Comparison of Glycemic Markers in Chronic Hemodialysis Using Continuous Glucose Monitoring
Abstract
Background. The proportion of diabetic patients undergoing haemodialysis is rapidly increasing. Glucose control among such patients is difficult to assess. We aimed to evaluate the clinical performance of a continuous glucose monitoring system (CGMS) in type 2 diabetic patients on chronic haemodialysis.
Methods. We used a 4-day CGMS to monitor glucose levels in 19 haemodialysed type 2 diabetic patients (HD T2) including 2 days with and 2 days without dialysis session, and 39 non-HD T2 in a double-centre study.
Results. The glucose concentration according to the glucose meter and CGMS were correlated in HD T2 patients (r = 0.90, P < 0.0001) and in non-HD T2 patients (r = 0.81, P < 0.0001). The relative absolute difference (RAD) between glucose determined by a glucose meter and glucose determined by the CGMS did not differ between HD T2 and non-HD T2 patients (9.2 ± 10.5 vs. 8.2 ± 7.6%; P = 0.165). Glycated haemoglobin (A1c) and mean glucose concentration were strongly correlated in non-HD T2 patients (r = 0.71; P < 0.0001) but weakly correlated in HD T2 patients (r = 0.47; P = 0.042). Fructosamine was correlated with the mean glucose concentration in non-HD T2 (r = 0.67; P < 0.0001) but not in HD T2 patients (r = 0.04; P = 0.88).
Conclusion. CGM is a validated marker of glycaemic control in HD diabetic patients. This tool showed that A1c and fructosamine, despite being good markers of glycaemic control in non-HD diabetic patients, are of poor value in HD diabetic patients.
Introduction
The number of diabetic patients with nephropathy has gradually increased over the past decade [1]. Indeed, type 2 diabetic patients with end-stage renal disease (ESRD) now represent a substantial proportion of patients undergoing chronic haemodialysis [2]. Diabetic patients have a high mortality rate, mainly attributed to cardiovascular disease. This is particularly marked in diabetic patients on chronic haemodialysis. These patients have a poorer outcome than non-diabetic subjects [3].
Glycaemic control is a risk factor for micro- and macrovascular diseases in diabetic people [4–7] as in the general population [8] with positive intervention trials regarding microvascular endpoints and controversial results regarding cardiovascular outcomes [6,9]. Some studies have previously explored the association between glycated haemoglobin (A1c) and clinical outcome in dialysis patients. However, the findings are not consistent, with some authors finding an association between high A1c and poor dialysis survival and others showing no association between A1c and survival [10–13].
A1c and fructosamine, which are routine markers of glycaemic control in the diabetic populations, were shown to inappropriately reflect glycaemic control, nearly 20 years ago, in dialysis of diabetic patients [14,15]. However, the impact of modern anaemia treatment and new devices for A1c determination on the relationship between glucose concentration and markers of glucose metabolism has not been extensively reported to our knowledge.
Devices that continuously monitor glucose levels have been developed to easily assess blood glucose patterns and improve management of diabetes. The continuous interstitial glucose monitoring systems (CGMS) can detect unrecognized hypoglycaemia [16,17] and other patterns requiring insulin adjustment, not detected with intermittent blood glucose monitoring [18,19]. Early uncontrolled studies [16,17] and a cross-over trial [20] have shown some benefit of using CGMS devices in terms of improved glycaemic control. This was recently confirmed using randomized controlled trials in several diabetic populations, particularly in pregnant women with diabetes [21] and in uncontrolled type 1 diabetes [22,23].
Our objective was to evaluate the clinical performance of a CGMS in type 2 diabetic patients with haemodialysis.
Patients
All haemodialysed type 2 diabetic (HD T2) patients regularly attending the dialysis centres in Poitiers and Corbeil-Essonnes were invited to have their glucose control assessed using a 4-day MiniMed CGMS (CGMS; Medtronic Diabetes, Northridge, CA, USA). Through a catheter inserted subcutaneously, interstitial glucose (IG) was recorded with the CGMS. The monitor recorded IG levels every 10 s, and then stored a smoothed average over 5 min. The range of IG detection was 40–400 mg/dL. CGMS results were obtained for 2 days with and 2 days without dialysis sessions.
Type 2 diabetes was diagnosed using ADA and WHO classification systems. Patients were on stable diabetes treatment for at least 4 months.
