Insulin Pumps

Abstract

Introduction

While many people with type 1 diabetes treated by insulin pump therapy (continuous subcutaneous insulin infusion, CSII) achieve good glycemic control, and usually better than on multiple daily insulin injections (MDI), blood glucose levels can worsen from time to time and most patients will need advice on how to re-establish strict blood glucose control. Some patients struggle to improve much, even from the start of CSII. Several factors restrict the ability to achieve optimal blood glucose levels on CSII, but three of the most important are exercise management, insulin delivery problems caused by various types of infusion set failure, and excessive meal-related hyperglycemia. We review here papers that cover all these areas. While much more work is needed, practical messages are emerging—that a greater basal rate reduction than commonly recommended may be needed to avoid hypoglycemia after exercise, “silent” occlusions of infusion sets are common but could be lessened by new types of infusion cannulas and the quality as well as the quantity of meal fat is important for determining the extent of postprandial hyperglycemia.

Sustainability of CSII efficacy in both type 1 and 2 diabetes is also reported in the papers chosen this year. The use of insulin pump therapy in type 2 diabetes continues attract attention, particularly following the results of the recent randomized control trial of CSII vs. MDI in poorly controlled type 2 diabetes (the OpT2mise Trial) (1). Two papers included in this article refer to the maintenance of improved control on CSII in type 2 diabetes, one is a 12 month extension of the OpT2mise study and the other is an observational study of CSII in type 2 diabetes for pump durations up to 9 years. Rather surprisingly, the long-term sustainability of reductions in severe hypoglycemia in type 1 diabetes treated by CSII have been little studied (as opposed to HbA1c). It is reassuring from a study of Spanish patients on insulin pump therapy included here that hypoglycemia reduction on CSII is maintained over many years for most patients.

Insulin pumps with predictive low-glucose insulin-suspend have now entered clinical practice in several countries and one of the first reports of real-life experiences of hypoglycemia avoidance using such devices and the patient responses is reported.

Finally, we include a paper on the variable uptake of CSII in children in different countries, and the contribution this makes to the differences in HbA1c seen in those populations.

Key Articles Reviewed for the Article

Insulin pump basal adjustment for exercise in type 1 diabetes: a randomised crossover study

McAuley SA, Horsburgh JC, Ward GM, La Gerche A, Gooley JL, Jenkins AJ, MacIsaac RJ, O'Neal DN

Diabetologia 2016; 59: 1636–1644

Reduced silent occlusions with a novel catheter infusion set (BD FlowSmart): results from two open-label comparative studies

Gibney M, Xue Z, Swinney M, Bialonczyk D, Hirsch L

Diabet Technol Therapeut 2016; 18: 136–143

Improving patient experience with insulin infusion sets: practical guidelines and future directions

Evert AB, Bode BW, Buckingham BA, Nardacci E, Verderese CA, Wolff-McDonagh P, Walsh J, Hirsch IB

Diabetes Educator 2016; 42: 470–484

Extra-virgin olive oil reduces glycemic response to a high-glycemic index meal in patients with type 1 diabetes: a randomized controlled trial

Bozzetto A, Alderisio A, Giorgini M, Barone F, Giacco A, Riccardi G, Rivellese AA, Annuzzi G

Diabetes Care 2016; 39: 518–524

Sustained efficacy of insulin pump therapy in type 2 diabetes: 9-year follow-up in a cohort of 161 patients

Morera J, Joubert M, Morello R, Rod A, Lireux B, Reznik Y

Diabetes Care 2016; 39: e74–e75

Sustained efficacy of insulin pump therapy compared with multiple daily injections in type 2 diabetes: 12-month data from the OpT2mise randomized trial

Aronson R, Reznik Y, Conget Y, Castañeda JA, Runzis S, Lee SW, Cohen O, for the OpT2mise Study Group

Diabetes, Obesity and Metabolism 2016; 18: 500–507

Long-term outcome of insulin pump therapy: reduction of hypoglycaemia and impact on glycaemic control

Quiros C, Gimenez M, Rıos P, Careaga M, Roca D, Vidal M, Conget I

Diabetic Med 2016; 33: 1422–1426

Hypoglycemia prevention and user acceptance of an insulin pump system with predictive low glucose management

