Internetbased pilot study comparing low fat with high fat evening snacks in children and adolescents

1
Internet-based pilot study comparing low fat with
high fat evening snacks in children and
adolescents with Type 1 Diabetes using continuous
glucose monitoring.
Darrell M. Wilson1, Kimberly Caswell1, H. Peter
Chase2, Michael Tansey3, Larry A. Fox4, Stuart
Weinzimer5, Dongyuan Xing6, Roy Beck6, Craig
Kollman6, William Tamborlane5, Katrina Ruedy6 and
the Diabetes Research in Children Network
(DirecNet) Study Group. 1Stanford, CA 2Denver,
CO 3Iowa City, IA 4Jacksonville, FL 5New
Haven, CT 6Tampa, FL
2
Abstract
Introduction Bedtime snacks are frequently
employed in the treatment of children with type 1
diabetes to prevent nocturnal hypoglycemia, and
the inclusion of protein and/or fat is commonly
recommended. Methods Ten subjects with T1DM
(age 6-18y, HbA1c 6.9 0.5) enrolled in a pilot
trial evaluating the Navigator glucose sensor
completed a study comparing the effect of bedtime
snack fat content on nocturnal hypoglycemia. On
12 separate nights, each subject received either
a low-fat (30 gm carbohydrate, 3 gm protein, 1 gm
fat 138 kCal) or high-fat (30 gm CHO, 2 gm
protein, 20 gm fat 320 kCal) snack in random
order. Subjects checked their glucose and via a
website, were assigned to a high or low fat
snack. Subjects used their usual evening snack
algorithm to determine the size (in 15 gram
carbohydrate increments). Data were from nights
with at least 5h (average of 8.11.3h) of sensor
data after the snack. Results Average BG
(mg/dl) on 128 valid study nights before snack
were similar in both groups (163 55, high fat
16453, low fat). The proportion of nights with
hypoglycemia (sensor glucose 70) was similar in
both groups (23 high fat vs 21 low fat).
Likewise, the proportion of nights with
hyperglycemia (sensor glucose 200) was similar
in both groups (68 high fat vs 61 low fat).
Conclusion Even very well controlled children
and adolescents with T1DM have a very high
frequency of high and low glucose levels.
However, the altering the amount of fat in the
bedtime snack had no impact on frequency of
hyperglycemia or hypoglycemia. This study also
highlights the feasibility of web based research
in patients home environment.
3
Background

• Hypoglycemia is a frequent and potentially
  serious complication of intensive insulin
  therapy, particularly in children with type 1
  diabetes
• Almost half of severe hypoglycemia episodes occur
  overnight, while children are not being actively
  supervised or monitored
• Common practice suggests the ingestion of a
  fat-containing bedtime snack may reduce the risk
  of nocturnal hypoglycemia, presumably via
  delaying carbohydrate absorption

4
Rationale

• The purpose of this outpatient study was to
  determine whether, among children with type 1
  diabetes (T1DM) utilizing insulin pump, a
  pre-bedtime snack with a relatively high fat
  content provides greater protection from
  nocturnal hypoglycemia than a snack containing
  the same amount of carbohydrate and protein, but
  a lower fat content.

5
Methods - Subjects

• Ten subjects 6-lt18 years old with T1DM for over a
  year and using an insulin infusion pump
• Mean hemoglobin A1C - 6.9
• Enrolled in a pilot trial evaluating the Abbott
  Navigator Continuous Glucose Monitoring System
• On 12 nights, subjects were randomized to a
  carbohydrate snack or a carbohydrate plus fat
  snack

6
Abbott FreeStyle Navigator Sensor
Sensor Delivery Unit
Receiver
Transmitter
Sensor
7
Snacks
LOW FAT 30 gm CHO 2.5 gm protein 1.3 gm
fat 138 kCal
HIGH FAT 30 gm CHO 2 gm protein 20 gm fat 320
kCal
8
Web-based Snack Assignment and Data Collection

• Subjects logged on the study website and
• entered the pre-snack glucose level
• entered the level of activity during that day
• used usual practices for determining the amount
  of carbohydrate (in 15-gram increments)
• entered the amount of insulin to be taken to
  cover the snack
• A minimization algorithm was used to determine
  the order of the snack types based on
• the pre-snack meter glucose level
• the self-reported amount of activity during that
  day

9
Web-based Snack Assignment and Data Collection
10
Web-based Snack Assignment and Data Collection
11
Outcome Measures

• The primary outcome was hypoglycemia
• at least one glucose value 70 mg/dL on either
  the Navigator or the FreeStyle meter
• Hyperglycemia was defined as at least one glucose
  value 200 mg/dL and at least 50 mg/dL above
  pre-snack glucose on either the Navigator or the
  FreeStyle meter.
• At least 12 nights with at least 5 hours of
  sensor data were collected

12
Results
13
Time to Hyperglycemia for the high-fat and
low-fat Nights. Hyperglycemia was defined as the
first glucose value 200 mg/dL and at least 50
mg/dL above pre-snack Navigator glucose. p-value
from permutation test to account for correlated
data from the same subject
14
Time to Hypoglycemia for the high-fat and low-fat
Nights. Hypoglycemia was defined as the first
glucose value 70 mg/dL. p-value from
permutation test to account for correlated data
from the same subject
15
The Mean Sensor glucose value at each hour in
each group after bedtime snack. BG at baseline
(hour zero) is the average sensor reading over
the 30 minutes prior to the snack. BG at each
hour after snack is the average of sensor
readings within 15 minutes of the
hour p-valuelt0.01 at 5 hours after snack
16
Results

• There were no meaningful differences between the
  high and low fat snacks on the frequency of
  hyperglycemia or hypoglycemia.
• Mean overnight glucoses were higher in the high
  fat snack
• There was no meaningful difference in the timing
  of hyper- or hypoglycemia
• This study highlights the feasibility of web-
  based research in patients home environment.

17
Discussion

• The addition of fat to a bedtime snack does not
  improve overnight glycemic control and may cause
  hyperglycemia
• This study demonstrates the utility of web-based
  randomization and data collection systems in
  examining clinically important outpatient
  problems in patients with chronic diseases.

18
(No Transcript)