Background Ìntensive insulin treatment is associated with an

increased risk of hypoglycemia, so strict glycemic monitoring
is essential. The best type of sample for identifying hypoglycemia
remains under debate.
Objectives To establish the number of hypoglycemic events
in intensive care patients relative to insulin administration
method and the method used to collect the blood sample.
Methods Retrospective descriptive study lasting 6 months.
Hypoglycemia was defined as a blood glucose level less than
80 mg/dL (mild: 50-79 mg/dL, severe: <50 mg/dL), measured
with a bedside glucometer and blood from the arterial catheter
or fingerstick, in critically ill patients who require insulin
administered subcutaneously (with sliding scales) or via continuous
intravenous perfusion (intense infusion protocol with
a nurse-managed insulin therapy algorithm).
ResuIts Analysis of the 6636 samples from 144 critically ill
patients revealed 188 mildly hypoglycemic samples (2.8%) and
3 severely hypoglycemic samples (0.04%). The prevalence of
mild hypoglycemia was greater when insulin was administered
intravenously (3.2%) rather than subcutaneously (2.3%; P .04).
Among patients receiving insulin intravenously, hypoglycemia
was found more often in arterial (4.5%) than in capillary (2.8%)
blood (P .01). The prevalence of hypoglycemia in capillary
blood samples did not differ significantly between subcutaneous
(2.3%) and intravenous (2.8%) insulin therapies (P .21).
oncIusions ith a target blood glucose level of 110 to 140
mg/dL, few hypoglycemic events are detected in critically ill
patients, regardless of whether insulin is administered intravenously
or subcutaneously. Analysis of solely arterial samples
may yield a higher prevalence of hypoglycemia than otherwise.
In a meta-analysis oI 29 studies (8432 patients),
Wiener et al16 assessed the risk/beneIit Iactor oI tight
glucose control in critically ill patients and was able
to draw a clear conclusion: the incidence oI hypoglycemia
was higher among patients receiving intensive
insulin therapy, and the increase in incidence
was in proportion to the tightness oI the glucose
control. In 2 later studies,7,13 hypoglycemia was
associated with mortality as an independent Iactor.
In addition, in an international clinical trial, Normoglycemia
in Intensive Care Evaluation Survival Using
Glucose Algorithm Regulation (NICE-SUGAR),17 investigators
Iound more deaths Irom cardiovascular causes
and more episodes oI hypoglycemia (·40 mg/dL; to
convert to millimoles per liter, multiply by 0.055)
in the IIP group and concluded that hypoglycemia
might simply be a strong marker oI illness severity.
In a subsequent consensus statement on inpatient
glycemic control, the American Association oI
Clinical Endocrinologists and the American Diabetes
Association18 recommended use oI an IIP to control
hyperglycemia in intensive care unit (ICU) patients,
with the protocol to begin being used when glucose
values are close to 180 mg/dL, maintaining an optimum
range between 140 and 180 mg/dL.
Published analyses to date do not diIIerentiate
by the type oI blood sample used Ior analysis oI
glucose level. The objective oI the present study was
to assess the number oI hypoglycemic events Iound
in ICU patients according to the type oI sample
analyzed (Iingerstick or arterial blood), as well as
the type oI rapid insulin administration used (subcutaneous
or intravenous).
MateriaIs and Methods
The retrospective study included glycemia tests
oI ICU patients hospitalized between April 1 and
September 30, 2008. Patients studied were in a multipurpose
10-bed ICU that serves all medical and
surgical specialties except cardiac, thoracic, and neurological
cases.
Our IIP was based on the Yale insulin inIusion
protocol.19 ThereIore, hypoglycemia was considered
as a blood glucose level less than 80 mg/dL, with a
value oI 50 to 80 mg/dL categorized as mild hypoglycemia
and less than 50 mg/dL as severe hypoglycemia20,21
because our IIP requires
intravenous insulin at this value to
prevent the neurological complications
associated with hypoglycemia,
even iI the patient is asymptomatic.
In our standard protocol, used in
this study, an Optium Xceed glucometer
(Abbott Diabetes Care, MediSense
Products, Doncaster, Australia) calibrated
per manuIacturer instructions is assigned to
each patient, and strips appropriate to that calibration
are documented and kept in that patient`s box
oI supplies in the ICU. Our unit requires 3 point-oIcare
measurements to conIirm hypoglycemia. When
the glycemia measurement is less than 80 mg/dL,
the measurement is repeated twice with the same
glucometer but with 2 diIIerent samples, to ensure
against any error in the Iingerstick technique or in
