profiles in groups using three different dialysate glucose concentrations. χ2 or Fisher exact tests were used to evaluate categorical data. Correlations between the CGM data and other clinical glycemic parameters were performed using linear regression analysis and multiple regression analysis. P < 0.05 was considered to indicate statistical significance.

Participants Characteristics

Nighty-eight participants were included in this study, of whom 68 (69%) were men; the average age was 62 ± 12 years, HbA1c was 6.4 ± 1.2% (45.9 ± 12.7 mmol/mol), and GA was 20.8 ± 6.8% (Table 1). Sixty-five patients (66%) were treated with insulin therapy, and six (6%) were treated with diet therapy only. The participants underwent 4-h maintenance HD three times per week and ate the next meal within 0.5–1 h after the completion of HD. Eighty-two patients started their HD in the morning (between 0800 and 1000 h), 11 in the afternoon (between 1300 and 1500 h), and five in the evening (between 1500 and 1700 h).

Table 1

Demographics of enrolled participants (n = 98)

DemographicValue
Male sex 68 (69.4) 
Age, years 62 ± 12 
Dry weight, kg 63.7 ± 14.5 
BMI, kg/m2 23.8 ± 4.5 
Duration of diabetes, years 20.5 (1–58) 
Duration of hemodialysis, months 2.0 (0.5–171.0) 
Dialysate glucose concentration, mg/dL  
 100 65 (66.3) 
 125 24 (24.5) 
 150 9 (9.2) 
HD time  
 Morning 82 (83.7) 
 Afternoon 11 (11.2) 
 Evening 5 (5.1) 
Blood flow, mL/min 198.9 ± 20.1 
Dialysis fluid flow, mL/min 453.1 ± 88.8 
Dialysis membrane area, m2 1.78 ± 0.29 
Mode of hemodialysis  
 HD 93 (94.9) 
 HDF 5 (5.1) 
Red blood cell count, × 104/μL) 353.4 ± 57.6 
Hemoglobin concentration, g/dL 10.6 ± 1.6 
Hematocrit, % 33.0 ± 5.2 
Serum albumin, g/dL 3.6 ± 0.5 
HbA1c  
 % 6.4 ± 1.2 
 mmol/mol 45.9 ± 12.7 
GA, % 20.8 ± 6.8 
GA-to-HbA1c ratio, % 3.27 ± 0.82 
Casual plasma glucose, mg/dL 141.0 ± 59.8 
Casual CPR, ng/mL 4.34 (≤0.03–32.30) 
CPR index 3.96 (0.01–34.80) 
Diet therapy, kcal/IBWkg 32.6 ± 3.1 
ESA dosage, units/week 5500 (0–30000) 
Treatment  
 Diet 6 (6.1) 
 OHA 20 (20.4) 
 Insulin ± OHA 56 (57.1) 
 Insulin + GLP-1A 9 (9.2) 
 GLP-1A 7 (7.1) 
Retinopathy  
 None 14 (14.3) 
 Simple 17 (17.3) 
 Preproliferative 4 (4.1) 
 Proliferative 60 (62.1) 
 Unknown 3 (3.1) 
Neuropathy 89 (90.8) 
48-h SGL  
 Mean, mg/dL 150.9 ± 36.9 
 SD, mg/dL 40.1 ± 15.6 
 CV, % 26.7 ± 8.2 
 Maximum, mg/dL 271 ± 54 
 Minimum, mg/dL 77 ± 23 
 Range, mg/dL 193.9 ± 52.7 
 GMI, % 6.9 ± 0.8 
 TIR, % 78.7 (3.6–100) 
 TBR70, % 0 (0–22.5) 
 TBR54, % 0 (0–9.5) 
 TAR, % 19.5 (0–96.4) 
HD day, CGM data  
 Mean, mg/dL 150.5 ± 32.9 
 SD, mg/dL 42.0 ± 17.3 
 CV, % 27.9 ± 9.3 
 Maximum, mg/dL 249 ± 58 
 Minimum, mg/dL 86 ± 26 
 Range, mg/dL 163.3 ± 57.7 
 Start of HD, mg/dL 177.6 ± 53.8 
 End of HD, mg/dL 119.2 ± 44.0 
 Change during HD, mg/dL −61 (−225 to 164) 
 TIR, % 77.4 (1.4–100) 
 TBR70, % 0 (0–28.0) 
 TBR54, % 0 (0–16.6) 
 TAR, % 18.2 (0–98.6) 
Non-HD day, CGM data  
 Mean, mg/dL 151.0 ± 35.9 
 SD, mg/dL 38.1 ± 14.2 
 CV, % 25.4 ± 8.1 
 Maximum, mg/dL 241 ± 57 
 Minimum, mg/dL 92 ± 31 
 Range, mg/dL 149.4 ± 52.5 
 TIR, % 79.9 (0–100) 
 TBR70, % 0 (0–23.2) 
 TBR54, % 0 (0–2.4) 
 TAR, % 17.8 (0–100) 
All-cause hypoglycemia 34 (34.7) 
HD-related hypoglycemia 21 (21.4) 
Post-HD hypoglycemia 20 (20.4) 
Hypoglycemia during HD 14 (14.3) 
DemographicValue
Male sex 68 (69.4) 
Age, years 62 ± 12 
Dry weight, kg 63.7 ± 14.5 
BMI, kg/m2 23.8 ± 4.5 
Duration of diabetes, years 20.5 (1–58) 
Duration of hemodialysis, months 2.0 (0.5–171.0) 
Dialysate glucose concentration, mg/dL  
 100 65 (66.3) 
 125 24 (24.5) 
 150 9 (9.2) 
HD time  
 Morning 82 (83.7) 
 Afternoon 11 (11.2) 
 Evening 5 (5.1) 
Blood flow, mL/min 198.9 ± 20.1 
Dialysis fluid flow, mL/min 453.1 ± 88.8 
Dialysis membrane area, m2 1.78 ± 0.29 
Mode of hemodialysis  
 HD 93 (94.9) 
 HDF 5 (5.1) 
Red blood cell count, × 104/μL) 353.4 ± 57.6 
Hemoglobin concentration, g/dL 10.6 ± 1.6 
Hematocrit, % 33.0 ± 5.2 
Serum albumin, g/dL 3.6 ± 0.5 
HbA1c  
 % 6.4 ± 1.2 
 mmol/mol 45.9 ± 12.7 
GA, % 20.8 ± 6.8 
GA-to-HbA1c ratio, % 3.27 ± 0.82 
Casual plasma glucose, mg/dL 141.0 ± 59.8 
Casual CPR, ng/mL 4.34 (≤0.03–32.30) 
CPR index 3.96 (0.01–34.80) 
Diet therapy, kcal/IBWkg 32.6 ± 3.1 
ESA dosage, units/week 5500 (0–30000) 
Treatment  
 Diet 6 (6.1) 
 OHA 20 (20.4) 
 Insulin ± OHA 56 (57.1) 
 Insulin + GLP-1A 9 (9.2) 
 GLP-1A 7 (7.1) 
Retinopathy  
 None 14 (14.3) 
 Simple 17 (17.3) 
 Preproliferative 4 (4.1) 
 Proliferative 60 (62.1) 
 Unknown 3 (3.1) 
Neuropathy 89 (90.8) 
48-h SGL  
 Mean, mg/dL 150.9 ± 36.9 
 SD, mg/dL 40.1 ± 15.6 
 CV, % 26.7 ± 8.2 
 Maximum, mg/dL 271 ± 54 
 Minimum, mg/dL 77 ± 23 
 Range, mg/dL 193.9 ± 52.7 
 GMI, % 6.9 ± 0.8 
 TIR, % 78.7 (3.6–100) 
 TBR70, % 0 (0–22.5) 
 TBR54, % 0 (0–9.5) 
 TAR, % 19.5 (0–96.4) 
HD day, CGM data  
 Mean, mg/dL 150.5 ± 32.9 
 SD, mg/dL 42.0 ± 17.3 
 CV, % 27.9 ± 9.3 
 Maximum, mg/dL 249 ± 58 
 Minimum, mg/dL 86 ± 26 
 Range, mg/dL 163.3 ± 57.7 
 Start of HD, mg/dL 177.6 ± 53.8 
 End of HD, mg/dL 119.2 ± 44.0 
 Change during HD, mg/dL −61 (−225 to 164) 
 TIR, % 77.4 (1.4–100) 
 TBR70, % 0 (0–28.0) 
 TBR54, % 0 (0–16.6) 
 TAR, % 18.2 (0–98.6) 
Non-HD day, CGM data  
 Mean, mg/dL 151.0 ± 35.9 
 SD, mg/dL 38.1 ± 14.2 
 CV, % 25.4 ± 8.1 
 Maximum, mg/dL 241 ± 57 
 Minimum, mg/dL 92 ± 31 
 Range, mg/dL 149.4 ± 52.5 
 TIR, % 79.9 (0–100) 
 TBR70, % 0 (0–23.2) 
 TBR54, % 0 (0–2.4) 
 TAR, % 17.8 (0–100) 
All-cause hypoglycemia 34 (34.7) 
HD-related hypoglycemia 21 (21.4) 
Post-HD hypoglycemia 20 (20.4) 
Hypoglycemia during HD 14 (14.3) 

Data are presented as n (%), mean ± SD, or median (minimum to maximum). CPR, C-peptide immunoreactivity; CPR index, CPR-to–plasma glucose ratio (CPR/plasma glucose × 100); ESA, erythropoietin-stimulating agent; GLP-1A, glucagon-like peptide 1 agonist; HDF, hemodiafiltration; IBW, ideal body weight (height [m]2 × 22); OHA, oral hypoglycemic agent.

