Medical Policy

 

Subject: Automated Insulin Delivery Devices
Document #: DME.00040 Publish Date:    11/09/2017
Status: Revised Last Review Date:    11/02/2017

Description/Scope

This document addresses the use of automated insulin delivery devices (for example, the Medtronic MiniMed 530G, 630G, and 670G).  Such devices integrate a continuous interstitial glucose monitor (CIGM) with an insulin infusion pump to enable automated control of pump action based on real-time glucose monitor readings.  Such devices may also be referred to as “artificial pancreas devices.”

Note: For additional information regarding the use of external insulin pumps and continuous interstitial glucose monitors, please see:

Note: This document does not address the use of bionic pancreas or dual-hormone automated insulin delivery devices. These types of devices involve the administration of both insulin and glucagon.

Note: This document does not address supplies related to the use of automated insulin delivery devices.

Position Statement

Medically Necessary:

Use of an open-loop or hybrid closed-loop automated insulin delivery device with a low glucose suspend feature is considered medically necessary for individuals who meet the following criteria:

  1. Age 7 or older; and
  2. Type 1 diabetes mellitus; and
  3. HbA1c value of 5.8% to 10%.

Replacement of an open-loop or hybrid closed-loop automated insulin delivery device with a low glucose suspend feature is considered medically necessary when the medically necessary criteria above have previously been met and all of the criteria below have been met:

  1. The device is out of warranty; and
  2. The device is malfunctioning; and
  3. The device cannot be refurbished.

Not Medically Necessary:

Replacement of an open-loop or hybrid closed-loop automated insulin delivery device with a low glucose suspend feature is considered not medically necessary when the criteria above have not been met, including but not limited to for the sole purpose of receiving the most recent insulin pump technology (commonly referred to as an “upgrade”).

Investigational and Not Medically Necessary:

Use of an open-loop or hybrid closed-loop automated insulin delivery device, including those with a low glucose suspend feature, is considered investigational and not medically necessary for individuals who have not met the criteria above.

Use of non-hybrid closed-loop or non-FDA-approved hybrid closed-loop automated insulin delivery devices is considered investigational and not medically necessary under all circumstances.

Rationale

At this time, several automated insulin delivery systems have been approved by the U.S Food and Drug Administration (FDA).  The Medtronic MiniMed 530G and 630G are open-loop devices with a threshold suspend feature.  The MiniMed 670G system is a hybrid closed-loop system that received FDA approval in September 2016. 

Automated insulin delivery systems integrate an external insulin pump and CIGM device to potentially provide tighter glucose control than is possible with these two devices alone, or together but not integrated.  Open-loop devices require manual adjustment of insulin administration rates based on CIGM data as well as manual calculation and administration of pre-meal insulin bolus doses.  Most such devices still require self-monitoring of blood glucose concentrations as well.  Open-loop devices may include a low glucose suspend feature that suspends insulin delivery for a set period of time when the CIGM device detects that glucose concentrations have reached a pre-set lower threshold.  Some open-loop devices may go a step further and involve a “predictive” low glucose suspend feature.  This feature uses a predictive algorithm to determine when glucose concentrations are headed towards a pre-set lower threshold and then decrease or suspend insulin delivery before the threshold is reached.  Hybrid closed-loop devices eliminate the requirement of routine manual adjustment of pump administration rates, with the insulin pump and CIGM devices working together to predict and calculate insulin dose requirements.  However, these types of devices still require the manual calculation and administration of pre-meal insulin bolus doses, hence the “hybrid” moniker.  Finally, closed-loop systems are fully automated and require little intervention or involvement of the individual beyond routine system calibration.

Open-loop Threshold Suspend Devices

There are currently a small number of well-designed and conducted studies addressing the use of the threshold suspend-type device.  The first and largest of these studies was reported by Bergenstal and others in 2013.  This industry-sponsored trial involved 247 subjects who were randomly assigned to treatment with combined insulin pump-continuous interstitial glucose monitor (CIGM) system with or without a threshold suspend function (experimental group, n=121; controls n=126, respectively).  Enrolled subjects were between 16 and 70 years old, had type 1 diabetes with glycated hemoglobin (HbA1c) levels between 5.8% and 10.0%, had been using an insulin pump for at least 6 months and experienced at least two nocturnal hypoglycemic events (≤ 65 mg/dL) lasting more than 20 minutes during a 2 week run-in phase.  Subjects in the experimental group were required to use the suspend feature at a minimum between 10 PM and 8 AM daily for the duration of the 3 month long trial period. In this group, the threshold value was initially set at 70 mg/dL and could be adjusted to a value between 70 to 90 mg/dL.