Most of the HD T2 patients were familiar with blood glucose monitoring. A dedicated study nurse carried out capillary blood glucose measurements for CGMS calibration. All patients' usual blood glucose meters were checked using control solutions provided by manufacturers (Accucheck Go Roche Diagnostics, Basel, Switzerland, and One Touch Ultra LifeScan Inc., Milpitas, CA, USA), both devices using the whole blood hexokinase method. Self-monitoring of capillary glucose performed before breakfast and before dinner was used to calibrate the CGMS, as recommended [24].
The CGMS was inserted in the late afternoon, for >4 h after meal on a weekday without dialysis. This system was blinded for glucose results; thus, diabetes treatment management was conducted as routinely by nephrologist and/or diabetologist and was not influenced by CGMS results. To distinguish between diurnal and nocturnal hypoglycaemias, we made a post hoc analysis of the time of hypoglycaemia and considered that nighttime was 23:00–7:00, and daytime was 7:00–23:00.
Non-HD T2 controls were a group of non-nephropathic type 2 diabetic patients. They were recruited for a CGMS study assessing post-meal glycaemic excursions.
This study was conducted according to Helsinki declaration principles. All participating subjects gave informed consent.
Biochemical measurements
Samples were drawn at the start of a dialysis session on the day preceding CGMS insertion.
A1c levels were determined using an HPLC DCCT/EDIC calibrated method: ADAMS A1C HA-8160 (Menarini, Florence, Italy) (normal values 4.0–6.0%).
Fructosamine levels were determined using a colorimetric method (Fruc, Roche Diagnostics, Meylan, France). Haemoglobin was routinely determined using a colorimetric method run on a Roche Sysmex XE 2100 analyser (Roche Diagnostics).
Statistical analysis
Calculations of the relative absolute difference (RAD) that evaluate the accuracy of the CGMS and the area under the curve (AUC) are presented in supplementary material.
Data were stored and analysed using the Statview 5.0 software (SAS Institute, Cary, NC, USA) program. Data are presented as means ± SD, or median (range). The performance of the CGMS was compared to that of capillary glucose self-monitoring using the Bland & Altman method [25]. Non-HD T2 controls and HD T2 patients were compared using the unpaired t-test. The mean glucose concentration was the mean of all glucose values recorded over the 96 h. Repeated measure analysis of variance was used to compare data obtained during the first 3 h of dialysis and with those obtained during the corresponding 3 h of a day without dialysis. We calculated Pearson's correlation coefficients and adjusted correlation coefficients performing multivariate linear regression analyses.
Results
We included 19 HD T2 and 39 non-HD T2 patients in our study. The clinical and biological characteristics of study participants are summarized in Table 1. There were no significant differences in gender, age or BMI between HD T2 and non-HD T2 controls. Diabetes treatment consisted in bedtime insulin added to oral therapy for non-HD T2 controls and in insulin for most of the HD T2 patients.
Table 1
Comparison of clinical and biological characteristics between type 2 diabetic patients on chronic (HD T2) and non-haemodialyzed type 2 diabetic patients (non-HD T2)
| HD T2 | Non-HD | ||
|---|---|---|---|
| patients | T2 patients | P | |
| Age (years) | 64 ± 10 | 65 ± 6 | 0.753 |
| Sex (male/female) | 8/11 | 25/14 | 0.112 |
| Diabetes duration (years) | 21 ± 11 | 17 ± 7 | 0.055 |
| Duration of dialysis treatment (months)a | 24 (13–35) | – | – |
| BMI (kg/m2) | 30.2 ± 4.9 | 29.6 ± 4.3 | 0.656 |