Glycemic Profiles on HD and Non-HD Days

The complete glycemic profiles for HD and non-HD days are shown in Fig. 1. The mean SGL did not significantly differ between the 2 days (150.5 ± 32.9 mg/dL on HD day vs. 151.0 ± 35.9 mg/dL on non-HD day; P = 0.7881). Although the maximum SGL did not significantly differ (249 ± 58 mg/dL on HD day vs. 241 ± 57 mg/dL on non-HD day; P = 0.0822), the minimum SGL was significantly lower on the HD day than on the non-HD day (86 ± 26 mg/dL on HD day vs. 92 ± 31 mg/dL on non-HD day; P = 0.0068). The SD and CV were significantly higher on the HD day than on the non-HD day (42.0 ± 17.3 mg/dL and 27.9 ± 9.3% on HD day vs. 38.1 ± 14.2 mg/dL and 25.4 ± 8.1% on non-HD day; P = 0.0042 and P = 0.0020, respectively), and the range of SGL was significantly higher on the HD day than on the non-HD day (163.3 ± 57.7 vs. 149.4 ± 52.5 mg/dL; P = 0.0045). TIR, TBR70, and TAR did not significantly differ between the 2 days (77.4% [1.4–100%], 0% [0–28.0%], and 18.2% [0–98.6%] on HD day vs. 79.9% [0–100%], 0% [0–23.2%], and 17.8% [0–100%] on non-HD day; P = 0.6901, P = 0.1471, and P = 0.7907, respectively), but TBR54 was significantly higher on the HD day than on the non-HD day (0% [0–16.6%] vs. 0% [0–2.4%]; P = 0.0299).

Figure 1

Continuous glucose monitoring of participants over 24 h. Mean ± SD SGL on HD day (A) and non-HD day (B).

Figure 1

Continuous glucose monitoring of participants over 24 h. Mean ± SD SGL on HD day (A) and non-HD day (B).

Close modal

The univariate linear regression analysis showed that the mean 48-h SGL was significantly correlated with the duration of diabetes, HbA1c, GA, and GA-to-HbA1c ratio (r = 0.23, P = 0.0235; r = 0.69, P < 0.0001; r = 0.53, P < 0.0001; and r = 0.20, P = 0.0435, respectively), whereas multiple linear regression analyses of these items revealed that the mean 48-h SGL significantly correlated with HbA1c, GA, and GA-to-HbA1c ratio (r2 = 0.48, β = 1.49, F = 30.26, P < 0.0001; r2 = 0.57, β = −1.37, F = 9.11, P = 0.0033; and r2 = 0.53, β = 1.34, F = 12.61, P = 0.0006, respectively) (Supplementary Table 1). The univariate analysis showed that 48-h SD was significantly correlated with HbA1c, GA, and GA-to-HbA1c ratio (r = 0.51, P < 0.0001; r = 0.53, P < 0.0001; and r = 0.29, P = 0.0035, respectively), and multivariate analysis of these items revealed that 48-h SD significantly correlated with HbA1c and GA (r2 = 0.35, β = 0.37, F = 9.15, P = 0.0032 and r2 = 0.29, β = 0.24, F = 13.41, P < 0.0001, respectively) (Supplementary Table 1). The univariate analysis showed that 48-h CV significantly correlated with GA and GA-to-HbA1c ratio (r = 0.23, P = 0.207 and r = 0.20, P = 0487, respectively), and multivariate analysis revealed that 48-h CV significantly correlated with GA only (r2 = 0.05, β = 0.21, F = 5.54, P = 0.0207) (Supplementary Table 1). In addition, HbA1c significantly correlated with GMI (r = 0.69, P < 0.0001), TIR (r = −0.61, P < 0.0001), and TAR (r = 0.69, P < 0.0001), but not with TBR70 (r = −0.14, P = 0.1629) or TBR56 (r = −0.14, P = 0.1742). Similarly, GA significantly correlated with GMI (r = 0.55, P < 0.0001), TIR (r = −0.48, P < 0.0001), and TAR (r = 0.50, P < 0.0001), but not with TBR70 or TBR54, and GA-to-HbA1c ratio correlated with GMI (r = 0.20, P = 0.0298) and TAR (r = 0.21, P = 0.0426). Furthermore, HbA1c significantly correlated with the mean SGL of HD and non-HD days (r = 0.66, P < 0.0001 and r = 0.64, P < 0.0001, respectively) and with SD of HD and non-HD days (r = 0.43, P < 0.0001 and r = 0.48, P < 0.0001, respectively), but not with CV of HD and non-HD days (r = 0.10, P = 0.3278 and r = 0.05, P = 0.6590, respectively). GA significantly correlated with the mean SGL of HD and non-HD days (r = 0.53, P < 0.0001 and r = 0.48, P < 0.0001, respectively), with SD of HD and non-HD days (r = 0.54, P < 0.0001 and r = 0.42, P < 0.0001, respectively), and with CV of the HD day (r = 0.23, P = 0.0207), but not with CV of the non-HD day (r = 0.19, P = 0.0594).

Prevalence of Hypoglycemia and SGL Nadir During Dialysis

Thirty-four (35%) of the 98 participants exhibited all-cause hypoglycemia during their 48-h CGM periods. In addition, 14 (14%) of the 98 showed hypoglycemia during HD, and 20 (20%) showed post-HD hypoglycemia. In total, 21 participants (21%) had HD-related hypoglycemia. All HD-related hypoglycemic episodes were asymptomatic. Furthermore, 18 participants (18%) showed nocturnal hypoglycemia, and four of these episodes (22% of the nocturnal hypoglycemic episodes) were symptomatic. In 14 patients who experienced hypoglycemia during HD, the SGL nadir during HD was 61.1 ± 10.2 mg/dL, and the SGL at the end of HD was 80.0 ± 40.3 mg/dL, which increased thereafter in five participants, and was not below the hypoglycemic threshold (<70 mg/dL) at end of HD in eight participants. In 20 participants who showed post-HD hypoglycemia, the SGL at end of HD was 86.6 ± 34.5 mg/dL, and the SGL was not below the hypoglycemic threshold in 12 of these. The SGL declined after the end of HD in 17 participants, and the SGL nadir after HD was 57.8 ± 9.2 mg/dL.

The SGL nadir was below the dialysate glucose concentration in 49 participants (50%), and this occurred irrespective of the dialysate glucose concentration; it occurred in 31 (48%) of 65 participants treated with dialysate containing 100 mg/dL glucose, in 14 (58%) of 24 treated with 125 mg/dL, and in four (44%) of nine treated with 150 mg/dL. The mean SGL profile during HD and during the post-HD period did not differ between the groups of participants using different dialysate glucose concentrations (Supplementary Fig. 1).

