Abstracts from ATTD 20147th International Conference on AdvancedTechnologies & Treatments for Diabetes Vienna, Austria, February 5–8, 2014
2014; Mary Ann Liebert, Inc.; Volume: 16; Issue: S1 Linguagem: Inglês
10.1089/dia.2014.1515
ISSN1557-8593
AutoresDiabetes is a condition of rising prevalence in Europe, which treatment is not only costly, but also very laborious. Thus, it creates major problems to European public health. In the era of limited resources in healthcare…,
Tópico(s)Pancreatic function and diabetes
ResumoDiabetes Technology & TherapeuticsVol. 16, No. S1 AbstractsFree AccessAbstracts from ATTD 20147th International Conference on AdvancedTechnologies & Treatments for DiabetesVienna, Austria, February 5–8, 2014Published Online:30 Jan 2014https://doi.org/10.1089/dia.2014.1515AboutSectionsPDF/EPUB ToolsPermissionsDownload CitationsTrack CitationsAdd to favorites Back To Publication ShareShare onFacebookTwitterLinked InRedditEmail ATTD 2014 Oral PresentationsO-1 7TH FRAMEWORK PROGRAM – FUNDED EHEALTH SYSTEMS FOR DIABETESKardas P.1, Beck P.2, Bromuri S.3, Chiarugi F.4, Enzmann M.5, Höll B.2, Keller O.6, Lane S.7, Mader J.8, Marchesini O.9, Mougiakakou S.10, Neubauer K.8, Pieber T.R.2,8, Plößnig M.11, Puricel S.G.12, Schaupp L.8, Spat S.21Medical University of Lodz, Poland2JOANNEUM RESEARCH Forschungsgesellschaft mbH, HEALTH – Institute for Biomedicine and Health Sciences, Graz, Austria3Haute Ecole Specialisee de Suisse Occidentale4Foundation for Research and Technology - Hellas, Institute of Computer Science, Computational Medicine Laboratory, Heraklion, Greece5Fraunhofer Institute for Secure Information Technology (SIT), Darmstadt, Germany6Deutsches Forschungszentrum für KünstlicheIintelligenz GMBH, Saarbrucken, Germany7Triteq Ltd, Hungerford, United Kingdom8Medical University of Graz, Department of Internal Medicine, Division of Endocrinology and Metabolism, Graz, Austria9Portavita B.V., Amsterdam, The Netherlands10ARTORG Center for Biomedical Engineering Research, University of Bern, Switzerland11Salzburg Research Forschungsgesellschaft, Salzburg, Austria12Centre Hospitalier Universitaire Vaudois, Lausanne, SwitzerlandDiabetes is a condition of rising prevalence in Europe, which treatment is not only costly, but also very laborious. Thus, it creates major problems to European public health. In the era of limited resources in healthcare – both in terms of workforce and budget – there is an urgent need to redesign approaches to diabetes treatment. This might be possible with employment of novel eHealth technologies to both prevention and treatment of diabetes.European Commission, being fully aware of this situation, decided to facilitate research and technological development in the field of diabetes. Thus, several RTD projects were funded under the 7th Framework Program. Of these, five complementary projects will be presented during this workshop: • AP@home (Artificial Pancreas at home) – project aiming at developing a functional artificial pancreas• REACTION Project – project aiming at developing an integrated ICT platform that supports improved long-term management of diabetes• COMMODITY12 – project aiming at developing a system providing COntinuous Multi-parametric and Multi-layered analysis Of DIabetes TYpe 1 & 2• EMPOWER – project aiming at supporting the self-management of diabetes patients through a modular and standards-based Patient Empowerment Framework• GoCarb – project aiming at designing a computational system which will support individuals with type 1 diabetes in automatically estimating the grams of carbohydrate in a meal in near real-timeDue to its large spectrum, the workshop will be a perfect forum to learn latest European eHealth initiatives for diabetes, and get in touch with newest eHealth technologies. Both academia and business are cordially invited to take part in this event.O-2 DO WE NEED A REMOTE MONITORING SYSTEM IN THE PRODUCT? REMOTE MONITORING - PRODanne T.11Kinder - und Jugendkrankenhaus, Auf der Bult, Hannover, GermanyDuring the initial outpatient pilot-trials of the DREAM project the sessions are performed under the supervision of a diabetes research team consisting of an onsite physician specialized in diabetes treatment and technical supporter engineer. The patients glucose levels (glucose sensor and SMBG), and technical alarms were