| Serum creatinine (μmol/L) | – | 76 ± 24 | – |
| Haemoglobin (g/dL) | 11.6 ± 0.8 | 13.6 ± 1.2 | <0.0001 |
| Mean corpuscular volume (μm3) | 93 ± 5 | 91 ± 5 | 0.0483 |
| EPO dose (IU/week) | 9300 ± 1800 | – | – |
| Treatment with ACE-I or ARB (yes/no) | 8/11 | 32/39 | 0.817 |
| Treatment with beta blockers (yes/no) | 2/17 | 4/35 | 0.975 |
| Diet/OHA/OHA + insulin/insulin (n) | 1/6/2/10 | 0/0/39/0 | – |
| Fructosamine (μmol/L) | 362 ± 46 | 285 ± 42 | <0.0001 |
| A1c (%) | 7.2 ± 1.1 | 7.7 ± 0.8 | 0.085 |
| Mean glucose (CGMS) (mg/dL) | 168 ± 40 | 151 ± 41 | 0.145 |
| HD T2 | Non-HD | ||
|---|---|---|---|
| patients | T2 patients | P | |
| Age (years) | 64 ± 10 | 65 ± 6 | 0.753 |
| Sex (male/female) | 8/11 | 25/14 | 0.112 |
| Diabetes duration (years) | 21 ± 11 | 17 ± 7 | 0.055 |
| Duration of dialysis treatment (months)a | 24 (13–35) | – | – |
| BMI (kg/m2) | 30.2 ± 4.9 | 29.6 ± 4.3 | 0.656 |
| Serum creatinine (μmol/L) | – | 76 ± 24 | – |
| Haemoglobin (g/dL) | 11.6 ± 0.8 | 13.6 ± 1.2 | <0.0001 |
| Mean corpuscular volume (μm3) | 93 ± 5 | 91 ± 5 | 0.0483 |
| EPO dose (IU/week) | 9300 ± 1800 | – | – |
| Treatment with ACE-I or ARB (yes/no) | 8/11 | 32/39 | 0.817 |
| Treatment with beta blockers (yes/no) | 2/17 | 4/35 | 0.975 |
| Diet/OHA/OHA + insulin/insulin (n) | 1/6/2/10 | 0/0/39/0 | – |
| Fructosamine (μmol/L) | 362 ± 46 | 285 ± 42 | <0.0001 |
| A1c (%) | 7.2 ± 1.1 | 7.7 ± 0.8 | 0.085 |
| Mean glucose (CGMS) (mg/dL) | 168 ± 40 | 151 ± 41 | 0.145 |
aData are mean ± SD or median (25–75 percentile).
EPO, erythropoietin; OHA, oral hypoglycaemic agents; A1c, glycated haemoglobin.
Table 1
Comparison of clinical and biological characteristics between type 2 diabetic patients on chronic (HD T2) and non-haemodialyzed type 2 diabetic patients (non-HD T2)
| HD T2 | Non-HD | ||
|---|---|---|---|
| patients | T2 patients | P | |
| Age (years) | 64 ± 10 | 65 ± 6 | 0.753 |
| Sex (male/female) | 8/11 | 25/14 | 0.112 |
| Diabetes duration (years) | 21 ± 11 | 17 ± 7 | 0.055 |
| Duration of dialysis treatment (months)a | 24 (13–35) | – | – |
| BMI (kg/m2) | 30.2 ± 4.9 | 29.6 ± 4.3 | 0.656 |
| Serum creatinine (μmol/L) | – | 76 ± 24 | – |
| Haemoglobin (g/dL) | 11.6 ± 0.8 | 13.6 ± 1.2 | <0.0001 |
| Mean corpuscular volume (μm3) | 93 ± 5 | 91 ± 5 | 0.0483 |
| EPO dose (IU/week) | 9300 ± 1800 | – | – |
| Treatment with ACE-I or ARB (yes/no) | 8/11 | 32/39 | 0.817 |
| Treatment with beta blockers (yes/no) | 2/17 | 4/35 | 0.975 |
| Diet/OHA/OHA + insulin/insulin (n) | 1/6/2/10 | 0/0/39/0 | – |
| Fructosamine (μmol/L) | 362 ± 46 | 285 ± 42 | <0.0001 |
| A1c (%) | 7.2 ± 1.1 | 7.7 ± 0.8 | 0.085 |
| Mean glucose (CGMS) (mg/dL) | 168 ± 40 | 151 ± 41 | 0.145 |
| HD T2 | Non-HD | ||
|---|---|---|---|
| patients | T2 patients | P | |
| Age (years) | 64 ± 10 | 65 ± 6 | 0.753 |
| Sex (male/female) | 8/11 | 25/14 | 0.112 |
| Diabetes duration (years) | 21 ± 11 | 17 ± 7 | 0.055 |
| Duration of dialysis treatment (months)a | 24 (13–35) | – | – |
| BMI (kg/m2) | 30.2 ± 4.9 | 29.6 ± 4.3 | 0.656 |
| Serum creatinine (μmol/L) | – | 76 ± 24 | – |
| Haemoglobin (g/dL) | 11.6 ± 0.8 | 13.6 ± 1.2 | <0.0001 |
| Mean corpuscular volume (μm3) | 93 ± 5 | 91 ± 5 | 0.0483 |
| EPO dose (IU/week) | 9300 ± 1800 | – | – |
| Treatment with ACE-I or ARB (yes/no) | 8/11 | 32/39 | 0.817 |
| Treatment with beta blockers (yes/no) | 2/17 | 4/35 | 0.975 |
| Diet/OHA/OHA + insulin/insulin (n) | 1/6/2/10 | 0/0/39/0 | – |
| Fructosamine (μmol/L) | 362 ± 46 | 285 ± 42 | <0.0001 |
| A1c (%) | 7.2 ± 1.1 | 7.7 ± 0.8 | 0.085 |
| Mean glucose (CGMS) (mg/dL) | 168 ± 40 | 151 ± 41 | 0.145 |
aData are mean ± SD or median (25–75 percentile).
