Each benefit plan, summary plan description or contract defines which services are covered, which services are excluded, and which services are subject to dollar caps or other limitations, conditions or exclusions. Members and their providers have the responsibility for consulting the member's benefit plan, summary plan description or contract to determine if there are any exclusions or other benefit limitations applicable to this service or supply. If there is a discrepancy between a Medical Policy and a member's benefit plan, summary plan description or contract, the benefit plan, summary plan description or contract will govern.
Blue Cross and Blue Shield of Montana (BCBSMT) may consider blood glucose monitors (BGMs) designed for home use self-monitoring of blood glucose levels medically necessary for the following:
- Insulin Dependent Diabetes Mellitus (IDDM),
- Non-Insulin Dependent Diabetes Mellitus (NIDDM), OR
- Gestational diabetes.
Professional (intermittent 72 hour) monitoring of glucose levels in interstitial fluid may be considered medically necessary for:
- patients with Type I or Type II insulin dependent diabetes prior to insulin pump initiation to determine basal insulin levels, OR
- patients with Type I or Type II insulin dependent diabetes whose diabetes is poorly controlled** despite current compliance with a regimen including four or more finger sticks each day, and either three or more insulin injections or use of an insulin pump.
**Poorly controlled diabetes includes but is not limited to the following clinical situations:
- unexplained hypoglycemic episodes;
- hypoglycemic unawareness;
- suspected postprandial hyperglycemia; OR
- recurrent diabetic ketoacidosis.
Personal continuous glucose (long-term) monitoring (CGM) in interstitial fluid, including real-time monitoring, as a technique of diabetic monitoring, may be considered medically necessary in patients with Type I insulin dependent diabetes who:
- are pregnant; OR
- have been compliant with a regimen that includes:
- four or more finger sticks and three or more insulin injections per day; OR
- use of an insulin pump;
AND have not achieved adequate metabolic control as evidenced by at least one of the following:
- hypoglycemic unawareness or frequent hypoglycemia; OR
- frequent nocturnal hypoglycemia, less than 50 mg/dL; OR
- wide fluctuations in blood sugar patterns over time (<50 mg/dL, or >150 mg/dL); OR
- discordant hemoglobin A1c (HbA1c) and fingerstick blood glucose levels (i.e., patient with consistent normal blood glucose levels at home but high HbA1C levels).
BCBSMT considers o
Data presented to the FDA advisory committee meeting consisted of studies validating the correlation between the measurements of glucose in interstitial fluid with the blood glucose measurements made with home monitoring devices. While the individual values between the two may vary, in general, the panel found that the overall trends in glucose levels detected by frequent measurements produced potentially clinically important information. However, there were no clinical data presented regarding improvements in HbA1c measurements or a decreasing incidence of hypoglycemic episodes in those whose antidiabetic medications were managed based on more frequent readings of interstitial fluid glucose. However, members of the advisory panel felt that more frequent measurements should extrapolate to improved diabetic management. For example, prior studies have shown that HbA1c levels are lowest among patients who have the highest frequency of daily blood glucose measurements. Nevertheless, the use of trends of daily glucose levels implies a different type of diabetic management compared to traditional methods of serial finger stick glucose methods. The following clinical applications were suggested by the FDA advisory panels:
- Hypoglycemic episodes can be identified more readily by the use of an alarm in the GlucoWatch device. This may be particularly helpful in patients with hypoglycemic unawareness or overnight hypoglycemia. In addition, patients with adequate glucose control, as measured by HbA1c levels, may undergo monitoring to ensure that this control does not come at the expense of unrecognized hypoglycemia.
- Unsuspected postprandial hyperglycemia may be detected, which contributes to elevated HbA1c concentrations in patients whose HbA1c levels are considered adequate. Postprandial hyperglycemia has been related to increased cardiovascular risks. Both fast-acting insulin (insulin lispro) and fast-acting oral hypoglycemics (i.e., repaglinide) may be particularly effective in treating postprandial hyperglycemia.