transferred from patient's home via the internet and the Remote safety and control diabetes management system (MDRS) to the command and control center within the clinic and externally to the other team members at other remote sites. Optionally, the study team can provide the patient's caregivers with a PC which will be connected to the MDRS system and will show only the information and alarms relating to that specific patient. This allows the patient's caregiver to view glucose levels and insulin delivery from the own bedside. If required, support will be provided over the phone by the study team. The MDRS system is composed from two modules, the Remote diabetes monitoring (RDM) module and the Safety Module for patient alerts. This software is based on the ZONTM Control Software Package of Galooli Ltd. It is designed to transfer all the data from the patient to a remote control & command center which is capable of controlling tens of patients simultaneously.Thus, the remote monitoring enables the supervising personnel to alert the patient and intervene in cases of impending hypoglycemia, long standing hyperglycemia and technical faults of any component of the AP system and provides an added benefit of the whole closed loop system.O-3 KEYNOTE SPEAKER: GLUCOSE MONITORING - AND BEYOND?Weitgasser R.1,21Dept of Internal Medicine, Diakonissen Hospital Salzburg, Salzburg, Austria21st Dept of Internal Medicine, Paracelsus Medical University, Salzburg, AustriaBack in the 1980s when first programmes for patient education were established and validated monitoring glucose control became recognized as being a valuable part of diabetes care. Technical improvement in BG monitoring like the development of more accurate, small und fast acting devices led to a broad use of SMBG. Besides technical improvement SMBG was focused to structured measurements providing information on daily glucose variability. This challenge got support by the development of CGMS. New systems became a treatment complement predominantly used in patients on intensive insulin treatment regimens. Glucose sensor augmented insulin pump treatment meanwhile approaches the long aimed-for closed-loop system. Additional steps like the low glucose suspend to prevent hypoglycaemic reactions as well as the development of self-learning algorithms will probably further enhance glycaemic control. Beyond these technical device oriented steps into the future measures to improve patient empowerment and compliance need to be enforced. According to ongoing research the near future will probably provide each diabetic patient with some kind of a CGMS replacing currently used conventional structured SMBG. The use of smart phones or related e-health technology providing easy and time-sparing measures for physical activity (e.g. kind of, duration, intensity), nutrition (e.g. carbohydrate content/counting, fat content) and stress (illness, accident, psychological stress) combined with SMBG/CGM as well as information on blood pressure, lipids, body weight, smoking habits, etc. could help to improve treatment quality when embedded into a continuous feedback loop between patients and caring medical personal.O-4 A RANDOMIZED TRIAL OF A HOME SYSTEM TO REDUCE NOCTURNAL HYPOGLYCEMIA IN TYPE 1 DIABETESChase H.P.11Pediatrics, University of Colorado Anschutz Medical Campus, Aurora, USAH. Peter Chase, MD, for the In-Home Closed Loop Study Group University of Colorado Anschutz Medical CampusBackground: Overnight hypoglycemia is common in individuals with type 1 diabetes (T1D) and has many ramifications, including being a major barrier for optimal glycemic control. It is likely the predicted low-glucose suspend (PLGS) feature will be the second component of the artificial pancreas (AP).Methods: Following in-hospital safety studies, we have participated in an in-home randomized trial to determine whether nocturnal hypoglycemia could be safely reduced by temporarily suspending insulin pump delivery when hypoglycemia was predicted by an algorithm based on continuous glucose monitor (CGM) glucose levels. Following an initial run-in phase, 45 individuals with T1D (age 15–45 years) participated in a 42-night trial (total 1,912 nights) of randomized PLGS vs. control nights.Results: Overnight hypoglycemia with at least one CGM value ≤60 mg/dl occurred in 196 of 942 (21%) intervention