Practical Aspects of Continuous
Glucose Monitoring
2008
Rosanna Fiallo-Scharer, MD
Laurel Messer, RN, BSN, CDE
Barbara Davis Center for Childhood Diabetes
Presentation Outline
Historical background
Accuracy & research
Insurance coverage for
pediatrics
Barbara Davis Center experience
Real-Time and retrospective use
of CGM
Practical child and family issues
CGM
Historical Background
CGMS®
– FDA- approved in 1999
– Retrospective review of downloaded data
– Principle: Glucose oxidase-coated subcutaneous
sensor
Glucowatch Biographer
– FDA- approved in 2001
– First approved real time device
– Principle: reverse iontophoresis through intact skin
Real Time Continuous Glucose
Monitoring
Freestyle Navigator
Paradigm Real Time System
Dexcom STS
Continuous Glucose Monitors
The DirecNet experience
Inpatient Accuracy Study
Principal Aim:
To assess the accuracy of the Medtronic
MiniMed CGMS and the GlucoWatch
Biographer II vs. gold standard plasma
glucose measurements in children with
T1DM
Subject Demographics
• 91 Children and
Adolescents
• 51% Female
• 43% Pumpers
• Mean HbA1c = 7.8%
Ages of Subjects
3 to < 7
12 to < 18
7 to < 12
Daily Glucose Variations
400
Procedure
•Regular meals and insulin
doses
•GS glucose q30-60 min
380
360
340
320
300
glucose (mg/dl)
280
260
240
220
200
180
160
140
120
100
80
60
40
2:00 PM 3:00 PM 4:00 PM 5:00 PM 6:00 PM 7:00 PM 8:00 PM 9:00 PM
10:00
PM
11:00 PM
12:00
AM
1:00 AM 2:00 AM 3:00 AM 4:00 AM 5:00 AM 6:00 AM 7:00 AM 8:00 AM 9:00 AM
10:00
AM
11:00
AM
12:00
PM
1:00 PM 2:00 PM
Results
Diabetes Technology &Therapeutics Vol 5 (5), 2003
n
r
Mean RAD
Median RAD Within ISO
GWB
3,672
0.86
22%
16%
80%
CGMS
(original)
5,658
0.77
26%
19%
53%
CGMS
(Modified)
1,120
0.90
16%
11%
72%
Ultra
2,068
0.97
6%
9%
94%
ISO Criteria: If reference glucose ≤ 75 mg/dL, sensor glucose within ± 15 mg/dL;
if reference glucose > 75 mg/dL, sensor glucose within ± 20%.
Factors NOT Impacting Accuracy
For Either the GW or CGMS
Age of the Subject
BMI (body mass index)
Sensor age (CGMS)
Location of GWB placement
– Upper vs lower arm
– Inner vs outer arm
Sensor Accuracy by Glucose Level
40%
GWB
Original CGMS
35%
M
30%
e
d
25%
i
a
20%
n
Modified CGMS
15%
R
A
10%
D
5%
0%
0
50
100
150
200
250
Reference Glucose (mg/dL)
300
350
400
CGMS Sensitivity and False Alarm
rate for detection of hypoglycemia
Alarm Setting
Sensitivity
False Alarm Rate
(mg/dl)
60
80
100
120
49%
84%
100%
100%
58%
64%
75%
84%
Down alert
Only 24% and 8% of truly hypoglycemic
incidents were detected by the simple
alarm during the hypoglycemia test and
overnight, respectively
Combining the simple alarm with the down
alert improves those sensitivity rates to
88% and 77%, respectively
Diabetes Technol Ther. 2004 Oct; 6(5): 559-66
Accuracy of the Freestyle
Navigator and Guardian RT
Diabetes Care. 2007 Jan; 30 (1):59-64
Results
n
Median RAD
Within ISO
Guardian RT
1,434
14%
64%
Navigator
1,811
12%
74%
Ultra
2,068
9%
94%
ISO Criteria: If reference glucose ≤ 75 mg/dL, sensor glucose within ± 15 mg/dL;
if reference glucose > 75 mg/dL, sensor glucose within ± 20%.