The authors selected area under the curve (AUC) for nocturnal hypoglycemia events as the primary efficacy outcome measure.  They calculated this by multiplying the magnitude (in milligrams per deciliter) and duration (in minutes) of each qualified hypoglycemic event.  The primary safety outcome was the change in HbA1c levels at the end of the trial period.  The mean AUC for nocturnal hypoglycemic events was 980 in the experimental group and 1568 in the control group, indicating a 37% reduction in nocturnal hypoglycemia events in the experimental group vs. controls (p<0.0001).  Combined daytime and nighttime hypoglycemic events was a secondary outcome measure, and the results likewise indicated a significant decrease in the intervention group (p<0.001).  In terms of overall event data, the intervention group experienced a mean of 3.3 hypoglycemic episodes per subject-week vs. 4.7 per subject-week in the control group (p<0.001).  The mean number of times the suspend feature was activated in the experimental group per subject was 2.08 per 24-hour period and 0.77 each nocturnal measuring period.  The mean sensor glucose value at the beginning of nocturnal events was the same for each group, 62.6 mg/dL.  However, after 4 hours, the mean sensor glucose value was 162.3 mg/dL in the experimental group and 140.0 mg/dL in the control group.  The authors reported that there was no statistically significant difference between groups with regard to change in HbA1c levels.  No severe hypoglycemic events were reported in the experimental group vs. four in the control group.  There were no deaths or serious device-related adverse events.  It should be noted that this study involved the use of the Medtronic Paradigm Veo System which was commercialized in Europe in 2010 after receiving a CE mark.

The second randomized controlled trial (RCT), published by Ly in 2013, also used the Medtronic Paradigm Veo System.  This study involved 95 subjects randomized to 6 months of treatment with either Veo system (n=46) or to insulin pump treatment alone (n=49).  Subjects were aged 4 to 50 years old with type 1 diabetes, had used an insulin pump for at least 6 months, had an HbA1c level of 8.5% or less, and had impaired awareness of hypoglycemia.  Impaired awareness of hypoglycemia was defined as a score of at least 4 on the modified Clarke questionnaire.  The automated insulin suspension threshold was 60 mg/dL.  The primary study outcome was combined incidence of severe hypoglycemic events (defined as hypoglycemic seizure or coma) and moderate hypoglycemic events (defined as an event requiring assistance from another person).  The authors noted that the baseline rate of severe and moderate hypoglycemia was significantly higher in the experimental group (129.6 vs 20.7 events per 100 subject-months).  After 6 months, the frequency of moderate to severe hypoglycemic events per 100 subject-months was 34.2 in the control group vs. 9.6 in the experimental group.  The authors reported the incidence rate ratio was 3.6 (p<0.001).  No episodes of ketoacidosis or hyperglycemia with ketosis were reported in either group.  The authors conducted a sensitivity analysis in subjects younger than 12 years (n=15 per group).  They noted that the high baseline hypoglycemia rates could be explained in part by 2 outliers, and when those subjects were excluded from the analysis, the primary outcome was no longer statistically significant.  The incidence rate ratio for moderate and severe events excluding the 2 children was 1.7 (p=0.08).  Mean HbA1c level, a secondary outcome, did not differ between groups at baseline or at 6 months.  Change in HbA1c levels during the treatment period was -0.06% in the control group and -0.1% in the experimental group (p=not significant).  

A retrospective analysis of the threshold suspend feature was reported by Agrawal (2015).  This cohort study involved 20,973 subjects using the Medtronic Paradigm Veo System.  Subjects were able to adjust the threshold suspend feature at their discretion and uploaded their pump and sensor data during a 40-week period.  The authors compared data from 758,382 subject-days when the suspend feature was activated to the 166,791 subject-days when it was not.  Overall 70% of subjects (n=14,673) had the suspend feature activated 100% of the time.  Conversely, 11% (n=2249) did not use that feature at all.  The remaining subjects used the feature some unspecified portion of the time.  The mean sensor threshold for the suspension feature was a glucose level of 62.8 mg/dL.  According to the authors, there was a mean of 0.82 suspend events per subject-day on days when the feature was active.  On days when the threshold suspend feature was on, sensor glucose values were reported to be 50 mg/dL or less 0.64% of the time vs. 2.1% of sensor glucose values 50 mg/dL or less on days when the feature was off.  The reduction in hypoglycemia was greatest at night.  They concluded that the use of an automated insulin delivery device with threshold suspend appeared to be associated with fewer and shorter hypoglycemic episodes.  However, data describing the length and severity of hypoglycemic episodes was not fully discussed in this article.

In 2017, Gómez (2017) published the results of a cohort study evaluating the safety and efficacy of sensor-augmented pumps with low-threshold suspend feature in 11 subjects with hypoglycemia unawareness.  All subjects used a combination system involving the Medtronic Paradigm 722 or Paradigm Veo pump connected to the MiniMed CGM device.  The mean follow-up time was 47 ± 22.7 months; the authors reported that the total daily dose of insulin improved from 0.89 ± 0.39 U/kg to 0.67 ± 0.25 U/kg at the last visit (p<0.001).  The mean number of basal doses increased from 4.7 ± 1.7 to 5.1 ± 1.4 at the last visit, and the number of boluses decreased from 5.1 ± 2.1 to 4.7 ± 1.5.  Sensor use over the course of the study did not change significantly (p=0.105).  The mean HbA1c concentrations improved from 8.1 ± 1.9% at baseline to 7.1 ± 0.8% at last follow-up (p<0.001).  At baseline, only 17% of subjects had achieved HbA1c ≤ 7.0%, whereas at last follow-up 43% of subjects had achieved HbA1c ≤ 7.0% (p<0.001).  Furthermore, at baseline 80% of subjects had had at least one episode of hypoglycemic awareness compared to 10.8% at last follow-up (p<0.001).  Similarly, episodes of severe hypoglycemia decreased from 66.6% to 2.7% (p<0.001).  This study demonstrated significant benefits to sensor-augmented insulin pump therapy with a low glucose suspend threshold.