- The devices may be used periodically to confirm the status of current antidiabetic therapy. Currently, some patients may perform seven to nine finger sticks a day on a periodic basis to confirm the success of diabetic management.
- Patients may use the devices in specific circumstances when the normal routine is upset, i.e., changes in work shifts or while traveling.
- The devices may be used to monitor changes in insulin therapy, i.e., the initiation of an insulin pump.
- The device may be used as an educational tool to more easily illustrate how glucose levels vary with activities and meal choices.
- Quality of life may be improved by decreasing the number of finger sticks.
The key clinical outcomes regarding the clinical utility of interstitial measurements of glucose, using either the CGMS or the GlucoWatch G2 Biographer, relates to their ability to provide either additional information on glucose levels leading to improved glucose control, or to improve the morbidity/mortality associated with clinically significant severe and acute hypoglycemic or hyperglycemic events. Because diabetic control encompasses numerous variables including the diabetic regimen and patient self-management, randomized controlled trials (RCTs) are important to isolate the contribution of interstitial glucose measurements to overall diabetic management.
Chase and colleagues reported the results of a trial of 40 children with poorly controlled Type 1 diabetes (HbA1c >8) who were randomized to diabetic management with or without glucose monitoring with a GlucoWatch device. Conventional glucose monitoring was performed four times daily in both groups. Those randomized to the treatment group were asked to wear the device four times per week for three months. After three months, all patients received Biographers and were followed up for six months. HbA1c values were determined at baseline and after 1, 3, 6, and 9 months. The median HbA1c level dropped from 8.9% to 8.4% in the treatment group, while in the control group the HbA1c increased from 8.6% to 9%. While this difference was statistically significant, it should be noted that the worsening of HbA1c in the control group was nearly as large in magnitude as the improvement in HbA1c seen in the Biographer group. There was no significant improvement in “fear of hypoglycemia” or quality of life between the two groups. In a second observational phase of the trial, all subjects were provided Biographer devices and observed over an additional six months. During this phase, the Biographer group maintained median HbA1c at 8.5%, and the control group improved median HbA1c to 8.6%, which was their original level. It was noted that the frequency of use of the Biographer declined over the course of the first phase of the study, which may be why the Biographer group did not show further improvement in HbA1c over the subsequent six months of use.
Baseline characteristics of the two randomized study groups were reported to be without statistically significant differences. However, the baseline median HbA1c levels for these two groups were different by 0.3%, which is almost as large as the 0.4%–0.5% change observed within groups after provision of the Biographer, and this difference may have clinical significance. Also at baseline, slightly more patients in the control group used insulin pumps or received three or more insulin injections per day compared with the Biographer group, which had slightly more subjects receiving only two insulin injections per day. It is unknown whether these slight imbalances were a result of the small sample or whether there were any problems with randomization.
The authors do not discuss whether such differences may have influenced the observed results, but additional analyses adjusting for differences in potentially confounding baseline characteristics and exploring whether outliers could have influenced the results would be of interest. In addition, it is unclear whether subjects in the Biographer group received more frequent or more intense contact with physicians and the diabetes clinic. Biographer subjects were required to visit the clinic each week to download Biographer data; whereas the control group was able to fax back conventional finger stick glucose meter data. This process may have provided more in-person opportunity for medical input in the Biographer group.
Interpretation of this study’s results should also take into consideration the observation that HbA1c levels may fluctuate over time, even without intervention, and variations of up to 1% may be observed clinically in the pediatric population. In this study, the control group’s HbA1c got worse during the intervention study, which partially contributed to the statistically significant difference between groups. Improvement in HbA1c has been observed in control groups in multiple other studies, most likely as a result of study effects (Hawthorne effect) in which participants in a trial achieve better compliance when results are being monitored. It is unclear why the control group got worse in this study, and this raises concerns over the reproducibility of the study.