reading the glucometer.
Two insulin therapy protocols were used: (1)
the unit`s normal standard, under which patients
with a blood glucose level less than 180 mg/dL
receive subcutaneous rapid insulin, with blood glucose
level checked every 6 hours and insulin administered
according to a predeIined sliding scale oI
blood glucose levels) and (2) intravenous insulin
therapy Iollowing a new IIP in the unit that uses
dynamic scales (perIusion velocity adjusted as the
blood glucose level changes), administered Ior
patients with blood glucose levels greater than 180
mg/dL or patients with a blood glucose level greater
than 150 mg/dL who meet at least 1 oI the Iollowing
clinical criteria: sepsis with Iailure oI 2 or more
organs, invasive mechanical ventilation Ior more
than 5 days and insulin-dependent diabetes.2,22
BrieIly, those protocols are deIined as Iollows:
Sliding-scale protocol: blood collection at 6-hour
intervals. With subcutaneous administration oI insulin,
blood samples Ior glucose testing are collected via
Iingerstick because these patients do not have an
arterial catheter. The guideline Ior subcutaneous
administration oI insulin is every
6 hours, and thereIore our protocol
requires testing oI blood glucose
level at 6-hour intervals; more
Irequent sampling (as in IIP
patients, next) would reIlect the
therapeutic eIIect oI the previously
administered dose oI insulin.
IIP with dynamic scales: hourly
blood collection. Whether the sample
is obtained Irom the arterial
catheter (Iirst 48 hours) or Iingerstick
(thereaIter), more Irequent
monitoring is required to determine the speed with
which intravenously administered insulin is inIused,
depending on the blood glucose level measured.
BeIore the arterial sample is collected Ior analysis,
2 mL oI diluted blood is discharged Irom the dead
space oI the arterial catheter.
The therapeutic protocols were interchangeable
throughout the study period: iI a patient receiving 1
treatment met the criteria Ior the other, the treatment
was changed (see Figure 1). The therapeutic objective
Ior subcutaneous insulin therapy was to maintain
a blood glucose level oI 180 mg/dL or less; the
IIP objective was to achieve a blood glucose level
within a range oI 110 to 140 mg/dL. The same type
oI insulin (Aspart, Novo Nordisk A/S, Copenhagen,
Denmark) was administered, whether subcutaneously
or intravenously.
Clinical histories were included Ior patients with
at least 4 blood glucose measurements recorded per
day, whether Irom capillary or arterial blood. Patients
were excluded iI they had sustained hypoglycemia
Ior 4 hours or longer despite discontinuation oI the
intravenous insulin therapy.
The variables analyzed were blood glucose values
(in milligrams per deciliter) measured in arterial
or capillary blood samples Irom patients treated
with intensive IIP and capillary samples oI patients
treated with subcutaneous insulin.
All patients included in the study, or their
guardians, provided inIormed consent, and the protocol
was approved by the hospital`s Committee on
Ethics and Clinical Research.
$tatisticaI AnaIysis
SPSS version 16.0 (SPSS, Inc, Chicago, Illinois)
was used Ior statistical analysis. A 2 test was used
to compare diIIerences in the incidence oI hypogly -
cemia between diIIerent groups. A P value less than
.05 is considered statistically signiIicant. Data are
presented either as a mean (SD) or as a percentage.
ResuIts
OI 144 patients admitted to the ICU during the
study period who required insulin therapy, 44
patients were excluded (42 had an insuIIicient number
oI recorded blood glucose levels |·4 per day|
and 2 had sustained hypoglycemia Ior more than 4
consecutive hours despite suspension oI the IIP perIusion).
For the 100 patients included, 6636 glycemia
tests were available Ior analysis.
Most oI the included patients (74°) had a blood
glucose level less than 180 mg/dL with subcutaneous
insulin and did not meet the criteria Ior inclusion
in the IIP group (see Figure 1). However, under the
dynamic therapeutic protocol, 21° received both
subcutaneous and intravenous insulin at diIIerent
points in their ICU stay, in general requiring intravenous
insulin therapy during the most serious period
oI the problem Ior which they were hospitalized.
Only 5° received intravenous insulin exclusively.
The mean (SD) age oI the study participants was
66.4 (17.5) years, 67° were male, and the mean (SD)
score on the Acute Physiology and Chronic Health
Evaluation (APACHE) II at admission was 16.3 (7.4).
Diagnoses on admission to the ICU included sepsis
(30.9°), heart disease (24.7°), acute respiratory
Iailure (15.9°), postoperative monitoring reIerred by
surgical specialties (10.6°), hepatopancreatic abnormalities