The studies described above demonstrate a significant benefit to individuals who utilized threshold suspend-type devices, with significant reduction in severe hypoglycemic events. 

Hybrid Closed-Loop Devices

At this time, the majority of the available evidence for automated insulin delivery devices other than the low glucose threshold suspend-type devices is very limited.  The best available data in this category addresses the clinical utility of a hybrid closed-loop system, specifically the MiniMed 670G system.  This system consists of a CIGM device, an insulin pump device, and a blood glucose meter used to calibrate the CIGM device.  The system is able to increase, decrease or stop insulin delivery automatically beyond pre-set infusion rates in response glucose concentration measurements by the CIGM.  The device has two modes, Manual and Automatic.  In Manual mode, the device operates in a similar fashion to a low glucose suspend threshold device, stopping insulin delivery in response to low glucose measurements by the CIGM.  In Automatic mode, the device can automatically adjust basal insulin infusion rates to increase, decrease, or suspend delivery based on CIGM data.  In either mode, the user must manually deliver insulin during meals.  This combination of an automatic basal insulin delivery mode combined with manual bolus insulin delivery prior to meals is referred to as a “hybrid closed-loop” system.  The critical difference between threshold suspend-type devices and the hybrid closed-loop system is the ability to automatically vary basal insulin infusion rates based on CIGM data.  Such automated closed-loop control of insulin administration is a new tool in the treatment of diabetes.

A small observational case series (de Bock, 2016) involved 8 subjects with type 1 diabetes and was designed to evaluate a hybrid closed-loop algorithm.  During the study, the investigators challenged the hybrid closed-loop system (MiniMed 670G) with hypoglycemic stimuli including exercise and an over-calibrated sensor set to read glucose concentrations as higher than actually present.  The authors reported no overnight or exercise-induced hypoglycemia during use of the device.  They noted that all recorded daytime hypoglycemia events were attributable to bolused post-prandial insulin in participants with aggressive carbohydrate factors.  They concluded that algorithm refinement was needed in preparation for long-term outpatient trials.

Bergenstal et al. (2016) published the results of a pivotal safety study of the MiniMed 670G system in a research letter in the Journal of the American Medical Association.  The study involved 123 subjects aged 14-75 years old who had type 1 diabetes mellitus for at least 2 years, HbA1c less than 10, and insulin pump therapy for a minimum of 6 months.  All subjects wore the 670G system for approximately 3.5 months.  The study involved three phases, including a 2-week run-in period, a 3-month at-home use period, and a 5-day/6-night hotel study.  The run-in period involved familiarization of the participants to the device.  The home-use period involved a 6-day period where the device was used in the non-auto mode, to allow for collection of insulin use and glucose sensor levels.  During the home study phase, subjects were required to have a companion with them during the night to respond to sensor alarms as needed.  Following that period, the participants were instructed to use the device in the closed-loop auto mode for the duration of the home phase.  During this phase, the high sensor glucose alert was set at 300 mg/dL and the low sensor glucose alert was set at 70 mg/dL.  The target glucose was 120 mg/dL, although a temporary target of 150 mg/dL could be used in certain scenarios (for example, exercise).  The hotel phase of the study occurred during the 3 month home study period, with at least 20 subjects participating in this phase each month.  The purpose of this portion of the study was to stress the subjects with sustained daily exercise and unrestricted eating to monitor the device’s response to significant physiological variations.  The authors reported that no episodes of severe hypoglycemia or ketoacidosis were noted during the study period.  There were 20 device-related adverse events reported during the study period, including skin irritation or rash (n=2), hyperglycemia (n=6), and severe hyperglycemia (defined as greater than 300 mg/dL, n=12).  All events were resolved at home.  The closed-loop auto function was used for a median of 87.2% of the study period.  HbA1c levels improved from 7.4% at baseline to 6.9% at the completion of the study period.  The daily dose of insulin changed from 47.5 U/d to 50.9 U/d, and mean weight changed from 76.9 kg to 77.6 kg.  The percentage of sensor glucose values within the target range changed from 66.7% at baseline to 72.2% at study end.  No statistical analysis was provided on these results.  The authors reported that their study demonstrated that hybrid closed-loop automated insulin delivery was associated with few serious or device-related adverse events in individuals with type 1 diabetes.  They noted, however, that their study had several limitations, including a lack of a control group, restriction to relatively healthy and well-controlled subjects, and a relatively short follow-up.  The authors caution that this study’s design was descriptive and its purpose was limited to the evaluation of the safe use of the 670G AutoMode function.  This study (IDE G140167) was not designed to determine the effectiveness of the device compared to conventional methods such as manual daily insulin injections or non-automated insulin pump therapy. 

