INFUSYSTEMS ASIA. INFUSYSTE INFUSYSTEMS ASIA - INFUSYSTEMS ASIA - INFUSYSTEMS ASIA - INFUSYSTEMS ASIA - INFUSYSTEMS ASIA - INFUSYSTEMS ASIA - INFU Episodic use of Real-Time Continuous Glucose Monitoring for endurance exercise by people with Type 1 diabetes, including a personal perspective Daniel Seller (1,2), David O. Neal (1), Mark Hargreaves (3), Alicia Jenkins (1) (1) University of Melbourne, Dept. of Medicine (St. Vincent’s Hospital), Melbourne, Australia; (2) Physiotherapy Dept., St. Vincent’s Hospital, Melbourne, Australia; (3) University of Melbourne, Dept. of Physiology, Melbourne, Australia R esearch studies have demonstrated that people with Type 1 diabetes derive greater glycaemic benefit, specifically lower HbA1c levels, and possibly less hypoglycaemia, with use of real-time Continuous Glucose Monitoring (RT-CGM) to facilitate insulin dosing delivered by either continuous subcutaneous insulin infusion (CSII) therapy or multiple daily injections (MDI) (1-5). However glycaemic benefit, at least in clinical trials, relates to the number of days per week the person with diabetes uses RT-CGM, usually with 60-70% or more time wearing and reacting to the RT-CGM system being required to derive significant benefit (1,2). In the JDRF-CGM Study, significant predictors of HbA1c reduction after 6-months RT-CGM use were adulthood, high RT-CGM usage time when first commenced on the technology and frequent pre-CGM-study blood glucose monitoring (3). Although people with recurrent severe hypoglycaemia have often been excluded from RTCGM-trials, RT-CGM is also associated with at least a trend to less severe hypoglycaemia (2) and less time with low interstitial fluid glucose (5). While there is strong clinical trial evidence indicating that continuous RT-CGM use is likely to improve glycaemia, because there are no subsidies for RT-CGM devices or sensors in our region (Australia), the few people with diabetes who use RT-CGM often do so episodically. However, we believe that a situation in which episodic RT-CGM use could be of benefit to people with Type 1 diabetes is for prolonged exercise such as long-distance running or cycling, and for safety critical situations in recreation (e.g. mountaineering, flying) or work (e.g. high altitude work or long-distance truck driving). There are few reports of such RTCGM use, no clinical trials of which we are aware, and few recommendations available for such use of RT-CGM. In this article, we discuss the physiological response to aerobic exercise in people with and without Type 1 diabetes, the glycaemic challenges of prolonged exercise, current guidelines for endurance exercise by people with Type 1 diabetes, and the potential advantages and disadvantages of RT-CGM in endurance sports events. In addition, we describe an author’s (DS) personal experience of his Type 1 diabetes and RT-CGM use during training for, and participation in a marathon. The physiological response to aerobic exercise in people without diabetes During an endurance event such as a marathon, performance is dependent on the mobilisation and utilisation of fuels by contracting skeletal muscle, delivery of oxygen to those muscles and the dissipation of heat. The major source of ATP for contracting skeletal muscle is the oxidation of glucose, derived either from intramuscular glycogen stores or from circulating plasma Vol.7 No.1 2012 INFUSYSTEMS ASIA 2492 Walnut Avenue, Suite 130 Tustin, Ca., 92780, USA Email: infuasia@yahoo.com EDITORIAL BOARD Editor in Chief J-L. Selam (USA) Associate Editor D. Selam (USA) Board Members Fergus Cameron (Australia) Arthur Charles (USA) Neale Cohen (Australia) Kyung Ah Han (Korea) Alicia Jenkins (Australia) Ryuzo Kawamori (Japan) Bruce King (Australia) Kisho Kobayashi (Japan) Boniface Lin (Taiwan) David McIntyre (Australia) Mitsuyoshi Namba (Japan) David O'Neal (Australia) Carmel Smart (Australia) Hiroshi Uchino (Japan) PUBLISHER Publiscripts 2492 Walnut Avenue, Suite 130 Tustin, Ca., 92780, USA Tel: +1 949 910 0991 Fax: +1 949 429 2160 www.publiscripts.com SPONSORED BY Medtronic Diabetes CONTENTS . Episodic use of Real-Time Continuous Glucose Monitoring for endurance exercise by people with Type 1 diabetes, including a personal perspective ...................... 1 . Strict Glycemic Control in Japanese Type 2 Diabetes Patients with Incretin-based Therapy – Efficacy of Continuous Glucose Monitoring for the secure transitransition and fine tuning .................... 