Original Paper Received: January 21, 2014 Accepted: April 12, 2014 Published online: October 7, 2014
Blood Purif 2014;38:68–73 DOI: 10.1159/000362863
Glycemic Pattern in Diabetic Patients on Hemodialysis: Continuous Glucose Monitoring (CGM) Analysis Massimo Gai a Ilenia Merlo a Sergio Dellepiane a Vincenzo Cantaluppi a Gianluca Leonardi a Fabrizio Fop a Cesare Guarena a Giorgio Grassi b Luigi Biancone a
a SCU Nephrology, Dialysis and Transplantation, b SCDU Endocrinology, Diabetology and Metabolism, Azienda Ospedaliera Città della Salute e della Scienza di Torino, Presidio ‘Molinette’, Torino, Italy
Key Words Continuous glucose monitoring system · Diabetes · Diabetic nephropathy · Glucose · Glycemic variability · Hemodialysis
Abstract Background/Aims: Recent evidences suggest that hemodialysis (HD) induces glycemic variations in diabetic patients. Continuous glucose monitoring (CGM) devices measure interstitial glucose in a ‘Holter-like’ manner thereby improving the glycemic control assessment method. Methods: A CGM device (Medtronic iPRO) was used on 12 diabetic patients with chronic HD for 6 days to assess intra- and extra-dialytic interstitial glucose. Results: In all enrolled patients, HD was associated with a decrease of interstitial glucose values. Intradialytic glucose nadir was 79 mg/dl and it was reached at the third hour after the beginning of the session. At the end of HD, interstitial glucose increased in all patients and a glycemic peak (187 mg/dl) occurred after an average time of 2.5 h. No episodes of nocturnal hypoglycemia occurred. Conclusion: HD is associated with significant intradialytic reduction of glycemia and postdialytic hyperglycemia. CGM devices result in better monitoring of glycemic trends in diabetic patients on chronic HD and could improve insulin management.
Introduction
Diabetic nephropathy is one of the leading causes of End Stage Renal Disease (ESRD) in western countries causing 21.5–27.4% of new hemodialysis (HD) accesses [1]. Since the kidney is a relevant site for glucose and insulin metabolism, glycemic control is particularly challenging in ESRD diabetic patients with a higher risk of both hypo- and hyperglycemia. Furthermore, recent findings suggest that dialysis may be per se a cause of glycemic variability; in fact, glucose is freely filtered during HD sessions and many types of hemodialyzers may adsorb insulin [2]. Moreover, an increasing body of evidence indicates glycemic variability is an independent risk factor of morbidity and mortality in diabetic patients [3]. Current diabetes monitoring is based on self-measured capillary blood glucose and glicated hemoglobin (HbA1c) levels [4]. In chronic HD, blood glucose is usually measured at the beginning and at the end of the session, but there are only few studies about glycemic trends during and after the session [5–7]. Moreover, in HD patients, glicated hemoglobin (HbA1c) may be an inadequate index of glycemic control due to ESRD-related anemia or frequent blood-transfusions. Furthermore, HbA1c
© 2014 S. Karger AG, Basel © 2014 S. Karger AG, Basel 0253–5068/14/0381–0068$39.50/0 E-Mail [email protected] www.karger.com/bpu
Massimo Gai, MD Azienda Ospedaliera S. Giovanni Battista di Torino S.C.U. Nefrologia, Dialisi e Trapianto Corso Bramante 88, IT–10126 Torino (Italy) E-Mail massimogai @ gmail.com
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 3/21/2015 9:14:15 PM
estimates mean plasmatic glucose but not glycemic instability. Continue glucose monitoring (CGM) is a new Holterlike system to monitor interstitial glucose which allows an analysis of several parameters such as peak, nadir, mean and variability. Interstitial glucose is the expression of blood glucose; only in the event of a rapid blood glucose variation, you can observe a delay (‘lag time’) of 10–15 minutes [8]. The aim of this study was to assess the glycemic control in diabetic patients on chronic HD using a CGM system and to monitor the interstitial glucose levels for 6 consecutive days.
Data Collection All patients were trained to measure capillary blood glucose 4 times/day and to record glucose values and administered insulin dosages and details of food consumption, physical activity and symptoms. The system must be calibrated, in data processing, through the insertion of at least four glucose values for each day of recording. Glucose values were obtained using a glucometer (GM); this is a device for the self-monitoring of capillary blood glucose. Each patient had its own GM checked before the study. Day 1 laboratory glucose levels were compared with data obtained from CGM and patients’ provided capillary-glucometer (Accu-Chek® Aviva, Roche Diagnostics). Further blood tests performed at day 1 after study admission were: hemoglobin, HbA1c, urea, albumin. Food consumption was not allowed during dialysis sessions. On the 6th day from the date of study admission, GGM devices were removed and data analyzed.