Interpretation of the clinical significance of reducing HbA1c by 0.5% has been explored and both magnitude and durability of the improvement are important factors to consider. Eastman and colleagues presented an abstract of a decision analysis model based on the above study and reported that “the model predicts that treating 100 subjects under Biographer-guided standard care, if maintained for the life of the cohort, would prevent 20 cases of proliferative retinopathy, four cases of macular edema, six cases of blindness, twelve cases of clinical albuminuria, eight cases of end-stage renal disease, six cases of neuropathy and one amputation.” However, this model makes a variety of assumptions regarding the durability of the improvement.
In summary, Chase and colleagues conducted a small, randomized controlled trial and reported a small but statistically significant difference in the median HbA1c levels between groups after three months. However, the relatively small magnitude of incremental improvement in HbA1c levels needs to be interpreted in the context of potentially different baseline statistics between subjects in the two groups, potential study effects (Hawthorne effect) in the Biographer group in this unblinded trial, and potential influences of receiving more intense medical attention in the Biographer group. It would be very helpful to see the results of this trial confirmed by another larger, multicenter randomized controlled trial and to have further studies explore the durability of HbA1c improvements over time
Continuous Glucose Monitoring Systems (CGMS)
Results of four randomized trials have been reported. The largest of them, which enrolled 128 adult patients with Type 1 diabetes, was initially available in abstract only. Among the 109 patients completing the three-month trial (the dropout rate was 15%), there was no statistically significant difference in HbA1c levels. Mean HbA1c levels in both the control and study groups declined from 9% at baseline to 8.3% at three months. Similarly, in another randomized study of 75 patients, there was no statistical difference in HbA1c levels after the three-month intervention. The other randomized studies included only 11 and 27 patients, respectively. In 2004, Tanenberg and colleagues reported on a study of 128 patients randomized to insulin therapy adjustments using data from either the CGMS or SMBG using a home blood glucose monitor over a 12-week period. At 12 weeks, HbA1c levels and hyperglycemic event frequency and duration did not differ with any statistical significance in the treatment groups. However, at 12 weeks, events of hypoglycemia (glucose < or = 60 mg/dL) were found to be significantly shorter in the CGMS group than in the SMBG group (49.4 +/- 40.8 vs. 81.0 +/- 61.1 minutes per event, p = .009). The authors concluded that durations of hypoglycemia can be further reduced by adjusting insulin therapy with data from the CGMS rather than using SMBG data alone. Nevertheless, the biochemically defined measurements of hypoglycemia (without accompanying evidence of symptoms and/or a clinically significant hypoglycemic event) are not compelling outcomes. The clinical significance of these test results has not been established, i.e., there is insufficient evidence showing the link between increased duration of asymptomatic hypoglycemia and subsequent clinical outcomes.
Lagarde and colleagues found a slight improvement in HbA1c levels using CGMS compared to controls in children with Type 1 diabetes. However, the difference did not reach statistical significance (p = 0.13). In a European study using a crossover design, Deiss and colleagues reported that CGMS did not decisively influence glycemic control of the total study cohort of children and adolescents with Type 1 diabetes. They suggested that more frequent use of CGMS at shorter intervals may be of greater value. A recent review raised questions about the accuracy of these systems.
Garg reported that in 91 patients with diabetes (75 were Type 1) real-time continuous glucose monitoring was able to reduce glycemic excursions by reducing hyperglycemia without increasing the risk of hypoglycemia. They also indicated that this type of monitoring may reduce long-term complications of diabetes. Recently, Deiss reported on a three-month study of 81 children and 81 adults with stable Type 1 diabetes who had HbA1c levels of 8.1% or greater. Patients were randomized to continuous real-time monitoring, continuous monitoring for three days every two weeks, or self-monitoring of blood glucose. At three months, 50% of patients with continuous real-time monitoring had a decrease in HbA1c of at least 1% compared to 37% of those with intermittent continuous monitoring, and 15 % of controls. These results suggest that continuous glucose monitoring may have potential for improving control in patients with diabetes; however, as the authors note, additional work is needed to determine long-term efficacy, clinical feasibility in patients with varying levels of glycemic control, and effect on rates of hypoglycemia.