Garg and colleagues (2017) published the results of an open-label safety study of the MiniMed 670G system involving 124 subjects (30 adolescents aged 14-21 years old and 94 adults).  All subjects underwent a 2 week in-home run-in phase using the 670G in open-loop mode followed by a 3 month hybrid closed-loop phase.  During the hybrid closed-loop phase, all subjects underwent a 6 day/5 night supervised hotel stay that included a 24-hour blood sampling period to compare glucose sensor measurements to lab-based venous blood glucose measurements.  The authors reported that sensor glucose readings during the hybrid closed-loop phase indicated that use of the 670G appeared to mitigate hyper- and hypoglycemia events in both the adolescent and adult groups.  The mean in-target glucose sensor reading in the adolescent group increased from 60.4% to 67.2% between the run-in to the hybrid closed-loop phase (p<0.001). For the adult group, the mean in-target glucose sensor reading went from 68.8% to 73.8% (p<0.001).  Similarly, time with glucose sensor readings of > 180 mg/dL decreased from 35.3% to 30.0% in the adolescent group (p<0.001) and 24.9% to 22.8% in the adult group (p<0.01045).  The mean time with sensor glucose readings < 70 mg/dL decreased from 4.3% to 2.8% in the adolescent group (p<0.000928) and 6.4% to 3.4% (p<0.001) in the adult group.  HbA1c concentrations decreased from a mean of 7.7% at baseline to 7.1% (p<0.001) at the end of the 3-month hybrid closed-loop phase in the adolescent group and from 7.3% to 6.8% (p<0.001) in the adult group during the same time frame.  The percent nighttime sensor glucose readings > 180 mg/dL decreased from 30.3% to 25.6% (p<0.001) in the adolescent group and 25.8% to 20.4% (p<0.001) in the adult group.  Similarly, mean nighttime sensor glucose readings < 50 mg/dL decreased from 1.3% to 0.6% in the adolescent group (p<0.001) and 1.1 to 0.7% (p<0.001) in the adult group.  These results demonstrated that within the study population, the hybrid closed-loop system was both safe and provided significantly better blood glucose control over treatment with an open-loop device. 

The FDA’s summary of safety and effectiveness data (SSED) for the MiniMed 670G system includes a description of the pivotal study described above (G140167), as well as a smaller Guardian CIGM sensor performance study (G140053).  The latter study was intended to determine the accuracy and precision of the Guardian sensor CIGM component of the 670G device in 93 subjects with Type I or Type II Diabetes Mellitus between the ages of 14-75 years.  Of this subject pool, 82 completed the study.  This prospective, single-sample correlational study did not involve a control group.  All subjects wore the Guardian sensor for a 7 day training period followed by a 7 day study period.  Subjects were randomized to one of two groups that determined when they participated in the in-clinic frequent sample testing; a day cohort (hours 1-12) and an evening cohort (hours 12-24).  There were five adverse events reported during the study, all which resolved without residual sequelae, including gastroenteritis, worsening of benign prostatic hypertrophy, rash at the IV site, upper respiratory symptoms, and a skin blister from skin tac used under tape.  No data were presented regarding the impact of the use of the Guardian sensor on diabetes-related health outcomes.

Nimri (2016), published the results of a small single-blind randomized controlled crossover trial involving 75 subjects with type 1 diabetes (25 adults and 50 children and adolescents).  Subjects were assigned to a 4-night monitoring period with either the MD-Logic Artificial Pancreas hybrid closed-loop device or control therapy with a sensor-augmented pump.  The MD-Logic System is composed of a MiniMed® Veo™ combined insulin pump and CIGM device, Enlite® glucose sensors, CONTOUR® LINK blood glucose meter, and a PC-based control algorithm.  Following a training period, subjects underwent a 4-day period of nocturnal testing with their assigned device.  After a 10-day washout period, the subjects underwent a second 4-day nocturnal testing period with the alternate device.  The authors reported that the intent-to-treat analysis demonstrated that percentage of time spent with sensor glucose < 70 mg/dL was significantly lower in the hybrid closed-loop group vs. the sensor-augmented group (2.07% vs. 2.6%, p=0.004).  Likewise, the percentage of time spent within normal range (90-140 mg/dL) was significantly greater in the hybrid closed-loop group vs. controls (75% vs. 50%, p=0.008).  The per-protocol analysis showed that the percentage of time spent with sensor glucose < 70 mg/dL in the hybrid closed-loop group was approximately half of controls (0.67% vs. 1.43%, p=0.005).  The authors concluded that this study demonstrated the safety and efficacy of the MD-Logic system for overnight use in children and adults.  However, additional investigation with larger populations is warranted to further understand the benefits of this system.

Several additional small studies involving other hybrid closed-loop devices have been published (Abraham, 2016; Anderson, 2016; Bally, 2017; Brown, 2015; DeBoer, 2017; Del Favero, 2015; Kovatchev 2014, 2017; Leelaranthna, 2014; Ly, 2014; Nimri, 2014a, 2014b; Pinsker, 2016; Sharifi, 2016; Stewart, 2016; Tauschmann, 2016a, 2016b, 2016c).  As with the studies described above, these studies also demonstrate improved control of glucose concentrations with fewer hypoglycemic events with a hybrid closed-loop delivery system.

