Abstract
Introduction
Chronic and post-prandial hyperglycemia are independent risk factors for diabetic complications. Glycemic patterns among hemodialysis end-stage-renal-disease (ESRD) differ as glucose metabolism changes with declining kidney function with more pronounced glycemic fluctuations. The objectives of this study are to determine glycemic patterns on hemodialysis days, the magnitude of post-hemodialysis rebound hyperglycemia (PHH) and their associated factors.
Methodology
148 patients on hemodialysis were analysed, 91 patients had end-stage-diabetic-renal disease (DM-ESRD), and 57 patients had end-stage-non-diabetic renal disease (NDM-ESRD). Glycemic patterns and PHH data were obtained from 11-point and 7-point self-monitoring blood glucose (SMBG) profiles on hemodialysis and non-hemodialysis days. PHH and its associated factors were analysed with logistic regression.
Results
Mean blood glucose on hemodialysis days was 9.33 [SD 2.7] mmol/L in DM-ESRD patients compared to 6.07 [SD 0.85] mmol/L in those with NDM-ESRD (p<0.001). PHH occurred in 70% of patients and was more pronounced in DM-ESRD compared to NDM-ESRD patients (72.5% vs 27.5%; OR 4.5). Asymptomatic hypoglycemia was observed in 18% of patients. DM-ESRD, older age, previous IHD, obesity, high HbA1c, elevated highly-sensitive CRP and low albumin were associated with PHH.
Conclusion
DM-ESRD patients experienced significant PHH in our cohort. Other associated factors include older age, previous IHD, obesity, high HbA1c, elevated hs-CRP and low albumin.
Keywords: renal dialysis, glycemic variability, diabetes complications, hyperglycemia, risk factors, Asians
INTRODUCTION
Over two million people worldwide currently on renal replacement therapy have Type 2 Diabetes Mellitus (T2DM), the leading cause of end-stage-renal-disease (ESRD).1,2 In Malaysia T2DM accounted for 61% of new dialysis patients.3
Glycemic patterns among diabetic patients with ESRD (DM-ESRD) whether or not on hemodialysis differ from those of diabetic patients without ESRD as glucose metabolism changes with decline in kidney function. Glycemic fluctuations are more pronounced among DM-ESRD as they may experience hemodialysis induced hypoglycemia and hyperglycemia.4,5 Furthermore, hemodialysis per se is an independent risk factor for glycemic fluctuations as glucose is freely filtered and insulin is absorbed during hemodialysis.6
Fluctuations in glucose metabolism have proven to be detrimental in DM-ESRD as they leads to poor survival mainly owing to cardiovascular complications.7 This was demonstrated by a six-year cohort study among DM-ESRD patients that showed a U-shape association between glycemic control (HbA1c <6% and >8%) and a decrease in overall survival.8 This U-shape association might indicate that chronic hyperglycemia is not the only indicator for morbidity and mortality, but also hypoglycemia and glucose fluctuations.7,8 Furthermore, many studies have shown glycemic variability (GV) as an independent risk factor for both morbidity and mortality among diabetic populations.9
Therefore, this study focused on glycemic patterns on hemodialysis days and non-hemodialysis days among ESRD patients. We also looked at post-hemodialysis rebound hyperglycemia (PHH), and its associated factors, as identifying these factors is hoped to optimize the management of this high-risk group.
METHODOLOGY
Study design and sample population
One hundred and fifty ESRD patients on maintenance hemodialysis (93 DM-ESRD patients and 57 NDM-ESRD patients), were recruited in this cross-sectional study. Sample size was calculated by multiple logistic regression using G Power software.10 Based on a study by Jin Y.P, 2015 we estimate the occurrence of post haemodialysis hyperglycemia among the DM-ESRD patients to be threefold that of NDM-ESRD patients.11 We considered a model with one binary covariate X with event rate under Ho, p1 =0.13 and the event rate under X = 1, p2 =0.40, giving the odds ratio of ~ 4.5. We further assumed R2 = 0.1, and an imbalanced design ratio of 2:1 between the 2 groups. The estimating sample size necessary to achieve a two-sided test with alpha of 0.05 and power of at least 80% was 102. Considering a 30% non-response rate, the final sample size was 146 rounded to 150.
Inclusion criteria were: adults age more than 18 with or without diabetes, patients on maintenance hemodialysis for at least three months. NDM-ESRD patients were included to observe the effect of hemodialysis on non-diabetic patients. Exclusion criteria were: Type 1 diabetes mellitus patients, presence of acute inflammatory state, hemoglobinopathy, history of blood transfusion or hospitalization for the last three months, and diagnosis of malignancy.
Socio-demographics, co-morbidities and laboratory data
Socio-demographic characteristics, clinical data, comorbidities and medication lists were obtained using a standardized questionnaire. At recruitment, baseline blood investigations were taken: glycated hemoglobin –A1c (HbA1c), lipid profile, renal profile, albumin, hemoglobin, inflammatory markers, iron studies and bone parameters.
Glucose monitoring on hemodialysis and non-hemodialysis days
Glucose values in our study were obtained from capillary glucose measurements using capillary glucometers (Bayer contour plus®). Patients were taught to self-measure the capillary glucose and were assisted during the hemodialysis sessions. Patients were advised to record the glucose values and medications taken and were also educated to recognize hypoglycemic signs and symptoms. On hemodialysis days, a 11-point capillary self-monitoring glucose (SMBG) profile was obtained with hourly capillary glucose taken during the hemodialysis sessions. On non-hemodialysis days, 7 points SMBG were taken by the patients. Patients were not required to fast during hemodialysis but were asked to report and food consumed that was out of their ordinary diet.
Statistical analysis
Analyses were performed using R studio version 1.0.153 using the STATS package for statistical analysis. In descriptive statistics, categorical data results were described as count and percentage while continuous data in mean and standard deviation (SD). Overall glycemic profile, specifically looking, as post hemodialysis glucose value was measured in terms of means glucose ± SD. The data was checked for normality visually by histogram and statistically using the Shapiro Wilk test. For bivariate analysis, a chi squared test was used for comparing categorical data. An independent t-test was used to compare means between the groups as the data was normally distributed. The assumption of equal variance was met using Levene’s test. The level of significance was set at 0.05. In order to determine the association between PHH with clinical and laboratory variables, simple logistic regression was done to derive the crude odd ratio. Subsequently, the variables which were significant at p < 0.15 were included in the final multivariate logistic regression analysis. All the crude and adjusted odds ratios were presented with 95% confidence intervals. For missing data, the Listwise deletion method was used.