Differences Between Participants Who Did and Did Not Experience HD-Related Hypoglycemia, Post-HD Hypoglycemia, and Hypoglycemia During HD

In comparing the groups of patients with type 2 diabetes undergoing HD who did or did not experience HD-related hypoglycemia and post-HD hypoglycemia, there were no significant differences in sex, age, dry weight, BMI, duration of diabetes, duration of HD, red blood cell count, hemoglobin concentration, or hematocrit (Table 2). Furthermore, there were no significant differences in markers of glucose status (HbA1c, GA, and GA-to-HbA1c ratio) or HD conditions (dialysate glucose concentration, HD time, blood flow, dialysis fluid flow, dialysis membrane area, and HD mode) (Table 2). However, in comparing the groups of participants who did or did not experience hypoglycemia during HD, red blood cell count, hemoglobin concentration, and hematocrit were significantly higher and serum albumin was significantly lower in the group with hypoglycemia during HD versus the group without (P = 0.0116, P = 0.0032, P = 0.0078, and P = 0.0121, respectively) (Table 2).

Table 2

Characteristics of participants who did or did not experience each type of hypoglycemia related to HD

HD-related hypoglycemiaPost-HD hypoglycemiaHypoglycemia during HD
 NoYesPNoYesPNoYesP
N of patients 77 21  78 20  84 14  
Sex, male/female 52/25 16/5 0.4454* 53/25 15/5 0.5416* 57/27 11/3 0.4206* 
Age, years 61 ± 12 65 ± 12 0.2511 61 ± 12 64 ± 13 0.3399 62 ± 12 64 ± 13 0.5632 
Dry weight, kg 64.4 ± 15.3 61.3 ± 11.0 0.3950 64.2 ± 15.3 61.7 ± 11.2 0.4887 63.7 ± 14.9 63.6 ± 12.2 0.9681 
BMI, kg/m2 24.0 ± 4.7 22.9 ± 3.8 0.3186 24.0 ± 4.7 22.9 ± 3.9 0.3526 23.8 ± 4.6 23.6 ± 4.4 0.8982 
Duration of diabetes, years 20 (4–58) 24 (1–46) 0.9620 20 (4–58) 25 (1–46) 0.9683 21 (1–58) 21 (6–45) 0.6257 
Duration of hemodialysis, months 1.0 (0.5–171.0) 5.0 (0.5–143.0) 0.5795 1.5 (0.5–171.0) 5.0 (0.5–143.0) 0.4360 2.5 (0.5–171.0) 1.0 (0.5–27.0) 0.1920 
Red blood cell count, × 104/μL 350 ± 54 364 ± 71 0.3226 351 ± 53 364 ± 73 0.3521 347 ± 53 389 ± 70 0.0116 
Hemoglobin, g/dL 10.4 ± 1.5 11.2 ± 1.7 0.0587 10.5 ± 1.5 11.2 ± 1.7 0.0723 10.4 ± 1.5 11.7 ± 1.7 0.0032 
Hematocrit, % 32.6 ± 4.9 34.7 ± 5.8 0.0929 32.6 ± 4.9 34.7 ± 6.0 0.1131 32.5 ± 4.8 36.4 ± 6.0 0.0078 
Albumin, g/dL 3.6 ± 0.5 3.5 ± 0.5 0.1198 3.6 ± 0.5 3.4 ± 0.6 0.0910 3.7 ± 0.4 3.3 ± 0.6 0.0121 
HbA1c   0.5342   0.5346   0.5999 
 % 6.4 ± 1.1 6.2 ± 1.2  6.4 ± 1.1 6.2 ± 1.3  6.3 ± 1.1 6.5 ± 1.4  
 mmol/mol 46.4 ± 12.6 44.4 ± 13.4  46.3 ± 12.5 44.4 ± 13.8  45.7 ± 12.3 47.6 ± 15.3  
GA, % 20.9 ± 5.8 20.5 ± 9.6 0.8194 20.8 ± 5.8 20.6 ± 9.9 0.9253 20.7 ± 5.8 21.1 ± 11.2 0.8627 
GA-to-HbA1c ratio 3.3 ± 0.7 3.3 ± 1.1 0.8988 3.3 ± 0.7 3.3 ± 1.1 0.9641 3.3 ± 0.8 3.2 ± 1.2 0.6626 
Casual plasma glucose, mg/dL 139.7 ± 48.2 145.4 ± 90.7 0.7023 139.4 ± 48.0 147.0 ± 92.7 0.6156 139.5 ± 48.6 149.5 ± 105.3 0.5670 
Casual CPR, ng/mL 4.28 (≤0.03–32.20) 4.52 (0.07–26.10) 0.4208 4.58 (≤0.03–32.20) 4.34 (0.07–26.10) 0.7089 4.87 (≤0.03–32.20) 4.34 (0.07–26.10) 0.6398 
CPR index 3.45 (0.01–26.83) 4.65 (0.02–34.80) 0.1148 3.58 (0.01–26.83) 4.65 (0.02–34.80) 0.2628 3.71 (0.01–26.80) 4.64 (1.25–34.80) 0.3454 
Dietary energy, kcal/IBWkg 32.3 ± 3.2 33.5 ± 2.2 0.1148 32.3 ± 3.2 33.7 ± 2.0 0.0707 32.4 ± 3.2 33.7 ± 1.9 0.1435 
ESA dosage, units/week 6,000 (0–30,000) 4,500 (0–24,000) 0.6295 6,000 (0–30,000) 4,500 (0–24,000) 0.9040 5,500 (0–30,000) 5,250 (0–24,000) 0.8896 
Treatment, n          
 Diet/OHA/insulin ± OHA/insulin + GLP-1A/GLP-1A 3/15/46/8/5 3/5/10/1/2 0.3712* 3/16/46/8/5 3/4/10/1/2 0.4574* 5/16/49/8/6 1/4/7/1/1 0.9410* 
Neuropathy 70 (91) 19 (90) 0.9514* 71 (91) 18 (90) 0.8873* 75 (91) 14 (100) 0.1987* 
Retinopathy, n          
 None/simple/prepro/pro/unknown 11/12/4/47/3 3/5/0/13/0 0.6349* 12/12/4/ 47/3 2/5/0/13/0 0.5537* 13/12/4/ 52/3 1/5/0/8/0 0.2903* 
HD time, n          
 Morning/afternoon/evening 68/5/4 14/6/1 0.0175* 69/5/4 13/6/1 0.0114* 71/8/5 11/3/0 0.3013* 
Dialysate glucose concentration, n          
 100/125/150 mg/dL 50/19/8 15/5/1 0.7121* 50/20/8 15/4/1 0.6155* 57/19/8 8/5/1 0.5706* 
Blood flow, mL/min 199 ± 18 197 ± 26 0.5732 199 ± 19 199 ± 24 0.9759 199 ± 19 199 ± 28 0.9514 
Dialysis fluid flow, mL/min 455 ± 90 448 ± 87 0.7530 454 ± 89 450 ± 89 0.8638 460 ± 92 414 ± 53 0.0773 
Dialysis membrane area, m2 1.77 ± 0.28 1.82 ± 0.34 0.5094 1.77 ± 0.28 1.84 ± 0.34 0.3610 1.77 ± 0.28 1.85 ± 0.34 0.3454 
Mode of hemodialysis, n          
 HD/HDF 72/5 21/0 0.2306 73/5 21/0 0.2451 79/5 14/0 0.3487 
48-h SGL data          
 Mean, mg/dL 154.3 ± 37.8 136.8 ± 29.9 0.0458 154.7 ± 37.6 136.5 ± 30.6 0.0482 153.0 ± 38.6 138.6 ± 20.4 0.1762 
 SD, mg/dL 38.1 ± 15.5 47.7 ± 14.0 0.0119 38.3 ± 15.6 47.1 ± 14.2 0.0240 38.4 ± 15.2 50.6 ± 14.8 0.0064 
 CV, % 24.4 ± 7.0 34.9 ± 7.3 <0.0001 24.6 ± 7.2 34.6 ± 7.4 <0.0001 25.1 ± 7.4 36.1 ± 6.9 <0.0001 
 Maximum, mg/dL 247.7 ± 60.1 253.6 ± 49.7 0.6811 271.2 ± 51.3 271.0 ± 65.6 0.9885 273.0 ± 53.1 259.9 ± 60.6 0.4028 
 Minimum, mg/dL 94.0 ± 22.0 54.8 ± 8.6 <0.0001 81.7 ± 22.2 59.9 ± 14.7 <0.0001 79.8 ± 22.8 62.1 ± 14.4 0.0062 
 Range, mg/dL 153.7 ± 55.6 198.8 ± 52.3 0.0012 189.5 ± 47.6 211.1 ± 67.7 0.1017 193.3 ± 51.9 197.8 ± 59.1 0.7668 
