Clinical Nephrology, Vol. 82 – No. 4/2014 (240-246)
Continuous glucose monitoring in hemodialyzed patients with type 2 diabetes: a multicenter pilot study Original ©2014 Dustri-Verlag Dr. K. Feistle ISSN 0301-0430 DOI 10.5414/CN108280 e-pub: August 27, 2014
Key words diabetes mellitus – continuous glucose monitoring – glycemic control – hemodialysis – insulin analogues
Received January 8, 2014; accepted in revised form May 8, 2014 Correspondence to Lori Képénékian Hôpital Civil, Clinique Médicale B, Service d’Endocrinologie, Diabète et Nutrition, 1, place de l’Hôpital, 67000 Strasbourg, France lori.kepenekian@ gmail.com
Lori Képénékian1, Agnieszka Smagala2, Laurent Meyer3, Olivier Imhoff3, Farideh Alenabi4, Liviu Serb5, Dominique Fleury6, François Dorey7, Thierry Krummel8, Jean-Pierre Le Floch9, François Chantrel4, and Laurence Kessler1 1Service
d’Endocrinologie, Diabète et Nutrition, CHU Strasbourg, Hôpital Civil, Strasbourg, 2Service de Médecine Interne, Hôpital Pasteur, Colmar, 3Clinique Sainte Anne, Strasbourg, 4Service de Néphrologie, Hôpital Pasteur, Colmar, 5Service de Diabétologie, Centre hospitalier de Mulhouse, Mulhouse, 6Service de Néphrologie, 7Service de Diabétologie, Centre hospitalier de Valenciennes, Valenciennes, 8Service de Néphrologie, CHU Strasbourg, Hôpital Civil, Strasbourg, and 9Clinique Médicale de Villecresnes, Villecresnes, France
Abstract. Aims: Hemodialyzed patients with diabetes face an increased cardiovascular risk. Optimal glycemic control can reduce morbidity and mortality, but it is difficult to achieve because of the alternation between dialysis and non-dialysis periods. This study evaluated the contribution of continuous glucose monitoring (CGM) to the management of insulin regimen. Methods: In this pilot prospective multicenter study, we performed CGM (Navigator®, Abbott, Rungis, France) for a total of 54 hours at baseline and for a 3-month follow-up period in a group of 28 hemodialyzed patients with type 2 diabetes treated by a basal-bolus detemir plus aspart insulin regimen. Insulin therapy was adapted to the CGM values. HbA1c and CGM parameters collected over the 3-month treatment period were compared using MANOVA for repeated measures. Results: After 3 months, HbA1c significantly decreased from 8.4 ± 1.0% (65 ± 1 mmol/mol) to 7.6 ± 1.0% (60 ± 11 mmol/mol; p < 0.01). Similarly, mean CGM glucose values significantly decreased from 9.9 ± 1.9 to 8.9 ± 2.1 mmol/L (p = 0.05). The frequency of glucose values > 10 mmol/L significantly decreased from 41.3 ± 21.9% to 30.1 ± 22.4% (p < 0.05), without a significant increase in the frequency of glucose values 6.5% (i.e., 47.5 mmol/mol), and 5) being on maintenance dialysis for more than three months. Patients were excluded when they met one of the following criteria: 1) severe anemia (hemoglobin levels were required to be stable
241
without variations in erythropoietin doses greater than 20% in the previous 3 months) or need for blood transfusions within 2 months before the beginning of the study; 2) life expectancy of less than 1 year; 3) presence of chronic inflammatory diseases; 4) presence of evolutive cancer requiring steroid treatment, chemotherapy, radiotherapy, or programmed surgery; 5) lack of compliance. Five French centers (Strasbourg University Hospital, Strasbourg Sainte Anne Hospital, Colmar Hospital, Mulhouse Hospital, and Valenciennes Hospital) participated in the trial. The study protocol was approved by the local ethics committee (East IV) and complied with the tenets of the Declaration of Helsinki. The trial was registered in clinicaltrials.gov (NCT01828970). and all patients provided written informed consent.
Study design Between January 1, 2010 and June 30, 2012, all patients who were admitted to the nephrology departments of the participating hospitals and fulfilled the inclusion criteria were consecutively enrolled. All of the participants underwent dialysis on high-flux dialyzers using a 100 mg/dL glucose solution. CGM (Navigator®; Abbott, Rungis, France) was used to analyze blood glucose excursions at baseline and after 3 months of treatment. After starting at the beginning of a dialysis session, CGM was first continued at home for the next 2 days under ambulatory conditions and then conducted during the subsequent dialysis session. Therefore, a total of 54 hours of CGM were available for the analysis, including two consecutive hemodialysis sessions at baseline and 3 months later. The probe for the system was subcutaneously inserted at the beginning of the first dialysis session to analyze interstitial glucose, and subsequently removed at the end of the second dialysis session. Due to the time required for CGM calibration, interstitial glucose levels were only partially monitored during the first dialysis session. The analogues were titrated to obtain an optimal glycemic control (i.e., a fasting glucose level between 6.7 and 7.8 mmol/L and a non-fasting glucose of less than 11.1 mmol/L at 2 hours after meals). A physician adapted the insulin doses according to the glucose
242
Képénékian, Smagala, Meyer, et al.