nights versus 322 of 970 (33%) control nights (odds ratio 0.52, 95% confidence interval 0.43 to 0.64, P < 0.001). Median hypoglycemia area under the curve was reduced by 81% and hypoglycemia lasting >2 hours was reduced by 78%. Median morning blood glucose was 129 mg/dl after control nights and 144 mg/dl after intervention nights (P < 0.001). In each arm, 6% of nights had morning blood glucose >250 mg/dl and morning ketosis was present <1% of the time.Conclusion: Use of a nocturnal PLGS system can substantially reduce overnight hypoglycemia, with only a slight increase in morning hyperglycemia and no increase in ketosis.O-5 PREDICTIVE LOW GLUCOSE MANAGEMENT WITH SENSOR AUGMENTED CSII IN RESPONSE TO EXERCISEDanne T.11Diabetes Centre for Children and Adolescents, Kinder- und Jugendkrankenhaus AUF DER BULT, Hannover, GermanyBackground: Predictive Low Glucose Management (PLGM) may help prevent hypoglycemia by stopping insulin pump delivery based on predicted sensor glucose (SG) values.Methods: Hypoglycemic challenges were simulated using the FDA-accepted glucose simulator with 100 virtual patients. PLGM was then tested with a system composed of a Paradigm insulin pump, an Enlite glucose sensor, and a Blackberry-based controller. Subjects (n = 22) on CSII [5f, 17m; age 15 (14–20) years, diabetes duration 7 (2–14) years; HbA1c 8.0 (6.7–10.4)%, (median(range)) exercised until the PLGM system suspended insulin delivery or until the reference blood glucose value (HemoCue®) reached the predictive suspension threshold setting.Results: PLGM reduced hypoglycemia (<70 mg/dL) in silico by 26.7% compared to no insulin suspension, as opposed to a 5.3% reduction in hypoglycemia with use of LGS. The median duration of hypoglycemia (time spent <70 mg/dL) with PLGM was significantly less than with LGS (58 min vs 101 min, respectively, p < 0.001). In the clinical trial the hypoglycemic threshold during exercise was reached in 73% of the patients and hypoglycemia was prevented in 80% of the successful experiments. The mean (±SD) sensor glucose at predictive suspension was 92 ± 7 mg/dL resulting in a post suspension nadir (HemoCue®) of 77 ± 22 mg/dL. The suspension lasted for 90 ± 35 (range: 30 to 120) min resulting in a sensor glucose at insulin resumption of 97 ± 19 mg/dL.Conclusions:In silico modeling and early feasibility data demonstrate that PLGM may further reduce the severity of hypoglycemia beyond that already established for algorithms that use a threshold-based suspension.O-6 SAFE CONTROL ALGORITHMS TO PERSONALIZE THE OUTPATIENT ARTIFICIAL PANCREASDoyle F.11Chemical Engineering, UC Santa Barbara, Santa Barbara, USAModel-based control algorithms for the artificial pancreas have been demonstrated in numerous clinical trials as effective methods to manage overnight periods, postprandial excursions, and even some forms of physical exertion (mild to moderate exercise). However, the requirements for patient safety are somewhat reduced in the in-clinic setting where medical professionals (and often engineers) are hovering over the patient for the duration of the trial.As we move to the outpatient testing for the artificial pancreas, patient safety is a paramount concern. Our experience points strongly to personalization as a key component of an algorithm for effective and safe feedback control. The overall approach in our studies includes three main elements: MPC with time-varying zones, a Health Monitoring System (HMS) overlay, and customization in the form of subject specific attributes (e.g., insulin sensitivity). Our zone MPC has been extended to deal with (personalized) varying safety concerns over diurnal cycles, and allows the design of safe strategies for the overnight window. Our Health Monitoring System (HMS) incorporates hypoglycemic alarming as well as meal detection, signaling to the subject for possible interventions (snack ingestion, missed bolus, etc.). Finally, we have explored a number of approaches for personalization, from the model parameters used in MPC to adaptation that takes place over hours or days to "learn" patient attributes. All of these elements enhance the safety of a closed-loop artificial pancreas. Clinical data from pilot outpatient studies will be discussed along with algorithmic details.O-7 21ST CENTURY TECHNOLOGY AND ITS USE IN DIABETIC FOOT CAREArmstrong