At this time, the data demonstrating the incremental benefit of automated hybrid closed-loop control of insulin administration is limited.  However, expert clinical opinion supports the use of these devices in light of the potential significant benefits available to the most at-risk individuals with type 1 diabetes. 

Closed-Loop Devices

At this time, there are no fully closed-loop devices available on the market, although several are under investigation.  Forlenza (2016) published the results of a small RCT involving 14 subjects randomized to treatment with either closed-loop treatment with the Medtronic ePID (external physiological insulin delivery) 2.0 controller vs. multiple daily injection therapy with blinded CIGM (n=7 in each group) for a 72-hour period.  The results indicated that mean serum glucose values were significantly lower in the closed-loop group vs. the controls (111 mg/dL vs. 130 mg/dL, p=0.003).  This was achieved without increased risk of hypoglycemia, as demonstrated by the percentage of time < 70 mg/dL being lower in the closed-loop group vs. controls (1.9% vs. 4.8%, p=0.46).  While the authors concluded that their results suggest that closed-loop therapy is superior to conventional therapy in maintaining euglycemia without increased hypoglycemia, additional investigation is warranted in larger studies.

Another small RCT published by Thabit (2017) involved 40 adult subjects with type 2 diabetes assigned to a 72-hour treatment period with either closed-loop treatment with the Florence D2W-T2 automated system (using a model-predictive control algorithm to direct subcutaneous delivery of rapid-acting insulin analogue without meal-time insulin boluses) or standard of care with subcutaneous insulin therapy.  The Florence D2W-T2 is composed of a tablet computer-based control algorithm linked to an Abbott Freestyle Navigator II CIGM and a Sooil DANA R Diabecare insulin pump.  In this study, the proportion of time spent in target sensor glucose range was significantly higher in the closed-loop group vs. the control group (59.8% vs. 38.1%, p=0.004).  The proportion of time spent with glucose concentrations > 10.0 mmol/L was significantly lower in the closed-loop group vs. controls (30.1 vs. 49.1, p=0.011).  No significant differences between groups was reported for mean sensor glucose concentrations or time spent with glucose concentrations lower than the target range.  Glucose variability was significantly reduced compared to controls (coefficient of variation [CV]=27.9 vs. 33.4, p=0.042), and nocturnal time spent within target range was significantly greater in the closed-loop group as well (68.9% vs. 48.8%, p=0.007).  No episodes of severe hypo- or hyperglycemia with ketonemia occurred in either group.  As with the previously described study, these results are promising, but additional investigation involving larger studies is needed.

A meta-analysis published by Weisman and others (2017) evaluated the existing data addressing the efficacy of closed-loop “artificial pancreas” devices compared to care with a conventional insulin pump.  A total of 24 studies were included, with 19 involving single hormone devices and 5 involving dual hormone systems.  Two studies involved both types of devices.  A total of 585 subjects were included in the analysis.  The authors reported a high degree of heterogeneity across studies, with significant range of mean differences reported (-6.3% to 26.68%).  Overall, the artificial “pancreas system” groups demonstrated a greater difference for time in target in overnight studies vs. standard care (p<0.0001).  Looking at time spent in hypoglycemia, 21 studies involving 463 subjects were included in the analysis.  The results showed that time in hypoglycemia was 2.45% in the artificial pancreas group vs. 4.88% in the standard care group (p<0.0001).  It was noted that this equates to a 50% relative risk reduction.  An analysis looking at change in insulin dose included 18 studies involving 389 subjects.  Overall, no differences between device groups were noted, but a sub-analysis looking at the closed-loop group only indicated that children using closed-loop devices had a significantly higher insulin dose vs. adults using this type of device. (p<0.0001).  Episodes of severe hypoglycemia were reported in 22 studies, with no significant differences reported between groups.  No sub-analysis was provided for single hormone-only devices, and it was not clear if data involving hybrid closed-loop devices was included in the analysis.

Other Information

The American Diabetes Association recommends the use of automated insulin delivery devices in the 2016 Standards of Medical Care.  Section 7 (Approaches to Glycemic Treatment) of that document states, “For patients with frequent nocturnal hypoglycemia, recurrent severe hypoglycemia, and/or hypoglycemia unawareness, a sensor-augmented low glucose threshold suspend pump may be considered.”  In support of this recommendation, the Association cites the ASPIRE trial results published by Bergenstal and colleagues in 2013 summarized above.

In addition to the threshold suspend-type devices, there are also “control-to-range” and “control-to-target” devices which operate on different principles (below).  At this time, there are no “control-to-range” automated insulin delivery devices which have been cleared to market in the US.  Additionally, the available evidence addressing their use involves mostly small case series which are insufficient to properly evaluate their safety and efficacy (Nimri, 2014a, 2014b).

There are other automated insulin delivery devices under development which attempt to more fully mimic the action of the pancreas.  One such device type is referred to as a bionic pancreas or dual-hormone artificial pancreas.  These systems involve the administration of both insulin and glucagon to maintain blood glucose within a targeted range.  These automated insulin delivery devices are not addressed in this document.