Materials and Methods
HD Sessions Patients’ HD prescription was not changed during the study and different filters and types of HD (i.e., acetate free biofiltration, standard bicarbonate HD, high flux HD) were used according to patients’ needs; standard HD parameters were: blood flow ≥300 ml/min, dialysis solution flow 500 ml/min, glucose concentration in dialysisate fluid 00 mg/dl and dialysis duration of 240 minutes. During all HD sessions, the clinical monitoring was performed according to the center best practice: in particular, hemodynamic parameters were recorded at the start of dialysis and then, each 60 minutes, blood glucose levels were measured twice (at the beginning and end of the HD session). Device Description Each CGM device comprises a recorder (iProTM 2 Professional CGM Medtronic MiniMed – Northridge, Calif., USA) and a sensor (ENLiTETM Glucose Sensor Medtronic MiniMed – Northridge, Calif., USA) with a platinum electrode working with the enzyme glucose oxidase. The enzyme-driven oxidation of glucose in the interstitial fluid generates a current recorded by the device every five minutes for six consecutive days. The device is applied in the subcutaneous tissue of the anterior abdominal wall at study admission. The system does not allow real-time reading of the glucose value; so the patients’ behavior is not affected from the CGM result. The accuracy of the CGM device, when compared with the levels of serum glucose, has proved satisfactory in several studies [8].
CGM Monitoring in Hemodialysis
Data Analysis Data were processed by an apposite software (CareLink® Pro Software). We first assessed the overall glycemic pattern of each patient. Aggregate data analysis was focused on intradialytic, postdialytic and nocturnal interstitial glucose description. Descriptive analysis was performed by SPSS (SPSS Inc., Chicago Ill., USA vers. 20.0.0). The distribution of continuous variables analyzed was assessed by skewness and kurtosis; in the case of non-normal distributions, such variables were described using the median (min–max). In the case of normal distributions, variables were described with mean and standard deviation. The CI was set for all tests at 95%. The longitudinal course of interstitial glucose was assessed by box plots and scatter plots.
Results
CGM Calibration Capillary glucose values measured by a glucometer were consistent with data analyzed by the Hospital Central Laboratory (p = 0.42, see table 1) and were used for CGM calibration. Patients’ Characteristics Patients’ characteristics are reported in detail in table 1. Average age was 61.8 ± 13.9 years, diabetes duration 39.6 (1.9–125.4) months, average HbA1c 55.07 ± 9.93 mmol/mol. All patients have a residual glomerular filtration rate
Blood Purif 2014;38:68–73 DOI: 10.1159/000362863
Glycemic Pattern in Diabetic Patients on Hemodialysis: Continuous Glucose Monitoring (CGM) Analysis Massimo Gai a Ilenia Merlo a Sergio Dellepiane a Vincenzo Cantaluppi a Gianluca Leonardi a Fabrizio Fop a Cesare Guarena a Giorgio Grassi b Luigi Biancone a
a SCU Nephrology, Dialysis and Transplantation, b SCDU Endocrinology, Diabetology and Metabolism, Azienda Ospedaliera Città della Salute e della Scienza di Torino, Presidio ‘Molinette’, Torino, Italy
Key Words Continuous glucose monitoring system · Diabetes · Diabetic nephropathy · Glucose · Glycemic variability · Hemodialysis
Abstract Background/Aims: Recent evidences suggest that hemodialysis (HD) induces glycemic variations in diabetic patients. Continuous glucose monitoring (CGM) devices measure interstitial glucose in a ‘Holter-like’ manner thereby improving the glycemic control assessment method. Methods: A CGM device (Medtronic iPRO) was used on 12 diabetic patients with chronic HD for 6 days to assess intra- and extra-dialytic interstitial glucose. Results: In all enrolled patients, HD was associated with a decrease of interstitial glucose values. Intradialytic glucose nadir was 79 mg/dl and it was reached at the third hour after the beginning of the session. At the end of HD, interstitial glucose increased in all patients and a glycemic peak (187 mg/dl) occurred after an average time of 2.5 h. No episodes of nocturnal hypoglycemia occurred. Conclusion: HD is associated with significant intradialytic reduction of glycemia and postdialytic hyperglycemia. CGM devices result in better monitoring of glycemic trends in diabetic patients on chronic HD and could improve insulin management.