EPO, erythropoietin; OHA, oral hypoglycaemic agents; A1c, glycated haemoglobin.
Validation of CGMS data in HD
The relationships between glycaemia determined by a glucose meter and mean glucose concentration determined by the CGMS were plotted as a function of HD status (Figure 1). We found a similar correlation between the glucose meter and mean glucose concentration in HD T2 patients (r = 0.90, P < 0.0001) and in non-HD T2 patients (r = 0.81, P < 0.0001).
Fig. 1
Correlation between mean glucose obtained through CGMS and glucose meters in HD T2 (blue circles) and non-HD T2 controls (red circles). Correlation in the whole patient group (r = 0.84; P < 0.0001). Correlation in HD T2 patients—blue line (r = 0.90; P < 0.0001). Correlation in non-HD T2 patients—red line (r = 0.81; P < 0.0001).
The RAD between glucose levels determined by the glucose meter and that determined by the CGMS did not differ significantly between HD T2 and non-HD T2 controls (9.2 ± 10.5 vs. 8.2 ± 7.6%; P = 0.165). The difference in glucose concentration between glucose meter and CGMS measurements was plotted using the Bland & Altman representation, using 624 paired points.
Comparison between HD T2 and non-HD T2 patients
A1c levels were lower in HD T2 patients than in non-HD T2 controls (7.2 ± 1.1 vs. 7.7 ± 0.8%; P = 0.051). Conversely, fructosamine levels were higher in HD patients than in non-HD T2 patients (362 ± 46 vs. 285 ± 42 μmol/L; P < 0.001).
The relationships between mean glucose determined with the CGMS and A1c on one hand and fructosamine on the other hand were plotted as a function of HD status (Figures 2 and 3).
Fig. 2
Correlation between mean glucose and A1c in HD T2 (blue circles) and non-HD T2 controls (red circles). Correlation in the whole patient group (r = 0.51; P < 0.0001). Correlation in HD T2 patients—blue line (r = 0.47; P = 0.042). Correlation in non-HD T2 patients—red line (r = 0.71; P < 0.0001).
Fig. 3
Correlation between mean glucose and fructosamine in HD T2 (blue circles) and non-HD T2 controls (red circles). Correlation in the whole patient group (r = 0.41; P = 0.0042). Correlation in HD T2 patients—no correlation line represented (r = −0.04; P = 0.88). Correlation in non-HD T2 patients—red line (r = 0.67; P < 0.0001).
Considering all patients together, we found a correlation between A1c and mean glucose (r = 0.51; P < 0.0001). This correlation had borderline statistical significance (r = 0.47; P = 0.042) in HD T2 patients, but was more strongly significant in non-HD T2 patients (r = 0.71; P < 0.0001).
Adjustment on haemoglobin did not modify the results: r adjusted = 0.52 (P = 0.0003), 0.42 (P = 0.08) and 0.68 (P = 0.0002) in the whole population, in HD T2 and in non-nephropathic controls, respectively.
Fructosamine levels were positively correlated with mean glucose for the whole group (r = 0.41; P = 0.0042). However, no correlation was found for HD T2 patients (r = −0.04; P = 0.88); the correlation observed for the whole group appeared to be due to non-nephropathic type 2 diabetic controls (r = 0.67; P < 0.0001).
Adjustment on haemoglobin did not modify the results: r adjusted = 0.39 (P = 0.04), −0.10 (P = 0.69) and 0.59 (0.003) in the whole population, in HD T2 and in non-nephropathic controls, respectively.