Background/Overview

According to the American Diabetes Association, diabetes is one of the most common chronic diseases in the United States, with approximately 30 million Americans with diagnosed disease.  Another 8 million are believed to have undiagnosed disease.  Diabetes mellitus, the fourth leading cause of death in the U.S., is a chronic condition, marked by impaired metabolism of carbohydrate, protein and fat, affecting nearly 21 million Americans.  The underlying problem in diabetes is in the production or utilization of insulin, the hormone secreted by the pancreas that controls the level of blood sugar by regulating the transfer of glucose from the blood into the cells.  Diabetes mellitus, if poorly controlled, can cause cardiovascular disease, retinal damage that could lead to blindness, damage to the peripheral nerves, and injury to the kidneys.  Management of diabetes mellitus involves normalization of blood sugar without potentially dangerous hypoglycemia, or low blood sugar.  Type 1 diabetes can occur at any age, but is most commonly diagnosed from infancy to late 30s.  In type 1 the pancreas produces little to no insulin, and the body’s immune system destroys the insulin-producing cells in the pancreas.  Type 2 diabetes typically develops after age 40, but has recently begun to appear with more frequency in children.  If a person is diagnosed with type 2 diabetes, the pancreas still produces insulin, but the body does not produce enough or is not able to use it effectively.

For some individuals with diabetes, the use of multiple daily insulin injection therapy is insufficient to provide adequate control of blood sugar levels.  In such cases, an external insulin pump may be recommended.  These devices are worn externally and are attached to a temporary subcutaneous insulin catheter placed into the skin of the abdomen.  The pump involves the use of a computer-controlled mechanism that can be set to administer the insulin at a set (basal) rate or provide injections (bolus) as needed.  The pump typically has a syringe reservoir that has a 2- to 3-day insulin capacity.  The purpose of the insulin pump is to provide an accurate, continuous, controlled delivery of insulin which can be regulated by the user to achieve intensive glucose control.

Whether an individual with diabetes uses injection therapy or an insulin pump, the individual needs to check blood glucose concentrations multiple times a day to make sure they are staying within normal blood glucose range.  As with injection therapy, sometimes self-monitoring blood glucose management is also insufficient.  In such circumstances, the use of a CIGM may be warranted.  These devices measure glucose concentrations in the fluid in between the body’s cells, also known as interstitial fluid.  They are designed to provide real-time glucose measurements, which have been found to accurately reflect blood glucose levels. 

The combined use of an insulin pump and CIGM, either with separate devices or using a device that incorporates both functions, has become more prevalent in clinical practice for individuals with difficult to control diabetes.  An evolution of this combination therapy has led to the development of “closed-loop” or “automated insulin delivery devices” systems.  Such devices combine the use of both an insulin pump and a CIGM device, but are designed to work automatically without the involvement of the individual to monitor glucose concentrations and the administration of insulin.   

The FDA has developed a guide to the three different types of automated insulin delivery devices, including:

Many different types of automated insulin delivery devices are currently available or under development.  Descriptions of each are provided in the beginning of the Rationale section above. 

Coding

The following codes for treatments and procedures applicable to this document are included below for informational purposes. Inclusion or exclusion of a procedure, diagnosis or device code(s) does not constitute or imply member coverage or provider reimbursement policy. Please refer to the member's contract benefits in effect at the time of service to determine coverage or non-coverage of these services as it applies to an individual member.

When services may be Medically Necessary when criteria are met for an open-loop or hybrid closed-loop automated insulin delivery device:

HCPCS

 

 

E0784

External ambulatory infusion pump, insulin [when specified as an open-loop or hybrid closed-loop automated insulin delivery device]

 

S1034

Artificial pancreas device system (e.g., low glucose suspend [LGS] feature) including continuous glucose monitor, blood glucose device, insulin pump and computer algorithm that communicates with all of the devices [when specified as an open-loop or hybrid closed-loop automated insulin delivery device]

 

 

 

 

ICD-10 Diagnosis

 

E10.10-E10.9

Type 1 diabetes mellitus

Z46.81

Encounter for fitting and adjustment of insulin pump

When services are Not Medically Necessary:
For replacement of an open-loop or hybrid closed-loop automated insulin delivery device when the replacement criteria are not met.

When services are Investigational and Not Medically Necessary:
When the code is specified as an open-loop or hybrid closed-loop automated insulin delivery device and criteria are not met or for all other diagnoses not listed.