RESULTS
Socio-demographics, co-morbidities and laboratory data analysis
From the total of 150 patients recruited, 148 patients data were included in the final analysis due to missing data from two patients. Table 1 demonstrates the baseline sociodemographics and co-morbidities of our cohort. Ninety-one (61.5%) patients had diabetes with a mean age of 57.6 years and mean duration of diabetes of 16.4 years. Mean duration of hemodialysis in DM-ESRD and NDM-ESRD patients were 3.8 and 4.5 years, respectively. Body mass index (BMI) in the majority of patients was Obese Class 1. The difference in socio-demographics and co-morbidities among both groups were not statistically significant apart from DM-ESRD patients having higher BMI and higher prevalence of smoking among NDM-ESRD patients. Ischemic heart disease (IHD) was present in one-fourth of the patients. Blood pressure control was poor with only 16 (10.7%) of DM-ESRD and 47 (31.3%) of NDM-ESRD patients achieving pre and post hemodialysis target blood pressure of less or equal to 130/80 mmHg.
Table 1.
Baseline socio-demographic and clinical characteristics of patients
| Characteristics | DM-ESRD n (%) | NDM-ESRD n (%) | Test statistic* | p value |
|---|---|---|---|---|
| Sex | 0.165 | 0.685 | ||
| Male | 51 (56.0) | 30 (52.6) | ||
| Female | 40 (44.0) | 27 (47.4) | ||
| Race | 1.078 | 0.583 | ||
| Malay | 79 (86.8) | 52 (91.2) | ||
| Chinese | 1 (1.1) | 1 (1.8) | ||
| Indian | 11 (12.8) | 4 (7.0) | ||
| BMI | 10.729 | 0.030*** | ||
| Underweight (<18.5) | 1 (1.1) | 4 (7.0) | ||
| Normal (18.5 – 22.9) | 11 (12.8) | 15 (26.3) | ||
| Overweight (23.0 – 24.9) | 15 (16.5) | 7 (12.3) | ||
| Obese Class 1 (25.0 – 29.9) | 41 (45.1) | 16 (28.1) | ||
| Obese Class 2 (>30.0) | 23 (25.3) | 15 (26.3) | ||
| BP Target Pre HD | 1.383 | 0.24 | ||
| Pre-HD BP (≤130/80) | 12 (13.2) | 4 (7.0) | ||
| Pre-HD BP (>130/80) | 79 (86.8) | 53 (93.0) | ||
| BP Target Pre HD | 3.426 | 0.064 | ||
| Post HD BP (≤130/80) | 34 (37.4) | 13 (22.8) | ||
| Post HD BP (>130/80) | 57 (62.6) | 44 (77.2) | ||
| Smoking | 7.074 | 0.008*** | ||
| Yes | 4 (4.4) | 10 (17.5) | ||
| No | 87 (95.6) | 47 (82.5) | ||
| Hypertension | 2.311 | 0.128 | ||
| Yes | 90 (98.9) | 54 (94.7) | ||
| No | 1 (1.1) | 3 (5.3) | ||
| IHD | 1.038 | 0.308 | ||
| Yes | 26 (28.6) | 12 (21.1) | ||
| No | 65 (71.4) | 45 (78.9) | ||
| Gout | 1.415 | 0.234 | ||
| Yes | 6 (6.6) | 7 (12.3) | ||
| No | 85 (93.4) | 50 (87.7) | ||
| Stroke | 0.307 | 0.58 | ||
| Yes | 5 (5.5) | 2 (3.5) | ||
| No | 86 (94.5) | 55 (96.5) | ||
| Hyperlipidaemia | 2.857 | 0.091 | ||
| Yes | 56 (61.5) | 27 (47.4) | ||
| No | 35 (38.5) | 30 (52.6) | ||
| Mean (SD) | Mean (SD) | Test statistic** | P value | |
| Age (years) | 57.6 (11.1) | 49.0 (11.2) | -4.59 | <0.001*** |
| Duration of HD (years) | 3.8 (3.1) | 4.5 (3.3) | 1.31 | 0.198 |
| BMI (kg/m2) | 27.1 (4.4) | 26.2 (6.6) | -0.862 | 0.391 |
Table 1 shows the baseline sociodemographic and clinical characteristics of patients (n=148), values expressed as mean ± standard deviation. DM-ESRD, diabetic-end stage renal disease; NDM-ESRD, non diabetic end-stage-renal-disease; HD, hemodialysis; BP, blood pressure; BMI, body mass index.
Chi square test
Independent t-test
P value<0.05
In terms of medications, 50 (54.9%) of diabetic patients were on insulin therapy, 18 (19.7%) on oral hypoglycemic agents (OHA), and 23 (25.2%) were not on regular medications. The majority of patients, i.e., 56 (82.3%) patients, would not take their medications on hemodialysis days. Patients on OHA alone would not take their OHA on hemodialysis days, while patient on basal-bolus insulin, would omit the insulin dose before their hemodialysis session.
Table 2 compares the baseline blood parameters between DM-ESRD and NDM-ESRD patients. Both groups had a non-significant difference in terms of blood parameters apart from HbA1c, phosphate and albumin. Mean HbA1c among DM-ESRD patients was 7.4% with 37% having HbA1c less than 6.5% and 30% falling between 6.5% to 8%. Analysing the highly sensitive C-reactive protein (hs-CRP) as a surrogate marker for cardiac disease showed both groups having high hS-CRP levels with means of 8.91mg/L and 7.03 mg/L among DM-ESRD and NDM-ESRD patients respectively. Albumin levels were significantly lower among DM-ESRD patients, while phosphate levels were higher among NDM-ESRD patients.
Table 2.