6 Page 2 glucose. Lipid oxidation is also an important source of energy for skeletal muscle, particularly during prolonged, low-intensity exercise (6-8). During exercise, increases in glucose delivery to muscle, secondary to skeletal muscle hyperaemia, sarcolemmal glucose transport due to glucose transporter (GLUT4) translocation and intracellular glucose disposal act in concert to enhance skeletal muscle glucose uptake (7,8). This occurs in an insulin-independent manner and, in fact, circulating insulin levels fall during exercise, in response to reductions in blood glucose. Hepatic glucose output increases in parallel with increasing muscle glucose uptake, although during prolonged strenuous exercise blood glucose levels can fall, stimulating the release of the hormones glucagon and adrenaline. Carbohydrate ingestion increases blood glucose availability and enhances endurance exercise performance. With regular exercise training, insulin sensitivity increases in both nondiabetic people and people with diabetes. The physiological response to aerobic exercise in people with Type 1 diabetes The metabolic response to exercise is essentially similar in people with Type 1 diabetes. However, with insulin therapy using CSII or MDI, once the exogenous insulin has been administered there is currently no means to lower circulating insulin levels. Since exercise and insulin exert additive effects on muscle glucose uptake, there is the potential for premature and possibly severe hypoglycaemia during exercise in people with Type 1 diabetes who have not appropriately adjusted their insulin dose, or ingested sufficient additional carbohydrate. Also, in Type 1 diabetes, particularly after years of the condition, there may be an impaired counter-regulatory hormone (glucagon and adrenaline) response to falling blood glucose levels. As already mentioned, exercise increases insulin sensitivity (6), which while seen as an overall benefit in health and diabetes management, may increase hypoglycaemia risk during exercise. As this heightened insulin sensitivity can be present for up to 72 hours postexercise (9), people with Type 1 diabetes can be at increased risk of clinically significant hypoglycaemia, up to and beyond 24 hours after exercise, a phenomenon also known as delayed onset or latent hypoglycaemia (6-9). Another temporary common phenomenon is that of post-exercise hyperglycaemia, due to an imbalance between glycogenolysis and the lower muscle demand for glucose. This usually corrects itself (9), sometimes to the point of hypoglycaemia, as insulin causes muscle and liver to replenish their glycogen stores from the plasma glucose (8). Vol.7 No.1 2012 Guidelines for managing Type 1 diabetes during prolonged exercise Evidence-based recommendations for both athletes with diabetes (whether elite, recreational, or novice) and their health-care team are limited. In 2004, the American Diabetes Association and the American College of Sports Medicine released a joint position statement (10) which outlined some general guidelines for exercise around three key points: 1) avoid physical activity if blood glucose (BG) is >13.9mmol/L with ketones, and use caution if BG is >16.7mmol/L without ketones; consume extra carbohydrates if BG is <5.5mmol/L; 2) identify when changes in insulin or food intake are necessary, and know the glycaemic response to different physical activities; 3) consume extra carbohydrates as needed to avoid hypoglycaemia, and have carbohydrate-based foods readily available during and after physical activity. A position statement from the National Athletic Trainers’ Association (9) makes 16 recommendations regarding athletes with Type 1 diabetes, organised into categories: diabetes care plans, training kit supplies, pre-participation examination, prevention and management of both hypoglycaemia and hyperglycaemia, insulin administration, travel recommendations, injuries, and glycaemic control. Although these guidelines contain a number of useful practical considerations not otherwise discussed in the exercise literature – such as air travel with diabetes supplies – they do not contain sport-specific recommendations for diabetes management during exercise, instead recommending that each athlete has their own specific diabetes care plan established. Two textbooks which deal specifically with people