Choudhary P, Olsen BS, Conget I, Welsh JB, Vorrink L, Shin JJ

Diabet Technol Therapeut 2016; 18: 288–291

Use of insulin pump therapy in children and adolescents with type 1 diabetes and its impact on metabolic control: comparison of results from three large, transatlantic paediatric registries

Sherr JL, Hermann JM, Campbell F, Foster NC, Hofer SE, Allgrove J, Maahs DM, Kapellen TM, Holman N, Tamborlane WV, Holl RW, Beck RW, Warner JT, for the T1D Exchange Clinic Network, the DPV Initiative, and the National Paediatric Diabetes Audit and the Royal College of Paediatrics and Child Health Registries

Diabetologia 2016; 59: 87–91

Exercise and CSII

Insulin pump basal adjustment for exercise in type 1 diabetes: a randomised crossover study

McAuley

1 2 Horsburgh

1 Ward

2 3 La Gerche

1 4 6 Gooley

1 Jenkins

1 2 5 MacIsaac

1 2 O'Neal

1 2 1 Department of Medicine, St Vincent's Hospital, University of Melbourne, Melbourne, Australia. 2 Department of Endocrinology & Diabetes, St Vincent's Hospital Melbourne, Melbourne, VIC, Australia. 3 Department of Pathology, University of Melbourne, Melbourne, VIC, Australia. 4 Department of Cardiology, St Vincent's Hospital Melbourne, Melbourne, VIC, Australia. 5 NHMRC Clinical Trials Centre, University of Sydney, Sydney, NSW, Australia. 6 Sports Cardiology, Baker IDI Heart and Diabetes Institute, Melbourne, VIC, Australia.

Diabetologia 2016; 59: 1636–1644

Background

The evidence base for current guidance on avoiding hypoglycemia during and after exercise in people treated by insulin pump therapy is weak. A commonly used approach when exercise is performed a long time after a meal insulin bolus is to reduce the basal rate by 50% before and after exercise. The aim of this study was to investigate changes in circulating insulin and glucose in patients undergoing exercise compared to rest using this strategy.

Methods

Adult type 1 diabetic patients on CSII (n=14) were randomized to either an exercise or rest study, with subsequent crossover to the alternative exercise or rest stage. After an overnight fast, the basal rate was reduced by half, starting 60 min before a 30 min period of moderately intense exercise on a stationary bicycle, and the reduced rate continued for 210 min. Circulating insulin and glucose levels were compared with the rest stage.

Results

Three subjects (21%) required oral or intravenous glucose to prevent or treat hypoglycemia during the exercise stage. In the other subjects, plasma glucose rose by a mean of 0.8 mmol/L. After basal-rate reduction, circulating free insulin levels decreased but then increased again during the 30 min exercise period, relative to rest. During the next 2 hours, there was an accelerated decline in insulin levels in those who had exercised.

Conclusions

While in most subjects a basal rate reduction of 50% is sufficient to prevent hypoglycemia, this does not do so in all patients exercising on CSII, particularly those with low to normal blood glucose levels at the start of exercise. This might be related to a transient increase in insulin levels associated with exercise, due in part to increased blood flow and insulin absorption. Greater basal rate reduction and/or carbohydrate supplementation may be needed to avoid exercise-related hypoglycemia.
Comment
The management of CSII in type 1 diabetes during exercise and the difficulty of avoiding exercise-induced hypoglycemia and/or hyperglycemia is one of the major concerns for patients. The existing guidelines are helpful but largely are not based on very robust evidence from randomized controlled trials. This is partly a result of the many factors that determine outcomes during such studies and during everyday life on CSII, including the duration, intensity, and timing of exercise (close to a meal and the meal insulin administered, or remote from the meal and during basal infusion), the fitness of the subject, previous history of hypoglycemia, the glycemic control, and level of insulinization at the time of exercise, the amount and timing of any carbohydrate supplement given, the timing and magnitude of any basal-rate reduction, and the duration of follow-up of hypoglycemic events (hypoglycemia many hours after exercise is common). These variables make trial design and suitable controls very complicated. The study of McAuley et al. addresses just one strategy: 50% basal-rate reduction implemented one hour before exercise occurring in the fasting state, with a 2-hour follow-up. A significant proportion of patients required intervention to treat or prevent hypoglycemia and possibly more would have done so if the study had been extended for, say, 24 hours. Another recent study (2) indicated that 50% basal rate reduction is inadequate for moderate exercise-induced hypoglycemia prevention – 80% reduction may be more appropriate, perhaps started 90 min before exercise rather than 60 min before, at least when blood glucose levels are low and the patient is therefore presumably well- or over-insulinized.

