Assesment 1 - Major report PDF

Preview

Summary

Major report
...

Description

Abstract:

In this narrative review, we summarize the current evidence on the accuracy and efficacy of the FreeStyle Libre for individuals with either type 1 or type 2 diabetes from three clinical studies. We found the Factory calibrated FreeStyle Libre sensor to act as an accurate and acceptable replacement to traditional blood glucose meters, often performing similar to more expensive continuous glucose monitoring sensors on the market. The use of the FreeStyle Libre interstitial glucose monitoring system was seen to co-operate and improve the quality of life for a broad range of diabetic individuals, observed by its high user satisfaction and low adverse events. However, there is the need for longer-term randomized studies with larger sample sizes and more adequate testing standards for subcutaneous glucose monitors before they replace accurate blood glucose meters. Introduction:

Diabetes is a disease which impacts on the body’s efficient regulation of blood glucose (BG) levels. Type 1 diabetes is caused by the autoimmune-mediated destruction of pancreatic β cell leading to insulin deficiency from a young age. Type 2 diabetes is the body’s resistance to insulin, resulting in the failure to naturally produce insulin. Both diseases are associated with a range of health complications, including blindness, heart and kidney disease, and amputations (Ndisang et al. 2017). Unfortunately, diabetes is the fastest growing chronic disease in Australia; increasing at a faster rate than diseases such as cancer and cardiovascular disease. In 2015, It was estimated that 1.2 million Australians had diabetes (AIHW 2016), of which most individuals had type 2 diabetes (85%), and only 15% had type 1 diabetes. Type 1 diabetes is a lifelong demanding condition requiring individuals to measure BG multiple times a day, aiding essential adjustments to insulin doses. Only 20% of individuals with type 2 diabetes use insulin therapy, whilst the majority can be managed through lifestyle modifications or other medications (AIHW 2016). When using traditional BG meters, many individuals may experience difficulties to their routine testing. These include pain and discomfort associated with the finger-stick blood samples. Another drawback associated with BG monitoring is that it only provides an intermittent measurement, not ‘real time’ blood glucose concentrations

(Fokkert et al. 2017). In recent decades, there has been a development of Continual Glucose Monitoring (CGM) systems. These devices display constant real time glucose values, with similar precision to traditional blood glucose meters However, the current CGM systems available on the market are expensive and require daily calibrating to allow for a reliable correlation between blood and interstitial glucose results (Bailey et al. 2015). The FreeStyle Libre Glucose Monitoring System is a Class 3 Medical Device, and was introduced to the Australian market by Abbott in 2014 (TGA 2017). It consists of a flash interstitial glucose monitoring system intended to replace painful BG meters and expensive CGMs. The system is comprised of a factory calibrated sensor along with a portable reader. The sensor is disposable after 14 days of use and does not need BG calibration during its use. The painless factory calibrated sensor helps prevent Irregular or incorrect calibration by patients. The device does not have constant alarms, which is favorable for individuals who complain of alarm fatigue (Bailey et al. 2015). It is important to assess the accuracy and efficiency of the FreeStyle Libre as a glucose monitor, in-comparison to existing CGMs and traditional BG tests.

Materials and Design

Diagram: (SEPARATE PAGES)

Fig. 1. Diagram of FreeStyle Libre system consisting of sensor and reader (FreeStyle Libre User’s Manual 2016)

Fig. 2. Glucose sensor filament measuring glucose in the interstitial fluid in the subcutaneous tissue layer (FreeStyle Libre User’s Manual 2016)

Biomaterials: The FreeStyle Libre system (Fig. 1.) has several features which differentiate it from other Continuous Glucose Monitoring (CGM) technology currently on the market:

-

The FreeStyle Libre flash sensor contains a wired enzyme sensor filament similar to those found within other CGMs on the market (Table 1.). These sensors work by inserting a glucose sensing filament into the subcutaneous tissue of the skin which measures the interstitial fluid glucose (Fig. 2).

-

Once the sensor is inserted onto the back of the upper arm, it requires an initial 1 hour equilibration to provide accurate glucose readings. The Flash sensor, unlike existing CGM sensor, requires no further calibrations with BG during the 14 days of wear since it is factory calibrated (Bailey et al. 2015). The FreeStyle Libre sensors are disposable after the 14 days of use.