Introduction
Diabetic nephropathy is one of the leading causes of End Stage Renal Disease (ESRD) in western countries causing 21.5–27.4% of new hemodialysis (HD) accesses [1]. Since the kidney is a relevant site for glucose and insulin metabolism, glycemic control is particularly challenging in ESRD diabetic patients with a higher risk of both hypo- and hyperglycemia. Furthermore, recent findings suggest that dialysis may be per se a cause of glycemic variability; in fact, glucose is freely filtered during HD sessions and many types of hemodialyzers may adsorb insulin [2]. Moreover, an increasing body of evidence indicates glycemic variability is an independent risk factor of morbidity and mortality in diabetic patients [3]. Current diabetes monitoring is based on self-measured capillary blood glucose and glicated hemoglobin (HbA1c) levels [4]. In chronic HD, blood glucose is usually measured at the beginning and at the end of the session, but there are only few studies about glycemic trends during and after the session [5–7]. Moreover, in HD patients, glicated hemoglobin (HbA1c) may be an inadequate index of glycemic control due to ESRD-related anemia or frequent blood-transfusions. Furthermore, HbA1c
© 2014 S. Karger AG, Basel © 2014 S. Karger AG, Basel 0253–5068/14/0381–0068$39.50/0 E-Mail [email protected] www.karger.com/bpu
Massimo Gai, MD Azienda Ospedaliera S. Giovanni Battista di Torino S.C.U. Nefrologia, Dialisi e Trapianto Corso Bramante 88, IT–10126 Torino (Italy) E-Mail massimogai @ gmail.com
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 3/21/2015 9:14:15 PM
estimates mean plasmatic glucose but not glycemic instability. Continue glucose monitoring (CGM) is a new Holterlike system to monitor interstitial glucose which allows an analysis of several parameters such as peak, nadir, mean and variability. Interstitial glucose is the expression of blood glucose; only in the event of a rapid blood glucose variation, you can observe a delay (‘lag time’) of 10–15 minutes [8]. The aim of this study was to assess the glycemic control in diabetic patients on chronic HD using a CGM system and to monitor the interstitial glucose levels for 6 consecutive days.
Data Collection All patients were trained to measure capillary blood glucose 4 times/day and to record glucose values and administered insulin dosages and details of food consumption, physical activity and symptoms. The system must be calibrated, in data processing, through the insertion of at least four glucose values for each day of recording. Glucose values were obtained using a glucometer (GM); this is a device for the self-monitoring of capillary blood glucose. Each patient had its own GM checked before the study. Day 1 laboratory glucose levels were compared with data obtained from CGM and patients’ provided capillary-glucometer (Accu-Chek® Aviva, Roche Diagnostics). Further blood tests performed at day 1 after study admission were: hemoglobin, HbA1c, urea, albumin. Food consumption was not allowed during dialysis sessions. On the 6th day from the date of study admission, GGM devices were removed and data analyzed.
Materials and Methods
HD Sessions Patients’ HD prescription was not changed during the study and different filters and types of HD (i.e., acetate free biofiltration, standard bicarbonate HD, high flux HD) were used according to patients’ needs; standard HD parameters were: blood flow ≥300 ml/min, dialysis solution flow 500 ml/min, glucose concentration in dialysisate fluid 00 mg/dl and dialysis duration of 240 minutes. During all HD sessions, the clinical monitoring was performed according to the center best practice: in particular, hemodynamic parameters were recorded at the start of dialysis and then, each 60 minutes, blood glucose levels were measured twice (at the beginning and end of the HD session). Device Description Each CGM device comprises a recorder (iProTM 2 Professional CGM Medtronic MiniMed – Northridge, Calif., USA) and a sensor (ENLiTETM Glucose Sensor Medtronic MiniMed – Northridge, Calif., USA) with a platinum electrode working with the enzyme glucose oxidase. The enzyme-driven oxidation of glucose in the interstitial fluid generates a current recorded by the device every five minutes for six consecutive days. The device is applied in the subcutaneous tissue of the anterior abdominal wall at study admission. The system does not allow real-time reading of the glucose value; so the patients’ behavior is not affected from the CGM result. The accuracy of the CGM device, when compared with the levels of serum glucose, has proved satisfactory in several studies [8].
CGM Monitoring in Hemodialysis
Data Analysis Data were processed by an apposite software (CareLink® Pro Software). We first assessed the overall glycemic pattern of each patient. Aggregate data analysis was focused on intradialytic, postdialytic and nocturnal interstitial glucose description. Descriptive analysis was performed by SPSS (SPSS Inc., Chicago Ill., USA vers. 20.0.0). The distribution of continuous variables analyzed was assessed by skewness and kurtosis; in the case of non-normal distributions, such variables were described using the median (min–max). In the case of normal distributions, variables were described with mean and standard deviation. The CI was set for all tests at 95%. The longitudinal course of interstitial glucose was assessed by box plots and scatter plots.
Results
CGM Calibration Capillary glucose values measured by a glucometer were consistent with data analyzed by the Hospital Central Laboratory (p = 0.42, see table 1) and were used for CGM calibration. Patients’ Characteristics Patients’ characteristics are reported in detail in table 1. Average age was 61.8 ± 13.9 years, diabetes duration 39.6 (1.9–125.4) months, average HbA1c 55.07 ± 9.93 mmol/mol. All patients have a residual glomerular filtration rate