with diabetes undertaking marathons are Colberg’s Diabetic Athlete’s Handbook (7) , and Nagi’s Exercise and Sport in Diabetes (6). Colberg recommends insulin pump users decrease their basal insulin on the morning of the event by 25100%, decrease pre-event meal bolus doses by 25-75%, and post-event meal boluses by 25-50%, and keep basal rates reduced by 10-25% for the rest of the day, and overnight following the event (7). Nagi recommends decreasing the basal insulin infusion rate by 50% or more, 30 minutes before the event. Similar dosage adjustments could be used by MDI users6. However, these recommendations represent consensus opinion given the limited data available. A personal perspective of endurance exercise with Type 1 diabetes The runner (DS) was diagnosed with Type 1 diabetes as a 10 year old boy in 1989, and has been using a Medtronic Paradigm (MMT 722, Medtronic, Minimed, Northbridge CA) insulin pump since November 2004. Prior to undertaking a marathon in 2010 at age 31 years, since 2003 he has trained for and undertaken a number of endurance events, including: a 24-hour team cycle relay six times, a 24hour team swimming relay three times, a five-day team kayaking relay, a number of 4-16km fun-runs, and one marathon. For the marathon in 2009 he wore a Medtronic Minilink RT-CGM (ParadigmTM Realtime system, Medtronic, Minimed, Northbridge CA), linked to his insulin pump – which he had worn previously on six occasions, including training runs, and one fun run. The preferred sensor insertion site was the anterolateral abdomen, chosen for ease of insertion, body contour and subcutaneous fat distribution, and to ensure that the transmitter was near enough to the pump, which he usually carried in a front pocket. Marathon training. The runner undertook a 16-week training program of three runs per week, which gradually increased in distance and intensity for 13 weeks, with a 3week taper before the marathon. Each training week consisted of a high-intensity “interval run” session and a moderate-high intensity “tempo run” session (each of 4060 minutes), and one low-moderate intensity “long run” for 90-180 minutes. Session intensities were based on the previous 10km race pace, rather than heart-rate. Based on experience in endurance exercise events, his insulin pump basal rates for all training sessions were decreased to 5% of the usual rate, commencing 30-90 minutes prior to the training run, depending on preceding BG trends. This basal rate adjustment regime was initially based on the guidance of his Diabetes Nurse Educator (DNE) involved with commencement of pump therapy: to reduce the basal rate by approximately 50%, 60 minutes prior to exercise. While consistent with published recommendations regarding basal rate adjustments (6,7), pharmacokinetic data on the effect of insulin pump basal rate adjustments on circulating insulin levels to guide these recommendations is currently extremely limited. Due to consistent hypoglycaemia early in exercise sessions, the temporary basal rate was titrated down until hypoglycaemia during exercise was avoided – to the current temporary basal rate of 5%. The target BG prior to training was 8mmol/L, with no session started with a BG less than 6mmol/L. For training sessions under 60 minutes, if the BG was below 8mmol/L, or if hypoglycaemia had been treated in the previous two hours, the pump was disconnected for the session. For long (16-32 km) training runs, the Vol.7 No.1 2012 pump basal rate, initially reduced to 5% of normal, was increased to 10% of normal in the later weeks of the training period due to a reduction in normal basal insulin doses related to increased insulin sensitivity with increased fitness. After running for approximately 90 minutes, the basal rate was increased to 15-25% of the normal basal rate (depending on BG) until the last 30 minutes, when 100% of the normal rate was resumed. While not suggested in any published recommendations, it minimises the author’s marked post-exercise hyperglycaemia: although still transiently rising up to 14-16mmol/L in the hour immediately after a long run, prior to adopting this approach it would commonly rise as high as 20mmol/L. The author is considering a temporary basal rate above 100%, or a small bolus upon completion of the activity, to further limit his post-exercise hyperglycaemia. Fingerprick BG testing was undertaken approximately two-hourly