 GMI, % 7.0 ± 0.8 6.6 ± 0.6 0.0388 7.0 ± 0.8 6.6 ± 0.6 0.0413 7.0 ± 0.8 6.7 ± 0.5 0.2773 
 TIR, % 79.0 (3.6–100) 76.1 (41.9–97.2) 0.3565 79.0 (3.6–100) 76.3 (41.9–97.2) 0.4098 79.0 (3.6–100) 76.3 (54.2–90.1) 0.3881 
 TBR70, % 0 (0–9.7) 3.8 (0.3–22.5) <0.0001 0 (0–9.7) 3.7 (0.3–22.5) <0.0001 0 (0–22.5) 3.7 (0.3–17.9) <0.0001 
 TBR54, % 0 (0–0.5) 0 (0–9.5) <0.0001 0 (0–4.2) 0 (0–9.5) <0.0001 0 (0–2.3) 0 (0–9.5) <0.0001 
 TAR, % 20.3 (0–96.4) 15.4 (0–57.2) 0.5077 20.4 (0–96.4) 14.9 (0–57.2) 0.4643 19.5 (0–96.4) 20.3 (0–40.7) 0.9232 
SGL data for HD day          
 Mean, mg/dL 152.7 ± 34.2 131.4 ± 17.7 0.0070 152.7 ± 33.9 130.4 ± 17.6 0.0057 150.1 ± 34.0 136.5 ± 18.3 0.1500 
 SD, mg/dL 40.2 ± 18.3 51.1 ± 14.5 0.0134 40.6 ± 18.6 49.9 ± 13.8 0.0399 40.6 ± 17.9 53.8 ± 15.3 0.0111 
 CV, % 26.0 ± 9.2 38.6 ± 7.7 <0.0001 26.3 ± 9.5 38.0 ± 7.5 <0.0001 27.0 ± 9.5 39.1 ± 8.3 <0.0001 
 Maximum, mg/dL 247.7 ± 60.1 253.6 ± 49.7 0.6811 249.2 ± 61.1 248.2 ± 44.3 0.9596 246.7 ± 58.1 262.7 ± 56.3 0.3396 
 Minimum, mg/dL 94.0 ± 22.0 54.8 ± 8.6 <0.0001 93.3 ± 22.7 55.5 ± 8.1 <0.0001 90.9 ± 23.5 53.7 ± 9.2 <0.0001 
 Range, mg/dL 153.7 ± 55.6 198.8 ± 52.3 0.0012 155.8 ± 58.4 192.7 ± 45.4 0.0101 155.7 ± 54.4 209.0 ± 58.2 0.0011 
 Start of HD, mg/dL 179.9 ± 53.4 169.1 ± 55.8 0.4175 179.9 ± 53.0 168.5 ± 57.1 0.4001 178.1 ± 53.3 174.3 ± 58.3 0.8095 
 End of HD, mg/dL 128.7 ± 41.4 84.4 ± 35.1 <0.0001 127.6 ± 42.4 86.6 ± 34.5 0.0001 125.7 ± 41.3 80.0 ± 40.3 0.0002 
 Change during HD, mg/dL −57 (−225 to 164) −71 (−205 to 49) 0.0568 −59 (−225 to 164) −70 (−205 to 49) 0.1096 −60 (−225 to 164) −87 (−205 to 16) 0.0570 
 TIR, % 78.9 (1.4–100) 73.0 (42.2–94.5) 0.1880 78.9 (1.4–100) 74.4 (42.2–94.5) 0.2356 78.9 (1.4–100) 72.5 (42.2–88.2) 0.1408 
 TBR70, % 0 (0–16.6) 6.6 (0.7–28.0) <0.0001 0 (0–16.6) 6.6 (0.7–28.0) <0.0001 0 (0–21.8) 7.1 (0.7–28.0) <0.0001 
 TBR54, % 0 (0–2.8) 0 (0–16.6) 0.0201 0 (0–2.8) 0 (0–16.6) 0.0143 0 (0–8.3) 0 (0–16.6) 0.1292 
 TAR, % 21.1 (0–98.6) 16.6 (0–56.4) 0.6001 21.1 (0–98.6) 16.3 (0–56.4) 0.5604 17.8 (0–98.6) 18.9 (0–56.4) 0.8310 
SGL data for non-HD day          
 Mean, mg/dL 156.9 ± 44.3 142.1 ± 51.3 0.1924 156.6 ± 44.1 142.5 ± 52.6 0.2221 155.9 ± 47.8 140.6 ± 31.9 0.2507 
 SD, mg/dL 36.0 ± 16.5 44.3 ± 16.8 0.0448 36.1 ± 16.4 44.4 ± 17.2 0.0475 36.2 ± 16.3 47.4 ± 17.1 0.0202 
 CV, % 22.9 ± 7.2 32.2 ± 10.2 <0.0001 23.0 ± 7.2 32.2 ± 10.5 <0.0001 23.4 ± 8.0 33.6 ± 8.4 <0.0001 
 Maximum, mg/dL 239.7 ± 53.4 245.2 ± 70.0 0.6967 239.4 ± 53.1 246.4 ± 71.6 0.6287 240.1 ± 55.9 245.4 ± 65.5 0.7500 
 Minimum, mg/dL 97.6 ± 31.1 69.2 ± 16.4 0.0001 97.3 ± 31.0 68.9 ± 16.7 0.0002 95.0 ± 31.4 70.7 ± 15.5 0.0057 
 Range, mg/dL 142.1 ± 45.2 176.0 ± 68.3 0.0080 142.1 ± 44.9 177.5 ± 69.7 0.0066 145.1 ± 49.6 174.7 ± 63.6 0.0507 
 TIR, % 82.0 (0–100) 78.9 (25.6–100) 0.7160 81.9 (0–100) 78.0 (25.6–100) 0.7309 81.4 (0–100) 77.2 (42.9–95.2) 0.8430 
 TBR70, % 0 (0–8.0) 0.3 (0–23.2) <0.0001 0 (0–8.0) 0.5 (0–23.2) <0.0001 0 (0–23.2) 0.2 (0–11.1) 0.0066 
 TBR54, % 0 (0–1.0) 0 (0–2.4) 0.0011 0 (0–1.0) 0 (0–2.4) 0.0007 0 (0–1.7) 0 (0–2.4) 0.0961 
 TAR, % 17.6 (0–100) 20.1 (0–74.4) 0.4056 17.8 (0–100) 16.5 (0–74.4) 0.3730 17.5 (0–100) 21.8 (0–56.7) 0.8390 
HD-related hypoglycemiaPost-HD hypoglycemiaHypoglycemia during HD
 NoYesPNoYesPNoYesP
N of patients 77 21  78 20  84 14  
Sex, male/female 52/25 16/5 0.4454* 53/25 15/5 0.5416* 57/27 11/3 0.4206* 
Age, years 61 ± 12 65 ± 12 0.2511 61 ± 12 64 ± 13 0.3399 62 ± 12 64 ± 13 0.5632 
Dry weight, kg 64.4 ± 15.3 61.3 ± 11.0 0.3950 64.2 ± 15.3 61.7 ± 11.2 0.4887 63.7 ± 14.9 63.6 ± 12.2 0.9681 
BMI, kg/m2 24.0 ± 4.7 22.9 ± 3.8 0.3186 24.0 ± 4.7 22.9 ± 3.9 0.3526 23.8 ± 4.6 23.6 ± 4.4 0.8982 
Duration of diabetes, years 20 (4–58) 24 (1–46) 0.9620 20 (4–58) 25 (1–46) 0.9683 21 (1–58) 21 (6–45) 0.6257 
Duration of hemodialysis, months 1.0 (0.5–171.0) 5.0 (0.5–143.0) 0.5795 1.5 (0.5–171.0) 5.0 (0.5–143.0) 0.4360 2.5 (0.5–171.0) 1.0 (0.5–27.0) 0.1920 
Red blood cell count, × 104/μL 350 ± 54 364 ± 71 0.3226 351 ± 53 364 ± 73 0.3521 347 ± 53 389 ± 70 0.0116 
Hemoglobin, g/dL 10.4 ± 1.5 11.2 ± 1.7 0.0587 10.5 ± 1.5 11.2 ± 1.7 0.0723 10.4 ± 1.5 11.7 ± 1.7 0.0032 
Hematocrit, % 32.6 ± 4.9 34.7 ± 5.8 0.0929 32.6 ± 4.9 34.7 ± 6.0 0.1131 32.5 ± 4.8 36.4 ± 6.0 0.0078 
Albumin, g/dL 3.6 ± 0.5 3.5 ± 0.5 0.1198 3.6 ± 0.5 3.4 ± 0.6 0.0910 3.7 ± 0.4 3.3 ± 0.6 0.0121 
HbA1c   0.5342   0.5346   0.5999 
 % 6.4 ± 1.1 6.2 ± 1.2  6.4 ± 1.1 6.2 ± 1.3  6.3 ± 1.1 6.5 ± 1.4  
 mmol/mol 46.4 ± 12.6 44.4 ± 13.4  46.3 ± 12.5 44.4 ± 13.8  45.7 ± 12.3 47.6 ± 15.3  
GA, % 20.9 ± 5.8 20.5 ± 9.6 0.8194 20.8 ± 5.8 20.6 ± 9.9 0.9253 20.7 ± 5.8 21.1 ± 11.2 0.8627 
GA-to-HbA1c ratio 3.3 ± 0.7 3.3 ± 1.1 0.8988 3.3 ± 0.7 3.3 ± 1.1 0.9641 3.3 ± 0.8 3.2 ± 1.2 0.6626 
Casual plasma glucose, mg/dL 139.7 ± 48.2 145.4 ± 90.7 0.7023 139.4 ± 48.0 147.0 ± 92.7 0.6156 139.5 ± 48.6 149.5 ± 105.3 0.5670 
Casual CPR, ng/mL 4.28 (≤0.03–32.20) 4.52 (0.07–26.10) 0.4208 4.58 (≤0.03–32.20) 4.34 (0.07–26.10) 0.7089 4.87 (≤0.03–32.20) 4.34 (0.07–26.10) 0.6398 
CPR index 3.45 (0.01–26.83) 4.65 (0.02–34.80) 0.1148 3.58 (0.01–26.83) 4.65 (0.02–34.80) 0.2628 3.71 (0.01–26.80) 4.64 (1.25–34.80) 0.3454 