Insulin Infusion Sets – Problems and Solutions

Reduced silent occlusions with a novel catheter infusion set (BD FlowSmart): results from two open-label comparative studies
Gibney M 1 Xue Z 2 Swinney M 3 Bialonczyk D 1 4 Hirsch L 1 1 Diabetes Care, BD Medical Affairs, Becton Dickinson and Company, Franklin Lakes, NJ. 2 Office of Science, Medicine and Technology, BD Corporate Clinical Development, Becton Dickinson and Company, Sparks, MD. 3 Diabetes Care, BD Research & Development, Becton Dickinson and Company, Andover, MA. 4 School of Pharmacy, MCPHS University, Boston, MA.

Diabet Technol Therapeut 2016; 18: 136–143

Background

Problems with infusion sets in CSII are common. Unexplained and unresponsive hyperglycemia without triggering an occlusion alarm suggests that silent occlusions may be occurring. This has led to the development of a novel 28 gauge, polymer infusion cannula with side and end ports (BD FlowSmart) that was designed to improve insulin delivery.

Methods

In-line pressure using the dual-port FlowSmart cannula was compared to a conventional infusion set (Medtronic Quick-set) in healthy subjects, using insulin diluent infused for up to 4.5 hours. Sets were inserted either manually or mechanically by trained nurses. Silent occlusion was defined as continuous pressure rise for ≥30 min without occlusion alarm.

Results

Silent occlusions were common with a traditional infusion set (>35% of sets). Significantly fewer silent occlusions occurred with the FlowSmart compared to the Quick-set for both manual and mechanical insertion (>75% reduction). Flow interruptions were also reduced.

Conclusions

Flow interruptions and silent occlusions were reduced using a new dual-port insulin infusion cannula.

Improving patient experience with insulin infusion sets: practical guidelines and future directions
Evert AB 1 Bode BW 2 Buckingham BA 3 Nardacci E 4 Verderese CA 5 Wolff-McDonagh P 6 Walsh J 7 Hirsch IB 1 1 Diabetes Care Center, University of Washington School of Medicine, Seattle, WA. 2 Atlanta Diabetes Associates / Emory University, Atlanta, GA. 3 Stanford School of Medicine, Palo Alto, CA. 4 St Peter's Health Partners Medical Associates, Albany, NY. 5 Diabetes Education Group, Lakeville, CT. 6 Florida Health Care Plan, Palm Coast, FL. 7 Advanced Metabolic Care and Research, San Diego, CA.

Diabetes Educator 2016; 42: 470–484

Background

Although insulin infusion sets are a crucial component of insulin pump therapy, establishing best practice in their use has been difficult because of the limited scientific study to date.

Methods

Diabetologists and diabetes educators met to review set-related complications and patient experience, to develop guidelines for best use and to recommend priorities for optimizing infusion set performance.

Results

Set failures because of cannula occlusion, kinking, leakage, detachment, and adhesion problems are common with CSII but the actual frequencies are unclear. A large number of sets are discontinued because of unexplained hyperglycemia (30%–50%), but also by pain at the site, accidental pulling out, and infection. Patients and caregivers should be alerted to common causes of set failure and in many cases issues can be resolved themselves, e.g., dislodgement, leakage, inadequate insertion, and air in the system. Help with recognizing problems and suggestions for patient and clinician action are given. A check list for establishing and maintaining healthy infusion set practices is included.