When services are also Investigational and Not Medically Necessary:

HCPCS

 

 

E0784

External ambulatory infusion pump, insulin [when specified as other type of automated insulin delivery device (for example, closed-loop)]

 

S1034

Artificial pancreas device system (e.g., low glucose suspend [LGS] feature) including continuous glucose monitor, blood glucose device, insulin pump and computer algorithm that communicates with all of the devices [when specified as other type of automated insulin delivery device (for example, closed-loop)]

 

 

 

 

ICD-10 Diagnosis

 

 

All diagnoses

References

Peer Reviewed Publications:

  1. Abraham MB, de Bock M, Paramalingam N, et al. Prevention of insulin-induced hypoglycemia in Type 1 diabetes with predictive low glucose management system. Diabetes Technol Ther. 2016; 18(7):436-443.
  2. Agrawal P, Zhong A, Welsh JB, et al. Retrospective analysis of the real-world use of the threshold suspend feature of sensor-augmented insulin pumps. Diabetes Technol Ther. 2015; 17(5):316-319.
  3. Anderson SM, Raghinaru D, Pinsker JE, et al.; Control to Range Study Group. Multinational home use of closed-loop control is safe and effective. Diabetes Care. 2016; 39(7):1143-1150.
  4. Bally L, Thabit H, Kojzar H, et al. Day-and-night glycaemic control with closed-loop insulin delivery versus conventional insulin pump therapy in free-living adults with well controlled type 1 diabetes: an open-label, randomised, crossover study. Lancet Diabetes Endocrinol. 2017; 5(4):261-270.
  5. Bergenstal RM, Garg S, Weinzimer SA, et al. Safety of a hybrid closed-loop insulin delivery system in patients with Type 1 diabetes. JAMA. 2016; 316(13):1407-1408.
  6. Bergenstal RM, Klonoff DC, Garg SK, et al.; ASPIRE In-Home Study Group. Threshold-based insulin-pump interruption for reduction of hypoglycemia. N Engl J Med. 2013; 369(3):224-232.
  7. Brown SA, Kovatchev BP, Breton MD, et al. Multinight "bedside" closed-loop control for patients with type 1 diabetes. Diabetes Technol Ther. 2015; 17(3):203-209.
  8. Buckingham BA, Bailey TS, Christiansen M, et al. Evaluation of a predictive low-glucose management system in-clinic. Diabetes Technol Ther. 2017; 19(5):288-292.
  9. de Bock M, Dart J, Roy A, et al. Exploration of the performance of a hybrid closed loop insulin delivery algorithm that includes insulin delivery limits designed to protect against hypoglycemia. J Diabetes Sci Technol. 2017; 11(1):68-73.
  10. DeBoer MD, Breton MD, Wakeman C, et al. Performance of an artificial pancreas system for young children with type 1 diabetes. Diabetes Technol Ther. 2017; 19(5):293-298.
  11. Del Favero S, Place J, Kropff J, et al.; AP@home Consortium. Multicenter outpatient dinner/overnight reduction of hypoglycemia and increased time of glucose in target with a wearable artificial pancreas using modular model predictive control in adults with type 1 diabetes. Diabetes Obes Metab. 2015; 17(5):468-476.
  12. Forlenza GP, Nathan BM, Moran AM, et al. Successful application of closed-loop artificial pancreas therapy after islet autotransplantation. Am J Transplant. 2016; 16(2):527-534.
  13. Garg SK, Weinzimer SA, Tamborlane WV, et al. Glucose outcomes with the in-home use of a hybrid closed-loop insulin delivery system in adolescents and adults with type 1 diabetes. Diabetes Technol Ther. 2017; 19(3):155-163.
  14. Gómez AM, Marín Carrillo LF, Muñoz Velandia OM, et al. Long-term efficacy and safety of sensor augmented insulin pump therapy with low-glucose suspend feature in patients with Type 1 diabetes. Diabetes Technol Ther. 2017; 19(2):109-114.
  15. Kovatchev B, Cheng P, Anderson SM, et al. Feasibility of long-term closed-loop control: a multicenter 6-month trial of 24/7 automated insulin delivery. Diabetes Technol Ther. 2017; 19(1):18-24.
  16. Kovatchev BP, Renard E, Cobelli C, et al. Safety of outpatient closed-loop control: first randomized crossover trials of a wearable artificial pancreas. Diabetes Care. 2014; 37(7):1789-1796.
  17. Kropff J, Del Favero S, Place J, et al.; AP@home Consortium. 2 month evening and night closed-loop glucose control in patients with type 1 diabetes under free-living conditions: a randomised crossover trial. Lancet Diabetes Endocrinol. 2015; 3(12):939-947.
  18. Leelarathna L, Dellweg S, Mader JK, et al.; AP@home Consortium. Day and night home closed-loop insulin delivery in adults with type 1 diabetes: three-center randomized crossover study. Diabetes Care. 2014; 37(7):1931-1937.
  19. Ly TT, Breton MD, Keith-Hynes P, et al. Overnight glucose control with an automated, unified safety system in children and adolescents with type 1 diabetes at diabetes camp. Diabetes Care. 2014; 37(8):2310-2316.
  20. Ly TT, Nicholas JA, Retterath A, et al. Effect of sensor-augmented insulin pump therapy and automated insulin suspension vs standard insulin pump therapy on hypoglycemia in patients with type 1 diabetes: a randomized clinical trial. JAMA. 2013; 310(12):1240-1247.
  21. Nimri R, Bratina N, Kordonouri O, et al. MD-Logic overnight type 1 diabetes control in home settings: a multicentre, multinational, single blind randomized trial. Diabetes Obes Metab. 2017; 19(4):553-561.
  22. Nimri R, Muller I, Atlas E, et al. Night glucose control with MD-Logic artificial pancreas in home setting: a single blind, randomized crossover trial-interim analysis. Pediatr Diabetes. 2014a; 15(2):91-99.
  23. Nimri R, Muller I, Atlas E, et al. MD-Logic overnight control for 6 weeks of home use in patients with type 1 diabetes: randomized crossover trial. Diabetes Care. 2014b; 37(11):3025-3032.
  24. Renard E, Farret A, Kropff J, et al.; AP@home Consortium. Day-and-night closed-loop glucose control in patients with type 1 diabetes under free-living conditions: results of a single-arm 1-month experience compared with a previously reported feasibility study of evening and night at home. Diabetes Care. 2016; 39(7):1151-1160.
  25. Sharifi A, De Bock MI, Jayawardene D, et al. Glycemia, treatment satisfaction, cognition, and sleep quality in adults and adolescents with Type 1 diabetes when using a closed-loop system overnight versus sensor-augmented pump with low-glucose suspend function: a randomized crossover study. Diabetes Technol Ther. 2016; 18(12):772-783.
  26. Stewart ZA, Wilinska ME, Hartnell S, et al. Closed-loop insulin delivery during pregnancy in women with Type 1 diabetes. N Engl J Med. 2016; 375(7):644-654.
  27. Tauschmann M, Allen JM, Wilinska ME, et al. Day-and-night hybrid closed-loop insulin delivery in adolescents with Type 1 diabetes: a free-living, randomized clinical trial. Diabetes Care. 2016a; 39(7):1168-1174.
  28. Tauschmann M, Allen JM, Wilinska ME, et al. Home use of day-and-night hybrid closed-loop insulin delivery in suboptimally controlled adolescents with Type 1 diabetes: a 3-week, free-living, randomized crossover trial. Diabetes Care. 2016b; 39(11):2019-2025.
  29. Tauschmann M, Allen JM, Wilinska ME, et al. Sensor life and overnight closed loop: a randomized clinical trial. J Diabetes Sci Technol. 2017; 11(3):513-521.
  30. Thabit H, Hartnell S, Allen JM, et al. Closed-loop insulin delivery in inpatients with type 2 diabetes: a randomised, parallel-group trial. Lancet Diabetes Endocrinol. 2017; 5(2):117-124.
  31. Weisman A, Bai JW, Cardinez M, et al. Effect of artificial pancreas systems on glycaemic control in patients with type 1 diabetes: a systematic review and meta-analysis of outpatient randomised controlled trials. Lancet Diabetes Endocrinol. 2017; 5(7):501-512.
  32. Zisser H, Renard E, Kovatchev B, et al.; Control to Range Study Group. Multicenter closed-loop insulin delivery study points to challenges for keeping blood glucose in a safe range by a control algorithm in adults and adolescents with type 1 diabetes from various sites. Diabetes Technol Ther. 2014; 16(10):613-622.

