Tip 1 diyabetli hastalarda karbonhidrat sayımının metabolik kontrol üzerine etkisi

PDF Dosyası: 

arastirmax_1628_pp_1-5.pdf

Introduction
Carbohydrate (CH) counting is a meal-planning
approach offering a number of influential advantages
and not a new concept, which was initially used in
the 1920’s in diabetic patients’ meal plan following
the discovery of insulin (1,2). It is a single nutrient
focused method providing a more specific method of
matching food and mealtime insulin resulting in improved blood glucose control (2,3).
Insulin requirements are mainly determined by the
CH content rather than the protein or fat content
of a meal (3,4). Following a CH-rich meal, most of
the absorbed CH enters bloodstream in the first 15
minutes and converted to glucose in approximately
2 hours (1,3). Therefore, postprandial blood glucose
levels are directly related with the amount of CH consumed with meals (1,5,6).
Carbohydrate counting has recently gained popularity with the widespread use of insulin infusion
pumps (6,9). In order to minimize the unpredictability of blood glucose levels for patients with stable
insulin dosage, it is recommended not to exceed the
predetermined daily CH amount (6). Consequently,
this may cause disruption in metabolic control and
problems with compliance to treatment in type 1 diabetic patients, who wish a compliant life style but
are unable to use an insulin pump due to social security and financial problems. In the present study,
we aimed to assess the effect of CH counting, which
is easily learned, of no additional cost and provides
flexibility while choosing food, on metabolic control
in patients using multiple insulin injection therapy.
Material and Methods
Nine volunteer adolescent patients with type 1 diabetes mellitus (DM), who have been followed up for at least
one year in the Department of Pediatric Endocrinology
and Adolescence, used multiple insulin injection therapy (premeal short acting insulin+intermediate acting
**Endocrine and Adolescence Unit, Department of Pediatrics, Dokuz Eylül
University Faculty of Medicine, Izmir, Turkey
Reprint request: Dr. Ayhan Abacı, Endocrine and Adolescence Unit,
Department of Pediatrics, Dokuz Eylül University Faculty of Medicine,
İnciralti-35340, İzmir, Turkey
E-mail: ayhanabaci@gmail.com
Date submitted: January 12, 2009 • Date accepted: February 17, 20092 • March 2009 • Gulhane Med J Abacı et al.
insulin twice a day neutral protamine hagedorn [NPH]),
and wanted a flexible multiple daily insulin (FMDI)
therapy were recruited into the study. Lispro insulin
(Humalog Pen®
, Lilly) or Aspart insulin (Novorapid®
,
Novo Nordisk Pharmaceuticals, Princeton, NJ, USA)
were used as short-acting insulin and NPH (Insulatard®
or Humulin–N®
) was used as intermediate-acting insulin. The mean diabetes duration and age of the study patients was 4.23±3.53 and 15.45±1.47 years, respectively,
and neither of them was in the honeymoon period.
Before CH counting, each patient and family were
given recommendations and education regarding
nutrition, meal planning, and the use of CH counting with FMDI regimen in a mean duration of three
months. Food labels, exchange lists, food models, and
restaurant reference guides were used as educational
tools. Each patient was evaluated by a nutritionist at
quarterly clinic visits. A 24-hour dietary recall with
assessment of CH counting skills was performed at
each clinic visit. All patients were taught insulin treatment algorithm for blood glucose levels above and
below the target ranges. The same members of the intramural diabetes team of our department, including
a physician, diabetic educational nurse, and dietitian
instructed and followed the study patients, who were
asked to monitor blood glucose levels for a minimum
of four times a day and to report their blood glucose
at regular intervals. During the CH counting, the target ranges of blood glucose levels were designated as
follows: preprandial 80-150 mg/dL, postprandial 120-
180 mg/dL and bedtime >110 mg/dL. Patients and
their families were requested to demonstrate their understanding about the concepts of CH counting and
their ability regarding the use of insulin to CH ratios.