Conclusions

Educational tools and guidelines are necessary to empower patients and caregivers to prevent, diagnose, and troubleshoot infusion set problems such as unexplained hyperglycemia.
Comment
There are many possible causes of silent occlusions of insulin infusion sets and unexplained hyperglycemia resulting from it, including cannula kinking, insulin aggregation, tissue obstruction, and fibrin formation at the infusion site. The new dual-port cannula in the study of Gibney et al. shows notable promise for reducing silent occlusions. However, the study has several limitations, especially the use of diluent rather than insulin infusion, short duration of study and the limited activity of the participants. Possibly, occlusions would have been more frequent (and perhaps the new cannula more effective) with infusion sets used for the usual 3 days or the longer times that some patients use cannulas, with infusion of a short-acting insulin analog rather than diluent (lipohypertrophy and the possibility of insulin aggregation when using insulin solution in the pump might influence delivery through the cannula) and if the subjects underwent the normal daily activities and periods of exercise typical of everyday CSII usage. It is also uncertain whether occlusions would differ in people with diabetes rather than healthy subjects, or if patients were inserting the cannulas rather than the perhaps more careful insertion by a health-care professional. Clearly, longer-term studies in people with diabetes are necessary.

The very useful expert discussions and practical guidelines on infusion set issues summarized in the paper of Evert et al. takes the view that the patient can or should be educated to recognize and deal with many set-related problems himself or herself. This might take the form of changing to a different type of cannula (steel rather than Teflon, or 90^o rather 45^o angled), replacing the set or rotating to a new anatomical site. It would be interesting to know how much glycemic control could be improved in CSII patients by instituting a rigorous, comprehensive and standardized infusion set education program for patients and caregivers and health-care professionals.

Meal-Time Boluses

Extra-virgin olive oil reduces glycemic response to a high-glycemic index meal in patients with type 1 diabetes: a randomized controlled trial
Bozzetto A 1 Alderisio A 1 Giorgini M 1 Barone F 1 Giacco A 1 Riccardi G 1 Rivellese AA 1 Annuzzi G 1 1 Department of Clinical Medicine and Surgery, University of Naples Federico II, Naples, Italy.

Diabetes Care 2016; 39: 518–524

Background

It is known that quality of carbohydrate at meals (i.e., the glycemic index) influences the postprandial glucose increase in diabetes and that high fat meals also increase post-meal blood glucose levels. The aim of this study was to investigate the influence of fat quality on increasing post-meal glucose levels and its interaction with high- and low-glycemic index meals.

Methods

Type 1 diabetic subjects (n=13) treated by CSII were randomly allocated to eat either high or low glycemic index meals of the same carbohydrate content for one week, followed by crossover to eating the alternative type of carbohydrate, but with low fat, high saturated fat (butter), or high monounsaturated fat (extra virgin olive oil) included in the meal. Continuous glucose monitoring (CGM) was performed for 6 hours to monitor glycemic responses.

Results

In the early postprandial period (0–3 hours), the area under the curve for blood glucose levels were lower for all combined low-glycemic index meals compared to high-glycemic index meals. The quality of fat had little influence on postprandial glucose in low-glycemic index meals, but with high-glycemic index meals there was a significant reduction in area under the curve for glucose when meals contained olive oil compared to butter or low fat.

Conclusions

The shape and the extent of postprandial blood glucose increases are influenced by the quality of both meal carbohydrate and fat content. Extra virgin olive oil attenuates meal-induced blood glucose increases associated with high-glycemic index foods and consideration of the type of fat in the diet of CSII patients may be a way of improving glycemic control.
Comment
Excessive postprandial blood glucose increases are a major contributor to poor control in many patients on CSII, and conventional carbohydrate counting for deciding an appropriate insulin bolus dose can be inadequate for limiting meal-induced hyperglycemia. It is well known that high fat meals delay gastric emptying and cause insulin resistance, thereby elevating and extending postprandial blood glucose increases (3), and the square-wave function of insulin pumps is designed to minimize this when high-fat meals are eaten. The study of Bozzetto et al. emphasizes that the type of fat (saturated vs. monounsaturated) is also important in determining meal-related glucose levels, at least with high-glycemic index food, and they introduce the interesting notion that inclusion of olive oil with high-glycemic index meals could help limit hyperglycemia. The mechanism for this needs further study but possibly monounsaturated fats stimulate glucagon-related peptide (GLP)-1 release more than saturated fat, thereby inhibiting gastric emptying and glucagon release and increasing insulin sensitivity (4).