(7.1°), and medication imbalance (4.4°).
Diabetes was diagnosed in 22° oI study participants.
OI the hypoglycemias detected, 62.1° were Irom
patients being treated with intravenous insulin therapy
when the sample was collected and 37.9° were
Irom patients receiving subcutaneous insulin therapy
when the sample was collected. Patients treated
with intravenous insulin therapy showed a mean
(SD) blood glucose level oI 150.6 (54.5) mg/dL,
and patients receiving subcutaneous insulin therapy
had a mean (SD) blood glucose level oI 153.3
(50.2) mg/dL (Figure 2).
Categorical analysis oI the blood glucose measurements
by insulin administration protocol is
detailed in Figure 3. In the case oI intravenous
insulin therapy, we observed a trend toward lower
values (·110 mg/dL) at the expense oI high values
(~140 mg/dL), although this diIIerence was not statistically
signiIicant.
We detected 188 (2.8°) mild hypoglycemic
events (130 in the intravenous group and 58 in the
subcutaneous group; 3.2° vs 2.3°; P ÷ .04). The
58 mild hypoglycemic events in the subcutaneous
group were spread across the 74 patients treated
with subcutaneous insulin only (APACHE score:
mean, 14.4; SD, 7.03). The 130 hypoglycemic events
in the intravenous group were split between 109
events (83.8°) in 21 patients who alternated between
both protocols (APACHE score: mean, 15.6; SD, 7.01)
and only 21 events (16.2°) in 5 patients treated
exclusively with intravenous insulin (APACHE score:
mean, 28; SD, 3.03), oI which 18 hypoglycemic events
were detected in capillary samples and 3 hypoglycemic
events were detected in arterial samples.
OI the 130 hypoglycemic events in patients
receiving insulin intravenously, 35 hypoglycemic
events were detected in 771 arterial blood samples
and only 95 events were detected in the 3353 capillary
blood samples, resulting in a signiIicant diIIerence
(4.5° vs 2.8°, P ÷ .01). In capillary samples,
obtained by Iingerstick, the numbers oI hypoglycemic
events detected did not diIIer between patients
receiving subcutaneous insulin therapy (2.3°) and
patients receiving intravenous insulin therapy (2.8°;
P ÷ .21, see Table). Severe hypoglycemic events were
detected in just 3 samples (0.04°), 2 oI these in
capillary samples Irom patients receiving intravenous
insulin therapy and the third in a capillary sample
Irom a patient receiving subcutaneous insulin therapy.
iscussion
The number oI mild hypoglycemias was slightly
higher in patients receiving intravenous insulin
therapy than in patients receiving subcutaneous
insulin, which makes sense iI one considers the
stricter therapeutic objective in patients treated
intravenously. However, the use oI inIusion protocols
with demonstrated saIety and eIIicacy results in
the detection oI low levels oI hypoglycemia.18 In our
ICU, the objective is to achieve blood glucose levels
oI 110 to 140 mg/dL, a bit higher than the target
values initially recommended, because the literature
suggests that intensive insulin therapy with lower
target levels Ior blood glucose is associated with a
higher incidence oI hypoglycemic
events and only scant improvement
in morbidity and mortality.
7,23 Our unit has chosen a
target blood glucose level Ior
patients receiving intravenous
insulin therapy that is within the
recommended guidelines oI the
American Association oI Clinical
Endocrinologists and the American Diabetes Association
in their consensus statement on inpatient
glycemic control.18
Our Iindings concur with results oI other studies
in which diIIerent blood glucose ranges were compared,
whether in a comparison oI subcutaneous
insulin therapy and IIP24,25 or between 2 IIPs with
diIIerent ranges.6,7,12,13,15,17
Other than our study, we Iound only 6 studies14,19,22,24,26,27
in which mild hypoglycemic events were
analyzed. In most clinical trials,6,7,10,12,13,17 only severe
hypoglycemia is considered. The deIining value oI
hypoglycemia varies in these studies, which makes
comparisons diIIicult. Braithwaite et al26 reported a
2.4° incidence oI hypoglycemias less than 70 mg/dL
and did not Iind any hypoglycemic events less than
50 mg/dL. Bland et al22 reported a 2.1° incidence
oI moderate hypoglycemia (40-60 mg/dL) in the
group treated with intensive insulin therapy, and a
0.1° incidence oI severe hypoglycemias, very similar
to our results in both cases. Goldberg et al19
also obtained similar results (0.2° hypoglycemia
·70 mg/dL), whereas Osburne et al27 reported a 6.9°
incidence oI hypoglycemia in the 60 to 80 mg/dL
range and only a 0.9° incidence Ior hypoglycemia
less than 60 mg/dL. In all oI these studies, and in
our study, percentages were calculated against the
total number oI samples taken. In other studies14,17
in which slight hypoglycemia was assessed, researchers