Mean (SD) blood parameter s comparing DM-ESRD a nd NDM-ESRD
| Characteristics | DM-ESRD Mean (SD) | NDM-ESRD Mean (SD) | T statistic | P value |
|---|---|---|---|---|
| Hemoglobin (g/dL) | 10.47 (1.7) | 10.38 (1.7) | -0.339 | 0.735 |
| HbA1c (%) | 7.40 (1.6) | 5.41 (0.5) | -10.845 | <0.001* |
| Total cholesterol (mmol/L) | 4.8 (1.3) | 5.0 (1.2) | 1.003 | 0.318 |
| LDL (mmol/L) | 2.97 (1.16) | 3.10 (1.18) | 0.664 | 0.508 |
| TG (mmol/L) | 2.30 (1.7) | 2.04 (1.4) | -1.03 | 0.304 |
| HDL (mmol/L) | 0.97 (0.22) | 1.08 (0.29) | 2.56 | 0.012 |
| HSCRP (mg/L) | 8.91 (10.2) | 7.03 (7.1) | -1.31 | 0.192 |
| Ferritin (ug/L) | 554.1 (402) | 665.2 (435) | 1.497 | 0.137 |
| Transferrin saturation (%) | 23.96 (11.2) | 24.82 (8.8) | 0.526 | 0.6 |
| Calcium (mmol/L) | 2.16 (0.22) | 2.19 (0.24) | 0.883 | 0.379 |
| Phosphate (mmol/L) | 1.85 (0.54) | 2.18 (0.73) | 2.893 | 0.005* |
| iPTH (pmol/L) | 73.6 (58.3) | 103.7 (105.1) | 1.889 | 0.063 |
| ALP (U/L) | 176.32 (173.3) | 143.37 (112.2) | -1.413 | 0.159 |
| Albumin (mmol/L) | 38.3 (4.2) | 40.1 (2.7) | 3.19 | 0.002* |
Table 2 shows the baseline blood parameters of patients (n=148), values expressed as mean ± standard deviation. HbA1c, glyclated hemoglobin A1c; LDL, low density lipoprotein; TG, triglycerides; HDL, high density lipoprotein; HSCRP, highly sensitive C-reactive protein; iPTH, intact parathyroid hormone; ALP, alkaline phosphatase.
p < 0.05.
In our DM-ESRD cohort, 13 (14.3%) took their OHA/insulin on hemodialysis days with 3 (3.3%) of patients reporting hypoglycemic symptoms during hemodialysis sessions. Almost all (94.7%) patients ate during hemodialysis.
Figure 1 illustrates mean blood glucose on hemodialysis days, which were significantly different (p<0.01) between DM-ESRD and NDM-ESRD patients. The mean (±SD) blood glucose was 9.33±2.7 mmol/L in DM-ESRD and 6.07±0.85 mmol/L in NDM-ESRD. The mean fasting blood glucose was 7.9 mmol/L and 5.1 mmol/L in DM-ESRD and NDM-ESRD patients, respectively.
Figure 1.
Glycemic patterns among DM-ESRD vs. NDM-ESRD during hemodialysis days and non-hemodialysis days.
Legend: DM-ESRD shows more marked glucose fluctuations during hemodialysis days with prominent post-hemodialysis hyperglycemia coupled with persistent hyperglycemia until the end of the day. Glycemic fluctuations were not prominent during non-hemodialysis days in both DM-ESRD and NDM-ESRD. Timing hemodialysis day: D1 = Fasting, D2 = Prior hemodialysis, D3 = 1st hour hemodialysis, D4 = 2nd hour hemodialysis, D5 = 3rd hour hemodialysis, D6 = 4th hour hemodialysis, D7 = 2 hours post hemodialysis, before meal, D8 = 2hrs post meal, D9 = before dinner, D10 = 2 hours post dinner, D11= before sleep. Timing non-hemodialysis: ND1 = Fasting – before breakfast, ND2 = 2 hours post breakfast, ND3 = before lunch, ND4 = 2 hours post lunch, ND5 = before dinner, ND6 = 2 hours post dinner, ND7 = before sleep.
During the intra-dialytic period, the mean blood glucose was 8.1 mmol/L and 5.9 mmol/L in DM-ESRD and NDM-ESRD groups, respectively. Among DM-ESRD, 61 (67.0%) of patients had readings within the suggested limits (4.4–8.5 mmol/L). Thirty (32.9%) patients recorded blood glucose more than 8.5 mmol/L. Among NDM-ESRD patients, the majority recorded values within the suggested intradialytic glucose limits. Eighteen (12%) patients i.e. 7 (7.6%) with DM-ESRD and 11 (19.3%) with NDM-ESRD, experienced intradialytic asymptomatic hypoglycemia. This phenomenon which was seen predominantly among NDM-ESRD patients occurred mainly during the first hour of hemodialysis with mean pre-hemodialysis blood glucose of 4.3 mmol/L among NDM-ESRD patients.
Post-hemodialysis trends showed a mean pre-prandial reading among DM-ESRD patients of 10.4 mmol/L and a mean post-prandial reading of 11.1 mmol/L. Among DM-ESRD patients, 70 (77%) had blood glucose of more and equal to 7.0 mmol pre-prandially, while 65 (72%) had blood glucose more than 8.0 mmol/L post-prandially. Figure 1 shows that the DM-ESRD group had persistently high glucose level post-hemodialysis until the end of the day, which was not seen in the NDM-ESRD group.
Post-hemodialysis hyperglycemia (PHH)
PHH (defined as blood glucose ≥ 11.1 mmol/L 2-hours after first meal) post hemodialysis was prominent among DM-ESRD patients, as almost 75 (82.4%) of them experienced more than 80% increase of blood sugar, compared to NDM-ESRD patients during the end of hemodialysis, odds ratio 4.5 (CI: 2.2 – 9.6) (Figure 2). However, among NDM-ESRD patients, almost half of the patients, i.e. 25 (43.8%) experienced a reduction in blood sugar post hemodialysis as compared to only 14 (15.4%) of DM-ESRD patients.
Figure 2.
Capillary glucose changes post hemodialysis sessions.
Legend: Figure 2 showing the capillary glucose changes occurs post hemodialysis session between DM-ESRD and NDM-ESRD. In DM-ESRD, 75 (82.4%) experienced more than 80% increased of blood sugar value, compared to NDM-ESRD during the end of hemodialysis with odds ratio of 4.5 (CI: 2.2 – 9.6).
Glycemic pattern on non-hemodialysis days
On non-hemodialysis days (Figure 1), there were significant differences between blood sugar profiles among DM-ESRD and NDM-ESRD patients, where the mean (SD) blood glucose was 9.85 ± 3.1 mmol/L and 6.0 ±0.88 mmol/L, respectively. In DM-ESRD patients, post-prandial hyperglycemia more or equal to 8.5 mmol/L were observed in 61 (67.0%) patients with 28 (30.8%) experiencing blood sugar more or equal to 11.1 mmol/L (mean preprandial 9.1 ± 3.1 mmol/L and post-prandial 10.4 ± 3.3 mmol/L).
Correlation between mean blood glucose and HbA1c
In our population, there was a strong correlation between mean blood glucose and HbA1c among all patients with R2 = 0.73. However, among DM-ESRD patients, the correlation was moderate with R2 = 0.59.