The notion of “fat counting” (modulating insulin boluses according to meal fat content) in insulin pump therapy has not been enthusiastically embraced to date because it is complicated, bolus calculators are not calibrated for fat, and it is difficult to gauge the fat content of meals (see Yearbook 2012); moreover, probably relatively few patients with diabetes even take note of the glycemic index of foods, which is a better established determinant glycemic control. Simple advice to eat a more Mediterranean diet may help improve control in pump patients rather than fat counting.

Sustained Efficacy of CSII

Sustained efficacy of insulin pump therapy in type 2 diabetes: 9-year follow-up in a cohort of 161 patients
Morera J 1 Joubert M 1 Morello R 2 Rod A 1 Lireux B 1 Reznik Y 1 1 Department of Endocrinology, University of Caen, Côte de Nacre Regional Hospital Center, Caen, France. 2 Department of Statistics, University of Caen, Côte de Nacre Regional Hospital Center, Caen, France.

Diabetes Care 2016; 39: e74–e75.

Background

CSII is emerging as a valuable treatment option for selected people with type 2 diabetes but the durability of the treatment is unclear. The aim of this study was to report on the long-term efficacy of CSII in people with type 2 diabetes treated between 1998 and 2012.

Methods

The cohort consisted of 161 people with type 2 diabetes, mean age at baseline 58.3 years, body mass index (BMI) 33.2 kg/m², HbA1c 9.0%±1.7%, and insulin dose 1.2±0.9 units/kg. Most (96%) were receiving insulin injections at baseline. The commonest reason for starting CSII was HbA1c >8.0% (70%), followed by insulin >1.5 units/kg (13.9%). Follow up was for up to 9 years of pump therapy, mean 5.1 years.

Results

The decrease in HbA1c from baseline was −1.3% at one year and this improvement was maintained throughout long-term follow up: −1.3% (n=105) at 3 years, −1.5% (n=70) at 5 years, and −1.4% at 9 years (n=17). Patients with a baseline HbA1c <8.0% showed no improvement. Insulin requirements decreased by 13% at one year and the subsequent insulin dosage was maintained through follow up. Body weight increased by a mean of 2.9 kg at one year but was stable over 7 years.

Conclusions

A retrospective analysis of a large number of type 2 diabetic patients with MDI failure who were transferred to CSII showed sustained efficacy of insulin pump therapy over periods of up to 9 years.

Sustained efficacy of insulin pump therapy compared with multiple daily injections in type 2 diabetes: 12-month data from the OpT2mise randomized trial
Aronson R 1 Reznik Y 2 Conget Y 3 Castañeda JA 4 Runzis S 5 Lee SW 6 Cohen O 7 for the OpT2mise Study Group 1 LMC Diabetes and Endocrinology, Toronto, ON, Canada. 2 Department of Endocrinology, University of Caen Côte de Nacre Regional Hospital Center, Caen, France. 3 Diabetes Unit, Endocrinology and Nutrition Department, University Hospital Clinic, Barcelona, Spain. 4 Medtronic Bakken Research Center, Maastricht, The Netherlands. 5 Medtronic International Trading Sàrl, Tolochenaz, Switzerland. 6 Medtronic Diabetes, Northridge, CA. 7 Institute of Endocrinology, Chaim Sheba Medical Center, Tel Hashomer, Israel.

Diabetes, Obesity and Metabolism 2016; 18: 500–507

Background

In the recently published OpT2mise trial, patients with type 2 diabetes who failed to achieve an HbA1c less than 8% on MDI after a 2 month optimization period were randomized to either continued MDI or CSII for 6 months. The aim of this study was to report on the 12 month results of the trial, where MDI patients had been switched to CSII after 6 months.

Methods

After the optimization run-in period, 331 patients with HbA1c ≥8% and insulin dosage ≥0.7 units/kg were randomized to CSII or MDI for 6 months; the MDI group was then switched to CSII for a further 6 month continuation phase.

Results

With a baseline of HbA1c 9.0%, at the end of the randomization phase the HbA1c was lower with CSII than MDI (mean difference −0.7%). HbA1c on CSII was maintained for 12 months. HbA1c for MDI patients switched to CSII after the first 6 months decreased by 0.8%, so that at the end of the continuation phase the final HbA1c was similar in both groups. Insulin dose was 20.4% less in the CSII vs. MDI group at 6 months; the insulin dosage on CSII was maintained over 12 months and for those patients switched from MDI to CSII the dosage decreased by a further 19%. There was no difference in weight or severe hypoglycemia rate between groups.