The effect of dialysis on glycaemic parameters
The comparison between CGMS data obtained during dialysis and those obtained on a day without dialysis session is presented in Table 2. Glucose concentration RAD, which evaluates the accuracy of the device, obtained on the day of dialysis did not differ significantly from that obtained on the day without dialysis.
Table 2
Comparison of CGMS data in type 2 diabetic patients on chronic haemodialysis according to the day with or without dialysis session
| Day with | Day without | ||
|---|---|---|---|
| dialysis session | dialysis session | P-valuea | |
| RAD (%) | 7.0 ± 5.2 | 9.0 ± 5.6 | 0.360 |
| Mean glucose CGMS (mg/dL) | 164 ± 43 | 176 ± 46 | 0.113 |
| AUC glucose (mg/dL × 24 h) | 3850 ± 1073 | 4113 ± 1130 | 0.062 |
| Time <70 mg/dL (h/day)b | 0.6 (0–5.2) | 0.3 (0–2.2) | 0.169 |
| Time >180 mg/dL (h/day)b | 7.8 (1.1–23.4) | 8.7 (0.4–24) | 0.487 |
| Glycaemic variability | |||
| SD of glucose (mg/dL) | 48 ± 24 | 52 ± 20 | 0.347 |
| Day with | Day without | ||
|---|---|---|---|
| dialysis session | dialysis session | P-valuea | |
| RAD (%) | 7.0 ± 5.2 | 9.0 ± 5.6 | 0.360 |
| Mean glucose CGMS (mg/dL) | 164 ± 43 | 176 ± 46 | 0.113 |
| AUC glucose (mg/dL × 24 h) | 3850 ± 1073 | 4113 ± 1130 | 0.062 |
| Time <70 mg/dL (h/day)b | 0.6 (0–5.2) | 0.3 (0–2.2) | 0.169 |
| Time >180 mg/dL (h/day)b | 7.8 (1.1–23.4) | 8.7 (0.4–24) | 0.487 |
| Glycaemic variability | |||
| SD of glucose (mg/dL) | 48 ± 24 | 52 ± 20 | 0.347 |
aPaired t-test.; bData are mean ± SD or (range) when specified.
RAD, relative absolute difference; CGMS, continuous glucose monitoring system.
Table 2
Comparison of CGMS data in type 2 diabetic patients on chronic haemodialysis according to the day with or without dialysis session
| Day with | Day without | ||
|---|---|---|---|
| dialysis session | dialysis session | P-valuea | |
| RAD (%) | 7.0 ± 5.2 | 9.0 ± 5.6 | 0.360 |
| Mean glucose CGMS (mg/dL) | 164 ± 43 | 176 ± 46 | 0.113 |
| AUC glucose (mg/dL × 24 h) | 3850 ± 1073 | 4113 ± 1130 | 0.062 |
| Time <70 mg/dL (h/day)b | 0.6 (0–5.2) | 0.3 (0–2.2) | 0.169 |
| Time >180 mg/dL (h/day)b | 7.8 (1.1–23.4) | 8.7 (0.4–24) | 0.487 |
| Glycaemic variability | |||
| SD of glucose (mg/dL) | 48 ± 24 | 52 ± 20 | 0.347 |
| Day with | Day without | ||
|---|---|---|---|
| dialysis session | dialysis session | P-valuea | |
| RAD (%) | 7.0 ± 5.2 | 9.0 ± 5.6 | 0.360 |
| Mean glucose CGMS (mg/dL) | 164 ± 43 | 176 ± 46 | 0.113 |
| AUC glucose (mg/dL × 24 h) | 3850 ± 1073 | 4113 ± 1130 | 0.062 |
| Time <70 mg/dL (h/day)b | 0.6 (0–5.2) | 0.3 (0–2.2) | 0.169 |
| Time >180 mg/dL (h/day)b | 7.8 (1.1–23.4) | 8.7 (0.4–24) | 0.487 |
| Glycaemic variability | |||
| SD of glucose (mg/dL) | 48 ± 24 | 52 ± 20 | 0.347 |
aPaired t-test.; bData are mean ± SD or (range) when specified.
RAD, relative absolute difference; CGMS, continuous glucose monitoring system.
The glucose time course in the first 3 h of dialysis, or at an equivalent time on the days without a dialysis session, is presented in Figure 4. The 3-h mean glucose concentration was significantly lower on the day of dialysis (P < 0.0001).