Factors associated with PHH
Table 3 demonstrates a simple logistic regression analysis of clinical characteristics and blood parameters associated with PHH. In this study, DM-ESRD, obesity, previous IHD, older age, high HbA1c, elevated hs-CRP and low albumin were associated with the risk of PHH. Table 4 shows the final multivariate logistic regression model, the model is fit with R2 of 0.258. Increasing age of the patient are significant with adjusted odds ratio of 1.04, while DM-ESRD had adjusted odds ratio of almost three times higher than NDM-ESRD.
Table 3.
Factors associated with post haemodialysis hyperglycaemia (PHH) using simple logistic regression
| PHH |
Test statistic | P Value | Crude Odds ratio (95% CI) | |||
|---|---|---|---|---|---|---|
| Increase | Decrease | |||||
| Diabetes | 3.989 | <0.01 | ||||
| Yes | 74 (72.5) | 17 (37.0) | 4.5 (2.2, 9.6) | |||
| No | 28 (27.5) | 29 (63.0) | Reference | |||
| Gender | 0.650 | 0.515 | ||||
| Female | 48 (47.1) | 19 (41.3) | 1.26 (0.63, 2.58) | |||
| Male | 54 (52.9) | 27 (58.7) | Reference | |||
| Smoking | -0.991 | 0.322 | ||||
| Yes | 8 (7.8) | 6 (13.0) | 0.56 (0.19, 1.82) | |||
| No | 94 (92.2) | 40 (87.0) | Reference | |||
| BMI (kg/m2) | 22.7 | <0.01 | ||||
| Normal (18.5 – 22.9) | 16 (15.7) | 10 (21.7) | Reference | |||
| Underweight (<18.5) | 2 (2.0) | 3 (6.5) | 0.877 | 0.380 | 2.40 (0.34, 20.77) | |
| Overweight (23.0 – 24.9) | 14 (13.7) | 8 (17.4) | 0.951 | 0.342 | 2.63 (0.36, 23.37) | |
| Obese 1 (25.0 – 29.9) | 41 (40.2) | 16 (34.8) | 1.404 | 0.160 | 3.84 (0.59, 31.31) | |
| Obese 2 (>30.0) | 29 (28.4) | 9 (19.6) | 1.592 | 0.111 | 4.83 (0.70, 41.37) | |
| IHD | 1.53 | 0.125 | ||||
| Yes | 30 (29.4) | 8 (17.4) | 1.98 (0.86, 5.01) | |||
| No | 72 (70.6) | 38 (82.6) | Reference | |||
| Gout | -0.600 | 0.549 | ||||
| Yes | 8 (7.8) | 5 (10.9) | 0.70 (0.22, 2.43) | |||
| No | 94 (92.2) | 41 (89.1) | Reference | |||
| Hyperlipidaemia | 0.999 | 0.318 | ||||
| Yes | 60 (58.8) | 23 (50.0) | 1.43 (0.71, 2.89) | |||
| No | 42 (41.2) | 23 (50.0) | Reference | |||
| Stroke | 0.147 | 0.883 | ||||
| Yes | 5 (4.9) | 2 (4.3) | 1.13 (0.23, 8.13) | |||
| No | 97 (95.1) | 44 (95.7) | Reference | |||
| Hypertension | 0.0808 | 0.419 | ||||
| Yes | 100 (98.0) | 44 (95.7) | 2.27 (0.27, 19.43) | |||
| No | 2 (2.0) | 2 (4.3) | Reference | |||
| Mean (SD) | ||||||
| Age (years) | 56.6 (12.0) | 49.2 (10.2) | 3.383 | <0.01 | 1.06 (1.03, 1.10) | |
| Duration HD (years) | 3.9 (3.1) | 4.5 (3.4) | -0.983 | 0.326 | 0.95 (0.86, 1.06) | |
| BMI (kg/m2) | 27.3 (5.1) | 25.6 (5.8) | 1.756 | 0.079 | 1.06 (0.99, 1.14) | |
| HbA1c (%) | 6.9 (1.6) | 6.1 (1.5) | 2.613 | <0.01 | 1.44 (1.12, 1.94) | |
| HSCRP (mg/L) | 8.4 (8.3) | 6.4 (6.1) | 1.42 | 0.155 | 1.04 (0.99, 1.10) | |
| Ferritin (μg/L) | 622.2 (413.1) | 560.2 (421.7) | 0.807 | 0.412 | 1.00 (0.99,1.00) | |
| Albumin (mmol/L) | 38.5 (4.0) | 39.9 (3.0) | -2.109 | 0.035 | 0.89 (0.80, 9.88) | |
| Clearance (%) | 69.5 (9.0) | 69.0 (9.4) | 0.260 | 0.795 | 1.01 (0.97, 1.05) | |
| LDL (mmol/L) | 3.0 (1.2) | 3.0 (1.0) | 0.115 | 0.908 | 1.02 (0.75, 1.39) | |
| HDL (mmol/L) | 0.99 (0.2) | 1.0 (0.3) | -1.174 | 0.240 | 0.44 (0.11, 1.74) | |
| HB (g/dL) | 10.44 (1.7) | 10.40 (1.7) | 0.168 | 0.866 | 1.02 (0.83, 1.25) | |
| TG (mmol/L) | 2.3 (1.8) | 1.9 (1.1) | 1.334 | 0.182 | 1.20 (0.94, 1.61) | |
| Transferrin saturation | 24.0 (10.2) | 25.2 (10.9) | -0.611 | 0.541 | 0.99 (0.96, 1.02) | |
| Calcium (mmol/L) | 2.17 (0.21) | 2.16 (0.24) | 0.336 | 0.737 | 1.31 (0.26, 6.34) | |
| Phosphate (mmol/L) | 1.9 (0.6) | 2.1 (0.7) | -1.408 | 0.160 | 0.68 (0.39,1.16) | |
| ALP (U/L) | 148.6 (108.0) | 175.0 (154.4) | -1.169 | 0.242 | 0.99 (0.99, 1.00) | |
| iPTH (pmol/L) | 78.1 (67.6) | 100.5 (102.4) | -1.465 | 0.162 | 0.99 (0.99,1.00) | |
Table 3 shows simple logistic regression analysis of sociodemographic, clinical co-morbidities and blood parameters among patients in cohort (n=148) with post hemodialysis hyperglycemia (PHH). Diabetes, Body Mass Index (BMI) category, Ischemic Heart Disease (IHD), age, glycated hemoglobin (HbA1c), highly sensitive C-reactive protein (hs-CRP) and albumin were significant at P<0.15 to be included in multiple logistic regression.
Table 4.