Conclusions

Insulin pump therapy achieves superior glycemic control in people with type 2 diabetes who are poorly controlled on MDI and this effect is sustained over 12 months.

Long-term outcome of insulin pump therapy: reduction of hypoglycaemia and impact on glycaemic control
Quiros C 1 Gimenez M 1 Rıos P 1 Careaga M 1 Roca D 1 Vidal M 1 Conget I 1 1 Diabetes Unit, Endocrinology Department, Hospital Clinic i Universitari, Barcelona, Spain.

Diabetic Med 2016; 33: 1422–1426

Background

The aim of this study was to report the long-term (at least 5 years) sustainability of the CSII effect on reduction in HbA1c and hypoglycemia in a group of patients started on CSII according to Catalan National Health Service indications, including failure to reduce HbA1c below 7.5%, recurrent severe hypoglycemia and preconception and pregnancy planning.

Methods

Data from adult type 1 diabetic patients (n=158) who were treated by CSII for at least 5 years were retrospectively analyzed, including the indication for starting pump therapy, annual HbA1c, and hypoglycemia occurrence.

Results

The main indication for starting CSII was suboptimal control (HbA1c) (60.9%), followed by recurrent severe hypoglycemia (25.5%) and preconception and pregnancy planning (6.6%). In those who started CSII because of hypoglycemia, this problem was resolved in 93%, without a significant increase in HbA1c: 7.3% vs. 7.4% (value at baseline vs. 5 years). The mean severe hypoglycemia rate in the 2 years before CSII was 1.33 vs. 0.08 episodes/patient-year in the last 2 years of follow up (P=0.003). For those starting CSII because of elevated HbA1c (mean HbA1c 8.4% at baseline), the level was significantly less at 5 years (mean 8.0%), with a maximal decrease in the first 2 years, followed by a slight increase. A higher baseline HbA1c was correlated with the decrease in HbA1c on CSII.

Conclusions

The efficacy of CSII in reducing severe hypoglycemia is maintained over 5 years.
Comment
Morera et al. and Aronson et al. provide evidence that the efficacy of CSII in selected people with type 2 diabetes is not short-lived but is maintained over several years. Interestingly, several studies show that the improvement in mean HbA1c in type 1 diabetes with CSII tends to worsen after 1–2 years (5), confirmed by Quiros et al., though some 90% still have better control at 5 years than at baseline (5). The ineffectiveness of CSII in type 2 diabetes below a baseline of about 8% that Morera et al. report is similar to the baseline dependence of HbA1c with insulin pump therapy in type1 diabetes (6). It is unclear in what other respects there are relative nonresponders for insulin pump therapy in type 2 diabetes.

The paper of Quiros et al. is of note because although there is increasing documentation of the HbA1c changes over several years of insulin pump therapy, little is known about whether the reduction in hypoglycemia seen with CSII is maintained in the long term. The authors rightly separate the data into that for patients started because of elevated HbA1c and those started because of recurrent hypoglycemia. Most previous studies have included all patients in studies of long-term HbA1c changes – clearly those with disabling hypoglycemia may have a low HbA1c at the start of CSII, which may increase during the study.

Predictive Low-Glucose Insulin-Suspend Pumps

Hypoglycemia prevention and user acceptance of an insulin pump system with predictive low glucose management
Choudhary P 1 Olsen BS 2 Conget I 3 Welsh JB 4 Vorrink L 5 Shin JJ 5 1 King's College London, London, UK. 2 Herlev Hospital, Herlev, Denmark. 3 Diabetes Unit, Clinic and University Hospital, Barcelona, Spain. 4 Medtronic International Trading Sàrl, Tolochenaz, Switzerland. 5 Medtronic, Inc. , Northridge, CA.

Diabet Technol Therapeut 2016; 18: 288–291

This manuscript is also discussed in the article on Continuous Glucose Monitoring in 2016 page S-11.

Background

The MiniMed 640G is a low-glucose insulin-suspend pump where the basal rate is automatically suspended when the sensor glucose value is predicted within 30 min to fall below a set hypoglycemic value. It is now commercially available in several countries. This study was a user evaluation designed to assess the ability of the technology to prevent hypoglycemia and the user acceptability.