post-run and once the glucose was decreasing the insulin basal rate is again decreased, usually to 60-70% of the normal basal rate. This rate was gradually increased by 10-15% every four-six hours, until insulin requirements returned to normal, usually within 18-24 hours, with regular fingerprick BG monitoring – including overnight testing. During a long run, 1-2 energy gels (~2530gm carbohydrate each) are consumed– depending on the distance of the session – without an insulin bolus. The bolus dose immediately following a session is reduced by ~25%, with no correction given for hyperglycaemia. Bolus doses after this are given at the normal insulin to carbohydrate ratios. BG testing is routinely performed approximately hourly during long runs, approximately 2-3 hourly following completion of the run, and usually once overnight following long training runs. Race day. The goal running time was four hours or less. In the three days pre-event, carbohydrate loading was undertaken – with a diet consisting of predominantly low glycaemic-index foods, such as pasta, and rice – with normal insulin boluses given. The RT-CGM sensor was inserted into the anterolateral abdomen approximately 36 hours prior to the event – allowing adequate time for settling in and calibration prior to the event, yet attempting to minimise skin irritation issues which had occurred with previous RT-CGM use due to the adhesive tapes. This site had been successfully used during a fun-run previously. Figure 1 shows two days of the author’s RT-CGM sensor results, from midnight prior to the race, alongside the fingerprick BG results, and absolute insulin dose (basal rate + bolus doses). Percentage of normal basal insulin infusion rate is also included to show relative insulin adjustments, as well as overall duration of temporary basal rate use. Overnight prior to the event, two BG tests were done – to facilitate target BG levels and to optimise sensor calibration at event commencement. The basal rate was reduced to 10% of normal 90 minutes before the race start (Figure 1(a)). The RTCGM was briefly disconnected to reinforce the adhesive tape at this time. Problems related to this disconnection and reconnec- Page 3 tion caused no RT-CGM signal for the following two hours, including the first 30 minutes of the race - as can be seen from Figure 1 (b). During this time, a fingerprick BG reading 20 minutes before the race start was 11.8 mmol/L, and one energy bar (50g carbohydrate) was consumed, with a reduced bolus of 2.7 units of NovoRapid (50% of predicted) given. After the first 5km, when the RT-CGM was still offline, a fingerprick BG was 17.4mmol/L. This BG reading was used to calibrate the RT-CGM which came back online 15 minutes later, approximately two hours after it had been disconnected – which is the normal delay cited by the manufacturer following connection prior to the first reading. A small correction bolus (1.5 units NovoRapid insulin) was given, and the temporary basal rate increased to 20%, for 30 minutes. As the then active RT-CGM showed a decreasing trend, the insulin basal rate was further decreased to 15%, and a BG reading was 6.3mmol/L – with a corresponding RTCGM reading of 10.2mmol/L. Approximately 20g rapid acting carbohydrate (jelly beans) was ingested. Energy gels (25g of moderate glycaemic index carbohydrate) were consumed after running 12km, 24km, 30km, and 36km, with no insulin boluses. The race was completed in 3:56:45 at an average running pace of 5 minutes 37 seconds per kilometre. The average heart rate during the race was 169 beats per minute – 90% of the predicted maximum heart rate. The median (range) BG reading from the four fingerprick tests during the race was 9.4mmol/L (6.3 – 17.4mmol/L), compared to a median (range) interstitial glucose reading of 12.9mmol/L (10.1 – Advantages of RT-CGM Disadvantages of RT-CGM 17.3mmol/L) from 39 data points from the RT-CGM. Delayed onset hypoglycaemia • Monitoring trends makes it easier to • Cost: ~$80 per sensor for up to six was avoided, although there was a marked hyperglycaemic excursion seen on RTproactively treat low or high BG levdays use CGM trace following completion of the run els • Discomfort: skin reaction to adhe(Figure 1 (c)), with an associated finger• Easier to monitor the effects of sive, both on sensor site, and tapes prick BG reading of 14.6mmol/L. This interventions (e.g. hypo. treatments) needed