calculated the incidence on the basis oI the number
oI patients who had at least 1 hypoglycemic episode.
With respect to severe hypoglycemia, published clinical
trials have reported Irequencies oI 5° to 18°,
much higher than our Iindings, as reIlected in a metaanalysis
done aIter NICE-SUGAR was completed.28
The diIIerences in results may be attributable to
diIIerences in methods.20,29 In addition, despite having
a similar study protocol, diIIerent centers in
other clinical trials had diIIerent routines Ior sampling
and analysis oI blood glucose levels10,17,30,31 or
considered capillary and arterial blood samples to
be interchangeable Ior purposes oI analysis.12,13,20,21,31,32
Our study addresses this gap in the literature by
comparing blood glucose determinations on the
basis oI the type oI sample analyzed. DiIIerences
between the 2 groups were Iound only when we
included arterial samples in the analysis (as in the
previous studies). This Iinding could have important
implications in treating critically ill patients because
it supports studies such as those by Scott et al,32
Lacara et al,33 and Slater-MacLean et al,34 who Iound
that capillary blood glucose values were higher than
the corresponding arterial blood glucose values, and
NICE-SUGAR,17 which recommended against use oI
Iingerstick samples Ior analysis oI blood glucose levels.
With respect to sampling procedures, the literature
establishes that blood glucose values obtained
Irom Iingerstick samples result in overestimates oI
blood glucose level in comparison with 'reIerence
standard¨ laboratory techniques.33,35-38 Bedside glucometer
readings in critically ill patients result in
overestimates oI blood glucose level39-42 and lack
precision2,18,32,34,43 Ior a variety oI reasons: Iluid accumulation
in the Iingertips, poor peripheral perIusion
due to shock or administration oI vasopressors,
anemia, sample volume insuIIicient Ior analysis, or
sampling method used.20 In addition, glucometers
must be calibrated Irequently, but such calibration
may be neglected in clinical practice. In our study,
each patient was provided with a box oI supplies
that included the patient`s assigned Optium-Xceed
glucometer, calibrated per the manuIacturer`s instructions,
44 and strips appropriate to that calibration; iI
additional strips were needed and came Irom a diIIerent
lot, the glucometer was recalibrated.
Limitations oI this study include its retrospective,
observational design and the Iact that the samples
are not simultaneous in these critically ill
patients. Case-control studies are needed to veriIy
that the results obtained are not the result oI any
clinical and/or metabolic instability in the included
patients. In addition, under our standard clinical
protocol, the hypoglycemia readings were conIirmed
by repeated measurements with point-oI-care devices
rather than by control laboratory measurements.
Even though blood samples were collected
Irom arterial catheters hourly, discarding 2 mL oI
diluted blood each time (which was returned to the
patient`s circulatory Ilow to avoid iatrogenic anemia),
little has been published about arterial catheter
sampling techniques that do not increase collateral
eIIects such as anemia, inIection, and ischemia/
pseudoaneurysm oI the radial artery. The complications
oI hypoglycemia move researchers to search
Ior systems with higher sensitivity and speciIicity to
detect hypoglycemia. In this sense, our group is conducting
a clinical trial to determine whether the use
oI an arterial catheter permits more precise monitoring
oI glycemia, as this system oIIers the important
advantage oI avoiding the hourly Iingersticks that
produce discomIort Ior both patients and nurses.19,22,45
oncIusions
With a target Ior blood glucose level oI 110 to
140 mm Hg, the number oI hypoglycemic events
detected in critically ill patients is low, regardless oI
the type oI administration oI insulin therapy (intravenous
IIP with a nurse-managed insulin therapy
algorithm or subcutaneous with sliding scales).
Nonetheless, a higher incidence oI hypoglycemias less
than 80 mg/dL was detected in arterial than capillary
blood samples with a bedside glucometer. ThereIore,
iI only arterial samples are analyzed in critically ill
patients, the incidence oI hypoglycemias could be
higher than iI capillary and arterial samples are used.
ACKNOLEDGMENTS
e thank the ÌCU nursing team for their collaboration in
collecting arterial blood samples according to the study
protocol. The authors appreciate the English language
review by Elaine Lilly, PhD.
FÌNANCÌAL DÌSCLOSURES
This project was funded in part by the Third Fellowship
in Health Sciences awarded by the Caixa de Manlleu
Foundation, Barcelona, Spain.