Guillod reported on a retrospective study that described findings from a group of 88 patients with Type 1 diabetes who underwent a CGMS exam. The prevalence of nocturnal hypoglycemia (NH) was 67% (32% of them unsuspected). A measured hypoglycemia (finger stick) at bedtime (22–24 hr [between 10PM and Midnight]) had a sensitivity of 37% to detect NH, while a single measure of interstitial blood glucose 4 mmol/l or less at 3 AM., had a sensitivity of 43%. After 6–9 months, suspicions of NH decreased from 60% to 14% (p<0.001). The authors concluded that NH was highly prevalent and often undetected. Self-monitoring blood glucose at bedtime, which detected hypoglycemia, had sensitivity almost equal to that of 3 AM. interstitial blood glucose and should be preferred because it is easier to perform. Tubiana-Rufi reported on an uncontrolled study of 182 patients (children and adults) with poorly controlled Type 1 diabetes. Using the Guardian RT system, which the authors indicated required three calibrations a day, resulted in improvement in HbA1c levels over three months. The DirecNet Study Group reported results of another non-comparative study of 30 patients with Type 1 diabetes that used an insulin pump with the FreeStyle Navigator CGM system for 13 weeks. During this time, the mean HbA1c levels improved from 7.1% to 6.8% and the percentage of glucose values between 71 and 180 mg/dl increased from 52% to 60%. Two patients had severe skin reactions related to the sensor mount adhesive.
Wilson and colleagues, as part of the Diabetes Research in Children Network (DirecNet), evaluated the accuracy and precision of the FreeStyle Navigator CGMS in 30 children with Type 1 diabetes (mean age 11.2 years). The Navigator glucose values were compared with reference serum glucose values of blood samples obtained in an inpatient clinical research center and measured in a central laboratory and in an outpatient setting with a FreeStyle meter. Median absolute difference (AD) and median relative absolute difference (RAD) were computed for sensor-reference and sensor-sensor pairs. The median AD and RAD were 17 mg/dl and 12%, respectively, for 1,811 inpatient sensor-reference pairs, and 20 mg/dl and 14%, respectively, for 8,639 outpatient pairs. The median RAD between two simultaneous Navigator measurements (n = 1,971) was 13%. Ninety-one percent of sensors in the inpatient setting and 81% of sensors in the outpatient setting had a median RAD of 20% or less. The authors concluded that the Navigator's accuracy does not yet approach the accuracy of current-generation home glucose meters, but it is sufficient to believe that the device has the potential to be an important adjunct to treatment of youth with Type 1 diabetes.
Several authors note that these results provide a compelling rationale for conducting a randomized controlled trial (RCT) of use of continuous glucose monitoring in Type 1 diabetes. Recent advances in technology now allow linkage between the CGM device and an insulin pump. Halvorson reported on an uncontrolled pilot trial of 10 children with Type 1 diabetes. The small size and lack of control group limit the ability to draw any conclusions from this study. Publications are also beginning to report on early trials of use of these devices in patients with Type 2 diabetes. Wolpert discussed the skills needed for diabetes management using real-time monitoring and commented specifically on the role of calibration as well as understanding the lag between capillary and interstitial glucose levels.
A recent systematic review of randomized studies identified seven studies with 335 patients that fulfilled their inclusion criteria. Study duration varied from 12 to 24 weeks. This review concluded that compared with self-monitoring, CGMS was associated with a non-significant reduction in HbA1c levels and that evidence is insufficient to support the notion that CGMS provides a superior benefit over self-monitoring for HbA1c reduction. There was some indication from this review of improved detection of asymptomatic nocturnal hypoglycemia in the CGMS group.