-

Unlike other sensors on the market which require a separate transmitter, the FreeStyle Libre transmitter is inbuilt. The sensor allows the reader to receive current and historical interstitial glucose levels with a 15-minute frequency for up to 8 hours. The reader displays visual trends and alerts (e.g. Trend arrows), but does not have real-time audio alarms.

-

The FreeStyle Libre reader also contains a built-in BG meter. The BG readings are separate and do not impact the sensor readings. BG readings are needed when symptoms do not match the sensor’s readings and during times of rapidly changing BG levels (FreeStyle Libre User’s Manual 2016).

Device comparisons:

Device

FreeStyle libre (Abbot) system

Dexcom G4 Platinum system

Medtronic MiniMed 640G system

Table 1. Device comparison with competing models from different manufacturers adapted from (FreeStyle Libre User’s Manual 2016), (DEXCOM G4® PLATINUM USER GUIDE 2014) and (MiniMed™ 640G System User Guide 2015)

Flash or Continuous glucose monitor

Flash (tap reading via NFC)

Continuous (every 5 minutes via Bluetooth)

Continuous (every 5 minutes via Bluetooth)

14 days

7 days

6 days

Requires capillary blood calibration every 12 hours

Requires capillary blood calibration every 12 hours

Sensor life

Calibration

Factory Calibrated

Yes (compatible with Animas Vibe®)

Yes (compatible with Enlite™ sensor)

Insulin Pump Integration.

No

System Alert

Visual

Visual and Audio

Transmitter

Inbuilt (disposable)

Dexcom G4 Transmitter (nondisposable)

Sensor Weight

5g

11.3g (Transmitter and sensor)

6.25g

Sensor glucose reading range

40-500 mg/dL

40 - 400 mg/dL

40 - 400 mg/dL

Water resistant sensor/transmitter

1 metre for a maximum of 30 minutes

2.44 meters for a maximum of 24 hours.

Visual and Audio Guardian™ 2 Link transmitter (nondisposable)

2.4 meters for up to 30 minutes

Clinical Safety and Efficacy Bailey and colleagues in 2015 evaluated the use of the Factory Calibrated FreeStyle Libre. The study of 72 patients with type 1 or type 2 diabetes, was undertaken at four U.S. clinical sites. Each participant wore the sensor on the back of each upper arm for 14 days. The sensor interstitial glucose results were compared with 8 capillary BG results per day using the BG meter built into the reader, as well as with venous BG readings using a Yellow Springs Instrument (YSI) at three 8 hour clinic visits (32

samples per visit). The first accuracy metric used was the Mean Absolute Relative Difference (MARD), which shows how far away the glucose sensor reading is from a blood glucose readings on average. The second metric used was the Consensus Error Grid, which demonstrate the clinical relevance of a glucose monitoring result. The grid has zones A-E and results in zones A-B are clinically acceptable. The results of the FreeStyle Libre sensor were highly correlated to capillary BG data. The overall MARD was 11.4% for sensor results with capillary BG reference. During the clinic visits, the MARD of the sensors’ results against capillary BG and with venous BG reference was 12.1% and 12% respectively. When using the Consensus Error Grid, 99.7% of sensors glucose results fell within Zone A and Zone B, and accuracy was not affected by factors such as BMI, age, diabetes type, insulin treatment and insertion sites. The study focused on the use of the FreeStyle Libre for 14 days, and is therefore highly relevant to the ‘real-time’ use of the system. The study also provided a range of reliable and clinically relevant statistical data to show any variation between interstitial glucose sensor results against capillary and venous blood glucose samples. This clinical study was highly reliable because of minimal variables. These include the masking of sensor results for patients to prevent a skew in results. The maintenance of strip and sensor lots were used to minimize lot-to-lot variations when measuring capillary and interstitial glucose. Accurate models where used to compute the correlation of variables including age and gender, and the impact of lag time between measurements. The experiment used a sound sample size of n=72, and provided transparent information on what happened to individuals who dropped from the clinical trial. The study was also conducted in 2015 making it relevant to today’s standards. However, as the Abbott Corporation is the company which developed the device and funded the clinical trial the study may contain bias. Although there is potential for bias, the article contains a conflict of interest statement identifying S.A. as an employee of Abbott, and there are no comparisons between competing devices. The study also recognizes experimental limitations, such as the potential need to test the sensor beyond its life span of 14 days, and the limited venous BG data due to practical limitations when obtaining blood.