Fig. 4
Glucose concentration in the first 3 h of the dialysis session (blue circles) or equivalent time of the following day without dialysis (red circles). Circles and error bars represent mean and SEM.
Fourteen patients remained free from hypoglycaemia when monitored during the day without a dialysis session. Only two patients spent some intra-dialytic time with IG <70 mg/dL (22 and 45 min, respectively). The time in hypoglycaemias was low with 8/17 patients spending some time <70 mg/dL during nighttime (median 3 min/h) and 6/17 during daytime (median 2 min/h).
Discussion
Our objective was to study the ability of the CGMS to evaluate the glycaemic control in HD T2 patients, in comparison with capillary glycaemia, fructosamine and HbA1c. We were able to correlate capillary blood glucose and mean glucose value determined by the CGMS, as already known from the general diabetic population [26] in non-nephropathic controls but also in HD T2 patients. Accordingly, the RAD in glucose concentration was not different between HD T2 and non-HD T2 patients. To our knowledge, this point was not validated previously. Our results show that the CGMS is able to correctly evaluate glycaemic control in HD T2 patients. Using the CGMS, we found that mean glucose concentration correlated weakly with A1c but did not correlate at all with fructosamine in HD T2. In contrast, both markers correlated well with mean glucose concentration in non-HD T2 patients. Thus, the CGMS could represent a major advance in assessing glycaemic control in HD T2 patients.
Our findings suggest that A1c and fructosamine are not as good markers of glucose control in haemodialysis as in non-HD T2 patients, even with modern treatment for anaemia and calibrated A1c determinations. This is consistent with a number of previous reports [27–30], albeit most of them in non-EPO-treated HD patients. Adjusting on blood haemoglobin did not modify the relationship between A1c or fructosamine and mean glucose concentration in our study, neither in the whole patient group nor in HD and non-HD patients considered separately. Glycated albumin is an attractive alternative marker of glucose control in HD T2 patients, as it seems to be only weakly affected by haemoglobin levels [31–33]. Unfortunately, we were not able to measure glycated albumin in our population. The correlation between mean glucose evaluated using a CGMS and glycated albumin now needs to be determined.
We showed that the glucose profile obtained during dialysis showed lower glucose values than that obtained at the same time on the days without a dialysis session, maybe due to the dialysis of plasma glucose. Improvement of blood glucose during dialysis was also reported almost 30 years ago, using a laborious system of continuous glycaemia monitoring in HD T2 patients. However, measurements were only taken during the haemodialysis period [34]. Similar to our study, a drop in glycaemia was observed during HD sessions [35]. This is also consistent with a more recent study, which used a CGMS to show that the use of dialysis fluids that do not contain glucose improved glycaemic control in eight continuous ambulatory peritoneal dialysis diabetic patients [36].
The clinical value of using a CGMS in HD diabetic patients must be questioned. In the present study, our aim was not to prove its utility to improve patients' metabolic control and/or to alter the ominous fate of HD in diabetic patients. However, the CGMS helped to identify poor metabolic control in our HD population, which was not suspected based on A1c levels. The improvement of hyperglycaemia is crucial in order to limit the progression of diabetic complications [4–7] and may decrease the mortality rate [10], particularly in diabetic HD patients. Poor glycaemic control before starting haemodialysis treatment is associated with increased cardiovascular morbidity and a shortened survival in diabetic patients with ESRD [32]. Many HD patients with diabetes have inadequate glycaemic control [37]. Whether the CGMS could be used to indicate necessary adjustments in diabetes treatment in HD T2 patients, potentially lowering mortality rates in these patients would require a specifically designed trial. However, encouraging results in non-HD patients with hard endpoints such as in pregnant women could stimulate the settlement of such a trial. The benefits of such interventions remain speculative but should be properly determined using suitable prospective large-scale studies.
Our study has a number of limitations. The results must be interpreted with caution due to the small number of patients. Moreover, A1c and fructosamine and CGMS measurements were not repeated for each patient. Indeed, these measurements were made during a relatively short period. However, these limitations did not affect the good correlation observed between mean glucose concentration and these markers in non-HD T2 patients. We considered CGMS-determined IG as a surrogate for blood glucose, validated in non-HD T2 patients only [38]. However, the glucose RADs were similar and acceptable in both HD T2 and non-HD T2 patients, strongly suggesting that the CGMS was an accurate indicator of glycaemic control in HD subjects.