Government Agency, Medical Society, and Other Authoritative Publications:

  1. American Diabetes Association. Standards of Medical Care in Diabetes- 2016. Diabetes Care. 2016; 39(Suppl 2):1-112. Available at: http://care.diabetesjournals.org/content/39/Supplement_1. Accessed on September 12, 2017.
  2. Blue Cross Blue Shield Association. Artificial Pancreas Device Systems. TEC Assessment. 2013; 28(14).
  3. U.S Food and Drug Administration. Types of Artificial Pancreas Device Systems. Updated December 10, 2014. Available at: http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/HomeHealthandConsumer/ConsumerProducts/ArtificialPancreas/ucm259555.htm. Accessed on September 12, 2017.
  4. U.S Food and Drug Administration. Summary of safety and effectiveness data for the MiniMed 670G system. September 28, 2016. Available at: http://www.accessdata.fda.gov/cdrh_docs/pdf16/P160017b.pdf. Accessed on September 12, 2017.
Websites for Additional Information
  1. American Diabetes Association. Type 1 diabetes. Available at: http://www.diabetes.org/diabetes-basics/type-1/. Accessed on September 12, 2017.
  2. American Diabetes Association. Type 2 diabetes. Available at: http://www.diabetes.org/diabetes-basics/type-2/?loc=db-slabnav/. Accessed on September 12, 2017.
Index

MiniMed 530G
MiniMed 630G
MiniMed 670G
Enlite Sensor

The use of specific product names is illustrative only. It is not intended to be a recommendation of one product over another, and is not intended to represent a complete listing of all products available.

Document History

Status

Date

Action

Revised

11/02/2017

Medical Policy & Technology Assessment Committee (MPTAC) review. The document header wording updated from “Current Effective Date” to “Publish Date.” Added note regarding supplies. Revised MN statement to add hybrid closed-loop devices. Revised NMN statement to address upgrades and to clarify intent. Updated Rationale, Coding and References sections.

Revised

05/04/2017

MPTAC review. Deleted MN criteria related to prior pump use and nocturnal hypoglycemia. Lowered MN age criteria from 16 years old to 7 years old. Added new MN and NMN statements regarding device replacement. Updated Scope, Rationale, Coding, and References sections.

Reviewed

02/02/2017

MPTAC review. Updated Rationale, Coding and References sections.

 

01/20/2017

Updated Coding section.

New

11/03/2016

MPTAC review. Initial document development.