The second article by Fokkert and colleagues was published in 2017 in the peer reviewed journal BMJ Open Diabetes Research and Care. The research was undertaken at the Department of Internal Medicine of the Isala hospital. The objective was to determine the effectiveness of the Flash FreeStyle Libre glucose monitoring system against common and laboratory methods. The study involved 12 type 2 diabetes patients and 8 type 1 diabetes patients. The first aim was to compare the FreeStyle Libre sensors over 14 days, against 4 capillary blood glucose test per day using the StatStrip and FreeStyle Libre inbuilt BG meters. The second aim was to compare the FreeStyle Libre sensor against 3 laboratory capillary BG methods and a commonly used CGM system (Medtronic iPro2), during a standardised glucose loading test. The accuracy of the results were clinically acceptable with 97.8% of readings falling within zones A and B of the consensus error grid. The FreeStyle Libre also had a MARD of 12.1% compared to a built in BG meter and a MARD of 13.7% against the StatStrip in BG meter. However, the FreeStyle Libre displayed lower results than expected, compared to the StatStrip, during lower BG ranges. A slower rise in glucose level was observed for FreeStyle Libre against other comparative capillary BG and CGMs methods during the standardized glucose loading test.

Clinically valid statistical metrics were used In the study for measuring the variation between capillary and interstitial glucose values. The experiment intentionally utilized a broad range of capillary blood and interstitial glucose measuring methods during the standardised glucose loading, increasing the reliability by providing a large range of comparative data. This clinical study was performed in 2015, making it highly relevant to current research. The article was transparent when explaining patient dropouts and posed no financial biases as the clinical trial was externally funded and had no competing interests. However, this study poses issues with the methods used. The non-blinded nature of the study allowed patients to act on low glucose concentrations. This would have reduced the number of low readings in the lower glucose ranges. Another limitation of this study was the comparison of the CGM iPro2 against the FreeStyle Libre. The article lacks a clear explanation that the iPro2 is calibrated with BG measurements, whilst the FGM FreeStyle Libre is factory calibrated, thus being a biased comparison. Finally, the use of a very limited sample size may impact the decision on whether the study is reliable. However, the results

from this study are consistent with the previous study critiqued, suggesting reliability and accuracy in the method tested.

The study ‘Evaluation of subcutaneous glucose monitoring systems under routine environmental conditions in patients with type 1 diabetes’, was performed at the clinical research centre at the Medical University of Graz in 2017 with 12 people with type 1 diabetes. This clinical study was used to compare 3 commercially available flash or continuous subcutaneous glucose monitors (Abbott FreeStyle Libre, Dexcom G4 Platinum and Medtronic MiniMed 640G) (Table 1.), against capillary and venous BG as a standard. This was done for 12 hours and mimicked real-life conditions using standardized meal plans/portions and exercise tests. The FreeStyle Libre sensor was factory calibrated, whilst the other 2 sensors needed capillary blood calibration every 12 hours. The study was used to observe the effect of metabolic variables on sensor accuracy, induced by physical exercise and food intake on, glucagon and lactate to name a few. The sensors’ performance was analysed by ISO 15197:2013 criteria fulfilment, MARD and its clinical acceptance on the consensus error grid. In this study, it was found the FreeStyle Libre was superior in sensor performance. The sensor fulfilled ISO 15197:2013 criteria by 73.2%, whilst the Dexcom and Medtronic only fulfilled criteria by 56.1% and 52.0%, respectively. The overall MARD of the sensors’ results were 13.2% (FreeStyle Libre), 16.8% (G4 Platinum) and 21.4% (MiniMed 640G). The clinical sensor performance shown by the consensus error grid values in Zones A and B for the FreeStyle Libre, G4 Platinum and MiniMed 640G were found to be 100%, 99.3% and 98.6%. It was also found that variations in metabolites from exercise and eating did not influence the performance of the sensors. This clinical study has both strengths and limitations within its methodology. This recent study is the first to create a head to head comparison of the FreeStyle Libre against other subcutaneous glucose sensors on the market. It has also brought attention of the limited specific requirements of subcutaneous measuring standards. Since no accepted assessment standard exists, these devices are compared against ISO 15197:2013 criteria requirements for in vitro capillary blood glucose monitoring systems (ISO 2013). Similar to the previous studies, this used clinically valid statistical metrics for measuring the clinical performance for these 3 sensors. The