Table 1. Baseline characteristics of the study participants (n = 28). Characteristics Females/males, n (%) Age (years) Body mass index (kg/m2) Duration of diabetes (years) Type of dialysis Hemodialysis duration (years) KT/V (Daudirgas) Complications, n (%) Coronary heart disease Stroke Amputation Retinopathy Neuropathy Comorbidity, n (%) Hypertension Dyslipidemia
9/19 (32/68%) 65.7 ± 9.4 33.3 ± 6.6 22.8 ± 9.8 3.6 ± 2.5 1.5 ± 0.3 11 (39%) 4 (14%) 4 (14%) 19 (68%) 12 (43%) 27 (96%) 23 (82%)
months. The primary endpoint was the change in HbA1c as measured by high-performance liquid chromatography. The secondary endpoints included changes in body weight, insulin requirements (IU/d), symptomatic hypoglycemia (i.e., the number of episodes with a glucose level < 3.3 mmol/L per patient and per month), and CGM parameters (i.e., mean CGM glucose values, within-subjects standard deviation (wSD) and coefficient of variation (wCV = wSD/mean), mean amplitude of glycemic excursion (MAGE), frequency of episodes with a glucose level 10 mmol/L)). All deaths and major cardiovascular events (i.e., myocardial infarction, stroke, and peripheral vascular disease) were carefully recorded. The mean values were compared using multivariate analysis of variance (MANOVA) for repeated measures, whereas frequencies were examined using the χ2-test.
Results
Figure 1. HbA1c values (A) and occurrence of symptomatic hypoglycemia (B) at baseline and after 3 months of treatment with a CGM-adapted detemir plus aspart insulin regimen (**p 7.8 mmol/L on two consecutive days, basal insulin (detemir) was increased by 10%. If fasting glucose was 11 mmol/L on two consecutive days, rapid insulin (aspart) was increased by 10%. If non-fasting glucose values were
Continuous glucose monitoring in hemodialyzed patients with type 2 diabetes: a multicenter pilot study Original ©2014 Dustri-Verlag Dr. K. Feistle ISSN 0301-0430 DOI 10.5414/CN108280 e-pub: August 27, 2014
Key words diabetes mellitus – continuous glucose monitoring – glycemic control – hemodialysis – insulin analogues
Received January 8, 2014; accepted in revised form May 8, 2014 Correspondence to Lori Képénékian Hôpital Civil, Clinique Médicale B, Service d’Endocrinologie, Diabète et Nutrition, 1, place de l’Hôpital, 67000 Strasbourg, France lori.kepenekian@ gmail.com
Lori Képénékian1, Agnieszka Smagala2, Laurent Meyer3, Olivier Imhoff3, Farideh Alenabi4, Liviu Serb5, Dominique Fleury6, François Dorey7, Thierry Krummel8, Jean-Pierre Le Floch9, François Chantrel4, and Laurence Kessler1 1Service
d’Endocrinologie, Diabète et Nutrition, CHU Strasbourg, Hôpital Civil, Strasbourg, 2Service de Médecine Interne, Hôpital Pasteur, Colmar, 3Clinique Sainte Anne, Strasbourg, 4Service de Néphrologie, Hôpital Pasteur, Colmar, 5Service de Diabétologie, Centre hospitalier de Mulhouse, Mulhouse, 6Service de Néphrologie, 7Service de Diabétologie, Centre hospitalier de Valenciennes, Valenciennes, 8Service de Néphrologie, CHU Strasbourg, Hôpital Civil, Strasbourg, and 9Clinique Médicale de Villecresnes, Villecresnes, France
Abstract. Aims: Hemodialyzed patients with diabetes face an increased cardiovascular risk. Optimal glycemic control can reduce morbidity and mortality, but it is difficult to achieve because of the alternation between dialysis and non-dialysis periods. This study evaluated the contribution of continuous glucose monitoring (CGM) to the management of insulin regimen. Methods: In this pilot prospective multicenter study, we performed CGM (Navigator®, Abbott, Rungis, France) for a total of 54 hours at baseline and for a 3-month follow-up period in a group of 28 hemodialyzed patients with type 2 diabetes treated by a basal-bolus detemir plus aspart insulin regimen. Insulin therapy was adapted to the CGM values. HbA1c and CGM parameters collected over the 3-month treatment period were compared using MANOVA for repeated measures. Results: After 3 months, HbA1c significantly decreased from 8.4 ± 1.0% (65 ± 1 mmol/mol) to 7.6 ± 1.0% (60 ± 11 mmol/mol; p < 0.01). Similarly, mean CGM glucose values significantly decreased from 9.9 ± 1.9 to 8.9 ± 2.1 mmol/L (p = 0.05). The frequency of glucose values > 10 mmol/L significantly decreased from 41.3 ± 21.9% to 30.1 ± 22.4% (p < 0.05), without a significant increase in the frequency of glucose values 6.5% (i.e., 47.5 mmol/mol), and 5) being on maintenance dialysis for more than three months. Patients were excluded when they met one of the following criteria: 1) severe anemia (hemoglobin levels were required to be stable
