Initial resuscitation guided by the Surviving Sepsis Campaign recommendations and early echocardiographic assessment of hemodynamics in intensive care unit septic patients: A pilot study*.

Abstract

: To compare therapeutic interventions during initial resuscitation derived from echocardiographic assessment of hemodynamics and from the Surviving Sepsis Campaign guidelines in intensive care unit septic patients.
: Prospective, descriptive study in two intensive care units of teaching hospitals.
: The number of ventilated patients with septic shock who were studied was 46. Transesophageal echocardiography was first performed (T1 < 3 hrs after intensive care unit admission) to adapt therapy according to the following predefined hemodynamic profiles: fluid loading (index of collapsibility of the superior vena cava ≥36%), inotropic support (left ventricular fractional area change <45% without relevant index of collapsibility of the superior vena cava), or increased vasopressor support (right ventricular systolic dysfunction, unremarkable transesophageal echocardiography study consistent with sustained vasoplegia). Agreement for treatment decision between transesophageal echocardiography and Surviving Sepsis Campaign guidelines was evaluated. A second transesophageal echocardiography assessment (T2) was performed to validate therapeutic interventions.
: Although transesophageal echocardiography and Surviving Sepsis Campaign approaches were concordant to manage fluid loading in 32 of 46 patients (70%), echocardiography led to the absence of blood volume expansion in the remaining 14 patients who all had a central venous pressure <12mm Hg. Accordingly, the agreement was weak between transesophageal echocardiography and Surviving Sepsis Campaign for the decision of fluid loading (κ: 0.37 [0.16;0.59]). With a cut-off value <8 mm Hg for central venous pressure, κ was 0.33 [-0.03;0.69]. Inotropes were prescribed based on transesophageal echocardiography assessment in 14 patients but would have been decided in only four patients according to Surviving Sepsis Campaign guidelines. As a result, the agreement between the two approaches for the decision of inotropic support was weak (κ: 0.23 [-0.04;0.50]). No right ventricular dysfunction was observed. No patient had anemia and only three patients with transesophageal echocardiography documented left ventricular systolic dysfunction had a central venous oxygen saturation <70%.
: A weak agreement was found in the prescription of fluid loading and inotropic support derived from early transesophageal echocardiography assessment of hemodynamics and Surviving Sepsis Campaign guidelines in patients presenting with septic shock.