/ 2/ 2.

/ 80.070 2.

/ 20.03.39-090038:-.-0/8/0:..2508//349/1107831..9.5..0.2508   .:84389.90907 47130789..420907.7907.3/ 80..034:8   7.4393:4:8 397.7907.3.0.3/-44/174290.79. .3/397.0/38:3907.8..8./2389070/8:-.02.../2389070/ 397.04154.9079.508 !   43.49 .08 47.33.5..8.814:3/2470419033.02.2508  %0570.9039870.034:8   38:3907.034:85071:843 39038031:84357494.304:8 98/38.02.304:8   .02.:9.034:8 54.02.5.8841908.709-44/:.:9.:9..38:-.5..9070338:3.041 2/54.25081742.4800...79.7 -44/8.3.   9.041 94  2.304:8  !   2435..870.4792  #08:983.8:70/ 9.7    -44/ !  %0570.03. 5.07054.3:780 2.338:3397.0/2/54. 3.9039870..903984706:7038:3 ...

3.7/08841090738:38.90/3.70/090.880880/9078.0/.2508 2..:9.9.8...79.0398. 5.304:8 3.03.034:8 478:-.209.02.8841840.07570.90398 70./2389070/397. .3490780  3.0.02.90398  030709./ 1054.04154..884189:/08 5.7907.

0 38:3907.70.3.3/903.480 .807.089..97.8.-030191.8357454794394909930884190:.981742.8 .90398.. 4724..9.5 .947  3..7/4.02.9474199 :.9478 14:3/2470/0..3/24700584/084154.02. 339038.:808 ..7.339038.02..7.0.0.9039870.:843903..90789:/08  54.943.3/.33/0503/0391.&83 :. .02.:..  2.4804792#0:.439743.90/92479.70./03..8.0.//943 3.943$:7.43.0 ..2435.480./03.-0 94/7.339073.79.8033.04154..884.:..43974 3..943  $&# 3..8.5.

43. 29825-0.94341 3..3/90207..9020394335.02.8:-806:039.02.079  3.897432.70:39 & 5.3/.0799422408507907 2:95-   390!74:5.42203/0/:8041.7344898.3.70741308880.9039 .02..90398  99057494.339038.3/4.43.3!94...:/0/9.954.0.494-03-03:80/03:.438038:889./94 .-0908 884.94370.43974 90207..48094 2.3884.43974 507.70.:08.480 ...

3/ 2.30-09003 ./ 2.39.34592:2 7.33.

9.47/39490950418.88088903:2-074154.0 %04-0.8841 :.3.90/4349/1107039.47.8 94./  !:-80/.304:8 47397.0..90 -9095041-44/8.4800.7907.8 90950417.034:8  .02.3..-44/ .250 .04190570803989:/..80894/.250:80/147./23897.3.5/38:3.90398.:9.943:80/ 8:-.80.039814:3/ 3&5.0/ 130789.

.:/0/.3/30:744..3/094/8 %070974850..3/ 8:7.9039889:/0/0703.8-.438/070/ ..038:331:843 57494.4800..08..907.9080..850.8.3 2.9807.8.903984859.. 947.4 %0701470 54.80/4390.8.-44/:.02.2:95:75480  -0/&9. .808  :7!.7/..00889. ..02..3/ $05902-07   !.0/-0900357.20/.9.059.90898 41&5.089:/3..

/ 9. ..:041 94 2.

 .82/54.3 2.02.3/0889.90470/../.

..02..9.9..90949.70 20.70 0/$0380 !74/:..998.8907 :897.8:702039894.5. .031905.3 59:2.9039 8-4 418:5508390&  :7:39706:7085439 41..  -0.425.3/05939..90398.9438 .-7.3/..:20390/. 03 90..884.02.-0908.557457..02.94388.3:1..3/89758..:804:7!706:708 397.95./.-7.7/57494.9039 .70/4.420907 --499.07054.20.98 43.90/954.3 2.034:838:3.8:70203980889.825942.02.90/ 5072.4 :80/3 9889:/ .0399030:744.00/:..  0.8830/94 0.943 .:094 570..  34:789.4317254.9:7073897:.880.

7/ :3/07./  9020.5.48070:80/   90:39 83472.420907  %438:3907.2508 94038:70 .09908.20 :.89.557494.8:70203987050./390:.90/9.00889.90398 9.307747390130789.389.3 2.4800.420907-:99/11070398.36:0473 70..-44/:.90.3/.

2.308 .480 0.47/394.9079.4800./2389070/147 5..0.3/  397.08 5071:843.4800.30!390:399.903989-44/:.5/38:3 9-44/:.890 -44/:.9.041 -44/:./2389070/ ..:9.0/0.074:78.08:-..8.304:87.04.0.08 ./ 70.570/0130/8/38../:890/..0.0870.3  2.034:838:3 907..9:808 /3.3/38:3.51443.4800.0.