Multiple logistic regression analysis for factors associated with post hemodialysis hyperglycemia (PHH) (N=148)
| Variable | β | SE | Wald | Adjusted OR (95% CI) | P-Value |
|---|---|---|---|---|---|
| Diabetes | 1.08 | 0.56 | 1.951 | 2.96 (1.01, 9.09) | 0.050* |
| Age (years) | 0.041 | 0.019 | 2.157 | 1.04 (1.00, 1.08) | 0.031* |
| IHD | 0.584 | 0.515 | 1.134 | 1.80 (0.68, 5.22) | 0.257 |
| HBA1c (%) | 0.004 | 0.176 | 0.026 | 1.00 (0.72, 1.44) | 0.980 |
| Albumin (mmol/L) | -0.093 | 0.065 | -1.429 | 0.91 (0.80, 1.03) | 0.153 |
| HSCRP (mg/L) | 0.0152 | 0.031 | 0.497 | 1.02 (0.96, 1.08) | 0.620 |
| BMI (kg/m2) | |||||
| Normal (Reference) | - | - | - | Reference | |
| Underweight | 0.588 | 1.05 | 0.558 | 1.80 (0.22, 16.89) | 0.577 |
| Overweight | 0.023 | 1.10 | 0.021 | 1.02 (0.12, 10.28) | 0.983 |
| Obese | 0.643 | 1.02 | 0.628 | 1.90 (0.25, 16.94) | 0.530 |
| Obese Class 1 | 1.12 | 1.06 | 1.051 | 3.06 (0.38, 29.24) | 0.293 |
Table 4 shows multiple logistic regression analysis of significant factors associated with post hemodialysis hyperglycemia among ESRD patients during hemodialysis (HD) day. OR, odd ratio; HbA1c, glyclated hemoglobin; HSCRP, highly sensitive C-reactive protein; Ischemic Heart Disease (IHD) and BMI class. R2= 0.258 (Nagelkerke)
P<0.05.
DISCUSSION
Cardiovascular disease is the leading cause of morbidity and mortality in ESRD particularly in diabetic patients, where the excessive cardiovascular risk may be attributed to underlying co-morbidities and population-based factors; however, these do not account for all the observed risk.4 Studies have shown that besides the average HbA1c (a marker for chronic hyperglycemia), short-term glycemic variation (GV) is also an independent risk factors for diabetic complications.12–14 GV, which describes glycemic fluctuations or oscillations around a mean value, is an independent risk factor for diabetes-associated morbidity and mortality.15 Post-prandial hyperglycemia; among the main contributors to GV, similar to PHH, was shown to play a significant role in the pathophysiology of diabetic complications in terms of inducing oxidative stress and the inflammatory process.16,17 Hence, it is crucial to evaluate the glycemic pattern in hemodialysis patients, especially emphasizing PHH, as it is one of the main contributors to glycemic fluctuations.
In our study, DM-ESRD patients had more significant glycemic fluctuations compared to those with NDM-ESRD. During hemodialysis days, we observed that PHH was fourfold higher in DM-ESRD compared to NDM-ESRD patients. An 80% rise in blood glucose post-hemodialysis occurred in 82% of diabetic patients. This hyperglycaemia was subsequently persistent throughout the day. On the other hand, this persistent hyperglycemia post hemodialysis was not seen among NDM-ESRD patients. This shows that hemodialysis predisposes DM-ESRD patients to constant hyperglycemia. Other patients exhibited lowered blood glucose levels in the first hour of hemodialysis, which subsequently became constant over the four hours of hemodialysis. This asymptomatic intradialytic hypoglycemia was observed in 18 (12.6%) patients, and more markedly seen in NDM-ESRD patients i.e., 11 (19.3%) patients. In the DM-ESRD patients, only 7 (7.8%) of out of the 91 patients developed intradialytic hypoglycemia; and the majority occurred in the first hour of hemodialysis with a mean blood glucose pre-hemodialysis of 5.4 mmol/L. Most of these patients were on insulin treatment. Although patients with NDM-ESRD had no significant glycemic fluctuations during hemodialysis, almost half experienced a reduction in blood sugar levels post-hemodialysis. Development of intra-dialytic hypoglycemia and reduction of blood sugar among ESRD patients even non-diabetic was an important observation, as currently there are no guidelines on management of insulin or OHA on hemodialysis days, which is left to the nephrologist’s discretion.7 However, we demonstrated that the number of patients developing intra-dialytic hypoglycemia was small compared to those developing PHH, which is similar to previous studies.18–20 Furthermore, among our population, all patients were encouraged to eat during hemodialysis to prevent episodes of hypoglycemia.
Our observations were similar to other studies on hemodialysis patients. Abe et al. showed that plasma glucose decreased with hemodialysis and hyperglycemic spikes were observed post hemodialysis which were attributed to decreased insulin due to hemodialysis clearance and/or the release of counter-regulatory hormones.6 Similar intradialytic glucose reduction and PHH were observed by Gai et al., which demonstrates PHH occurs 150 minutes post-hemodialysis.20 Kazempour-Ardenilli et al., showed glycemic readings were lower on hemodialysis days as compared to non-hemodialyis days.18 Both Mirani et al., and Jin et al., showed that GV was more pronounced on hemodialysis days, however, mean blood glucose was lower on hemodialysis days compared to non-hemodialysis days.11,19
This observation of intra-dialytic hypoglycemia coupled with significant PHH should prompt the nephrologist to adjust glycemic management of DM-ESRD patients. Possible administration of additional insulin or less hypoglycemic agents e.g., dipeptidyl-peptidase 4 inhibitors or glucagon-like-peptide analogs post-hemodialysis or on hemodialysis days will eventually help reduce glycemic fluctuations.6,21–23 Development of intradialytic hypoglycemia should be taken seriously because, as shown in our study and previous studies, a majority of these events were asymptomatic (role of autonomic neuropathy in long-standing diabetes).5,24 Events of intradialytic hypoglycemia were mostly asymptomatic in our study and previous studies, suggesting underlying autonomic neuropathy from long-standing diabetes. These might suggest the possibility of other undetected hypoglycemic events which may further aggravate glycemic fluctuations. Therefore, the role of additional intra-dialytic glucose monitoring should be further studied.