standardised real-life conditions utilized in this study attempted to control the variables as much as possible to provide a fair comparison. However, this could also be viewed as a disadvantage as the study was only evaluated under artificial conditions at the clinical centre and therefore might not reflect sensor performance during routine use in daily life. e.g. sensor accuracy reduced as result from pressure during sleep. Since the study was not randomized or blinded, there is the potential for bias. This study is not explicit in the devices used to analyse capillary and venous blood glucose to compare to the sensors accuracy and to calibrate the devices. The brief study period and very small sample size are major limitations since statistically significant results cannot be concluded. Although the study was externally funded, a few authors of the article have received speaker honoraria from medical companies including Abbott which could pose as bias in the study.

Conclusion: In conclusion, based on these three clinical studies, there is sufficient evidence to confidently approve the success of the FreeStyle Libre. These current clinical studies show the FreeStyle Libre as a device which produces acceptable accuracy when producing on‐demand glucose measurements. Although the device is intended as an adequate and accurate substitute of blood glucose tests for a broad range of diabetic individuals in everyday life, in our opinion the device is not a ‘complete’ replacement

to blood glucose tests. This is due to the limited current standards for subcutaneous sensors against blood glucose. Although all three-studies showed similar results for the accuracy of the FreeStyle Libre sensor, suggesting reliability, it Is recommended future studies incorporate larger sample sizes, as well as randomized and blinded methodologies to reduce potential experimental bias.

References Aberer, F., Hajnsek, M., Rumpler, M., Zenz, S., Baumann, P.M., Elsayed, H., Puffing, A., Treiber, G., Pieber, T.R., Sourij, H. & Mader, J.K. 2017, 'Evaluation of subcutaneous glucose monitoring systems under routine environmental conditions in patients with type 1 diabetes', Diabetes Obes Metab, vol. 19, no. 7, pp. 1051-5. Australia’s health 2016 - 3.7 Diabetes 2016, Australian Institute of Health and Welfare.

Bailey, T., Bode, B.W., Christiansen, M.P., Klaff, L.J. & Alva, S. 2015, 'The Performance and Usability of a Factory-Calibrated Flash Glucose Monitoring System', Diabetes Technol Ther, vol. 17, no. 11, pp. 787-94. DEXCOM G4® PLATINUM USER GUIDE 2014, Dexcom, . Fokkert, M.J., van Dijk, P.R., Edens, M.A., Abbes, S., de Jong, D., Slingerland, R.J. & Bilo, H.J. 2017, 'Performance of the FreeStyle Libre Flash glucose monitoring system in patients with type 1 and 2 diabetes mellitus', BMJ Open Diabetes Res Care, vol. 5, no. 1, pp. 1-8 FreeStyle Libre User’s Manual - Flash Glucose Monitoring System 2016, Abbott Diabetes Care, . ISO 15197:2013 2013, International Organization for Standardization, . Leelarathna, L. & Wilmot, E.G. 2018, 'Flash forward: a review of flash glucose monitoring', Diabetic Medicine, vol. 35, no. 4, pp. 472-82. MiniMed™ 640G System User Guide 2015, Medtronic, . Ndisang, J.F., Vannacci, A. & Rastogi, S. 2017, 'Insulin Resistance, Type 1 and Type 2 Diabetes, and Related Complications 2017', J Diabetes Res, vol. 2017, p. 1478294. Public Summary - FreeStyle Libre Flash Glucose Monitoring System Sensor 2017, Australian Department of Health, The Therapeutic Goods Administration....