In conclusion, a CGMS is a new validate tool to evaluate glycaemic control in HD T2. Further studies are necessary to assess whether the CGMS can improve glycaemic control and related endpoints in this population as in other populations of diabetic subjects [21–23].
We thank all participating patients. We greatly appreciate the help provided by staff at the Department of Diabetology of CHU Poitiers and Centre Hospitalier Sud Francilien in conducting this study. We thank Cecile Demer (Poitiers) for her secretarial assistance. Sylviane Parenteau (Poitiers) is acknowledged for her nurse skills. The association GEMMS (Poitiers) is acknowledged for its financial support. Alex Edelmann & coll edited the English of the text.
Conflict of interest statement. None declared.
References
1
, , , et al.
Prevalence of chronic kidney disease in the United States
,
JAMA
,
2007
, vol.
298
(pg.
2038
-
2047
)
2
U.S. Renal Data System
USRDS 2007 Annual Data Report National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, 2007
3
, , , et al.
Interdialytic weight gain and survival in hemodialysis patients: effects of duration of ESRD and diabetes mellitus
,
Kidney Int
,
2000
, vol.
57
(pg.
1141
-
1151
)
4
The Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications Research Group
Retinopathy and nephropathy in patients with type 1 diabetes four years after a trial of intensive therapy
,
N Engl J Med
,
2000
, vol.
342
(pg.
381
-
389
)
5
, , , et al.
Multifactorial intervention and cardiovascular disease in patients with type 2 diabetes
,
N Engl J Med
,
2003
, vol.
348
(pg.
383
-
393
)
6
, , , et al.
Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes
,
N Engl J Med
,
2008
, vol.
358
(pg.
2560
-
2572
)
7
The DCCT Research Group
Effect of intensive therapy on the development and progression of diabetic nephropathy in the Diabetes Control and Complications Trial
,
Kidney Int
,
1995
, vol.
47
(pg.
1703
-
1720
)
8
, , , et al.
Glycated haemoglobin, diabetes, and mortality in men in Norfolk cohort of European Prospective Investigation of Cancer and Nutrition (EPIC-Norfolk)
,
BMJ
,
2001
, vol.
332
(pg.
996
-
999
)
9
, , , et al.
Effects of intensive glucose lowering in type 2 diabetes
,
N Engl J Med
,
2008
, vol.
358
(pg.
2545
-
2559
)
10
, , , et al.
A1c and survival in maintenance hemodialysis patients
,
Diabetes Care
,
2007
, vol.
30
(pg.
1049
-
1055
)
11
, , , et al.
Glycemic control is a predictor of survival for diabetic patients on hemodialysis
,
Diabetes Care
,
2001
, vol.
24
(pg.
909
-
913
)
12
, , , et al.
Impact of glycemic control on survival of diabetic patients on chronic regular hemodialysis: a 7-year observational study
,
Diabetes Care
,
2006
, vol.
29
(pg.
1496
-
1500
)
13
, , , et al.
Hemodialyzed type I and type II diabetic patients in the US: Characteristics, glycemic control, and survival
,
Kidney Int
,
2006
, vol.
70
(pg.
1503
-
1509
)
14
, , .
Serum fructosamine in uraemia, myeloma and acute inflammatory disorders—relationship to serum glucose and albumin levels
,
Ann Clin Biochem
,
1989
, vol.
26
Pt 1): 63–68
15
, , , et al.
Glycated proteins as indices of glycaemic control in diabetic patients with chronic renal failure
,
Diabet Med
,
1996
, vol.
13
(pg.
514
-
519
)
16
, , , et al.
Limitations of conventional methods of self-monitoring of blood glucose: lessons learned from 3 days of continuous glucose sensing in pediatric patients with type 1 diabetes
,
Diabetes Care
,
2001
, vol.
24
(pg.
1858
-
1862
)
17
, , , et al.
Nocturnal hypoglycemia detected with the continuous glucose monitoring system in pediatric patients with type 1 diabetes
,
J Pediatr
,
2002
, vol.
141
(pg.
625
-
630
)
18
, , , et al.
A pilot study of the continuous glucose monitoring system: clinical decisions and glycemic control after its use in pediatric type 1 diabetic subjects
,
Diabetes Care
,
2001
, vol.
24
(pg.
2030
-
2034
)
19
, , , et al.