Conclusions
These devices have sufficient accuracy to
allow tracking of glucose values
However, neither device is as accurate as
meters presently available on the market
Particularly useful for detecting postprandial glycemic excursions and
overnight glucose trends
GWB2 Randomized Controlled
Trial
Diabetes Care. 2005 May; 28(5):1101-6
HbA1c
12
HbA1c (%)
10
8
6
4
2
0
Baseline
3 Months
Usual Care
6 Months
GW2B
Ultra Daily Use
First Month:
5.1 ± 1.6 in Usual
Care group, 5.6 ± 1.7 in GWB group
Third Month:
5.1 ± 1.8 in Usual
Care group, 5.3 ± 1.6 in GWB group
Sixth Month:
4.8 ± 1.7 in Usual
Care group, 5.1 ± 1.7 in GWB group
GW2B use
First Month: 2.1 ± 0.8 uses per week (64%
= at least 2 sensors/ week)
Third Month:
GW2B use)
1.6 ± 0.7 uses per week (7 of the 99 subjects discontinued
Sixth Month: 1.5 ± 0.6 uses per week (26 of the 98 subjects discontinued
GW2B use)
Questionnaire regarding non-use of the GW2B (55 subjects)
76% = “skin irritation”
56% = “skips too frequently”
47% = “alarms too frequently”
33% = “readings not accurate”
31% = “too busy to use it”
22% = “forget to use it”
18% = “did not help with diabetes
management”
Lesson learned:
For CGM to help with diabetes
management, patients must use
them!
Freestyle Navigator Pilot trial
Journal Pediatr. 2007 Oct; 151 (4): 388-93
Summary
57 pediatric subjects enrolled
– 30 Pumpers
– 27 MDI
A1c data collected at 13 weeks
Voluntary sensor use offered after 13 weeks
73% of pump users continued to use sensor at
12 months and 78% of MDI users continued at
10 months (median hrs/wk 98 and 85
respectively)
A Randomized Clinical Trial to
Assess the Efficacy of RealTime Continuous Glucose
Monitoring in the Management
of Type 1 Diabetes
funded by a grant from
the Juvenile Diabetes Research
Foundation
Clinical Trial Design
 450 subjects age >=8 years
• 330 with A1c >7.0% and 120 with HbA1c <7.0%
• 1/3 in each age group: 8-<15, 15-<25, >25
• 50% pump users, 50% MDI
 Randomization to RT-CGM or Usual Care
• Navigator, DexCom, Paradigm/Guardian REAL-Time
 Primary outcome at 6 months
 Months 7-12: both groups use RT-CGM
 Outcomes: HbA1c, hypoglycemia, quality of life
 Cost-effectiveness ancillary study
Current status
JDRF RCT
6 month data collection already completed
12 month data collection still in progress.
Insurance coverage for
pediatrics
Currently only one device approved for
pediatric use
Real time technology not universally
covered
Some success with approval on case by
case basis
Usually approved on appeals process
Presentation Outline
RT- CGM systems for pediatrics
Accuracy & research
Insurance coverage for
pediatrics
Barbara Davis Center experience
Real-Time and retrospective use
of CGM
Practical child and family issues
Barbara Davis Center
experience with CGM
Currently: 100+ pediatric pts on CGM
5 years old through adulthood
Commercially: Paradigm REAL-Time, Dexcom
SEVEN
Research: Navigator
Varied experiences with devices
Using a sensor on daily basis
Insert sensor (every 3-7 days)
Warm up period with no glucose readings
Entering fingerstick BG for calibrations
Device starts reading REAL-TIME
information
Occasional downloading
of device for
RETROSPECTIVE
information
Real Time CGM Use
SENSOR glucose levels
Different from BG levels due to lag time
Updates every 1-5 minutes
Arrows
Alarms
Trend information
Real Time CGM Use
EDUCATION POINT:
– Must always do a BG for insulin,
treatment and management decisions
WHY?
• No device currently FDA
approved for replacement
therapy
• Sensor may not be
reading accurately
Retrospective data
Downloaded at home or in clinic
(anyone have experience with this?)