The 2007 Standards of Medical Care by the American Diabetes Association (ADA) does not mention this technology in the section on assessment of glycemic control. Recommendations in this section are for self-monitoring of blood glucose three or more times daily for patients using multiple insulin injections. The 2008 Standards of Care from the ADA include a recommendation that “CGMS may be a supplemental tool to SMBG for selected patients with Type 1 diabetes, especially those with hypoglycemia unawareness.”
In December 2007 the Juvenile Diabetes Research Foundation (JDRF) completed recruitment for a six-month trial at ten centers of real-time CGMS in patients with Type 1 diabetes. Results of this study, that randomly assigned 322 adults and children with Type 1 diabetes to continuous glucose monitoring or self (home) monitoring, were released in 2008. With HbA1c as the primary outcome measure, there was a significant difference among patients 25 years of age or older that favored continuous monitoring (mean HbA1c difference 0.53%), while the difference between groups was not statistically significant for those age 15 to 24 years or 8 to 14 years. Unlike many prior studies, this study was sufficiently large to detect a meaningful change in HbA1c levels between groups. The population in this study had relatively well controlled diabetes in that entry criterion was glycated Hb of 7% to 10% but about 70% had levels between 7% and 8%; in addition, over 70% of patients were using an insulin pump. No significant differences were noted in rates of hypoglycemic events, but the study was likely not sufficiently large to detect potential differences. The authors also reported that monitor use was greatest in those patients age 25 or older where 83% of patients used the monitor six or more days per week.
A 2009 study by the JDRF studied the potential benefits of continuous glucose monitoring (CGM) in the management of adults and children with well-controlled Type 1 diabetes. In this study, 129 adults and children with intensively treated Type 1 diabetes (age range 8-69 years) and HbA1c <7.0% were randomly assigned to either continuous or standard glucose monitoring for 26 weeks. The main study outcomes were time with glucose level at or below 70 mg/dL, HbA1c level, and severe hypoglycemic events. At 26 weeks, biochemical hypoglycemia (at or below 70 mg/dL) was less frequent in the CGM group than in the control group (median 54 vs. 91 min/day), but the difference was not statistically significant (P = 0.16). Time out of range (70 or less or greater than 180 mg/dL) was significantly lower in the CGM group than in the control group (377 vs. 491 min/day, p=0.003). There was a significant treatment group difference favoring the CGM group in mean HbA1c at 26 weeks adjusted for baseline values. One or more severe hypoglycemic events occurred in 10% and 11% of the two groups, respectively (p not significant). The authors concluded that the weight of evidence suggests that CGM is beneficial for individuals with Type 1 diabetes who have already achieved excellent control with HbA1c <7.0%.
In a randomized study of 132 adults and children from France, Raccah and colleagues reported improved HbA1c levels (change in A1c of 0.96% vs. 0.55%) in patients who were fully protocol compliant for use of an insulin pump integrated with CGMS compared to those using a pump with standard glucose self monitoring.
In a one-year, multicenter, randomized controlled trial of 485 adults and children, Bergenstal et al. compared efficacy of sensor-augmented pump therapy with that of a regimen of multiple daily insulin injections with inadequately controlled Type 1 diabetes. In this study, patients were eligible with the following criteria:
- Type 1 diabetes,
- between the age of seven and 70,
- received multiple daily injections that included long-acting analogue insulin during previous three months,
- glycated hemoglobin was between 7.4% and 9.5%.
The patients were randomly assigned to one of two groups: pump therapy or injection therapy. The pump therapy group used a device that integrates an insulin pump with continuous glucose monitoring (MiniMed Paradigm Real-Time System). Before randomization, all patients received intensive diabetes management. The pump therapy group used insulin aspart. The injection therapy group used both insulin glargine and insulin aspart under guidance of their treating physician. Sensor glucose values were collected at one-week periods at baseline, six months, and one-year in the two study groups. The injection therapy group collected data, but did not display data with Guardian REAL-Time.