In our study, we report that PHH, as reported by SMBG, is more significant in patients with DM-ESRD, older age, obesity and previous IHD. Other associated blood parameters include high HbA1c, elevated hs-CRP and lower albumin. The association between HbA1c levels and glycemic fluctuations and mean blood glucose had been heavily investigated previously with conflicting results. Conversely, studies have shown that HbA1c has a weak correlation with glycemic fluctuations but has a significant relationship with chronic hyperglycemia and mean blood glucose.25–28 Interestingly, recent studies among Asian populations showed similar findings to our study where HbA1c correlates with glycemic variability indices.29,30 Notably, most of these studies exclude ESRD patients where HbA1c is a less reliable surrogate for glycemic control as it may falsely increase or decrease due to factors related to ESRD, e.g. anemia and uremia.7,31 Anemia present in 42% of our population may have confounded our findings of 68% of patients with HbA1c less than 8%, with mean of 7.4% (reasonable control). The correlation between mean blood glucose and HbA1c in our population was moderate with R2 of 0.59, similar to other studies on DM-ESRD where the R2 is not more than 0.50 compared to NDM-ESRD with R2 more than 0.80.27,32 Nonetheless, HbA1c level more than 8.5% in DM-ESRD patients was related to increased mortality and should not deter clinicians from controlling the glucose level.31
PHH is also associated with older age in which there is pancreatic beta cell dysfunction with limited capability to generate coupled with insulin resistance.33,34 Studies have shown that beta-cell dysfunction plays a significant role in explaining dysglycemia, where insufficient insulin secretion for accurate glycemic regulation may lead to glucose-related metabolic disorders, resulting in increase glucose fluctuations and sustained hyperglycemia.35,36 Other studies also showed similar findings where glucose fluctuations are more marked in the older age population.29,30
Previous IHD, high hs-CRP along with low albumin and relative obesity can be explained by the malnutrition-inflammatory complex syndrome (MCIS), which a term coined to describe the chronic inflammatory state in hemodialysis patients, which is usually accompanied by malnutrition or protein-energy wasting (PEW).5 Oxidative stress and high inflammatory levels are associated with endothelial dysfunction and subsequently, micro and macro-angiopathy in diabetic patients, particularly resulting in cardiovascular complications.37
We specifically looked at the inflammatory biomarker hs-CRP to add prognostic information on cardiovascular risk in our population. A previous study evaluated the role of hs-CRP and showed that there was a linear relationship with vascular risk; a value of less than 1mg/L (lower risk), 1 to 3 mg/L (moderate risk) and more than 3mg/L (higher risk).38 In hemodialysis patients, although elevated hs-CRP at a single time point is an important predictor of cardiovascular events, the values are not static and may reflect the chronic inflammatory process due to hemodialysis, intercurrent clinical events, decreased residual renal functions and PEW.39–41 A study done in hemodialysis patients showed that serum CRP levels increased annually during the follow up period.42 In our study, the mean hs-CRP in DM-ESRD and NDM-ESRD patients were 8.91 mg/L and 7.03 mg/L respectively, with only 41 (27.7%) of patients with level less 3 mg/L. This suggests that hemodialysis patients, regardless of their diabetic status, were in a constant state of inflammation, which predisposes them to a higher risk of a cardiovascular event. PEW, on the other hand, represented by low albumin, is common among patients undergoing maintenance hemodialysis and is by far the strongest risk factor for adverse outcomes and death.43
A 10-year cohort study which evaluated serum albumin, C-reactive protein, and carotid atherosclerosis as predictors of 10-year mortality in hemodialysis patients showed that serum albumin concentration was superior as a predictor of mortality.44 However, in our cohort, the difference between 38 mmol/L among patients experiencing PHH compared to 40 mmol/L maybe hard to appreciate in clinical setting. Nonetheless, by addressing the issue of malnutrition and chronic inflammation among hemodialysis patients, we may improve the occurrence of glycemic fluctuations and subsequently improve outcomes in these patients.
The first limitation of this study was its cross-sectional design and one-off blood sugar monitoring during hemodialysis and non-hemodialysis days. This design may not accurately represent the overall picture of the patients, as many factors can influence single snapshot monitoring. Another limitation was the usage of SMBG instead of continuous glucose monitoring system (CGMS) in assessing PHH. CGMS is preferable as SMBG can miss specific peaks and nadirs in glucose values.45,46 However, it is challenging to perform CGM in daily practice, given discomfort, cost, and the need for calibration compared to the SMBG. The practical aspect of SMBG in terms of easy availability, monitoring and interpretation, and lower cost makes it the preferred method in our population. We did not limit or measure dietary intake of patients during the study period, which makes it an additional confounding factor in the glycemic profile of the patients. Some previous studies restricted dietary intake or asked the patient to fast during hemodialysis to reduce confounding. However, doing so is not reflective of the normal day-to-day glucose fluctuations of patients, and allowing normal dietary intake reflects real-life data and consequently will allow meaningful alterations in management.
CONCLUSION
DM-ESRD patients experienced more significant fluctuations of glucose level, in particular, PHH on hemodialysis day compared to those with NDM-ESRD. Other associated factors for PHH include older age group, previous IHD, obese patient, high HbA1c, and hs-CRP coupled with low albumin. Malnutrition-inflammatory-complex syndrome, together with protein-energy wasting commonly seen among chronic hemodialysis patients, should be cornerstones in managing this group of patients as improving this phenomenon may improve glucose control and subsequently improve diabetes-related complications and mortality. Regular glucose monitoring via SMBG may provide valuable insights for treating physicians among this high-risk group.
Acknowledgments
The authors thank all the staff and patients from Pusat Perubatan Dialysis Bangi, Dengkil, Semenyih, Farah Mahami Dialysis Center and Bangi Dialysis Centre.
Statement of Authorship
All authors certified fulfillment of ICMJE authorship criteria.
Author Disclosure
The authors declared no conflict of interest.
Funding Source
Putra Grant IPS University Putra Malaysia supported this study (UPM/700-1/3/Geran Putra).
Authors are required to accomplish, sign and submit scanned copies of the JAFES Author Form consisting of: (1) Authorship Certification, that authors contributed substantially to the work, that the manuscript has been read and approved by all authors, and that the requirements for authorship have been met by each author; (2) the Author Declaration, that the article represents original material that is not being considered for publication or has not been published or accepted for publication elsewhere, that the article does not infringe or violate any copyrights or intellectual property rights, and that no references have been made to predatory/suspected predatory journals; (3) the Author Contribution Disclosure, which lists the specific contributions of authors; and (4) the Author Publishing Agreement which retains author copyright, grants publishing and distribution rights to JAFES, and allows JAFES to apply and enforce an Attribution-Non-Commercial Creative Commons user license. Authors are also required to accomplish, sign, and submit the signed ICMJE form for Disclosure of Potential Conflicts of Interest. For original articles, authors are required to submit a scanned copy of the Ethics Review Approval of their research as well as registration in trial registries as appropriate. For manuscripts reporting data from studies involving animals, authors are required to submit a scanned copy of the Institutional Animal Care and Use Committee approval. For Case Reports or Series, and Images in Endocrinology, consent forms, are required for the publication of information about patients; otherwise, appropriate ethical clearance has been obtained from the institutional review board. Articles and any other material published in the JAFES represent the work of the author(s) and should not be construed to reflect the opinions of the Editors or the Publisher.