Retrospective data
EDUCATION POINTS
Must know the WHY before knowing WHAT
to change!
Make dosing changes if BG/SG is out of
range 2 out of 3 days
Important to look at trends
Look at most recent week
Questions before changing: missed bolus?
menses? Illness? Bad pump set? Mistake in dose?
Sports?
Retrospective data
Trend graph
(sensor daily overlay, modal day)
Coming soon: JDRF online CGM
school!
Pediatric issues
Expectations
Sensor sticking
Sports
Alarms
Calibrations
Family dynamics
Pediatric issues
Expectations
Expectations
No fingerstick BGs
Will read from the moment you put on
Alarms will prevent all highs and lows
Reality:
4-8 BGs per day
Periods where not calibrated, not reading
Sensor errors for no reason
Alarms annoying
Pediatric issues
Sensor sticking
Three main problems:
1) Sensor does not stick
Try different types of preps (IV prep, skin prep, Skin
Tac, Mastisol, tincture of benzoine
2) Tape or preps causes skin reactions
Try different preps or tapes
Use IV3000/Tegaderm FIRST, and cut hole for
sensor to insert through
3) Not enough skin “real estate”
Try different sensors
Pinch up even if not indicated
Chart on handout
Pediatric issues
Sensor sticking
Pediatric issues
Sports
PROBLEM: CGM sensors fall off, sweat off, get broken
Placement
Consider where least impact and movement
Arm– can cover with ace bandage for extra
support (take off at night)
Sweating
Use antiperspirant under tape
Breaking
Never take CGM receiver out into game
Pediatric issues
Alarms
PROBLEM: Alarms can become overwhelming/
annoying Why?....
Calibration reminder * Calibration
error * Replace sensor * High
Glucose * Low Glucose * Projected
Low Glucose * Projected High
Glucose * Meter BG Now * Sensor
end * Weak signal * Disconnected *
Low transmitter * Bad transmitter *
Sensor error * Bad sensor
Pediatric issues
Alarms
GOAL: Set alarms that are
MEANINGFUL and will NOT drive the
child crazy!
HIGH ALARM: 250-300 mg/dl*
LOW ALARM: 70-80 mg/dl
Consider higher with hypoglycemia unawareness
Can gradually “tighten” alarms as glucose levels get
tighter
*My opinion only!
Pediatric issues
Alarms
Can I change the type of alarm notice?”
Navigator:
YES: vibrate or sound at three different volumes
Medtronic:
YES: vibrate or sound at different volumes
Dexcom:
NO: Will first vibrate then sound at increasing
volumes
Pediatric issues
Overnight alarms
PROBLEM: Parent/child has difficulty hearing alarms at
night
Solution: Improvise!
On bedside table (in a glass) or under pillow on
vibrate
In hallway outside of room
Baby monitor
Sibling in the room!
Spy equipment
Paradigm Real-Time: More difficult because
usually under blankets– new tech coming soon!
Pediatric issues
Calibrations
PROBLEM: Calibration alarms annoying, inconvenient, and
when fail, are frequent
SOLUTION: Calibrate before meals or 2 hours
after food or insulin shot
– Less likely to get a failed calibration alarm
– More likely to get accurate sensor readings
GOAL: Be the BOSS of the Calibration: Only
enter a calibration if BGs are STABLE
Family issues and dynamics
Adaptation: “It is just part of our life now. I wear it all the
time and don’t think much about it anymore.” 12 y.o.
female
Anxiety: “ I was worried that my mom was mad at my blood
sugars all the time so I didn’t want to show her” 11 y.o.
male
Ambivalence: “I like it when it works but I worry it is going to
fall out and I will have to put a new sensor in again”
14 y.o. male
Family issues and dynamics
THE BOTTOM LINE: They have to live with
diabetes, not with CGM
Interval wear (example: 1 sensor/week)
Occasional wear (example: during finals)
Try again later
The future of CGM
Smaller
Longer sensor wears
Less calibrations
More insurance coverage
Closed loop system (insulin pump
responds to sensor glucose levels)
Implantable sensors
Questions?