to secure transmitter • Can guide targeted BG testing – • Some difficulty with sensor insertion decreased shortly after, following the lunchtime insulin bolus. The hypoglymay be more useful than routine test- - due to angle, fragility of sensor caemic episodes late in the study period ing • Sensors prone to dislodge during (Figure 1(d)) may have been due to the ear• Instant information activity, due to weight of transmitter lier bolus (e) in the setting of increased • Downloadable: analysis of results • Time lag: ~11 min. difference insulin sensitivity. helps to refine diabetes management between BG, and interstitial glucose During this marathon, the plan was to perstrategies when used during event levels, which are exacerbated with form BG tests at approximately 14km intertraining rapidly rising or falling BG vals. Aside from the BG test at 5km, this • Low glucose suspend function help- • Difference between absolute interplan was followed – otherwise relying on ful if RT-CGM used with a compatible stitial and BG levels the RT-CGM trend data to alert to the need • Dropouts – both technological (e.g. for additional fingerprick BG tests, insulin insulin pump signal strength), and practical (e.g. basal rate adjustments, or carbohydrate disconnection to re-secure dressing) ingestion. When there was discordance between the RT-CGM data and the BG – which may partly reflect lag time – actions Table 1: Advantages and Disadvantages of RT-CGM for endurance or safety critical were based on the BG reading with the trend arrows providing additional perspecexercise. tive. Performing BG tests “on the run” did Vol.7 No.1 2012 Page 4 Figure 1: - RT-CGM trace, fingerprick blood glucose tests and insulin doses (basal & bolus) for 2010 Gold Coast Airport Marathon (shaded area). require much more concentration and dexterity in a field of thousands of runners than it had in training – with the BG meter even being dropped at one point, during a test. While the high and low interstitial glucose alarms of the RT-CGM were set at 9mmol/L and 4.5mmol/L respectively, these were ignored during the event – as the readings were consistently triggering the high alarms throughout the entire event, despite a fingerprick BG reading as low as 6.3mmol/L. Discussion Based on our reading of the literature, experiences of the author (DS) and of other athletes with Type 1 diabetes we believe that episodic use of RT-CGM could be helpful in endurance or safety critical exercise or occupations. In the 2006 Race Across America (RAAM), RT-CGM technology (FreeStyle Navigator, Abbott, Abbott Park IL) was used to augment fingerprick BG readings by Team Type 1 (www.teamtype1.org) – a team of eight elite cyclists with Type 1 diabetes11. Team Type 1 won the open team event in a record time of 5 days, 16 hours and 4 minutes. There was not a statistically significant change in median glucose readings between either pre-RAAM masked or unmasked phases of their RT-CGM trial compared to the RAAM phase. However, RT-CGM use was associated with significantly less hypoglycaemia: in the masked RT-CGM period (preRAAM) 5.5% of interstitial fluid glucose readings were below 60mg/dL (3.3mmol/L), compared to 3.7% in the preRAAM unmasked period, and 2.7% in the unmasked RAAM (race) period11. While this reduction may equate to relatively few hypoglycaemic episodes in daily life, the detrimental impact of even one hypoglycaemic episode during a competitive athletic event such as RAAM could be quite profound. Therefore, this reduction may provide a significant performance improvement to the athlete with Type 1 diabetes if it were made possible by the use of a RT-CGM system. For people both with and without Type 1 diabetes, endurance events typically require an extended period of committed training. In addition to the increased aerobic fitness from this training, other physiological adaptations occur – such as increased insulin sensitivity – which have implications for the athlete’s diabetes management both while training and in day-to-day life. Adjustments to the runner / author’s diabetes management strategy throughout training were developed largely through trial and error: as well as the limited evidence-based guidelines regarding diabetes management strategies during exercise, wide variability between people also limits the applicability of those recommendations which do exist. Some changes were