903989.907 9.070./475.4800.3 2.-44/:.

3/907014704:757494.:7041472470 47./42009.0.8..94190570.304:8 .3/.2508147:...7907.70/0130/.9 4:7 3907.:9..81448 $/3 8.990 907.90/.8058891.90398 309 4:/7010...4 706:7089089341-44/:.-0.3..084:7 .70./23897.0.0110.50:9.48.-0908   701 948057494.480 0.7907..07 4:78 ./2389070//4804138:3  !9/3..253 .9434138:380.3 .9 4:73907.943./23897..38 3.020.8 98:-.40.8.40.304:8.9431472470 9..90907 %0:/0301478:-. 130789..:9.8.039.2.48090893.3/38:3 /0503/039/.83! 5..:8090805..4-44/.90398/4349.057494.3.82470 1706:0398.9434138:3  -44/8.90.8941901443 ..4:8 .

1907 24701706:039 24394738706:70/94/0907230908500/9 .-44/:. .3 ..4-0.7907.041 2.943 0907908.3 970./2389070/38:3831:80/  /0503/34390-44/:.3.8.30/ 800:70 %0907.8.90907  %0907.147904907 90970.04190.9203920990.30..4800.8942.57494.50:9.-44/.92039 .0 1478:-.50:9. 9070..7907.-0 974:4:99089:/5074/1.8:70/  0147090..40.250 84-9./ 85.304:838:3907.90907 17894:78 47130789..30/174290.5.39.:9...88  241/:90/-44/8/8.034:8.397.903970.020.7907.5..7907.2508.90/147.0.70/174290/0.40.4800.9.480703907.

-44/:.4800.0 93.7.0.894./4708890 !4-0.3041 94 2.9..0.0.

447/8.20950 4138:3 85.79 4./ %08.

8:80/ 94.3/.8.3/.90.039.5.04:78/085908:85038434190!5071:843  4790 5.747.4393:.39 .28507/0.8:70/3..5.30/54./8:89.250817425.70/110703.903983.8..4800.083903.7907.0393:2-07 4170.04154  .425.:9.4800.:/0/ . 90898070.89-44/:.90398 0700..:/0/39089:/ 47907 :.304:838:3  5.438/070/89.8.903983.30/54..  $9.034:8  3.47/0/-44/:.831.8947080703.8:80/14789.-08.02.$ 4503.3/3.5  5..:/0/1475./2389070/ 09078:-.( .338:11..90398970.438039 .7 ..#080.9434190 397.8.5  %0.:00889.7.3.989.5574.9.78.-09003/110703974:58 !.14724709.4 . 8422990043 9.:/0/5.47/0/507 /./2990/9490&/:7390 89:/5074/4706:70/38:3907..:9./.38 574./03.. 1474:78474307/08590/8. 09071742.304:8 47397..989.88 $!$$.48020.907 20.7-44/8.:/0/190..507.88  48941903./8:89.7/.08507/.0!./.3 .0/070-44/:. .90398  ..90/98:-..3 $ 47.903980700.07843  $!$$ 3.3.2508415.7. 47.02.3.:08 327.-0147..8:702039870.480.-44/ :.:9..0/-904859.0  #08:98 15.903989 .4380.034:838:3907.5.4 348  .7907./0/314720/.90398 970..8.:/0/ .90398.-44/ !..3 2.02.03  032.20.3.02.3/9057494.90/ 939038..3 8.00889.70 57080390/0907.989.88  9089...

:843 390!74:5 800:70 40.:9.900704859.2.0.3///349200990.:9.50:9...304:8.5/:739024898074:85074/ 4190574-02147.1473.034:838:3.4 70.0/-49 8:-.0  ./98:-.907.034:838:30.0/397.304:8 38:3.07 :3/0790 /3.7907.706:73397.9/1107039 543983907&89.3/397.57494..034:8 38:3907.:8. 30307.0.0/  370.

:9070857.90/.9.90/ !.-0908.250.850.947 1.80   .0    .3/743.3/9020.%020.908   .8/.3480/34189:/5.90398-03970.20.3 $ -44/:.3 $  8.70.40.9039870.3/ 070 17425..3 $ .034:838:3907.02..3/20/.0..024394737010770/- 8:7.79/80.40.78 0702.250.9.3480843.8.4704390.0 .3.8/090.38:-.034:838:3907.908   05...304:838:3907.90/  0701742 5.79.-3472.5.:70   54894507.398.8     .5.0.945.:90!844.:9..:/0/80588   0..90/ 9397.5840/.79.:.5 03908.041    2.8     0.9432-.9..90398970./28843.4800.8./288439490&3.3..398  19054.5 03908.9 .90/9397.0419089:/5.943 ! .