During the transition to CH counting, the CH/insulin
ratios and insulin sensitivity indexes of each patient
were calculated as 500/total insulin and 1800/total insulin, respectively. Each patient and family were instructed to adjust unit per CH ratio and correction dose
based on 2-hour postprandial blood glucose. Basal and
premeal bolus doses were modified if fasting and postprandial blood glucose levels were above target ranges,
respectively.
For each patient, metabolic and clinical data were
collected for one year prior to FMDI initiation and
for the first year of FMDI therapy. Among the clinical and metabolic variables of the patients, who were
regularly examined with 3-month intervals, height,
weight, HbA1C, insulin dose, lipid profile, preprandial and postprandial blood glucose levels, basal-bolus
insulin ratios, presence of ketoacidosis were recorded
prospectively at regular intervals throughout the 12
months before and after institution of FMDI therapy.
Data of home blood glucose meters were downloaded, and the frequency of checks in the 14 days before each visit was evaluated along with the number
of hypoglycemic events (defined by a blood glucose
level below 65 mg/dL). HbA1c levels were measured
with Cobra Integra 400 plus (Roche Diagnostics®,
Switzerland). Normal HbA1c levels were accepted as
4.5-5.7 g/dL.
Anthropometric data [body mass index (BMI)]
were converted to standard deviation scores (SDS) by
using data from the National Health and Nutrition
Examination Survey (7). Insulin dosages were expressed as the total daily dose (TDD) in units per kilogram
of body mass.
Patients who were taught to count carbohydrates
but did not contact with the treatment team for at least six months were excluded. Among nonparametric
tests, Wilcoxon test was used to compare the clinical
and metabolic variables before and after CH counting.
The reported values were expressed as median (minimum-maximum). A p value of <0.05 was considered
statistically significant.
Results
With respect to before CH counting state, the mean
HbA1c levels of the patients showed a decrease of 1%,
which did not differ statistically (Table I). No statistically significant differences were detected between the
before and after CH counting values of total insulin
doses (IU/kg/day), HDL cholesterol, total cholesterol,
and triglycerides (p=0.086, p=0.183, p=0.173, p=0.813,
respectively) (Table I). The decrease in LDL cholesterol
with CH counting was statistically significant (p=0.036)
(Table I). No statistically significant differences were
detected between the before and after CH counting
assessments regarding the hypoglycemia frequency
and BMI SDS (p=0.575, p=0.109, respectively) (Table
I). None of the patients developed severe hypoglycemia or diabetic ketoacidosis during the CH counting.
At the end of one year, a decrease more than 1% in
HbA1c levels was observed in four patients (44%).
Three patients’ HbA1c levels (33.3%) declined below
8%. No statistically significant differences were found
regarding bolus-basal insulin ratios although the ratio
increased with CH counting (p=0.051). In terms of metabolic control, post-CH counting levels of morning
fasting blood glucose were statistically significantly
lower than the pre-CH counting state (p=0.028) (Table
II). No statistically significant differences were observed regarding the one year after CH counting levels
of other blood glucose measurements throughout the
day although they were lower than the year before’s
(p>0.05) (Table II).Volume 51 • Issue 1 Carbohydrate counting and type 1 diabetes mellitus • 3
Discussion
Nutritional education programs play an important
role in the management of metabolic control in diabetes mellitus (3,6,8). It has been shown that metabolic
control is better when nutritional therapy methods
are applied successfully (6,8,9). The primary goals of
nutritional treatment are to maintain healthy growth,
achieve near normal blood glucose levels, reduce the
risk of severe hypoglycemia, improve quality of life,
and avoid vascular complications (6,10,11). Recently,
important advances have taken place in the management and treatment of type 1 DM. Among these, the
most important one is the increasing use of CH counting method together with various intensive insulin
treatment regimens (4,12,13). It was shown in numerous studies that the use of CH counting, particularly
in conjunction with insulin pump provided better metabolic control (12-14).