241
without variations in erythropoietin doses greater than 20% in the previous 3 months) or need for blood transfusions within 2 months before the beginning of the study; 2) life expectancy of less than 1 year; 3) presence of chronic inflammatory diseases; 4) presence of evolutive cancer requiring steroid treatment, chemotherapy, radiotherapy, or programmed surgery; 5) lack of compliance. Five French centers (Strasbourg University Hospital, Strasbourg Sainte Anne Hospital, Colmar Hospital, Mulhouse Hospital, and Valenciennes Hospital) participated in the trial. The study protocol was approved by the local ethics committee (East IV) and complied with the tenets of the Declaration of Helsinki. The trial was registered in clinicaltrials.gov (NCT01828970). and all patients provided written informed consent.
Study design Between January 1, 2010 and June 30, 2012, all patients who were admitted to the nephrology departments of the participating hospitals and fulfilled the inclusion criteria were consecutively enrolled. All of the participants underwent dialysis on high-flux dialyzers using a 100 mg/dL glucose solution. CGM (Navigator®; Abbott, Rungis, France) was used to analyze blood glucose excursions at baseline and after 3 months of treatment. After starting at the beginning of a dialysis session, CGM was first continued at home for the next 2 days under ambulatory conditions and then conducted during the subsequent dialysis session. Therefore, a total of 54 hours of CGM were available for the analysis, including two consecutive hemodialysis sessions at baseline and 3 months later. The probe for the system was subcutaneously inserted at the beginning of the first dialysis session to analyze interstitial glucose, and subsequently removed at the end of the second dialysis session. Due to the time required for CGM calibration, interstitial glucose levels were only partially monitored during the first dialysis session. The analogues were titrated to obtain an optimal glycemic control (i.e., a fasting glucose level between 6.7 and 7.8 mmol/L and a non-fasting glucose of less than 11.1 mmol/L at 2 hours after meals). A physician adapted the insulin doses according to the glucose
242
Képénékian, Smagala, Meyer, et al.
Table 1. Baseline characteristics of the study participants (n = 28). Characteristics Females/males, n (%) Age (years) Body mass index (kg/m2) Duration of diabetes (years) Type of dialysis Hemodialysis duration (years) KT/V (Daudirgas) Complications, n (%) Coronary heart disease Stroke Amputation Retinopathy Neuropathy Comorbidity, n (%) Hypertension Dyslipidemia
9/19 (32/68%) 65.7 ± 9.4 33.3 ± 6.6 22.8 ± 9.8 3.6 ± 2.5 1.5 ± 0.3 11 (39%) 4 (14%) 4 (14%) 19 (68%) 12 (43%) 27 (96%) 23 (82%)
months. The primary endpoint was the change in HbA1c as measured by high-performance liquid chromatography. The secondary endpoints included changes in body weight, insulin requirements (IU/d), symptomatic hypoglycemia (i.e., the number of episodes with a glucose level < 3.3 mmol/L per patient and per month), and CGM parameters (i.e., mean CGM glucose values, within-subjects standard deviation (wSD) and coefficient of variation (wCV = wSD/mean), mean amplitude of glycemic excursion (MAGE), frequency of episodes with a glucose level 10 mmol/L)). All deaths and major cardiovascular events (i.e., myocardial infarction, stroke, and peripheral vascular disease) were carefully recorded. The mean values were compared using multivariate analysis of variance (MANOVA) for repeated measures, whereas frequencies were examined using the χ2-test.
Results
Figure 1. HbA1c values (A) and occurrence of symptomatic hypoglycemia (B) at baseline and after 3 months of treatment with a CGM-adapted detemir plus aspart insulin regimen (**p 7.8 mmol/L on two consecutive days, basal insulin (detemir) was increased by 10%. If fasting glucose was 11 mmol/L on two consecutive days, rapid insulin (aspart) was increased by 10%. If non-fasting glucose values were