Dietary energy, kcal/IBWkg 32.3 ± 3.2 33.5 ± 2.2 0.1148 32.3 ± 3.2 33.7 ± 2.0 0.0707 32.4 ± 3.2 33.7 ± 1.9 0.1435 
ESA dosage, units/week 6,000 (0–30,000) 4,500 (0–24,000) 0.6295 6,000 (0–30,000) 4,500 (0–24,000) 0.9040 5,500 (0–30,000) 5,250 (0–24,000) 0.8896 
Treatment, n          
 Diet/OHA/insulin ± OHA/insulin + GLP-1A/GLP-1A 3/15/46/8/5 3/5/10/1/2 0.3712* 3/16/46/8/5 3/4/10/1/2 0.4574* 5/16/49/8/6 1/4/7/1/1 0.9410* 
Neuropathy 70 (91) 19 (90) 0.9514* 71 (91) 18 (90) 0.8873* 75 (91) 14 (100) 0.1987* 
Retinopathy, n          
 None/simple/prepro/pro/unknown 11/12/4/47/3 3/5/0/13/0 0.6349* 12/12/4/ 47/3 2/5/0/13/0 0.5537* 13/12/4/ 52/3 1/5/0/8/0 0.2903* 
HD time, n          
 Morning/afternoon/evening 68/5/4 14/6/1 0.0175* 69/5/4 13/6/1 0.0114* 71/8/5 11/3/0 0.3013* 
Dialysate glucose concentration, n          
 100/125/150 mg/dL 50/19/8 15/5/1 0.7121* 50/20/8 15/4/1 0.6155* 57/19/8 8/5/1 0.5706* 
Blood flow, mL/min 199 ± 18 197 ± 26 0.5732 199 ± 19 199 ± 24 0.9759 199 ± 19 199 ± 28 0.9514 
Dialysis fluid flow, mL/min 455 ± 90 448 ± 87 0.7530 454 ± 89 450 ± 89 0.8638 460 ± 92 414 ± 53 0.0773 
Dialysis membrane area, m2 1.77 ± 0.28 1.82 ± 0.34 0.5094 1.77 ± 0.28 1.84 ± 0.34 0.3610 1.77 ± 0.28 1.85 ± 0.34 0.3454 
Mode of hemodialysis, n          
 HD/HDF 72/5 21/0 0.2306 73/5 21/0 0.2451 79/5 14/0 0.3487 
48-h SGL data          
 Mean, mg/dL 154.3 ± 37.8 136.8 ± 29.9 0.0458 154.7 ± 37.6 136.5 ± 30.6 0.0482 153.0 ± 38.6 138.6 ± 20.4 0.1762 
 SD, mg/dL 38.1 ± 15.5 47.7 ± 14.0 0.0119 38.3 ± 15.6 47.1 ± 14.2 0.0240 38.4 ± 15.2 50.6 ± 14.8 0.0064 
 CV, % 24.4 ± 7.0 34.9 ± 7.3 <0.0001 24.6 ± 7.2 34.6 ± 7.4 <0.0001 25.1 ± 7.4 36.1 ± 6.9 <0.0001 
 Maximum, mg/dL 247.7 ± 60.1 253.6 ± 49.7 0.6811 271.2 ± 51.3 271.0 ± 65.6 0.9885 273.0 ± 53.1 259.9 ± 60.6 0.4028 
 Minimum, mg/dL 94.0 ± 22.0 54.8 ± 8.6 <0.0001 81.7 ± 22.2 59.9 ± 14.7 <0.0001 79.8 ± 22.8 62.1 ± 14.4 0.0062 
 Range, mg/dL 153.7 ± 55.6 198.8 ± 52.3 0.0012 189.5 ± 47.6 211.1 ± 67.7 0.1017 193.3 ± 51.9 197.8 ± 59.1 0.7668 
 GMI, % 7.0 ± 0.8 6.6 ± 0.6 0.0388 7.0 ± 0.8 6.6 ± 0.6 0.0413 7.0 ± 0.8 6.7 ± 0.5 0.2773 
 TIR, % 79.0 (3.6–100) 76.1 (41.9–97.2) 0.3565 79.0 (3.6–100) 76.3 (41.9–97.2) 0.4098 79.0 (3.6–100) 76.3 (54.2–90.1) 0.3881 
 TBR70, % 0 (0–9.7) 3.8 (0.3–22.5) <0.0001 0 (0–9.7) 3.7 (0.3–22.5) <0.0001 0 (0–22.5) 3.7 (0.3–17.9) <0.0001 
 TBR54, % 0 (0–0.5) 0 (0–9.5) <0.0001 0 (0–4.2) 0 (0–9.5) <0.0001 0 (0–2.3) 0 (0–9.5) <0.0001 
 TAR, % 20.3 (0–96.4) 15.4 (0–57.2) 0.5077 20.4 (0–96.4) 14.9 (0–57.2) 0.4643 19.5 (0–96.4) 20.3 (0–40.7) 0.9232 
SGL data for HD day          
 Mean, mg/dL 152.7 ± 34.2 131.4 ± 17.7 0.0070 152.7 ± 33.9 130.4 ± 17.6 0.0057 150.1 ± 34.0 136.5 ± 18.3 0.1500 
 SD, mg/dL 40.2 ± 18.3 51.1 ± 14.5 0.0134 40.6 ± 18.6 49.9 ± 13.8 0.0399 40.6 ± 17.9 53.8 ± 15.3 0.0111 
 CV, % 26.0 ± 9.2 38.6 ± 7.7 <0.0001 26.3 ± 9.5 38.0 ± 7.5 <0.0001 27.0 ± 9.5 39.1 ± 8.3 <0.0001 
 Maximum, mg/dL 247.7 ± 60.1 253.6 ± 49.7 0.6811 249.2 ± 61.1 248.2 ± 44.3 0.9596 246.7 ± 58.1 262.7 ± 56.3 0.3396 
 Minimum, mg/dL 94.0 ± 22.0 54.8 ± 8.6 <0.0001 93.3 ± 22.7 55.5 ± 8.1 <0.0001 90.9 ± 23.5 53.7 ± 9.2 <0.0001 
 Range, mg/dL 153.7 ± 55.6 198.8 ± 52.3 0.0012 155.8 ± 58.4 192.7 ± 45.4 0.0101 155.7 ± 54.4 209.0 ± 58.2 0.0011 
 Start of HD, mg/dL 179.9 ± 53.4 169.1 ± 55.8 0.4175 179.9 ± 53.0 168.5 ± 57.1 0.4001 178.1 ± 53.3 174.3 ± 58.3 0.8095 
 End of HD, mg/dL 128.7 ± 41.4 84.4 ± 35.1 <0.0001 127.6 ± 42.4 86.6 ± 34.5 0.0001 125.7 ± 41.3 80.0 ± 40.3 0.0002 
 Change during HD, mg/dL −57 (−225 to 164) −71 (−205 to 49) 0.0568 −59 (−225 to 164) −70 (−205 to 49) 0.1096 −60 (−225 to 164) −87 (−205 to 16) 0.0570 
 TIR, % 78.9 (1.4–100) 73.0 (42.2–94.5) 0.1880 78.9 (1.4–100) 74.4 (42.2–94.5) 0.2356 78.9 (1.4–100) 72.5 (42.2–88.2) 0.1408 
 TBR70, % 0 (0–16.6) 6.6 (0.7–28.0) <0.0001 0 (0–16.6) 6.6 (0.7–28.0) <0.0001 0 (0–21.8) 7.1 (0.7–28.0) <0.0001 
 TBR54, % 0 (0–2.8) 0 (0–16.6) 0.0201 0 (0–2.8) 0 (0–16.6) 0.0143 0 (0–8.3) 0 (0–16.6) 0.1292 
 TAR, % 21.1 (0–98.6) 16.6 (0–56.4) 0.6001 21.1 (0–98.6) 16.3 (0–56.4) 0.5604 17.8 (0–98.6) 18.9 (0–56.4) 0.8310 
SGL data for non-HD day          
 Mean, mg/dL 156.9 ± 44.3 142.1 ± 51.3 0.1924 156.6 ± 44.1 142.5 ± 52.6 0.2221 155.9 ± 47.8 140.6 ± 31.9 0.2507 
 SD, mg/dL 36.0 ± 16.5 44.3 ± 16.8 0.0448 36.1 ± 16.4 44.4 ± 17.2 0.0475 36.2 ± 16.3 47.4 ± 17.1 0.0202 
 CV, % 22.9 ± 7.2 32.2 ± 10.2 <0.0001 23.0 ± 7.2 32.2 ± 10.5 <0.0001 23.4 ± 8.0 33.6 ± 8.4 <0.0001 
 Maximum, mg/dL 239.7 ± 53.4 245.2 ± 70.0 0.6967 239.4 ± 53.1 246.4 ± 71.6 0.6287 240.1 ± 55.9 245.4 ± 65.5 0.7500 
 Minimum, mg/dL 97.6 ± 31.1 69.2 ± 16.4 0.0001 97.3 ± 31.0 68.9 ± 16.7 0.0002 95.0 ± 31.4 70.7 ± 15.5 0.0057 
 Range, mg/dL 142.1 ± 45.2 176.0 ± 68.3 0.0080 142.1 ± 44.9 177.5 ± 69.7 0.0066 145.1 ± 49.6 174.7 ± 63.6 0.0507 
 TIR, % 82.0 (0–100) 78.9 (25.6–100) 0.7160 81.9 (0–100) 78.0 (25.6–100) 0.7309 81.4 (0–100) 77.2 (42.9–95.2) 0.8430 
 TBR70, % 0 (0–8.0) 0.3 (0–23.2) <0.0001 0 (0–8.0) 0.5 (0–23.2) <0.0001 0 (0–23.2) 0.2 (0–11.1) 0.0066 
 TBR54, % 0 (0–1.0) 0 (0–2.4) 0.0011 0 (0–1.0) 0 (0–2.4) 0.0007 0 (0–1.7) 0 (0–2.4) 0.0961 
 TAR, % 17.6 (0–100) 20.1 (0–74.4) 0.4056 17.8 (0–100) 16.5 (0–74.4) 0.3730 17.5 (0–100) 21.8 (0–56.7) 0.8390 