Adolescence is a period, in which metabolic control
is more hardly achieved with respect to young children
and adulthood. The increased metabolic disturbance
during puberty is attributed not only to insulin resistance, growth hormone secretion and sex steroids, but
also to psychological and behavioral changes, which
may result in considerable non-adherence to regimens
during this period. In the present study, an improvement in metabolic control was accomplished with a
shift to the flexible insulin regimen and CH counting
despite these hormonal changes encountered in adolescence. Therefore, we think that the flexible insulin
regimen with CH counting is an important treatment
method regarding providing flexibility in meal-planning in the particular period of frequent behavioral
alterations and eating disorders.
In the medical literature, most of the studies regarding CH counting were performed on patients using
Table I. Clinical characteristics of the patients before and one year after carbohydrate counting
Variables Before carbohydrate counting* After carbohydrate counting* p value**
Body mass index standard deviation scores 1.05 (-0.96-1.65) 1.16 (-0.23-1.46) 0.066
Hemoglobin A1c (%) 9.26 (7.02-12.31) 8.26 (7.31-9.30) 0.110
Triglyceride (mg/dL) 81 (46-199) 94 (69-177) 0.813
Cholesterol (mg/dL) 172.5 (139-280) 171 (129.5-183) 0.173
HDL (mg/dL) 57 (47-69.5) 62 (48.5-120) 0.183
LDL (mg/dL) 92.75 (59-192) 71.7 (47-105) 0.036
Hypoglycemic episodes/patient/year 51 (24-151.92) 66 (27.96-144) 0.575
Total insulin dose (IU/kg/day) 1.07 (0.6-1.46) 1.04 (0.7-1.49) 0.086
Bolus dose (IU/kg/day) 0.65 (0.27-0.99) 0.55 (0.35-0.72) 0.214
Basal dose (IU/kg/day) 0.40 (0.30-0.57) 0.54 (0.35-0.78) 0.015
Bolus/basal insulin ratio 1.46 (0.69-2.31) 1.11 (0.62-1.53) 0.051
*: median (minimum-maximum), **: Wilxocon test
Table II. The effect of carbohydrate counting on metabolic control
Glucose (mg/dL) Before carbohydrate counting* After carbohydrate counting* p value**
Prebreakfast 204.15 (152.25-219.5) 246.6 (157-347) 0.028
Postbreakfast 219.5 (150-292) 167 (109-197.75) 0.465
Prelunch 171 (108-213) 145.17 (95-235) 0.237
Postlunch 134.99 (64-181) 145.17 (95.5-235) 0.917
Predinner 194.75 (118-320.41) 156.62 (124-215.64) 0.021
Postdinner 148 (121.68-267.16) 179.12 (110-259.33) 0.715
Bedtime 198.31 (119.83-272.88) 163.5 (143.75-204.69) 0.051
Midnight (3:00 am) 214 (118-238) 185.65 (8144.95-244) 0.735
Mean 214 (118-238) 170 (104-202.54) 0.021
*: median (minimum-maximum), **: Wilxocon test4 • March 2009 • Gulhane Med J Abacı et al.
premeal regular insulin+NPH, premeal lispro+bedtime
glargine, premeal lispro+ultralente insulin and insulin pump (12-15). On the other hand, we did not encounter such a study evaluating the outcome of CH
counting method on patients using two doses of NPH
insulin and short-acting insulin analogues (lispro or
aspart), and that is why our study is important.
The most significant complication of CH counting,
which is frequently in conjunction with intensive insulin therapy regimen, is the increase in weight and
frequency of hypoglycemia (12,13,16,17). It is reported
that the risk of severe hypoglycemia and obesity is increased by three-fold and two-fold, respectively, with
intensive insulin therapy during adolescence (17).
Nonetheless, in other studies, it has been detected that
intensive insulin therapy regimen does not increase
the risk for hypoglycemia in patients with type 1 DM
(12,18). Especially in some studies, it is also stated
that insulin pump application in type 1 DM patients
does not increase the frequency of hypoglycemia and
does not cause weight gain (12,18-21). But very few of
these studies were randomized and all used regular insulin (20,21). Alemzadeh et al. found 52.3% decrease
in the rate of hypoglycemia with morning-evening
doses of basal ultralente insulin (12). Alemzadeh et
al., in a study with CH counting patients comparing
two different basal insulins, have detected that there
is a statistically significant decrease in the risk for hypoglycemia in the insulin glargine group when compared to ultralente (morning and evening) group (13).