Full text

Page 1

Crit Care Med 2012 Vol. 40, No. 10 2821
I
for the need for optimal evaluation and
monitoring of the hemodynamic status of
patients with severe sepsis or septic shock.
The Surviving Sepsis Campaign (SSC) has
implemented graded recommendations
to optimize hemodynamics in early
ntricate
contribute to the development of
circulatory failure in septic shock,
including hypovolemia, vasoplegia,
and cardiac dysfunction (1). This accounts
mechanisms may
sepsis (2). These guidelines are mainly
based on a randomized trial conducted
in an emergency department prior to
initial resuscitation, which showed that
goal-directed management with targeted
values of central venous pressure (CVP),
mean arterial pressure, and central venous
oxygen saturation (ScVO2) significantly
decreased hospital mortality in patients
with sepsis-induced circulatory failure (3).
This approach remains controversial in the
intensive care unit (ICU) setting (4). CVP
is a poor predictor of fluid responsiveness
(5). The detection of cardiac failure
could be challenging because septic
cardiomyopathy is not associated with
markedly increased filling pressures (6)
and tissue extraction abnormalities could
lower the sensitivity of ScVO2 in detection
of low flow states (7).
Over the last 20 yrs, echocardiography
has gained a pivotal role in the hemody-
namic assessment of ICU patients with
cardiopulmonary compromise (8). It
rapidly provides unparalleled informa-
tion on both the volume status and car-
diac function of unstable patients (9), and
results in immediate therapeutic impact
using standardized algorithms, especially
in patients with septic shock (10).
We sought to compare the therapeutic
interventions derived from early echocar-
diographic assessment of hemodynamics
Objective: To compare therapeutic interventions during ini-
tial resuscitation derived from echocardiographic assessment of
hemodynamics and from the Surviving Sepsis Campaign guide-
lines in intensive care unit septic patients.
Design and Setting: Prospective, descriptive study in two inten-
sive care units of teaching hospitals.
Methods: The number of ventilated patients with septic shock
who were studied was 46. Transesophageal echocardiography
was first performed (T1 < 3 hrs after intensive care unit admission)
to adapt therapy according to the following predefined hemody-
namic profiles: fluid loading (index of collapsibility of the superior
vena cava ≥36%), inotropic support (left ventricular fractional area
change <45% without relevant index of collapsibility of the supe-
rior vena cava), or increased vasopressor support (right ventricular
systolic dysfunction, unremarkable transesophageal echocardiog-
raphy study consistent with sustained vasoplegia). Agreement for
treatment decision between transesophageal echocardiography
and Surviving Sepsis Campaign guidelines was evaluated. A sec-
ond transesophageal echocardiography assessment (T2) was per-
formed to validate therapeutic interventions.
Results: Although transesophageal echocardiography and Sur-
viving Sepsis Campaign approaches were concordant to manage
fluid loading in 32 of 46 patients (70%), echocardiography led to the
absence of blood volume expansion in the remaining 14 patients
who all had a central venous pressure <12 mm Hg. Accordingly, the
agreement was weak between transesophageal echocardiography
and Surviving Sepsis Campaign for the decision of fluid loading (κ:
0.37 [0.16;0.59]). With a cut-off value <8 mm Hg for central venous
pressure, κ was 0.33 [−0.03;0.69]. Inotropes were prescribed based
on transesophageal echocardiog raphy assessment in 14 patients but
would have been decided in only four patients according to Surviv-
ing Sepsis Campaign guidelines. As a result, the agreement between
the two approaches for the decision of inotropic support was weak
(κ: 0.23 [−0.04;0.50]). No right ventricular dysfunction was observed.
No patient had anemia and only three patients with transesophageal
echocardiography documented left ventricular systolic dysfunction
had a central venous oxygen saturation <70%.
Conclusions: A weak agreement was found in the prescription
of fluid loading and inotropic support derived from early trans-
esophageal echocardiography assessment of hemodynamics and
Surviving Sepsis Campaign guidelines in patients presenting with
septic shock. (Crit Care Med 2012; 40:2821–2827)
Key Words: echocardiography; septic shock; surviving sepsis
Campaign guidelines
initial resuscitation guided by the Surviving Sepsis campaign
recommendations and early echocardiographic assessment of
hemodynamics in intensive care unit septic patients: a pilot study*
Koceila Bouferrache, MD; Jean-Bernard Amiel, MD; Loïc Chimot, MD; Vincent Caille, MD; Cyril Charron, MD;
Philippe Vignon, MD, PhD: Antoine Vieillard-Baron, MD, PhD
*see also p. 2911.
From the Intensive Care Unit (KB, LC, VC, CC,
AV-B), Section Thorax – Vascular diseases – Abdomen –
Metabolism, University Hospital Ambroise Paré,
Boulogne, France; Faculté de Médecine Paris Ile de
France Ouest (KB, LC, VC, CC, AV-B), Université de
Versailles Saint Quentin en Yvelines, Versailles, France;
Intensive Care Unit (JBA, PV), CHU de Limoges, Limoges,
France; Center of Clinical Investigation (JBA, PV), Inserm
0801, University of Limoges, Limoges, France.
Supplemental digital content is available for this
article. Direct URL citations appear in the printed text
and are provided in the HTML and PDF versions of this
article on the journal’s Web site (http://journals.lww.
com/ccmjournal).
The authors have not disclosed any potential con-
flicts of interest.
For information regarding this article, Email: antoine.
vieillard-baron@apr.aphp.fr
Copyright © 2012 by the Society of Critical Care
Medicine and Lippincott Williams and Wilkins
DOI: 10.1097/CCM.0b013e31825bc565