Data are presented as n (%), mean ± SD, or median (minimum to maximum), unless indicated otherwise. Data were analyzed using unpaired Student t test, unless indicated otherwise. Bold font indicates significance. CPR, C-peptide immunoreactivity; CPR index, CPR-to–plasma glucose ratio (CPR/plasma glucose × 100); ESA, erythropoietin-stimulating agent; GLP-1A, glucagon-like peptide 1 agonist; HDF, hemodiafiltration; IBW, ideal body weight (height [m]2 × 22); OHA, oral hypoglycemic agent; prepro, preproliferative; pro, proliferative.

*

Fisher exact test.

Mann-Whitney U test.

In the 48-h CGM data, the mean SGL and GMI were significantly lower and the SD, CV, and range of SGL were significantly higher in participants who experienced HD-related hypoglycemia than in those who did not. The results were similar for groups with and without post-HD hypoglycemia and with and without hypoglycemia after HD, but the range of SGL was not significantly different between patients with and without post-HD hypoglycemia, and the mean SGL, GMI, and range of SGL were not significantly different between patients with and without hypoglycemia during HD. Although TIR and TAR did not differ between the groups with and without each type of hypoglycemia related to HD, TBR70 and TBR54 were significantly higher in participants who experienced each type of hypoglycemia related to HD than in those who did not (Table 2). On the HD day, the mean SD, CV, and range of SGL significantly differed between the groups who did and did not experience HD-related hypoglycemia and post-HD hypoglycemia, and the SD, CV, and range of SGL significantly differed between the groups who did and did not experience hypoglycemia during HD. The SGL at the start of HD did not significantly differ between the two groups with and without each type of hypoglycemia related to HD, but the SGL at the end of HD was significantly lower in those who experienced each type of hypoglycemia related to HD than in those who did not. Furthermore, there were significant differences in TBR70 and TBR54, but not in TIR or TAR, on the HD day between the groups who did and did not experience each type of hypoglycemia related to HD (Table 2). On the non-HD day, the SD, CV, and TBR70 were significantly higher in participants who experienced each type of hypoglycemia related to HD than in those who did not. The range of SGL and TBR54 on the non-HD day were significantly different only between the groups who did and did not experience HD-related hypoglycemia and post-HD hypoglycemia (Table 2).

When examined with regard to diabetes treatment details, 14 (19.7%) of 71 patients treated with insulin, sulfonylureas, and/or glinides and seven (25.9%) of 27 patients treated with diet and medications that do not cause hypoglycemia experienced HD-related hypoglycemia, with no significant difference in the frequency of hypoglycemia (P = 0.5034). Only patients receiving insulin therapy were analyzed for differences in insulin doses between HD and non-HD days; four (10.5%) of 38 patients receiving the same doses of insulin, two (16.7%) of 12 treated with higher insulin doses on the HD day, and five (25.0%) of 20 treated with higher insulin doses on the non-HD day experienced HD-related hypoglycemia, and there was no difference in the hypoglycemia frequency (P = 0.3530). In addition, regarding basal, bolus, and total daily insulin dosages on the HD day, there was no significant difference between patients with and without HD-related hypoglycemia.

Glycemic profiles on the HD day were shown in participants undergoing HD in the morning, afternoon, and night (Supplementary Fig. 2). In a comparison of the mean SGL on the HD day between patients undergoing HD in the morning who did or did not experience HD-related hypoglycemia, the nocturnal SGL did not significantly differ between the two groups. However, the mean SGL in participants who experienced HD-related hypoglycemia was significantly lower than that in those who did not during the period between 1135 and 1405 h (P < 0.05) (i.e., from the middle of dialysis to after the end of HD). In a comparison of the mean SGL on the HD day between patients undergoing HD in the afternoon who did or did not experience HD-related hypoglycemia, the mean SGL in participants who experienced HD-related hypoglycemia was significantly lower than that in those who did not during the period between 0235 and 0350 h (P < 0.05) and between 1545 and 1920 h (P < 0.05 for 1545 to 1750 h, P < 0.01 for 1755 to 1840 h, and P < 0.05 for 1845 to 1920 h), (i.e., natural and from the middle of dialysis to after the end of HD). Comparison of the mean SGL in participants undergoing night HD was impossible because of the small number of cases.

Figure 2

Comparison of time courses of changes in SGL during and after HD between patients who did and did not experience HD-related hypoglycemia. Mean ± SD SGL during 5 h from start of HD in 21 patients who showed HD-related hypoglycemia (A) and 77 who did not (B). All patients underwent HD for 4 h and ate within 0.5–1 h of end of HD.

Figure 2

Comparison of time courses of changes in SGL during and after HD between patients who did and did not experience HD-related hypoglycemia. Mean ± SD SGL during 5 h from start of HD in 21 patients who showed HD-related hypoglycemia (A) and 77 who did not (B). All patients underwent HD for 4 h and ate within 0.5–1 h of end of HD.