In both of the above-mentioned studies, no statistically significant differences were observed between the
groups regarding tendency to increase in BMI (12,13).
In the study by Weintrop et al, statistically significant
difference was not detected concerning increase in frequency of hypoglycemia in patients with type 1 DM,
multiple daily injection therapy (regular insulin+NPH)
using and CH counting group, when compared with
the state prior to CH counting (14). Also in our study,
we did not find a statistically significant increase in
BMI SDS and hypoglycemia frequency when CH
counting was combined with intensive insulin therapy regimen. In several studies, it is reported that NPH
insulin increases the risk of hypoglycemia in diabetic
patients (22,23). On the other hand, most of the trials
reporting a decrease in hypoglycemia are studies with
insulin glargine, which shows no peak. The reasons
why no increased risk for hypoglycemia is observed in
our study, even though the patients have been using
NPH insulin, might be considered as the low number
of patients and the same basal insulin used before and
during the study.
In a number of studies, it has been shown that total insulin requirement is diminished with the transition from multiple insulin injection therapy to insulin
pump (13,14,24). In a study by Weintrop et al., total
insulin requirement of patients using insulin pump
was found to be significantly lower than that of those
using multiple insulin injection therapy (14). In the
same study, no statistically significant difference was
detected when patients using multiple insulin injection therapy were compared between before and after
CH counting states regarding total insulin requirement (14). Also in our study, no statistically significant difference was detected regarding an increase in
total insulin requirement although CH counting was
implemented. In a study comparing multiple insulin
injection therapy and insulin pump, Alemzadeh et al.
detected that bolus-basal insulin ratio increased with
CH counting (13). In another study by Alemzedah
et al. assessing the benefit of FMDI therapy in type 1
diabetic adolescents and children, bolus-basal insulin
ratio was detected lower in pubertal group when compared with prepubertal group (12). In our study, when
before and after CH counting bolus-basal ratios were
compared, bolus-basal ratio was found to be insignificantly reduced after CH counting. During the CH
counting, basal rate of the patients, rather than the
bolus doses, was found to be increased. Based on these
data, it might be considered that basal insulin requirement further increases in pubertal stage, in which an
increase in counter hormone effect takes place. In this
study, it is considered that the increased basal insulin
requirement might be explained with puberty, which
all of the patients were experiencing meanwhile.
Although a quality of life questionnaire was not filled
up by the patients in our study, we learned that all of
them were satisfied with CH counting and none wanted
to use the previous treatment method. Carbohydrate
counting plays a significant role in maintaining metabolic control for patients with type 1 DM as well as
introducing a flexible life style and increasing quality
of life. In the study of Alemzadeh et al., in which flexible multiple insulin therapy with morning and night
doses of ultralente insulin was carried out, a statistically significant reduction (1.3%, from 9.3% to 8.0%)
was observed in mean HbA1c levels with CH counting
(12). In the same study, they also stated that in 57%
of the patients, HbA1c levels decreased by 1% with
the therapy, which was considered significant regarding the potential future microvascular complications
(12). It is stated in DCCT report that a 1% decrease in
HbA1c levels diminishes the potential diabetic complications by 21% to 49% (25,26). Although statistical
analyses did not show significant differences between Volume 51 • Issue 1 Carbohydrate counting and type 1 diabetes mellitus • 5
the groups, we considered the decrease in HbA1c level
in our study important because of the above-mentioned findings.
In conclusion, the present study, despite the low
number of cases, suggests that treatment with CH
counting and two doses of intermediate-acting NPH
plus short-acting insulin analogues may be an alternative treatment method for type 1 DM patients who are
unable to use insulin pump due to financial problems
also providing flexibility in meal-planning especially
for adolescent type 1 DM patients.