technical – such as practicing the skill of performing BG tests and making pump adjustments while running, or determining the best method for carrying enough carbohydrate both for nutrition and hypoglycaemia treatment without a bag. Some successful modifications in the diabetes management strategy, such as increasing the basal insulin during the later stages of a run were somewhat unexpected. RT-CGM use during training allowed confirmation of glucose trends observed with BG testing, as well as fine-tuning of the post-training basal insulin management, to minimise the risks of delayed-onset hypoglycaemia. While fingerprick BG testing is episodic during training or an event, the RT-CGM data are usually always available, with the interstitial glucose value and trend arrows being updated every five minutes. This allows increased confidence of safety, and earlier management of glycaemic fluctuations – even if that consists of only a BG test to confirm the RT-CGM trend data. RT-CGM also facilitates monitoring of the effects of insulin basal rate changes, carbohydrate intake, and hypoglycaemia treatments. Another major benefit of RT-CGM is the ability for trend analysis during times when regular BG monitoring is impractical, such as overnight. This is particularly helpful when gradually returning basal insulin rates back to normal after a training session, to minimise the risk of delayed onset hypoglycaemia. Furthermore, the ability to upload RT-CGM data from the insulin pump (or standalone RT-CGM system) to a webbased program allows detailed post-event review, as well as evaluation and adaptation of management strategies, or formulation of new strategies. While more frequent RT-CGM use would likely provide even greater benefits when preparing for an exercise event this is precluded for most by cost as there are no Government or health insurance fund rebates. Currently in Australia, RT-CGM sensors which last six days cost approximately $80 each, and a transmitter for use with a compatible insulin pump costs $1250. A stand-alone RT-CGM device for MDI users or those with a non-compatible insulin pump is approximately twice this, with the disposable glucose sensor costs being the same. Additional technological limitations exist – such as the inherent time lag between blood and interstitial glucose levels, reading inaccuracies, temporary signal drop-outs, and sensor failure. Inaccuracies in readings may be caused by poor calibration, and are known to be worse at extremely high or low BG levels, and Vol.7 No.1 2012 with rapidly changing BG, such as may occur during exercise. Whilst a number of current-generation RT-CGM systems are equipped with trend alarms, the model used for this event had only high and low glucose alarms – of limited value during events which may precipitate rapid glycaemic change such as in this study, and these were largely ignored during the event. Recently however, Iscoe et al reported that setting the low glucose alarm at a higher level (5.5mmol/L) significantly reduced the incidence of exercise-induced hypoglycaemia compared to the routine low glucose alarm setting of 4mmol/L, without triggering false alarms (12) – so a similar strategy may be worth considering for events such as this. Signal drop-outs may be sensor-related, or due to other practical issues – such as dislodgement due to the athlete’s movements, or need to remove the transmitter to reinforce adhesive tapes. The size and weight of most earlier and current generation RTCGM transmitters is greater than that of an insulin pump infusion set, and coupled with the fragility and sensitivity to movement RT-CGM sensors require a larger and stronger adhesive dressing than that used for CSII therapy. This need is further exacerbated during physical activity – such as running – which causes the transmitter to bounce. A RT-CGM worn by the author in his first marathon malfunctioned after the first 500 metres of the event, most likely due to the bouncing causing the weight of the transmitter to pull the sensor out. This problem could be lessened by a combined insulin delivery and glucose sensing line. Prolonged RT-CGM use may also lead to skin irritation from adhesive film dressings, particularly during warmer months or from perspiration during physical activity. Our experience using different products to secure the sensor during a variety of events has found tapes more