3.pdf
3.pdf
U
M
D
A
T
v
p
m
u
i
p
i
G
l
G
u
i
0
d
5
se of the Continuous Glucose Monitoring System in Goettingen
inipigs, With a Special Focus on the Evaluation of Insulin-Dependent
iabetes
. Strauss, C. Tiurbe, I. Chodnevskaja, A. Thiede, S. Timm, K. Ulrichs, and V. Moskalenko
ABSTRACT
Objectives. Adult pig islet isolation has greatly improved in the past few years. Islet
grafts may now be tested in large animals. Continuous Glucose Monitoring System
(CGMS) was applied to diabetic Goettingen Minipigs (GMP) to improve the management
of hyperglycemia and hypoglycemia and their welfare before transplantation.
Methods. GMP (25–35 kg) received a minipig diet once daily. Diabetes was induced by
streptozotocin (STZ; 150 mg/kg intravenous [IV]; n 5) or by surgical pancreatectomy
(PGMP; n 3). Interstitial glucose concentration (IGC) was monitored continuously with
an implanted sensor; CGMS was calibrated using conventional blood glucose tests 3–4
times per day; CGMS data were fed into the monitor memory and analyzed using CGMS
software.
Results. Glucose sensors were handled accurately. Diabetes occurred 2–3 days after STZ
or immediately after pancreatectomy with basal C-peptide secretion of 0.4 ng/mL
(measured using intravenous glucose tolerance test) and prompt loss of body weight.
Insulin substitution was necessary to keep the GMP in good condition for up to 5–6
months, with stable body weight and normal behavior. Some GMP became hypoglycemic,
which was only documented by CGMS, but not by conventional glucose assays. Tight
glucose control and substitution of exocrine enzymes (Creon 25,000 E/d) reduced
morbidity of the PGMP, which was then comparable with that of STZ-GMP.
Conclusions. The CGMS, developed for humans, is equally suitable for the 2 GMP
diabetes models. Close-meshed glucose monitoring and insulin treatment improved the
general condition of the diabetic GMP, ie, the islet graft recipients, and will thus greatly
add to posttransplantation success.
e
s
R
A
A
i
2
W
O
HE LARGE animal model of insulin-dependent dia-
betes is an essential tool in preclinical investigation of
arious therapeutic approaches, such as transplantation of
ancreatic islets and bioartificial devices (macrodevice,
icroencapsulated islets, and so on). The pig seems to be a
seful model of diabetes because the function of its gastro-
ntestinal tract, the morphology of its pancreas, the many
hysiological responses, and the pharmacokinetics are sim-
lar to those of human beings.1 Due to their small size, the
oettingen Minipigs (GMP) are particularly suitable for
ong-term laboratory studies. To date, the Continuous
lucose Monitoring System (CGMS) has been successfully
sed in clinical practice to compare glycemic profiles after
slet and pancreas transplantation, and to evaluate glucose u
041-1345/08/$–see front matter
oi:10.1016/j.transproceed.2008.02.003
36xcursions during intensive insulin therapy.2 The aim of this
tudy was to test the CGMS in 2 diabetic animal models.
ESEARCH DESIGN AND METHODS
nimals
dult GMP, 11–17 months at age; weighing 20–35 kg, were housed
n single pens under controlled conditions (temperature was 19°–
3°C, and relative air humidity 40%–70%) with a 12:12 hour
From the Surgical Clinic I, University of Wuerzburg Hospital,
uerzburg, Germany.
Address reprint requests to Dr. K. Ulrichs, Surgical Clinic I,
berduerrbacher Str. 6, D-97080 Wuerzburg, Germany. E mail:
lrichs@chirurgie.uni-wuerzburg.de.
© 2008 by Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010-1710
Transplantation Proceedings, 40, 536–539 (2008)
l
e
m
a
c
I
D
C
d
a
a
m
c
w
m
s
P
T
g
i
t
l
o
d
l
a
T
t
v
F
G
T
c
2
s
p
a
C
T
q
a
c
g
o
c
c
5
c
I
t
T
w
s
I
W
s
L
P
g
1
R
T
a
c
l
T
m
w
t
m
c
G
F
G
1
m
F
i
4
c
CONTINUOUS GLUCOSE MONITORING SYSTEM 537ight/dark cycle and quarantined for 4 weeks before the start of the
xperiment. The animals were fed once daily with a standard
inipig diet and allowed free access to water. GMP care was in
ccordance with “good laboratory practice, regulations for non-
linical laboratory studies.”
