Albumine: le retour?
Jean-Paul Mira
CHU Cochin / GH Paris-centre
jean-paul.mira@cch.aphp.fr
Liens d’intérêt
• LFB
• Fresenius
• Baxter
Dose-dependent effects
Patel A. BMJ 2014; 349; 4561
Rochberg B. Ann Intern Med 2014; 161: 347
Prospective, randomized, double-blinded trial
16 ICU (Australia, New Zeland)
Intravascular fluid resuscitation by 4% albumin infusion
(treated group) or saline NaCl 0.9% infusion (control
group)
6997 patients
Treated group: 3497 patients
Control group: 3500 patients
Primary outcome:
death from any cause at 28-day period after randomization
N Engl J Med 2004, 350:2247-56
SAFE study subgroup analysis
Treated %
13.6
Control %
10.0
P
0.06
Severe sepsis patients
30.7
35.3
0.09
ARDS patients
39.3
42.4
0.72
Trauma patients
N Engl J Med 2004, 350:2247-56
Adjusted Odds ratio for mortality
Product
20% Albumin
Primary aims
↓Mortality D28
↓Mortality D28 and D90
Secondary aims
↓Mortality D90
↓SOFA score, LOS,…
↓SOFA score, LOS, …
Patients
Septic shock
N=800
Severe Sepsis/ S Shock
N=1800
Timing administration
First 6H after shock
D1
Stratification
Community vs
Nosocomial
Early (<H6) vs Late (H6H24)
20gx3/D for 3D vs saline
20gx3 at D1 then maintain
albuminemia>30g/L until
D28
Administration
Caironi P. N Engl J Med 2014
Caironi P. N Engl J Med 2014
Baseline characteristics
Albumin group
Control group
(n = 399)
(n= 393)
Age, median [IQR], years
66 [55;76]
66 [55;77]
SAPS2, median [IQR]
51 [40;66]
51 [41;64]
SOFA inclusion, median [IQR]
10 [8;12]
10 [8;12]
Mechanical ventilation, No. (%)
327 (83)
332 (86)
Renal replacement therapy, No. (%)
90 (23)
90 (23)
Lactate levels, median [IQR], mg/dL
2.1 [1.4;3.7]
2.3 [1.5;4]
Baseline characteristics
Albumin group
Control group
(n = 399)
(n= 393)
Age, median [IQR], years
66 [55;76]
66 [55;77]
SAPS2, median [IQR]
51 [40;66]
51 [41;64]
SOFA inclusion, median [IQR]
10 [8;12]
10 [8;12]
Mechanical ventilation, No. (%)
327 (83)
332 (86)
Renal replacement therapy, No. (%)
90 (23)
90 (23)
Lactate levels, median [IQR], mg/dL
2.1 [1.4;3.7]
2.3 [1.5;4]
Albuminemia D0, median [IQR], g/L
17.8 [14;21]
18.1 [14;22]
Caironi P. N Engl J Med 2014
Albuminemia
Albumin group
g/L
Control group
40
35
30
25
20
15
10
5
0
*
*
D3
D4
*
*
H-12
D1
D2
% of patients with Albuminenia < 25 g/L
H-12
D1
D2
D3
D4
Albumin group
88
60
38
23
36
Control group
88
91
93
93
93
Caironi P. N Engl J Med 2014
Risk of Death at 90 Days, according to Subgroup Analysis
Caironi P. N Engl J Med 2014
Survival Day 28 (ITT)
Survival
1.0
Albumin group
0.8
Mortality :
Control group
Albumin group: 24.1 %
0.6
Control group: 26.3%
P = 0,43
Absolute mortality reduction: 2.2%
Relative mortality reduction: 8.4 %
0.4
0
5
10
15
20
25
days
Survival Day 28
per protocol (n=682)
Death
Mortality :
Albumin group: 18.8 %
Control group: 23.7%
P = 0,1
Absolute mortality reduction: 4.9%
Relative mortality reduction: 20.6%
Control group
Albumin group
Caironi P. N Engl J Med 2014
Secondary outcomes
2000
P= 0,37
1800
Cristalloïds
1600
P= 0,03
1400
Albumin
Contrôle
1200
1000
P= 0,02
800
P= 0,22
600
P= 0,48
400
200
0
H-12
D1
D2
D3
D4
Secondary objectives
Albumin group Control group
Mortality at D 90
(n = 399)
(n= 393)
p
138 (34,7)
138 (35,1)
0,94
15 [0;23]
13 [0;22]
0,24
No of days, median [IQR]
MV-free within D28
Catecholamine-free within D28
24 [13,7;26]
Nosocomial infection incidence, No (%) 93 (23,3)
23 [8,7;25]
0,038*
90 (23)
0,67
Length of stay, median [IQR], days
ICU
11 [5;20]
10,5 [5;21]
0,61
Hospital
24 [13;38]
23 [11;37]
0,45
Renal injury
Creatininemia
evolution
Albumin group
Control group
Creatininemia
µmol/L
Patient RRT
%
300
250
25
200
150
100
20
50
0
5
15
10
D1
D2
D3
D4
0
D1
D2
D3
D4
Albumin group Control group
RRT-free within D28
28 [16;28]
28 [12;28]
0,49
Fluid Therapy
1. We recommend crystalloids as the initial fluid of choice (grade
1B)
2. We recommend against the use of hydroxyethyl starches (HES)
for fluid resuscitation of severe sepsis and septic shock (grade
1B)
3. We suggest the use of albumin in the fluid resuscitation of severe
sepsis and septic shock when patients require substantial
amounts of crystalloids (grade 2C).
Dellinger RP. Intensive Care Med 2013; 39: 165
Albumin in the management of sepsis?
• Albumin administration should be prescribed in complicated
cirrhotic patients
• Albumin administration might be prescribed in septic shock
–
after initial crystalloid infusion (>50/60/70/…ml/kg?) AND
– when albumin level is below 25 g/L AND
– when hemodynamic condition remains instable
• Prefer albumin 20% and low infusion rate
• Target: albuminemia at 30g/L
• Duration of treatment: short