Close modal

In the comparison of the mean SGL profile on the non-HD day between participants with and without HD-related hypoglycemia (Supplementary Fig. 3), the mean SGL in participants with HD-related hypoglycemia was significantly lower than that in those without during the night, between 0135 and 0350 h (P < 0.05 for 0135 to 0205 h, P < 0.01 for 0210 to 0250 h, P < 0.005 for 0255 to 0335 h, P < 0.01 for 0340 to 0410 h, and P < 0.05 for 0415 to 0430 h), and between 2305 and 2400 h (P < 0.05). A comparison of the mean SGL profile during the 5-h period after starting HD between participants who did and did not experience HD-related hypoglycemia is shown in Fig. 2. The mean SGL in participants who did experience HD-related hypoglycemia was significantly lower between 1 h 20 min and 5 h after the start of HD than that in those who did not experience HD-related hypoglycemia.

In the current study, we investigated glucose variability and hypoglycemia using CGM in 98 patients with type 2 diabetes undergoing HD. First, the mean glucose level did not differ between HD and non-HD days, but the glycemic variability was greater on the HD day than on the non-HD day. Previous studies provided no consensus on mean glucose levels or glycemic variability on HD and non-HD days (16,2224). One reason is the small number of cases in these studies; we were able to evaluate the difference in glycemic dynamics in more patients.

Second, mean glucose level correlated with both HbA1c and GA, but glycemic variability correlated with GA only. Similarly, we previously reported that HbA1c correlated with the mean SGL, more closely than GA did. In contrast, although GA itself reflected the mean SGL without correction, it did so less accurately than HbA1c, but it could serve as an indicator of glycemic variability (16). We also demonstrated previously that the GA-to-HbA1c ratio reliably reflects glycemic fluctuation in patients with diabetes without end-stage kidney disease (30). A recent study showed that HbA1c is no more variable and less biased than GA in patients with type 2 diabetes and mild renal dysfunction not undergoing HD (31). Our results support that HbA1c reflects mean glucose level and GA reflects glycemic fluctuations in the population of patients with type 2 diabetes undergoing HD. Additionally, the current study is the first to examine the relationship between existing glycemic markers and the new standardized CGM metrics in participants with diabetes undergoing HD. It was shown that HbA1c and GA were correlated with GMI, TIR, and TAR, but not with TBR, in these patients.

Third, the most important finding in this study was that >20% of the participants experienced HD-related hypoglycemia, and all of them were asymptomatic. Many factors have been suggested to be associated with the development of hypoglycemia, such as caloric intake loss (32), diabetic autonomic neuropathy (33), diabetes treatment, and chronic inflammation (34,35); however, it is still unclear which factors play an important role in HD-related hypoglycemia. Previous studies showed that HD-related hypoglycemia and HD-associated hyperglycemia were detected by CGM in these patients, but the characteristics of those who experience this glucose disarray remain unclear (22,27,36). In this study, there was no difference in the clinical characteristics of patients who did or did not experience hypoglycemia. In particular, there was no difference in the frequency of HD-related hypoglycemia with regard to diabetes treatment, insulin regimen, or insulin dosage. Furthermore, HD-related hypoglycemia was seen even in patients with diabetes undergoing HD who were treated with diet and medications that do not cause hypoglycemia. This suggests that patients with diabetes undergoing HD are at risk for HD-related hypoglycemia, regardless of diabetic treatments. Additionally, conventional glycemic markers, such as HbA1c, GA, and GA-to-HbA1c ratio, could not satisfactorily detect hypoglycemia. Markers of glycemic variabilities obtained from CGM, such as SD, CV, and range of SGL data calculated during a 48-h period including HD and non-HD days, showed higher values in patients who experienced HD-related hypoglycemia than in those who did not. Additionally, although there was no significant difference in the SGL at the start of HD between the two groups, the SGL at the end of HD was significantly lower in patients who experienced HD-related hypoglycemia than that in those who did not. Treatment to suppress daily glycemic fluctuation may help patients avoid this type of hypoglycemia. Therefore, further investigation should be undertaken to identify appropriate interventions to reduce glycemic variability and HD-related hypoglycemia. However, our results reveal that hypoglycemia during HD may be associated with higher red blood cell counts and a lower serum albumin level. A previous study suggested that glucose diffuses from the plasma into the erythrocytes (37); our results may support this phenomenon. A lower serum albumin level may reflect chronic inflammation in patients undergoing HD (38), but whether chronic inflammation is associated with HD-related hypoglycemia remains unclear. In addition, our study showed a sustained decrease in the mean SGL during HD, irrespective of the dialysate glucose concentration. In particular, the SGL decreased toward the end of HD, and its nadir was well below the dialysate glucose concentration in 49 participants (50%).

One strength of our results is that the sample size was the largest compared with previous CGM studies in dialysis patients. However, our study has some limitations as well. Because this was an observational study, we could not investigate the relationship between diabetic biomarkers and CGM data and mortality or progression of complications. We investigated the effects of differences in dialysate glucose concentration on glycemic profiles and HD-related hypoglycemia, but the ratio of patients using the dialysate high glucose concentration was small. In addition, we analyzed the CGM data from 2 consecutive days, both when on and off HD, to equalize the effects of both days, but the consensus guideline recommends >10 days of analyzable CGM data (39). Furthermore, the duration of HD in our participants was short. It is possible that many patients had a residual renal function, and GA may decrease as a result of the effect of proteinuria or other conditions. Therefore, GA is recommended as an indicator for long-term follow-up for patients undergoing maintenance HD in the Japanese guideline (40).

In conclusion, HD-related hypoglycemia unawareness is far more common in patients with type 2 diabetes undergoing HD than currently recognized, potentially putting them at risk for cardiovascular disease and mortality. Physicians should be aware of the high risk of asymptomatic hypoglycemia in those with diabetes undergoing HD, irrespective of the use of glucose-containing dialysate.

This article contains supplementary material online at https://doi.org/10.2337/figshare.14372123.

Acknowledgments. The authors thank Mark Cleasby from Edanz Group for editing a draft of this manuscript.

Funding. This work was supported in part by a Grant-in-Aid for Young Scientists (B) from the Japan Society for the Promotion of Science to A.H. (17K18081) and by Kitasato University Shogaku-Kifu research support to M.S.

The funders had no role in the study design, data collection or analysis, decision to publish, or preparation of the manuscript.

Duality of Interest. A.H. reports speaker honoraria from Medtronic, Abbott, Terumo, and Roche and grants from Abbott outside the submitted work. N.S. reports speaker honoraria from Medtronic outside the submitted work. A.O. reports speaker honoraria from Abbott and Terumo outside the submitted work. No other potential conflicts of interest relevant to this article were reported.

Author Contributions. A.H., K.T., N.K., and M.S. were responsible for the conception and design of the study. A.H., N.S, A.S, K.M., A.O., and I.M. identified eligible patients for the study and collected and evaluated the CGM data and clinical information. A.H., A.M., and T.M. undertook the statistical analysis of the data. A.H., N.K., and M.S. reviewed the data and were responsible for the drafting and editing of the manuscript. A.H. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