D.G.1, Giovinco N.1, Najafi B.11Surgery / Southern Arizona Limb Salvage Alliance (SALSA), University of Arizona College of Medicine, Tucson, USAAs Auguste Comte said two centuries ago, "Demography is destiny". Even he, however, couldn't foresee the massive changes occurring at breakneck speed in our world.Over the past generation, significant advances in care have led to incremental improvements in healing worldwide. However, it may be argued that the most potent advances in healing have been in organization of care. Technologies are now emerging that may allow further enhancements of organization and integration of care while also bringing in much needed bedside, chairside, and in-home diagnostics to identify key points in healing and potential early warning signs for recurrence. This symposium reviews what are believed to be several key areas of change over the next generation all yielding specific advances in wound diagnostics. The authors believe that devices will be organized into personal health servers in cloud-synchronized devices already existing in the home (eg, a scale), the clinic, and on (or in) the patient. This talk will explore the intersection of consumer technology with medical devices and their collision– or perhaps synergy – with our aging species.O-8 THE ROLE OF THE INTERVENTIONAL RADIOLOGIST IN DIABETIC FOOT MANAGEMENTReekers J.11Interventional Radiology, AMC, Amsterdam, NetherlandsThe role of the radiologist in diabetic foot management.Diabetic foot (DF) is recognized as one of the most serious complications of diabetic disease. About 5% off all patients with diabetes type II will develop an arterial diabetic foot problem. Up to 70% of all lower-leg amputations are performed in patients with diabetes and up to 85% of all amputations are preceded by an ulcer. Ulcer prevention is therefore recognized to be the best way to prevent amputation. However when an ulcer is present, the primary need is to achieve fast ulcer healing. If there is also concomitant infection the ulcer healing is often more difficult. Optimal wound care, antibiotics, off loading and other techniques should all be applied in daily practise to achieve ulcer healing. Active revascularisation plays a crucial role in achieving ulcer healing. Non-surgical revascularisation options for DF have expanded over the last decade and have become a prominent tool to prevent amputationThe radiologist plays in important role in the management of the diabetic foot. Both imaging and treatment are closely related. The definition and diagnosis of diabetic foot syndrome is not the same as CLI, and new diagnostic parameters can play an important role.Traditional imaging with duplex, CTA or MRA will guide the endovascular treatment options. Both open and endo treatment have the same outcome regarding limb salvage. Decision on all available treatment options is best done in a Multidisciplinary team.O-9 CURRENT STRATEGIES FOR THE TREATMENT OF THE SEVERELY INSULIN RESISTANT PATIENTHirsch I.11School of Medicine, University of Washington, Seattle, USAHistorically, the most severely insulin resistant patients had high levels of insulin antibodies from animal insulins. This resulted in the development of U-500 regular insulin in the early 1950s. Over time, this was also used for the more insulin resistant patients with type 2 diabetes. More recently, the use of U500 insulin has become extremely common due to the explosion of the severely insulin resistant patient. Many don't appreciate that our currently available U100 insulin glargine has minimal effect with doses greater than 1 u/kg. As a way to improve absorption, we are learning that splitting the large depot into more than one site can improve glycemia, but robust trials examining this strategy are lacking. Since the main rationale for using insulin analogues is to reduce hypoglycemia risk, it is reasonable and perhaps preferable to use NPH as the basal insulin for these patients. As for mealtime insulins, there are few studies comparing human insulin to a rapid-acting analogue for these patients, but the fast-acting insulins will have a prolonged activity with higher doses. The other strategy that often helps is adding a GLP-1 agonist to these patients. Small trials have shown improvements in glycemia and weight.For patients who still require high-doses of