All patients were seen at 3, 6, 9 and 12 months and used diabetes management software. At follow-up clinic visits, glucose data was reviewed, therapy was adjusted, glycated hemoglobin was measured, data on adverse events was collected. From January 2007 to December 2008, of 485 patients, follow-up data was missing on 10 patients, four patients were lost to follow-up, 32 discontinued the study or were withdrawn, and six did not provide one-year results, leaving 433 patients in the end analysis. At one year, the baseline mean glycated hemoglobin level (8.3% in the two study groups) had decreased to 7.5% in the pump-therapy group, as compared with 8.1% in the injection therapy group. In the post hoc analysis that used glycated hemoglobin targets recommended by the ADA for children between ages six and 12 years (<8%) and adolescents between ages of 13 and 19 years (<7.5%), a total of 35 of the 80 children and adolescents (44%) in the pump therapy and 16 of the 80 (20%) in the injection group reached these targets in one year.
The American Diabetes Association’s (ADA) 2010 standards of care for the treatment and management of diabetes mellitus include the following recommendations for CGM:
- “Continuous glucose monitoring (CGM) in conjunction with intensive insulin regimens can be a useful tool to lower A1C in selected adults (age ≥ 25 years) with Type 1 diabetes.
- Although the evidence for A1C lowering is less strong in children, teens, and younger adults, CGM may be helpful in these groups. Success correlates with adherence to ongoing use of the device.
- CGM may be a supplemental tool to SMBG in those with hypoglycemia unawareness and/or frequent hypoglycemic episodes”.
The ADA also lists the initiation of CGM as a treatment option for individuals when treatment goals are not met.
The 2007 American Association of Clinical Endocrinologists (AACE) lists CGMS as a clinical consideration for Type 1 diabetics with unstable glucose control and patients who cannot achieve an acceptable HbA1c. The AACE also states that CGMS is “particularly valuable in detecting both unrecognized nocturnal hypoglycemia and postprandial hyperglycemia”.
The 2010 AACE Consensus Statement on Continuous Glucose Monitoring states: “Personal CGM devices are owned by the patient. Glucose values are visible continuously, allowing for immediate therapeutic adjustments on the basis of “real time” glucose results.” The AACE recommends CGMS as a clinical consideration for the following patients:
- “Those with Type 1 DM and the following characteristics:
- Hypoglycemic unawareness or frequent hypoglycemia,
- Hemoglobin A1c (HbA1c) over target, or with excess glycemic variability (eg, judged to be excessive, potentially disabling, or life-theatening hypoglycemia),
- Requiring HbA1c lowering without increased hypoglycemia,
- During preconception and pregnancy
- Children and adolescents with Type 1 DM who have achieved HbA1c levels less than 7.0% (these patients and their families are typically highly motivated)
- Youth with Type 1 DM who have HbA1c levels of 7.0% or higher and are able to use the device on a near-daily basis.
The following patients might be good candidates for personal CGM, and a trial period of 2 to 4 weeks is recommended:
- Youth who frequently monitor their blood glucose levels,
- Committed families of young children (younger than 8 years), especially if the patient is having problems with hypoglycemia.
Intermittent use of professional CGM may be useful for youth with Type 1 DM who are experiencing changes to their diabetes regimen or have problems with:
- Nocturnal hypoglycemia/dawn phenomenon
- Hypoglycemia unawareness
- Postprandial hyperglycemia.”
In conclusion, data supports personal (continuous long-term) monitoring of glucose levels in interstitial fluid, including real time monitoring, for patients of all ages.
Disclaimer for coding information on Medical Policies
Procedure and diagnosis codes on Medical Policy documents are included only as a general reference tool for each policy. They may not be all-inclusive.
The presence or absence of procedure, service, supply, device or diagnosis codes in a Medical Policy document has no relevance for determination of benefit coverage for members or reimbursement for providers. Only the written coverage position in a medical policy should be used for such determinations.
Benefit coverage determinations based on written Medical Policy coverage positions must include review of the member’s benefit contract or Summary Plan Description (SPD) for defined coverage vs. non-coverage, benefit exclusions, and benefit limitations such as dollar or duration caps.