304:838:3907.5 .3/5./  .38:-.3 $ -44/:.20.0.:9.041    2.4800./.9039870.

7/407 .0/3:70 390.48020...0/.9047.034:8 38:3907.8041397..8:7020398 -38:3.48 /09.5 04-807.:08  2.94357494./23897.9703/94.884190-44/:./ :70  ...3.

./ .:08  2.99005038041.

-44/8.78.07054.0..5.834989.250 1742..304:838:343 !8.2508.0.034:874:5.0.5.02.9039870.338:3397..02.09397..02.0398/090.3/430.304:838:3907.90/3.335.3  $   %0 54.3/9097/3.38:-.903970.02.0.38:-.0.0398070/090.5.7907.:8843 %03:2-07412/54.47020.0398   35.02.0..034:838:3 907.0. 0.94:98/110703.90/3.90/3.:9.3 $   41.02.:9.:9.:9.02.08803801430.78.831.0.90/ 0.9039870.9039870.7 -44/8.38:-./ .7907.0398  390397..90/3:898.39/110703.54. 0.989.0.039835.0398 390397.0398 070/090.304:8 74:50708570.2508 708:933.:8.0398070/090. 903:2-0784154.0.5.470 20.30/-130789.2508  4-9. 0.90/-09003 -4957494. 0..2508 ..0398070/090.438/07890 .8 !  %0 2/54.90398970.5  8.034:8 54.9039870.304:8 38:3 .90398970./.0.3  $    .3397.3/54.889 0735.48 !8.3/430.02.8  !  3.2...:9.90398 70.3397.034:874:5070859-09003  0.5   .034:8 38:3907.034:838:3 !8..8..90/   2/54...0398070 /090.02.90/ 98:-. 831.90/390.7488905.90398 70.2508  190 54..0398   35.39  0/090.034:838:3907.90///349/1107-090035.9073.78.5   !  800%..5.3/390 8:-.3397.2508  4190803 .8.0  .03983908:-.-0 $0.0.3/ 5.0.250817425.02.304:838:3907.59.78.903984..304:874:5 .5.5.470 20.:9.

.48 9/0243897.90/ 397.4800.50:9..4-0.034:8 40.084154.3/011.9.07 90:804131:84357494.9434140.90/8.0-44/:.0894. 34:7 & 904-0.0.907907.708:983 90/090..02.90398970.9.897.109.035..08 41 94 2.

4808.0398.:0839.3/90207.4803.9.425.08 .9009..5..9    :7:39.43..7090.9039870.3909..02.038:3907.3/2479.42203/0/ -0.3397...70.4807..3.02039 3247-/9.:8090907.308       9079. 0.304:8 38:3907.:041 54.04154.43974  :713/38..2/54.-0908884.0.070 54.425.3884.438/070/ %0/0133.78438/11.3.308070.34:789:/ 014:3/4389:/08      3.8.0/ 32489.:9 7.7083908089:/08 .0.425.8      4380.02.438038:889. 9.3/438.9:70 8:08989.59407 9./03.943413..70/  09073.3 2.884./03. 073.392574.97.0398070 .7843418:-.90/9..02.:79708:9841490789:/08 3./1107039-44/:.939038.034:8 38:3907.709-44/:.2..709 .9039 .0147 5.943 3907.08147-44/:.02.59.  3.3..3/! 47-09003!89 /11070397./ .7344898.4800.02.:9.42203/0/:/03084190 207.04154.9020394335.989390 70..8.-9079..02.7054790/..80889.. 3/4.

354.0.3///34913/.02./ .3  2.03980889.

7054790/./03. 3.9054./ .   2.0 4124/07.3/09.02.

90/939038.30/82.0782./ 390 74:5970.038:3907.  2.7708:98 54. 3.3/./03.5 .808 4/-0709.02.8 .07054.04180.844-9..02. .7 944:7708:983-49.

02./03.04154.8 8-:73009.390 94 2.7054790/./ 070.  3.

 0889. 3.02./ 7./03.30.014754.3/43.3 2.