Methods

Type 1 diabetic subjects (n=40) used the system for 4 weeks with retrospective download of pump and sensor data, and user acceptance assessed using questionnaires.

Results

On average, there were 2.1 “suspends-before-low” per person/day. The most commonly used value used by participants for the low blood glucose limit was 3.0 mmol/L (54 mg/dL). The mean sensor glucose at the start of suspend was 4.0 mmol/L (71.8 mg/dL). The mean duration of suspend was 56.4 min, most commonly occurring in the early afternoon. Two hourly suspends mostly occurred at night. Sensor glucose values after 83.1% of the predicted suspends did not reach the preset low-glucose value. All subjects felt confident in using the system, and most agreed that the system helped to prevent hypoglycemia.

Conclusions

Sensor-augmented pump therapy using a predictive low-glucose insulin suspend system helps to prevent hypoglycemia in type 1 diabetes without increasing hyperglycemia.
Comment
This paper helps to provide evidence for the effectiveness of predictive low-glucose insulin-suspend pumps at reducing hypoglycemia in everyday clinical practice. However, there are several limitations with the study, including the relatively short duration of 4 weeks and the observational (nonrandomized) trial design. Moreover, few details of the patient characteristics are given, such as previous history of hypoglycemia. In clinical practice, the patients most likely to be transferred to sensor-augmented pump therapy (and those for whom funding is most likely to be available) are those with persistent, frequent severe hypoglycemia and/or hypoglycemia unawareness during conventional CSII, and it is a priority to evaluate the system in this group.

CSII Use in Different Countries

Use of insulin pump therapy in children and adolescents with type 1 diabetes and its impact on metabolic control: comparison of results from three large, transatlantic paediatric registries
Sherr JL 1 Hermann JM 2 Campbell F 3 Foster NC 4 Hofer SE 5 Allgrove J 6 Maahs DM 7 Kapellen TM 8 Holman N 9 Tamborlane WV 1 Holl RW 2 Beck RW 4 Warner JT 10 for the T1D Exchange Clinic Network, the DPV Initiative, and the National Paediatric Diabetes Audit and the Royal College of Paediatrics and Child Health Registries 1 Pediatric Endocrinology, Children's Diabetes Program, Yale School of Medicine, New Haven, CT. 2 ZIBMT, Institute of Epidemiology and Medical Biometry, University of Ulm, Ulm, Germany; affiliated with the German Center for Diabetes Research. 3 General Paediatrics and Diabetes, Leeds Children's Hospital, Leeds, UK. 4 Jaeb Center for Health Research, Tampa, FL. 5 Department of Paediatrics, Medical University of Innsbruck, Innsbruck, Austria. 6 Department of Paediatric Endocrinology and Diabetes, Royal London Children's Hospital, Barts Health NHS Trust, London, UK. 7 Barbara Davis Center for Childhood Diabetes, Aurora, CO. 8 Department of Endocrinology and Diabetology, University of Leipzig, Hospital for Children and Adolescents, Leipzig, Germany. 9 Institute of Cardiovascular and Medical Sciences, University of Glasgow, Glasgow, UK. 10 Paediatric Endocrinology and Diabetes, Children's Hospital for Wales, Cardiff, UK.

Diabetologia 2016; 59: 87–91

This manuscript is also discussed in Article on Diabetes Technology and Therapy in the Pediatric Age Group, page S-105.

Background

While the use of CSII in children and young people with diabetes has increased substantially in recent years, support for pump therapy and uptake varies widely between countries. This study aimed to compare data on frequency of use and impact of CSII uptake on HbA1c using three large multicenter registries in Germany/Austria, the United States, and England and Wales.

Methods

Data from 54,410 children and adolescents included in the German/Austrian Prospective Diabetes Follow-up Registry (DPV), the U.S. Type 1 Diabetes Exchange (T1DX) and the English/Welsh National Paediatric Diabetes Audit (NPDA) were studied for the period 2011–2012. Mode of insulin delivery, demographic details, ethnicity, and HbA1c were recorded.