nduction and Management of Diabetes
iabetes was induced in 5 GMP using streptozotocin (STZ) (Sigma
hemie, Muenchen, Germany); 150 mg/kg STZ dissolved in so-
ium citrate buffer (1 g STZ/10 mL) was infused for 10 minutes via
central venous catheter. Then 200 mL of 5% glucose was
dministered for 1 hour after infusion of STZ to avoid hypoglyce-
ia due to the release of insulin from destroyed islets. Hypergly-
emia occurred within 2–3 days after STZ administration. The pig
as considered diabetic when blood glucose levels reached 200
g/dL in 3 consecutive readings on 3 days, and basic C-peptide
ecretion was 0.4 ng/mL by intravenous glucose tolerance test.
ancreatectomy
otal pancreatectomy (PGMP) was performed in 3 GMP under
eneral anesthesia as previously described.3 After the abdominal
ncision, the pancreas tail was carefully dissected from surrounding
issues; the small branches of splenic vessels and portal vein were
igated. The spleen was preserved and remained intact until the end
f the operation. Afterwards the pancreas body and its head were
etached from the portal vein, vena cava, and duodenum by
igating the small vessels while preserving the pancreaticoduodenal
rtery and vein. The pancreatic duct was dissected, ligated, and cut.
he abdominal cavity was closed, and at the end of the anesthesia
he animal was transferred to its pen. The analgesia was performed
ia intramuscular injection of 1 g Novalgin (Aventis Pharma,
rankfurt/Main, Germany) and 3 times Tramal (Gruenenthal
mbH, Aachen, Germany) per day for 5 days after the operation.
he antibacterial prophylaxis was performed with long-term peni-
illin. Supplementation of insulin and exocrine enzymes (Creon
5,000 IE/d) started with the beginning of the normal diet on the
econd postoperative day. The real-time glucose monitoring was
erformed after complete recovery of the animals from the oper-
tion.
ig 1. Real-time glucose monitoring in normal nondiabetic
MP. The GMP show much lower glucose profiles (IGC of 53
2 mg/dL) as compared with profiles in human beings (70–120(g/dL).ontinuous Glucose Monitoring
he CGMS (Medtronic Minioled, Northridge, Calif, USA) ac-
uires and stores signals from a subcutaneous sensor connected to
glucose monitor. Via MiniMed’s Com-station the stored signals
an be downloaded into a personal computer and converted to
lucose levels. The sensor function is based on the reaction of
xidase with glucose; the signal is proportional to the glucose
oncentration in the interstitial fluid. The system registers glucose
oncentrations every 10 seconds, and stores the average level every
minutes. Three to four capillary glucose levels were used to
alibrate the device.
mplantation of the Sensor and Fixation of the Device on
he Animal
he sensor was inserted into the subcutaneous tissue. The CGMS
as put into a bag, which was fixed on the back of the pig with
traps.
nsulin Substitution
hen hyperglycemia occurred the substitution of insulin was
tarted as following: group 1 (STZ-GMP n 2; PGMP n 1):
antus insulin at 19:00-22:00 hours; group 2 (STZ-GMP n 2,
GMP n 1): Lantus insulin at 19:00-22:00 hours Altinsulin;
roup 3 (STZ-GMP n 1, PGMP n 1): Lantus insulin at
9:00-22:00 hours Novomix insulin.
ESULTS
he daily profiles of CGMS of normal nondiabetic minipigs
re shown in Figure 1. The average interstitial glucose
oncentration (IGC) was 53 12 mg/dL, which is much
ower compared with AIGC in humans (70–120 mg/dL).
he glucose sensors performed accurately according to the
ean absolute difference and the correlation coefficient,
hich are indicated in each figure; punctual blood glucose
ests confirmed the values documented by real-time glucose
onitoring. The animals tolerated the device well, and all
alibrations were performed without sedating the pigs. All
MP became diabetic 2–3 days after the STZ injection
ig 2. Real-time glucose monitoring in a diabetic GMP without
nsulin substitution over 3 consecutive days. The IGC was 287
0 mg/dL, the mean absolute difference was 5.9%, and the
orrelation coefficient was 0.92.AIGC 285 40 mg/dL, basic C- peptide secretion 0.4
n
i
e
T
t
f
a
d
p
S
g
a
2
c
(
m
L
h
t
i
g
c
b
D
D
F
o
2
t
w
F
c
w
t
N
r
F
T
p
T
c
538 STRAUSS, TIURBE, CHODNEVSKAJA ET ALg/mL by IVGTT) and lost body weight promptly when
nsulin was not substituted. A large amplitude of glucose
xcursions was shown, particularly after feeding (Fig 2).