insulin, we currently utilize U-500 regular insulin that has a longer duration of action than U-100 regular insulin, and thus we use it both as a basal and a prandial insulin. Indeed, many use it in an insulin pump with good success.O-10 PATIENT NARRATIVES ON THEIR EXPERIENCES OF CGMPickup J.11Guy's Hospital, King's College London School of Medicine, London, United KingdomRelatively little is known about the patient experience of real-time continuous glucose monitoring (CGM) in type 1 diabetes when used in everyday practice and when described in patients' own words. We therefore conducted an online survey of patient narratives about continuous CGM, with analysis of the first 100 responses by qualitative framework analysis. There were 50 adults and 50 children, median CGM use 1.9 years, with 87% using it in conjunction with CSII. We identified 6 themes, with various subthemes: metabolic control, life on CGM, GGM procedures, technical issues, financial issues and attitudes to CGM. Most patients had an overwhelmingly positive experience with reduced HbA1c and hypoglycaemia and improved quality of life. Patients tended to recognize and accept limitations such as sensor inaccuracies. Some healthcare professionals were reported to have a very negative attitude to the technology. Many patients said that CGM was life-changing.O-11 PROMOTING DIABETES SELF CARE: WHAT WORKS AND WHAT DOESN'TSeley J.J.11Division of Endocrinology, New York Presbyterian Hospital/ Weill Cornell Medical Center, New York, USALiving with diabetes can be very challenging. We ask patients to perform multiple tasks day after day including monitoring blood glucose at frequent intervals, planning meals, calculating insulin doses based on current blood glucose and carbohydrate intake and balancing meals and medication with physical activity to achieve and maintain glycemic targets. It is no wonder that many patients have difficulty doing all that they are asked to do on a daily basis.Research has shown that knowledge is not enough to promote diabetes self-care and behavior change. A number of behavioral change theories offer the clinician guidance in preparing, motivating and supporting patients in diabetes self-care. These include patient empowerment, health belief model, transtheoretical model and motivational interviewing. The patient empowerment model puts the patient in control of their self-care and promotes informed decision-making. In the health belief model, benefits and barriers to performing self-care behaviors are identified and potential strategies to reduce barriers are generated. The transtheoretical model views behavior change as an ongoing process of stages ranging from precontemplation where the patient is unaware of a problem to maintenance where the patient has the ability to perform self-care over time. Motivational interviewing is an approach where the clinician uses active listening and encourages and supports self-efficacy. With the availability of more and more technological tools to manage diabetes, it is more important than ever to provide comprehensive education and support to our patients in order for them to succeed in controlling their diabetes without compromising quality of life.O-12 HOW TO SAFELY AND EFFECTIVELY TRAIN PATIENTS TO USE INSULIN PUMPS AND STAY ON THE DEVICERogers H.11Diabetes, King's College Hospital NHS Foundation Trust, London, United KingdomHow to Assist Patients to use an Insulin Pump Safely and Effectively and to Stay on the DevicePeople with diabetes wish to be able to self-manage their diabetes in order to achieve biomedical outcomes within target and they also desire to have the burden of self-management reduced to the extent that their quality of life is improved. Insulin Pump therapy can assist, but this is not always the case. A certain approach needs to be in place in order to ensure that Insulin Pump therapy is not just another treatment given to patients with an expectation that they will both master it and achieve improved outcomes. Education is the key to ensuring improvements in biomedical outcomes, in self-management and in burden reduction. And not just any education programme - structured education that is underpinned by facets that lead to mastery and maintenance is required. Health Care Professionals (HCPs) are accustomed to providing education that includes