Results

Insulin pump use was threefold higher in the DPV (41%) and TIDX (47%) compared to NPDA (14%) registries. In contrast, HbA1c for subjects using any treatment mode was highest in NPDA (8.9% ±1.6%) vs. 8.0% ±1.6% (DPV) and 8.3% ±1.4% (T1DX). Pump use was associated with a lower mean HbA1c (8.0% ±1.2% vs. 8.5% ±1.7%, CSII vs. insulin injections, P<0.001). In all three registries, subjects of ethnic minority status were less likely to be treated with a pump (P<0.001) and boys were treated with a pump less often compared with girls (P<0.001).

Conclusions

In spite of similar patient characteristics, substantial differences in HbA1c occurred across patients in the three registries in the different countries, which may be due in part to differences in the frequency of use of CSII or to other, unknown factors.
Comment
It is tempting to ascribe the higher HbA1c level in patients from the England and Wales registry to the lower frequency of CSII uptake in those countries, but HbA1c was also higher in the insulin injection group in England and Wales. While pump use helps to lower HbA1c in patients from all of the registries, other factors must be involved in setting metabolic control in different countries. One of these factors might be socioeconomic status (SES) where there is clear evidence from several studies for a correlation between low SES and high HbA1c (7). Although the proportion of patients of ethnic minority was similar in the registries, those of ethnic minority are also more likely to have low SES. Other studies have shown that low pump uptake is also related to low SES (8). Lower pump use in England and Wales may also be because of more restrictive guidelines for CSII use in older children than in some other countries (those aged 12 years or over must have failed to achieve an HbA1c below 8.5% and/or have continued, frequent and disabling hypoglycemia on MDI), or because there is arguably less enthusiasm for CSII from health-care professionals in England and Wales compared to some other countries.

Footnotes

Author Disclosure Statement

J. C. P. has received speaker and/or consultancy fees from Medtronic, Roche, Eli Lilly, Novo Nordisk, Insulet and CeQur.

References

Reznik

, Cohen

, Aronson

, Conget

, Runzis

, Castaneda

, Lee

; OpT2mise Study Group. Insulin pump treatment compared with multiple daily injections for treatment of type 2 diabetes (OpT2mise): a randomised open-label controlled trial. Lancet, 2014; 384: 1265–72.

Franc

, Daoudi

, Pochat

, Petit

M-H

, Randazzo

, Petit

, Duclos

, Penfornis

, Pussard

, Not

, Heyman

, Koukoui

, Simon

, Charpentier

. Insulin-based strategies to prevent hypoglycaemia during and after exercise in adult patients with type 1 diabetes on pump therapy: the DIABRASPORT randomized study. Diabet Obes Metab, 2015; 17: 1150–57.

Lodefalk

, Åman

, Bang

. Effects of fat supplementation on glycaemic response and gastric emptying in adolescents with Type 1 diabetes. Diabet Med, 2008; 25: 1030–35.

Thomsen

, Storm

, Holst

, Hermansen

. Differential effects of saturated and monounsaturated fats on postprandial lipemia and glucagon-like peptide 1 responses in patients with type 2 diabetes. Am J Clin Nutr, 2003; 77: 605–11.

Nixon

, Folwell

, Pickup

. Variations in the quality and sustainability of long-term glycaemic control with continuous subcutaneous insulin infusion. Diabet Med, 2014; 31: 1174–77.

Pickup

, Sutton

. Severe hypoglycaemia and glycaemic control in type 1 diabetes: meta-analysis of multiple daily insulin injections versus continuous subcutaneous insulin infusion. Diabetic Med, 2008; 25: 765–74.

Collier

, Ghosh

, Hair

, Waugh

. Impact of socioeconomic status and gender on glycaemic control, cardiovascular risk factors and diabetes complications in type 1 and 2 diabetes: a population based analysis from a Scottish region. Diabetes Metab, 2015; 41: 145–51.

Lin

, Connor

, Ruedy

, Beck

, Kollman

, Buckingham

, Redondo

, Schatz

, Haro

, Lee

, Tamborlane

, Wood

; Pediatric Diabetes Consortium. Race, socioeconomic status, and treatment center are associated with insulin pump therapy in youth in the first year following diagnosis of type 1 diabetes. Diabet Technol Therapeut, 2013; 15: 929–34.