he PGMP showed intestinal paralysis immediately after
he operation, and the substitution of fluids was necessary
or 4–5 days. No other complications (bleeding, infection,
nd so on) were observed postoperatively. The pigs became
iabetic immediately after the operation, and the glucose
rofiles during CGMS were comparable with those of the
TZ-GMP. Lantus insulin alone (group 1) kept the GMP in
ood condition up to 5–6 months, with stable body weight
nd normal behavior, yet suboptimal glucose control (IGC
75 mg/dL) with high postprandial glucose levels. The
ombination of Lantus insulin with short-term insulin
group 2) improved the IGC (150–250 mg/dL), although
ore hypoglycemic episodes were documented than with
ig 3. Real-time glucose monitoring (RTGM) in a diabetic GMP
ver 4 consecutive days with insulin supplementation. AIGC was
50 97 mg/dL, the mean absolute difference was 8.0%, and
he correlation coefficient was 0.99. Hypoglycemic events are
ell documented using real-time glucose monitoring.
ig 4. Real-time glucose monitoring in a diabetic GMP over 3
onsecutive days with optimized insulin supplementation. IGC
as 152 82 mg/dL, mean absolute difference was 16.9%, and
he correlation coefficient was 0.95. Combination of Lantus and
ovomix insulin keeps the animal’s glucose profile within normalaange for several days.antus insulin alone (Fig 3). The hypoglycemic episodes
ad no clinical relevance, and the animals survived during
he whole observation period. The combination of Lantus
nsulin with Nomomix insulin, before feeding, improved the
lycemic control (Fig 4). STZ-GMP showed better IGC
ompared with PGMP, which was probably due to residual
eta cells (Fig 5).
ISCUSSION
ue to poor compliance one can consider a diabetic pig
ig 5. (A) Immunohistochemistry of a normal GMP pancreas.
he insert shows a well-shaped islet of Langerhans. (B) The
ancreas of a diabetic GMP (diabetes was induced with STZ).
he pancreas is “empty,” ie, only a very few residual beta cells
an be detected in the tissue section (insert).s a patient with “null desire for diabetes,” and, conse-
q
i
i
a
c
p
d
b
a
i
m
o
a
n
i
g
d
s
t
G
a
A
T
W
e
S
t
#
R
d
i
2
b
p
p
e
O
n
CONTINUOUS GLUCOSE MONITORING SYSTEM 539uently, with difficulties in controlling glycemia and
nsulin substitution. That is why intensive insulin therapy
s impossible, or at least very difficult in the diabetic
nimal. Some cases of animal death due to hypoglycemia
aused by insulin substitution have been described in the
resent literature.4 The goal of insulin substitution in the
iabetic animal consists of lowering preprandial glucose
200 mg/dL to keep the animals in good condition
efore transplantation. Insulin substitution in diabetic
nimals can be performed by injecting insulin or implant-
ng minipumps with a sustained release of insulin. The
easurement of HbA1c due to the limited permeability
f erythrocytes to glucose in diabetic pigs was considered
n unsuitable indicator of glycemia. The CGMS, origi-
ally developed for humans, is equally suitable for mon-
toring glycemia in the 2 GMP diabetes models. Hypo-
lycemic episodes and glucose excursions can be tightly
ocumented using CGMS after insulin injection. In
ummary, close-meshed glucose monitoring and insulin
reatment improved the general condition of the diabetic
MP, ie, the islet graft recipients, and will thus greatlydd to posttransplantation success. iCKNOWLEDGMENTS
he authors wish to thank Mr. A. Reichert, Mr. P. Heuler, Mr. J.
einberger, Ms. S. Endres, and Ms. A. Prappacher for their
xcellent technical skills in the management of the GMP; Mrs. B.
chneiker and Ms. S. Gahn for technical excellence in the labora-
ory; and Mrs. L. Stevenson for critically reading the scientific text.
This work was generously supported by the Project BARP,
NMP3-CT-2003- 505614 of the European Union.
EFERENCES
1. Larsen M, Wilken M, Gotfredsen C, et al: Mild streptozotocin
iabetes in the Goettingen minipig. A novel model of moderate
nsulin deficiency and diabetes. Am J Physiol Endocrinol Metab
82:1342, 2002
2. Kessler L, Passemard R, Oberholzer J, et al: Reduction of
lood glucose variability in type 1 diabetic patients treated by
ancreatic islet transplantation. Diab Care 25:2256, 2002
3. Kobayashi K, Kobayashi N, Okitsu T, et al: Development of a
orcine model of type 1 diabetes by total pancreatectomy and
stablishment of a glucose tolerance evaluation method. J Artif
rg 28:1035, 2004
4. Vo L, Tuch B, Wright D, et al: Lowering of blood glucose to
ondiabetic levels in a hyperglycemic pig by allografting of fetal pig
slet-like cell clusters. Transplantation 71:1671, 2001
...