3 .3.90/903.//943 /08590.7907. 5:-80/.8.38990 949..99480.0584/0  970850.25089.0390783 4907.8841-44/:.:.789:/57494.04390-..8..5.88 /430.08   47 .90/3.90/.//70880898.5390907..3.438/070/.8954.880880/ 7080.480/090723.078 .8.903984.82.-094 /110703.209...41908089:/08 .94384390 .250894 -03907.-44/8..1907 $&#.8954./ 3.084194  2:..3/3 4:789:/ 507.253 ..-0.03 3490789:/08  3./03. 97.:.34:713/38 .425.88       :789:/.079.-01475:75480841.3/.4800.02.4 /1107039.8841903:2-07 415.0832094/8  3.90.7.997-:9.3.//110703974:93081478..07054790/1706:03.8.02..08070.425090/  %0/110703.083708:982.039.9:70- .30.3/.02.73-44/:.97..3.7.87010.3../.07054.3:2-07418.

070892.:/0/..//943 :.08  970850.80 .843    147.3 90.79.3..4800.7-44/:.943 2.253574.480.90/1706:039 -:98:.84381:/.36:08   0/8/0:. 414:3/ 9.7094170.557457.420907 70.99.49909..5./083.7034990708:941.302.5.5.420907 ..93.2508 .425.42203/0/.30/ 1742130789.90/507902.9..89758070300/0/.94383970.70.9431 .88 .93.0 34:789:/  0.-.4:2038:11. 5.8574.:80 98:55479889:/088:.8841-44/:.7907.3/.90/  29.070892.039147.8830/ 59:2 .0.3/.90398708:93 4.2508390..9.78439 7010703.. .08 -090039074:5807014:3/43030 3.00/ :.0/:708 90907.3/...420907.3.8457088478  .70300/0/94.201742.-7..3:1.-7.0 .0  4-807.2508147.7034982:9..09.2508708:934.990708:984-9.3.3.025479...3/901..9.907 ..5071:843 /:09484.:084-9.480.8.9948.3/$.250.8.90949.90841-44/:.. 8.4800.9438   .3/./23897.  .8.943.79..4:/.309.90841 -44/:.304:839080.9:70 089.389:8041 130789...90398 .3/  $&# .94790./0/9.:/0/905.071 9.:2:...8390 570.90398-0.250.7907.9-44/:.30/..3/89758.-7.47708543/3.9438419889:/3.-8089..7/ .03.2532094/:80/  3.4209078 2:89-0.943 3901307958 5447507507.-0300.-44/:...0 89.9039 8..39 25. 570.9.-7.4800.9:707 83897:.4:889:/08 %813/3.-47.89480-$.79.7..3..//943.:08 .5..:/09870974850.94341.4397489:/08.3..-7.:08070079.3 ..884190950418.870.480.-4418:5508 9./1107039 49 90:./383..90/3..88 47 8..57.47.9039..9908.0/ 110703..

-93903..7/3241 /:90/-44/0.0394:-44/8.90907 8.3. 0110.7907.920 . 57494.8:702039895439 41 .-4.97403.907.43974.08 7.70.:..  990.302..7/.70/0. 310..2508070.80.4.40...3-.8-0035:-80/..8:7020398  .9471494..47209.90398 3.389.943 .//943 :3/074:789.7907.-4:9.02..36:089.25390.988:.8709:730/9490 5.431720/ -7050.02.4/..9039 8.4 9054.:/0/ 5.9090784:7 /8.8...90/20.3/8.94720.7.70.90/ 1742.302.-47.9/43493./38070.3/..9079.

07080.709147-44/:.39.3.9.425./23897.421479147-495.7.0398 /090.9.9039884 70..02..5 397.041..3/850.7/08841 9095041.30:78241907.94/0907230090790:80 41.43/:.9434138:3907.802439473 41.909075072982470570.7.7907..3. 39880380 4:774:58.0789480.93 .90/3.02..:8438 9.994 /090.9 574/:./.7907 %0.79.1.89.9438 4154.954. . 580:/4.4800.5..0.034:8 .041 94  22 903:2-074154.97.0/8.4/3904:7130789../..24.39 .90398.3/3:7808    43.. 14788902890780389.02..02.898889024110789025479.

3.!9.8088 9.08  43090088 .073.3 2.02.3:780 2.04154.304:898/38..4792478:-.0/38:3907./03.5 .:9.

/.8/090.5.7907.5.90398 903.47/3949089:/ 57494.8.40.943 .. $5.70.3/.390&3:78390.8.31.3.3  .93. 5.2147907.7907.-44/8.30  !  $ $&#$ %85740.0/3.5570.81:3/0/35.2508.7.4 %0.3:.-0/8/0:.909038.04154.8./0.70:80/   %$ 09.03.7/0/-90.79-90%7/0485 30..2508.90/3./03.25089.9$.-47.7...9433 ..08.043..7 -44/8.0 70.4.:9478..7907.2508.79.4:/-0 079.9.420907 %0701470  143.30: 4:3/..9.3.02.7907.0-.