knowledge and self-management skills, however HCPs are beginning to recognise that these alone are not enough. If the twin goals of self-management and reduced burden are to be achieved then the structured education also needs to incorporate ways to improve confidence, participation in goal setting and decision making, coping skills, and self-efficacy.Insulin pump pathwayChoice of pumpTrial using salineEducation - bite-sized chunksAvailability of HCPs and Pump ExpertiseWe will examine the attributes above to establish how the knowledge and skills needed for Insulin Pump therapy can be best presented by HCPs.O-13 HOW TO SAFELY AND EFFECTIVELY TRAIN PATIENTS TO USE CGM AND STAY ON THE DEVICEGianini A.11Department of Endocrinology Diabetes and Metabolic Diseases, Children's Hospital, Ljubljana, SloveniaEducation and the use of continous glucose monitoringAna GianiniModern technology entered diabetes treatment with continuous subcutaneous insulin infusion (CSII) and was frequently leading to improved matabolic control. In the last 10 years next to CSII, systems for continuous glucose monitoring (CGM) gave new information about glucose excursions in different situations. According to different databases such as Type 1 Diabetes Exchange more than 10% of adults are using CGM routinely, next to them 3% in pediatric cohorts.In Slovenia children and adolescents use CSII in more than 80% (530 from 660). The use of sensors was reimbursed for children in February 2010 and since then the number of young patients that continuously use CGM is increasing steadily, reaching 10% of pump users in 2013.Structured education for patients, families and often professional caregivers about the sensor use is of extreme importance at the CGM introduction. In the first month patients can be confused by the number of informations and alarms from CGMS leading frequently to disappointment and discontinuation of CGM use.Topics discussed at CGM introduction are• Proper insertion of the sensor• Importance of good calibration• Alarms, troubleshooting• ISIG signal• Practical use of CGM (profiles, correction boluses, food boluses, sport activity, sick days, school or kindergarten regimen … )Sometimes parents are advised to shut down the alarms for the first sensor or even for the first month of sensor use. In this case they simply follow CGM curve and values on the screen.24/7 telephone support can help to support the patient.O-14 THE SWEET-PROJECT - USE OF TECHNOLOGY FOR LONGITUDINAL BENCHMARKING OF INTERNATIONAL PEDIATRIC DIABETES CENTRES 2006 TO 2013: DATA ANALYSIS FROM 122.853 VISITS FROM 10.767 PATIENTSWitsch M.1, SumnÍk Z.2, Veeze H.3, de Beaufort C.1, Robert J.J.5, Forsander G.6, Szypowska A.7, Allgrove J.8, Waldron S.9, Rosu M.10, Gerasimidou-Vazeou A.11, Lange K.12, Kordonouri O.4, Maffeis C.13, Raposo J.F.14, Pankowska E.15, Madacsy L.16, Klee K.4, Aschemeier B.4, Danne T.4, for the SWEET Group1DCCP- Clinique pédiatrique de Luxembourg, Luxembourg, Luxembourg2University Hospital Motol, Department of Paediatrics, Prague, Czech Republic3Stichting Diabetes, Rotterdam, Netherlands4Kinderkrankenhaus auf der Bult, Hannover, Germany5Hopital des Enfants-Malades, Department Diabete de l'enfant, Paris, France6Sahlgrenska University Hospital, Gothenburg, Sweden7The Medical University of Warsaw, Department of Pediatric Diabetology, Neonatology and Birth Defects, Warsaw, Poland8Royal London Hospital, Whitechapel, Barts and the London NHS Trust, London, United Kingdom9Dorset County Hospital, Dietetic Department, Dorset, United Kingdom10Clinical Medical Center "Cristian Serban" for the Evaluation and Rehabilitation of Children and Adolescents Buzias, Buzias, Romania11Panagioti and Aglalia Kyriakou Children's Hospital, Department of Pediatrics and Diabetes Center, Athens, Greece12Hannover Medical School Medical School, Department of Medical Psychology, Hannover, Germany13University of Verona, Pediatric Diabetes Unit, Verona, Italy14Associação Protectora dos Diabéticos de Portugal, Lisboa, Portugal15Instytut Matki I Dziecka, Warsaw, Poland16Semmelweis University, Budapest, HungaryObjectives: "SWEET" is an acronym derived from "Better control in Pediatric and Adolescent diabeteS: Working to crEate cEnTers of Reference" and is based on a partnership of established national and European diabetes organizations (www.sweet-project.eu) led