Variability of insulin sensitivity during the first 4
days of critical illness
C. Pretty, A. Le Compte, J. G. Chase, G. Shaw, S. Penning, J-C Preiser, T. Desaive
Insulin
Introduction
Brain
Insulin sensitivity defines the metabolic balance between insulin concentration and
insulin-mediated glucose disposal.
→
Insulin losses
(liver, kidneys)
Blood
Glucose
When using exogenous insulin, variable insulin sensitivity can cause highly
variable outcome glycaemia.
Plasma
Insulin
Effective
insulin
Pancreas
Glucose
Insulin sensitivity is affected by the acute stress response to critical injury.
This study quantifies and compares the evolution of insulin sensitivity level and
variability for critical care patients receiving glycaemic control during their first 4 days
of ICU stay.
Liver
Other
cells
G   pG .G (t )  SI .G (t )
Q(t )
P(t )  EGP  CNS

1   G Q(t )
VG
Q  nI ( I (t )  Q(t ))  nC
Q(t )
1   G Q(t )
I  nK I (t )  nL
uex (t )
uen (G )
I (t )
 nI ( I (t )  Q(t )) 
 (1  xL )
1   I I (t )
VI
VI
Patients
Subjects & Methods
N
Age (yrs)
Gender (M/F)
APACHE II score
APACHE II ROD (%)
Operative/Non-Operative
Hospital mortality
ICU mortality
ICU length of stay (hrs)
Diabetic history: Type I/Type II
A retrospective analysis of patient data from the SPRINT tight glycaemic control (TGC)
study in the Christchurch Hospital ICU.
All patients commenced TGC within 12 hours of ICU admission and spent at least 24
hours on the SPRINT protocol.
Model-based insulin sensitivity (SI) was identified each hour for every patient.
Absolute level and hour-to-hour percent changes in SI were assessed on cohort and perpatient bases.
Levels and variability of SI were compared over time on 24-hour and 6-hour timescales
for the first 4 days of ICU stay.
164
65 [56-74]
102/62
19 [16-25]
32 [17-52]
66/98
25%
18%
142 [70-308]
10/22
24-hour analysis
Cohort level analysis
Per-patient level analysis
1
Results
Cohort analysis
0.9
Per-patient analysis
0.8
Cohort and per-patient median SI levels increased by
34% and 33% (p<0.001) between days 1 and 2 of ICU stay.
SI Level
Cohort and per-patient SI variability reduced by 32% and
36% (p<0.001).
Days 1-2
34
<0.0001
33
0.0004
Days 2-3
16
<0.0001
21
0.2559
Days 3-4
Analysis of the first 24 hours using 6-hour blocks of SI
data showed that most of the improvement in insulin
sensitivity level and variability seen between days 1 and 2
occurred during the first 12-18 hours of day 1.
6
p-value
% Increase
at median
0.0013
4
0.7
p-value
0.6
F(x)
% Increase
at median
0.5
0.4
0.3
0.2
0.6306
0-24hrs (3936 hours)
24-48hrs (3376 hours)
48-72hrs (2568 hours)
72-96hrs (2187 hours)
0.1
0
0
0.5
1
1.5
S [L/mU.min]
0-24hrs (164 patients)
24-48hrs (155 patients)
48-72hrs (112 patients)
72-96hrs (95 patients)
0
0.5
x 10
I
1
1.5
S [L/mU.min]
-3
-3
x 10
I
Cohort variability analysis
Per-patient variability analysis
1
Cohort analysis
This rapid improvement was likely due to the decline of
counter-regulatory hormones as the acute phase of
critical illness progressed.
0.9
Per-patient analysis
0.8
p-value
%
Reduction
at median
0.7
p-value
0.6
F(x)
SI Variability
%
Reduction
of IQR
0.5
0.4
Days 1-2
Days 2-3
Days 3-4
ICU patients have significantly lower and more variable
insulin sensitivity on day 1 than later in their ICU stay
and particularly during the first 12 hours.
32
20
14
<0.0001
0.0028
0.0269
36
18
17
<0.0001
0.0091
0.0369
0.3
0.2
0.1
0
-100
Clinically, these results suggest that while using TGC
protocols with patients during their first few days of ICU
stay, extra care should be afforded.
0-24hrs (3772 hours)
24-48hrs (3221 hours)
48-72hrs (2456 hours)
72-96hrs (2092 hours)
-50
0
Percentage change (%)
50
100
0-24hrs (162 patients)
24-48hrs (155 patients)
48-72hrs (112 patients)
72-96hrs (94 patients)
0
50
100
150
200
250
300
50% range of hour-to-hour variability
6-hour analysis
Cohort level analysis
Per-patient level analysis
1
0.9
0.8
SI Level
0-6 vs. 6-12hrs
6-12 vs. 12-18hrs
12-18 vs. 18-24hrs
18-24 vs. 24-48 hrs
%
Increase
at median
42
28
1
9
p-value
<0.0001
<0.0001
0.0335
0.0452
Per-patient analysis
%
Increase
at median
40
26
3
7
0.6
F(x)
Cohort analysis
0.75
0.7
p-value
0.5
0.4
0.25
0.3
0.0007
0.0123
0.4829
0.3776
0-6hrs (984 hours)
6-12hrs (984 hours)
12-18hrs (984 hours)
18-24hrs (984 hours)
24-48hrs (3376 hours)
0.2
0.1
0
0
0.5
1
1.5
S [L/mU.min]
0-6hrs (164 patients)
6-12hrs (164 patients)
12-18hrs (164 patients)
18-24hrs (164 patients)
24-48hrs (155 patients)
0
0.5
x 10
I
1
0
50
100
IQR2
-3
IQR1
x 10
I
Cohort variability analysis
-50
1.5
S [L/mU.min]
-3
-100
Per-patient variability analysis
1
0.9
0-6 vs. 6-12hrs
6-12 vs. 12-18hrs
12-18 vs. 18-24hrs
18-24 vs. 24-48hrs
Per-patient analysis
%
Reduction p-value
at median
36
< 0.0001
28
0.0673
9
0.1032
14
0.1075
0.8
% Reduction of IQR =
0.7
0.6
F(x)
SI Variability
Cohort analysis
%
Reduction p-value
of IQR
40
0.0017
24
0.0628
0
0.0931
18
0.1682
0.5
0.4
0.3
0-6hrs (820 hours)
6-12hrs (820 hours)
12-18hrs (820 hours)
18-24hrs (820 hours)
24-48hrs (3221 hours)
0.2
0.1
0
-100
-50
0
Percentage change (%)
50
100
0-6hrs (149 patients)
6-12hrs (161 patients)
12-18hrs (163 patients)
18-24hrs (163 patients)
24-48hrs (155 patients)
0
50
100
150
200
250
300
50% range of hour-to-hour variability
350
400

IQR1  IQR 2 
100 
IQR1
350
400