Study of glucose profiles with continuous glucose monitoring in adolescents with poorly controlled type 2 diabetes mellitus
,
J Pediatr Endocrinol Metab
,
2008
, vol.
21
(pg.
729
-
736
)
20
, .
Continuous subcutaneous glucose monitoring improved metabolic control in pediatric patients with type 1 diabetes: a controlled crossover study
,
Pediatrics
,
2003
, vol.
111
(pg.
933
-
938
)
21
, , , et al.
Effectiveness of continuous glucose monitoring in pregnant women with diabetes: randomised clinical trial
,
BMJ
,
2008
, vol.
337
pg.
a1680
22
, , , et al.
Improved glycemic control in poorly controlled patients with type 1 diabetes using real-time continuous glucose monitoring
,
Diabetes Care
,
2006
, vol.
29
(pg.
2730
-
2732
)
23
, , , et al.
Continuous glucose monitoring and intensive treatment of type 1 diabetes
,
N Engl J Med
,
2008
, vol.
359
(pg.
1464
-
1476
)
24
, , .
Real-time continuous glucose monitoring using GuardianRT: from research to clinical practice
,
Diabetes Metab
,
2007
, vol.
33
(pg.
415
-
420
)
25
, .
Statistical methods for assessing agreement between two methods of clinical measurement
,
Lancet
,
1986
, vol.
1
(pg.
307
-
310
)
26
, , , et al.
The continuous glucose monitoring system is useful for detecting unrecognized hypoglycemias in patients with type 1 and type 2 diabetes but is not better than frequent capillary glucose measurements for improving metabolic control
,
Diabetes Care
,
2003
, vol.
26
(pg.
1153
-
1157
)
27
, , .
Glycosylated haemoglobin in chronic renal failure
,
Br Med J
,
1979
, vol.
1
(pg.
1183
-
1184
)
28
, , .
Glycosylated haemoglobin in renal failure
,
Diabetologia
,
1980
, vol.
18
(pg.
437
-
440
)
29
, , , et al.
Long-term glycemic control measurements in diabetic patients receiving hemodialysis
,
Am J Kidney Dis
,
2002
, vol.
39
(pg.
297
-
307
)
30
, , , et al.
The impact of glycaemic control on outcomes in patients with end-stage renal disease and type 2 diabetes
,
Nephrology (Carlton, Vic)
,
2008
, vol.
13
(pg.
124
-
127
)
31
, , , et al.
Indicators for blood glucose control in diabetics with end-stage chronic renal disease: GHb vs. glycated albumin (GA)
,
J Med Invest
,
2006
, vol.
53
(pg.
223
-
228
)
32
, , , et al.
Association between markers of glycemic control, cardiovascular complications and survival in type 2 diabetic patients with end-stage renal disease
,
Intern Med (Tokyo, Japan)
,
2007
, vol.
46
(pg.
807
-
814
)
33
, , , et al.
Comparison of glycated albumin and hemoglobin A(1c) levels in diabetic subjects on hemodialysis
,
Kidney Int
,
2008
, vol.
73
(pg.
1062
-
1068
)
34
, .
Continuous blood glucose monitoring and characteristics of diabetes in patients on maintenance haemodialysis treatment
,
Scand J Urol Nephrol
,
1979
, vol.
13
(pg.
309
-
312
)
35
, , .
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
, vol.
11
(pg.
280
-
287
)
36
, , , et al.
Glycemic control in diabetic CAPD patients assessed by continuous glucose monitoring system (CGMS)
,
Kidney Int
,
2003
, vol.
64
(pg.
1480
-
1486
)
37
, , , et al.
Adequacy of glycemic control in hemodialysis patients with diabetes
,
Diabetes Care
,
2006
, vol.
29
(pg.
2247
-
2251
)
38
, .
Can interstitial glucose assessment replace blood glucose measurements?
,
Diabet Technol Ther
,
2000
, vol.
2
(pg.
461
-
472
)
Author notes
* Both authors contributed equally to this work.
© The Author [2009]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org
Oxford University Press
© The Author [2009]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org
Submit a comment
You have entered an invalid code
Thank you for submitting a comment on this article. Your comment will be reviewed and published at the journal's discretion. Please check for further notifications by email.
Source: https://academic.oup.com/ndt/article/24/9/2866/1855442
0 Response to "Comparison of Glycemic Markers in Chronic Hemodialysis Using Continuous Glucose Monitoring"
Post a Comment