References
- 1.Couser WG, Remuzzi G, Mendis S, Tonelli M. The contribution of chronic kidney disease to the global burden of major noncommunicable diseases. Kidney Int. 2011; 80(12):1258–70. PMID: 10.1038/ki.2011.368. [DOI] [PubMed] [Google Scholar]
- 2.Collins AJ, Foley RN, Herzog C, et al. Excerpts from the US Renal Data System 2009 Annual Data Report. Am J Kidney Dis. 2010; 55(1 Suppl 1):S1-420. PMID: PMCID: 10.1053/j.ajkd.2009.10.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Wong H, Goh B. Twenty Third Report of The Malaysian Dialysis and Transplant 2015 [Internet]. Kuala Lumpur 2017; 2017. Available from http://www.msn.org.my.
- 4.Kalantar-Zadeh K, Derose SF, Nicholas S, Benner D, Sharma K, Kovesdy CP. Burnt-Out Diabetes: Impact of chronic kidney disease progression on the natural course of diabetes mellitus. J Ren Nutr. 2009; 19(1):33–7. PMID: PMCID: 10.1053/j.jrn.2008.11.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Abe M, Kalantar-Zadeh K. Haemodialysis-induced hypoglycaemia and glycaemic disarrays. Nat Rev Nephrol. 2015; 11(5):302–13. PMID: PMCID: 10.1038/nrneph.2015.38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.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(4):280–7. PMID: 10.1111/j.1744-9987.2007.00491.x [DOI] [PubMed] [Google Scholar]
- 7.Rhee CM, Leung AM, Kovesdy CP, Lynch KE, Brent GA, Kalantar-Zadeh K. Updates on the management of diabetes in dialysis patients. Semin Dial. 2014; 27(2):135–45. PMID: PMCID: 10.1111/sdi.12198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.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(3):708–15. PMID: PMCID: 10.2337/db11-1015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Monnier L, Colette C, Owens D. The glycemic triumvirate and diabetic complications: Is the whole greater than the sum of its component parts? Diabetes Res Clin Pract. 2012; 95(3):303–11. PMID: 10.1016/j.diabres.2011.10.014. [DOI] [PubMed] [Google Scholar]
- 10.Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007; 39(2):175–91. PMID: 10.3758/bf03193146. [DOI] [PubMed] [Google Scholar]
- 11.Jin YP, Su XF, Yin GP, et al. Blood glucose fluctuations in hemodialysis patients with end stage diabetic nephropathy. J Diabetes Complications. 2015; 29(3):395–9. PMID: 10.1016/j.jdiacomp.2014.12.015. [DOI] [PubMed] [Google Scholar]
- 12.Home P. Contributions of basal and post-prandial hyperglycaemia to micro- and macrovascular complications in people with type 2 diabetes. Curr Med Res Opin. 2005; 21(7):989–98. PMID: 10.1185/030079905x49662. [DOI] [PubMed] [Google Scholar]
- 13.Colette C, Monnier L. Acute glucose fluctuations and chronic sustained hyperglycemia as risk factors for cardiovascular diseases in patients with type 2 diabetes. Horm Metab Res. 2007; 39(9):683–6. PMID: 10.1055/s-2007-985157. [DOI] [PubMed] [Google Scholar]
- 14.Weber C, Schnell O. The assessment of glycemic variability and its impact on diabetes-related complications: An overview. Diabetes Technol Ther. 2009; 11(10):623–33. PMID: 10.1089/dia.2009.0043. [DOI] [PubMed] [Google Scholar]
- 15.Monnier L, Colette C, Owens DR. Glycemic variability: The third component of the dysglycemia in diabetes. Is it important? How to measure it? J Diabetes Sci Technol. 2008; 2(6):1094–100. PMID: PMCID: 10.1177/193229680800200618. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hu G, Qiao Q, Tuomilehto J. Glucose tolerance and cardiovascular mortality. Cardiovasc Rev Reports. 2001; 22(11):649–54. https://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=14103450. [Google Scholar]
- 17.Ceriello A. Postprandial hyperglycemia and diabetes complications: Is it time to treat? Diabetes. 2005; 54(1):1–7. PMID: 10.2337/diabetes.54.1.1. [DOI] [PubMed] [Google Scholar]
- 18.Kazempour-Ardebili S, Lecamwasam VL, Dassanyake T, et al. Assessing glycemic control in maintenance hemodialysis patients with type 2 diabetes. Diabetes Care. 2009; 32(7):1137–42. PMID: PMCID: 10.2337/dc08-1688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Mirani M, Berra C, Finazzi S, et al. Inter-day glycemic variability assessed by continuous glucose monitoring in insulin-treated type 2 diabetes patients on hemodialysis. Diabetes Technol Ther. 2010; 12(10):749–53. PMID: 10.1089/dia.2010.0052. [DOI] [PubMed] [Google Scholar]
- 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(1):68–73. PMID: 10.1159/000362863. [DOI] [PubMed] [Google Scholar]
- 21.Rizzo MR, Barbieri M, Marfella R, Paolisso G. Reduction of oxidative stress and inflammation by blunting daily acute glucose fluctuations in patients with type 2 diabetes: Role of dipeptidyl peptidase-IV inhibition. Diabetes Care. 2012; 35(10):2076–82. PMID: PMCID: 10.2337/dc12-0199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Shyangdan DS, Royle P, Clar C, Sharma P, Waugh N, Snaith A. Glucagon-like peptide analogues for type 2 diabetes mellitus. Cochrane Database Syst Rev. 2011;2017(12). PMID: PMCID: 10.1002/14651858.CD006423.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Mori Y, Taniguchi Y, Sezaki K, Yokoyama J, Utsunomiya K. Liraglutide narrows the range of circadian glycemic variations in Japanese type 2 diabetes patients and nearly flattens these variations in drugnaive type 2 diabetes patients: A continuous glucose monitoringbased study. Diabetes Technol Ther. 2011; 13(11):1139–44. PMID: 10.1089/dia.2011.0137. [DOI] [PubMed] [Google Scholar]