by ISPAD. Data in participating centres were directly extracted from 2006 ongoing from local electronic health records.Methods: The SWEET Online platform allows presently nineteen centres from fifteen countries to connect to one unified anonymized diabetes database. Aggregate data are de-identified and exported for longitudinal health and economic data analysis.Results: The number of patients and patient visits increased from 2006 (n = 921) to 2012 (n = 7633), currently including 10,767 patients and 122,853 patient-visits overall. For example, patients with a valid HbA1c in the database rose from 744 (mean H bA1c: 8,1%) in 2006, to 1161 (8,1%) in 2007, 1412 (8,2%) in 2008, 1972 (8,2%) in 2009, 3320 (8,0%) in 2010, 5952 (7,9%) in 2011 and 6372 (7,9%) in 2012. The percent of patients within the target HbA1c range <7.5% increased steadily: 33% (2010), 35% (2011), 40% (2012). Over time the completeness of data increased from 82% to 98% (HbA1c), 78% to 83% (height), 78% to 84% (weight), 50% to 60% (blood pressure), 12% to 22% (microalbuminuria screening) and 30% to 45% (hyperlipidemia screening).Conclusions: Ongoing collection of benchmarking data motivates centres to improve data collection and reflects improving glycemic control in most participating European pediatric diabetes centres. While the degree of completeness is close to 90% or above for HbA1c, weight and height, the assessment of diabetes-associated co-morbidities leaves much room for improvement. Thus, information technology allows transparent analysis of real life diabetes treatment data as a basis for local center improvement, scientific studies and health economic analyses.O-15 INTERNATIONAL ASSESSMENT OF DIABETES MANAGEMENT, GLYCEMIC CONTROL AND DIABETES-RELATED BURDEN IN YOUTH WITH TYPE 1 DIABETES (T1D): THE TEENS STUDYLaffel L.1, Peterkova V.2, Domenger C.3, Dain M.P.3, Pilorget V.4, Candelas C.4, Danne T.5, Phillip M.6, Mazza C.7, Anderson B.8, Hanas R.91Pediatric Adolescent and Young Adult Section, Joslin Diabetes Center, Boston, USA2Pediatric Endocrinology, Endocrinology Research Center, Moscow, Russia3Global Diabetes Division, Sanofi, Paris, France4Clinical Sciences and Operations, Sanofi, Chilly Mazarin, France5Diabetes Centre for Children and Adolescents, Kinder und Jugendkrankenhaus "Auf der Bult", Hannover, Germany6National Center for Childhood Diabetes, Institute for Endocrinology and Diabetes at Schneider Children's Medical Center of Israel, Petah Tikva, Israel7Nutrition Department, Hospital de PediatrÍa J P Garrahan, Buenos Aires, Argentina8Department of Pediatrics Section of Psychology, Baylor College of Medicine, Houston, USA9Department of Pediatrics NU Hospital Group, Uddevalla and Sahlgrenska Academy at Gothenburg University, Gothenburg, SwedenBackground: TEENS, a cross-sectional study in 21 countries, collected data on management, HbA1c, acute complications, and burden from ∼6200 youth with T1D of ≧1 year and onset <18 y/o. Results from 1000 participants from 2 countries completing the study provide direction for future post-hoc analyses.Objectives: To examine treatments and HbA1c in T1D youth from USA/Russia in three predefined age groups and relate HbA1c to acute complications and diabetes burden.Methods: 25 US/20 Russian centers uniformly collected data by interview, record review and survey. A recruitment ratio of 25%/50%/25% was implemented in 3 age groups (8–12/13–18/19–25 y/o). Participants were sampled sequentially to avoid bias. HbA1c was measured uniformly; targets: <7.5% (58 mmol/mol) <18 y/o (ISPAD); <7% (53 mmol/mol) ≧18 y/o (ADA). Burden was assessed with the PAID (20-item) for patients (≧13 y/o) and PAID-PR (18-item) for parents of youth (≤18 y/o). Acute complications were assessed (% with DKA, severe hypoglycemia [seizure/coma]).Results: Median T1D duration was similar: 7.0 years (0.6–22.1) in US; 6.3 years (0.8–23.5) in Russia. Both syringes/pens were used by 36%/80% of US/Russian patients; pumps were used by 63%/19%, respectively. In both countries, only a small minority of participants achieved HbA1c targets (24%/16% in USA/Russia). Parents from both countries perceived greater diabetes burden than patients across all ages. Outcomes (DKA/hypoglycemia) are shown by age group (table).Conclusion: Diabetes burden appears to be universal; patients experience subopti
Referência(s)