1.
Afkarian
M
,
Sachs
MC
,
Kestenbaum
B
, et al
.
Kidney disease and increased mortality risk in type 2 diabetes
.
J Am Soc Nephrol
2013
;
24
:
302
308
2.
Alicic
RZ
,
Rooney
MT
,
Tuttle
KR
.
Diabetic Kidney Disease: Challenges, Progress, and Possibilities
.
Clin J Am Soc Nephrol
2017
;
12
:
2032
2045
3.
de Jager
DJ
,
Grootendorst
DC
,
Jager
KJ
, et al
.
Cardiovascular and noncardiovascular mortality among patients starting dialysis
.
JAMA
2009
;
302
:
1782
1789
4.
Kalantar-Zadeh
K
,
Kopple
JD
,
Regidor
DL
, et al
.
A1C and survival in maintenance hemodialysis patients
.
Diabetes Care
2007
;
30
:
1049
1055
5.
Kovesdy
CP
,
Park
JC
,
Kalantar-Zadeh
K
.
Glycemic control and burnt-out diabetes in ESRD
.
Semin Dial
2010
;
23
:
148
156
6.
Genuth
S
,
Eastman
R
,
Kahn
R
, et al.;
American Diabetes Association
.
Implications of the United kingdom prospective diabetes study
.
Diabetes Care
2003
;
26
(
Suppl. 1
):
S28
S32
7.
Hill
CJ
,
Maxwell
AP
,
Cardwell
CR
, et al
.
Glycated hemoglobin and risk of death in diabetic patients treated with hemodialysis: a meta-analysis
.
Am J Kidney Dis
2014
;
63
:
84
94
8.
Adler
A
,
Casula
A
,
Steenkamp
R
, et al
.
Association between glycemia and mortality in diabetic individuals on renal replacement therapy in the U.K
.
Diabetes Care
2014
;
37
:
1304
1311
9.
Freedman
BI
,
Andries
L
,
Shihabi
ZK
, et al
.
Glycated albumin and risk of death and hospitalizations in diabetic dialysis patients
.
Clin J Am Soc Nephrol
2011
;
6
:
1635
1643
10.
Ricks
J
,
Molnar
MZ
,
Kovesdy
CP
, et al
.
Glycemic control and cardiovascular mortality in hemodialysis patients with diabetes: a 6-year cohort study
.
Diabetes
2012
;
61
:
708
715
11.
Williams
ME
,
Lacson
E
 Jr
,
Teng
M
,
Ofsthun
N
,
Lazarus
JM
.
Hemodialyzed type I and type II diabetic patients in the US: characteristics, glycemic control, and survival
.
Kidney Int
2006
;
70
:
1503
1509
12.
Desouza
CV
,
Bolli
GB
,
Fonseca
V
.
Hypoglycemia, diabetes, and cardiovascular events
.
Diabetes Care
2010
;
33
:
1389
1394
13.
Akmal
M
.
Hemodialysis in diabetic patients
.
Am J Kidney Dis
2001
;
38
(
Suppl. 1
):
S195
S199
14.
Jackson
MA
,
Holland
MR
,
Nicholas
J
,
Lodwick
R
,
Forster
D
,
Macdonald
IA
.
Hemodialysis-induced hypoglycemia in diabetic patients
.
Clin Nephrol
2000
;
54
:
30
34
15.
Simic-Ogrizovic
S
,
Backus
G
,
Mayer
A
,
Vienken
J
,
Djukanovic
L
,
Kleophas
W
.
The influence of different glucose concentrations in haemodialysis solutions on metabolism and blood pressure stability in diabetic patients
.
Int J Artif Organs
2001
;
24
:
863
869
16.
Hayashi
A
,
Takano
K
,
Masaki
T
,
Yoshino
S
,
Ogawa
A
,
Shichiri
M
.
Distinct biomarker roles for HbA1c and glycated albumin in patients with type 2 diabetes on hemodialysis
.
J Diabetes Complications
2016
;
30
:
1494
1499
17.
Burmeister
JE
,
Scapini
A
,
da Rosa Miltersteiner
D
,
da Costa
MG
,
Campos
BM
.
Glucose-added dialysis fluid prevents asymptomatic hypoglycaemia in regular haemodialysis
.
Nephrol Dial Transplant
2007
;
22
:
1184
1189
18.
Sangill
M
,
Pedersen
EB
.
The effect of glucose added to the dialysis fluid on blood pressure, blood glucose, and quality of life in hemodialysis patients: a placebo-controlled crossover study
.
Am J Kidney Dis
2006
;
47
:
636
643
19.
Leinung
M
,
Thompson
S
,
Nardacci
E
.
Benefits of continuous glucose monitor use in clinical practice
.
Endocr Pract
2010
;
16
:
371
375
20.
Gai
M
,
Merlo
I
,
Dellepiane
S
, et al
.
Glycemic pattern in diabetic patients on hemodialysis: continuous glucose monitoring (CGM) analysis
.
Blood Purif
2014
;
38
:
68
73
21.
Jung
HS
,
Kim
HI
,
Kim
MJ
, et al
.
Analysis of hemodialysis-associated hypoglycemia in patients with type 2 diabetes using a continuous glucose monitoring system
.
Diabetes Technol Ther
2010
;
12
:
801
807
22.
Kazempour-Ardebili
S
,
Lecamwasam
VL
,
Dassanyake
T
, et al
.
Assessing glycemic control in maintenance hemodialysis patients with type 2 diabetes
.
Diabetes Care
2009
;
32
:
1137
1142
23.
Riveline
JP
,
Teynie
J
,
Belmouaz
S
, et al
.
Glycaemic control in type 2 diabetic patients on chronic haemodialysis: use of a continuous glucose monitoring system
.
Nephrol Dial Transplant
2009
;
24
:
2866
2871
24.
Vos
FE
,
Schollum
JB
,
Coulter
CV
,
Manning
PJ
,
Duffull
SB
,
Walker
RJ
.
Assessment of markers of glycaemic control in diabetic patients with chronic kidney disease using continuous glucose monitoring
.
Nephrology (Carlton)
2012
;
17
:
182
188
25.
Chantrel
F
,
Sissoko
H
,
Képénékian
L
, et al
.
Influence of dialysis on the glucose profile in patients with diabetes: usefulness of continuous glucose monitoring
.
Horm Metab Res
2014
;
46
:
810
813
26.
Vigersky
R
,
Shrivastav
M
.
Role of continuous glucose monitoring for type 2 in diabetes management and research
.
J Diabetes Complications
2017
;
31
:
280
287
27.
Abe
M
,
Kalantar-Zadeh
K
.
Haemodialysis-induced hypoglycaemia and glycaemic disarrays
.
Nat Rev Nephrol
2015
;
11
:
302
313
28.
Chow
E
,
Bernjak
A
,
Williams
S
, et al
.
Risk of cardiac arrhythmias during hypoglycemia in patients with type 2 diabetes and cardiovascular risk
.
Diabetes
2014
;
63
:
1738
1747
29.
Battelino
T
,
Danne
T
,
Bergenstal
RM
, et al
.
Clinical targets for continuous glucose monitoring data interpretation: recommendations from the international consensus on time in range
.
Diabetes Care
2019
;
42
:
1593
1603
30.
Ogawa
A
,
Hayashi
A
,
Kishihara
E
,
Yoshino
S
,
Takeuchi
A
,
Shichiri
M
.
New indices for predicting glycaemic variability
.
PLoS One
2012
;
7
:
e46517
31.
Zelnick
LR
,
Batacchi
ZO
,
Ahmad
I
, et al
.
Continuous glucose monitoring and use of alternative markers to assess glycemia in chronic kidney disease
.
Diabetes Care
2020
;
43
:
2379
2387
32.
Arem
R
.
Hypoglycemia associated with renal failure
.
Endocrinol Metab Clin North Am
1989
;
18
:
103
121
33.
Ewing
DJ
,
Winney
R
.
Autonomic function in patients with chronic renal failure on intermittent haemodialysis
.
Nephron
1975
;
15
:
424
429
34.
Aparicio
M
,
Cano
N
,
Chauveau
P
, et al.;
French Study Group for Nutrition in Dialysis
.
Nutritional status of haemodialysis patients: a French national cooperative study
.
Nephrol Dial Transplant
1999
;
14
:
1679
1686
35.
Kaysen
GA
.
Inflammation nutritional state and outcome in end stage renal disease
.
Miner Electrolyte Metab
1999
;
25
:
242
250
36.
Jin
YP
,
Su
XF
,
Yin
GP
, et al
.
Blood glucose fluctuations in hemodialysis patients with end stage diabetic nephropathy
.
J Diabetes Complications
2015
;
29
:
395
399
37.
Abe
M
,
Kaizu
K
,
Matsumoto
K
.
Plasma insulin is removed by hemodialysis: evaluation of the relation between plasma insulin and glucose by using a dialysate with or without glucose
.
Ther Apher Dial
2007
;
11
:
280
287
38.
Sun
CY
,
Lee
CC
,
Wu
MS
.
Hypoglycemia in diabetic patients undergoing chronic hemodialysis
.
Ther Apher Dial
2009
;
13
:
95
102
39.
Danne
T
,
Nimri
R
,
Battelino
T
, et al
.
International consensus on use of continuous glucose monitoring
.
Diabetes Care
2017
;
40
:
1631
1640
40.
Nakao
T
,
Inaba
M
,
Abe
M
, et al.;
Japanese Society for Dialysis Therapy
.
Best practice for diabetic patients on hemodialysis 2012
.
Ther Apher Dial
2015
;
19
(
Suppl. 1
):
40
66
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at https://www.diabetesjournals.org/content/license.