- 24.Tesfaye S, Boulton AJM, Dyck PJ, et al. Diabetic neuropathies: Update on definitions, diagnostic criteria, estimation of severity, and treatments. Diabetes Care. 2010; 33(10):2285–93. PMID: PMCID: 10.2337/dc10-1303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kohnert KD, Augstein P, Heinke P, et al. Chronic hyperglycemia but not glucose variability determines HbA1c levels in well-controlled patients with type 2 diabetes. Diabetes Res Clin Pract. 2007; 77(3):420–6. PMID: 10.1016/j.diabres.2007.01.021. [DOI] [PubMed] [Google Scholar]
- 26.Borg R, Kuenen JC, Carstensen B, et al. Associations between features of glucose exposure and A1C: The A1C-Derived Average Glucose (ADAG) study. Diabetes. 2010; 59(7):1585–90. PMID: PMCID: 10.2337/db09-1774. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Nathan DM, Kuenen J, Borg R, et al. Translating the A1C assay into estimated average glucose values. Diabetes Care. 2008; 31(8):1473–8. PMID: PMCID: 10.2337/dc08-0545. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Sartore G, Chilelli NC, Burlina S, et al. The importance of HbA1c and glucose variability in patients with type 1 and type 2 diabetes: Outcome of continuous glucose monitoring (CGM). Acta Diabetol. 2012; 49(Suppl 1):S153-60. PMID: 10.1007/s00592-012-0391-4. [DOI] [PubMed] [Google Scholar]
- 29.Fang FS, Li ZB, Li CL, Tian H, Li J, Cheng XL. Influence of glycemic variability on the HbA1c level in elderly male patients with type 2 diabetes. Intern Med. 2012; 51(22):3109–13. PMID: 10.2169/internalmedicine.51.8077. [DOI] [PubMed] [Google Scholar]
- 30.Tanaka C, Saisho Y, Tanaka K, et al. Factors associated with glycemic variability in Japanese patients with diabetes. Diabetol Int. 2014; 5(1):36–42. https://doi.10.1007/s13340-013-0129-8. [Google Scholar]
- 31.Coelho S. What is the role of HbA1c in Diabetic Hemodialysis Patients? Semin Dial. 2016; 29(1):19–23. PMID: 10.1111/sdi.12408. [DOI] [PubMed] [Google Scholar]
- 32.Hoshino J, Mehrotra R, Rhee CM, et al. Using hemoglobin A1c to derive mean blood glucose in peritoneal dialysis patients. Am J Nephrol. 2013; 37(5):413–20. PMID: PMCID: 10.1159/000349929. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Gu Z, Du Y, Liu Y, et al. Effect of aging on islet beta-cell function and its mechanisms in Wistar rats. Age (Dordr). 2012; 34(6):1393–403. PMID: PMCID: 10.1007/s11357-011-9312-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Tschen SI, Dhawan S, Gurlo T, Bhushan A. Age-dependent decline in β-cell proliferation restricts the capacity of β-cell regeneration in mice. Diabetes. 2009; 58(6):1312–20. PMID: PMCID: 10.2337/db08-1651. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.DeFronzo RA, Abdul-Ghani MA. Preservation of β-cell function: The key to diabetes prevention. J Clin Endocrinol Metab. 2011; 96(8):2354–66. PMID: 10.1210/jc.2011-0246. [DOI] [PubMed] [Google Scholar]
- 36.Vella A, Zinsmeister AR. Predicting diabetes using measures of β-cell function. Diabetes. 2012; 61(3):562–3. PMID: PMCID: 10.2337/db11-1785. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.FLAT-SUGAR Trial investigators, Probstfield JL, Hirsch I Design of FLAT-SUGAR: Randomized trial of prandial insulin versus prandial glp-1 receptor agonist together with basal insulin and metformin for high-risk type 2 diabetes. Diabetes Care. 2015; 38(8):1558–66. PMID: 10.2337/dc14-2689. [DOI] [PubMed] [Google Scholar]
- 38.Ridker PM. A test in context: High-sensitivity C-reactive protein. J Am Coll Cardiol. 2016; 67(6):712–23. PMID: 10.1016/j.jacc.2015.11.037. [DOI] [PubMed] [Google Scholar]
- 39.De Mutsert R, Grootendorst DC, Axelsson J, et al. Excess mortality due to interaction between protein-energy wasting, inflammation and cardiovascular disease in chronic dialysis patients. Nephrol Dial Transplant. 2008; 23(9):2957–64. PMID: 10.1093/ndt/gfn167. [DOI] [PubMed] [Google Scholar]
- 40.Van Tellingen A, Grooteman MPC, Schoorl M, et al. Intercurrent clinical events are predictive of plasma C-reactive protein levels in hemodialysis patients. Kidney Int. 2002; 62(2):632–8. PMID: 10.1046/j.1523-1755.2002.00470.x. [DOI] [PubMed] [Google Scholar]
- 41.Snaedal S, Heimbürger O, Qureshi AR, et al. Comorbidity and acute clinical events as determinants of C-reactive Protein variation in hemodialysis patients: implications for patient survival. Am J Kidney Dis. 2009; 53(6):1024–33. PMID: 10.1053/j.ajkd.2009.02.008. [DOI] [PubMed] [Google Scholar]
- 42.Shafi T, Jaar BG, Plantinga LC, et al. Association of residual urine output with mortality, quality of life, and inflammation in incident hemodialysis patients: The Choices for Healthy Outcomes in Caring for End-Stage Renal Disease (CHOICE) study. Am J Kidney Dis. 2010; 56(2):348–58. PMID: PMCID: 10.1053/j.ajkd.2010.03.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Kovesdy CP, Kalantar-Zadeh K. Why is protein-energy wasting associated with mortality in chronic kidney disease? Semin Nephrol. 2009; 29(1):3–14. PMID: PMCID: 10.1016/j.semnephrol.2008.10.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Kato A, Takita T, Furuhashi M, Maruyama Y, Hishida A. Comparison of serum albumin, C-reactive protein and carotid atherosclerosis as predictors of 10-year mortality in hemodialysis patients. Hemodial Int. 2010; 14(2):226–32. PMID: 10.1111/j.1542-4758.2009.00432.x. [DOI] [PubMed] [Google Scholar]
- 45.Rodbard D. Interpretation of continuous glucose monitoring data: Glycemic variability and quality of glycemic control. Diabetes Technol Ther. 2009; 11(Suppl 1):S55-67. PMID: 10.1089/dia.2008.0132. [DOI] [PubMed] [Google Scholar]
- 46.Service FJ. Glucose variability. Diabetes. 2013; 62(5):1398–404. PMID: PMCID: 10.2337/db12-1396. [DOI] [PMC free article] [PubMed] [Google Scholar]