effective than adhesive film dressings to secure the sensor and transmitter to prevent movement and also minimise build-up of perspiration. An additional advantage of tapes is that they can be applied in different orientations across the sensor and transmitter (i.e. vertically or diagonally) to rest the skin. Table 1 summarises the advantages and disadvantages of RT-CGM use during preparation for and participation in events such as a marathon. Conclusion Most non-elite athletes who run marathons or participate in other similar endurance events rarely undertake more than one or two per year – making refining diabetes management strategies based on experience alone a challenge. Despite some limitations, RT-CGM technology enables diabetes management strategies to be developed and refined with a far higher level of evidence and precision than with BG testing alone. In day-to-day life with Type 1 diabetes RT-CGM has been found to provide the greatest glycaemic benefit with more frequent use (1,2), but these highly physiologically challenging but infrequent endurance sporting events are a potentially useful opportunity for episodic RT-CGM use. RT-CGM use, sharing of the experiences, controlled clinical trials and the refinement of evidence based guidelines for exercise by intensively treated people with Type 1 diabetes may be very beneficial to participants and their diabetes-management teams. Better glycaemic control during and around exercise, and reduced fear of glycaemic excursions may encourage more people with Type 1 diabetes to reap the enjoyment and health benefits of exercise. References 1. O'Connell MA, Donath S, O'Neal DN, Colman PG, Ambler GR, Jones TW, Davis EA, and Cameron FJ, Glycaemic impact of patient-led use of sensor-guided pump therapy in type 1 diabetes: a randomised controlled trial. Diabetologia, 2009. 52(7): p. 1250-1257. 2. Bergenstal RM, Tamborlane WV, Ahmann A, Buse JB, Dailey G, Davis SN, Joyce C, Peoples T, Perkins BA, Welsh JB, Willi SM, and Wood MA, Effectiveness of SensorAugmented Insulin-Pump Therapy in Type 1 . Page 5 Diabetes. New England Journal of Medicine, 2010. 363(4): p. 311-320. 3. Tamborlane WV, Beck RW, Bode BW, Buckingham B, Chase HP, Clemons R, Fiallo-Scharer R, Fox LA, Gilliam LK, Hirsch IB, Huang ES, Kollman C, Kowalski AJ, Laffel L, Lawrence JM, Lee J, Mauras N, O'Grady M, Ruedy KJ, Tansey M, Tsalikian E, Weinzimer S, Wilson DM, Wolpert H, Wysocki T, and Xing D, Continuous glucose monitoring and intensive treatment of type 1 diabetes. New England Journal of Medicine, 2008. 359(14): p. 1464-1476. 4. Hermanides J, Norgaard K, Bruttomesso D, Mathieu C, Frid A, Dayan CM, Diem P, Fermon C, Wentholt IM, Hoekstra JB, and DeVries JH, Sensor-augmented pump therapy lowers HbA(1c) in suboptimally controlled Type 1 diabetes; a randomized controlled trial. Diabet Med, 2011. 28(10): p. 1158-1167. 5. Battelino T, Phillip M, Bratina N, Nimri R, Oskarsson P, and Bolinder J, Effect of continuous glucose monitoring on hypoglycemia in type 1 diabetes. Diabetes Care, 2011. 34(4): p. 795-800. 6. Nagi D, Exercise and sport in diabetes. 2nd ed. Diabetes in practice. 2005, Chichester, England ; Hoboken, NJ: Wiley. xiv, 218 p. 7. Colberg S, Diabetic athlete's handbook. 2009, Champaign, IL: Human Kinetics. xi, 284 p. 8. Lumb AN and Gallen IW, Diabetes management for intense exercise. Curr Opin Endocrinol Diabetes Obes, 2009. 16(2): p. 150155. 9. Jimenez CC, Corcoran MH, Crawley JT, Guyton Hornsby W, Peer KS, Philbin RD, and Riddell MC, National athletic trainers' association position statement: management of the athlete with type 1 diabetes mellitus. J Athl Train, 2007. 42(4): p. 536-545. 10. Zinman B, Ruderman N, Campaigne BN, Devlin JT, and Schneider SH, Physical activity/exercise and diabetes. Diabetes Care, 2004. 27 Suppl 1: p. S58-62. 11. Demma L HC, Bode B, The Utility of a Continuous Glucose Monitoring System for Elite Cyclists During the Race Across America. Infusystems USA, 2007. 4(1): p. 4-8. 12. Iscoe KE, Davey RJ, and Fournier PA, Increasing the low-glucose alarm of a continuous glucose monitoring system prevents exercise-induced hypoglycemia without triggering any false alarms. Diabetes Care, 2011. 34(6): p. e109. Evaluating investigational medications from pharmaceutical companies on diabetic subjects under FDA-approved study protocols. . For Phase 1-4 Clinical Trials . Tel: +1 714 734 7944 . Website: www.uciinc.net