Page 2

2822 Crit Care Med 2012 Vol. 40, No. 10
and from SSC guidelines during the initial
resuscitation of patients presenting with
septic shock on ICU admission. We spe-
cifically hypothesized that discrepancies
could exist between the two approaches
in the initiation of fluid loading and ino-
tropic support.
MATERIAL AND METHODS
Because transesophageal echocardiog-
raphy (TEE) and central venous and arterial
catheterizations are routinely performed in all
patients with severe sepsis and septic shock
in the two participating ICUs, our study was
considered as part of routine clinical prac-
tice by the Ethics Committee of the “Société
de Réanimation de Langue Française” (study
number 08-237/CE-SRLF) and no informed
consent was required from the patients or
their next of kin.
The study was prospectively conducted
over an 18-month period in the ICUs of two
teaching hospitals. Patients were included in
the study if they presented directly to the ICU
with septic shock, without previous resusci-
tation in the emergency department, and if
they required invasive mechanical ventilation.
Septic shock was defined as sepsis associated
with persistent hypotension after fluid loading
(11, 12) or with the presence of clinical signs of
tissue hypoperfusion confirmed by biochemi-
cal markers, such as elevated blood lactate or
base deficit (13). Exclusion criteria were a con-
traindication to TEE, ongoing pregnancy, age
<18 yrs, or moribund state.
All studied patients underwent a diagnostic
workup and therapeutic management of sepsis
according to our standard of care. Before the
third hour following ICU admission, the patient
was sedated and put on mechanical ventilation.
A central venous catheter was inserted in the
internal jugular or subclavian vein, and blood
pressure was monitored invasively. The correct
position of the tip of the central venous catheter
was confirmed by an anteroposterior chest ra-
diography performed at the bedside. The carina
was used as a landmark for detecting appropri-
ate positioning of the central venous catheter at
the junction of the superior vena cava and the
right atrium (14). All patients received at least
20 mL/kg of crystalloid in <1 hr. In the presence
of a diastolic blood pressure <40 mm Hg (15),
norepinephrine was immediately administered
simultaneously to volume expansion at a start-
ing dose of 0.125 μg/kg/min. Intravenous an-
tibiotic therapy was started within the first 2
hrs after admission, after bacteriological sam-
pling, and the entry site of infection was eradi-
cated when indicated. Ventilatory settings were
standardized as follows: tidal volume 8 mL/kg,
positive end-expiratory pressure according to
oxygenation variables and lung compliance
(usually ≤10 cm H2O), and plateau pressure
limited to 30 cm H2O.
TEE was performed under the ongoing
sedation used for mechanical ventilation, and
muscle relaxants if necessary, to have the pa-
tient perfectly adapted to the respirator (T1).
No change in sedation was made during the
study period. The airway pressure and the elec-
trocardiogram were displayed on the screen
of the upper-end systems (Sequoia C 256,
Siemens, PA and CX50 Philips Healthcare,
Andover, MA), which were connected with a 5
or 7 MHz multiplane esophageal probe. In the
four-chamber transesophageal view, we mea-
sured both the left ventricular (LV) and right
ventricular (RV) end-diastolic area (EDA) and
calculated the ratio of RV end-diastolic area
and LVEDA. In the transgastric short-axis view
at the midpapillary muscle level, we measured
the LVEDA and LV end-systolic area and com-
puted LV fractional area change (LVFAC) as
(LVEDA−LV end-systolic area)/LVEDA. From
this view, the probe was electronically rotated
110°–120° to measure pulse wave Doppler
velocities of the ejection flow within the LV
outflow track (16). We then measured the
velocity-time integral immediately above the
aortic annulus and calculated the LV stroke
volume by multiplying the velocity-time in-
tegral by the cross-sectional area of the LV
outflow tract measured at the midesophageal
view at 120° (17). Cardiac output was obtained
by multiplying the LV stroke volume by the
heart rate. Finally, from a transesophageal 90°
longitudinal examination of the superior vena
cava, we measured the maximal diameter of
the vessel during expiration and the minimal
diameter during inspiration and calculated the
superior vena cava collapsibility index (ΔSVC)
as, (maximal diameter of the vessel during
expiration−minimal diameter during inspira-
tion)/maximal diameter of the vessel during
expiration (18).
The interpretation of the TEE examination
performed at T1 led us to perform three thera-
peutic interventions based on distinct hemo-
dynamic profiles, as a standard of care in our
ICUs (Fig. 1). Fluid loading was performed in
the presence of marked respiratory variations
of the superior vena cava, as reflected by a
ΔSVC ≥36% (18). Inotropic support was initi-
ated or increased in the presence of septic car-
diomyopathy characterized by an LVFAC <45%
without relevant superior vena cava respiratory
variations (19). Increased vasopressor infusion
was performed in the presence of RV systolic
dysfunction defined as an RV end-diastolic
area/LVEDA ratio >1 (20) and when persistent
vasoplegia was the suspected mechanism for
the circulatory failure. Sustained vasoplegia
was suspected in the absence of TEE criteria
for fluid loading and inotropic support, and in
the absence of any other echocardiographic ab-
normality that could account for the circulato-
ry failure associated with ongoing septic shock
(e.g., acute valvular regurgitation secondary to
an infective endocarditis, tamponade related to
a purulent pericarditis).
During the TEE examination, systolic
blood pressure, diastolic blood pressure, mean
arterial pressure, CVP, and heart rate were re-
corded. Blood gases, base deficit, and lactate
were measured in an arterial blood sample,
and ScVO2 was measured in a sample of blood
obtained from the distal lumen of the central
venous catheter. The physicians in charge
of the patients were blinded to the values of
ScVO2 and CVP measured by the nurses un-
til the patient completed the study period.
Potential treatment interventions according to
therapeutic targets of the SSC guidelines were
virtually determined offline by an independent
figure 1. Therapeutic interventions in the 46 included septic patients according to echocardiographic
evaluation and Surviving Sepsis Campaign guidelines. ΔSVC, superior vena cava collapsibility index;
LVFAC, left ventricular fractional area contraction, CVP, central venous pressure; MAP, mean arterial
pressure; ScVO2, central venous oxygen saturation.

Page 3

Crit Care Med 2012 Vol. 40, No. 10 2823
investigator who had no access to TEE results
and interpretation (Fig. 1).
A new TEE assessment (T2) was performed
90–180 mins after the hemodynamic optimiza-
tion, which followed the initial TEE examina-
tion, to validate the therapeutic interventions
decided at T1. Blood pressure, heart rate, CVP,
blood gases, base deficit, lactate, and ScVO2
were measured again.
In each patient, the Simplified Acute
Physiology Score II (21) and the Sequential
Organ Failure Assessment score (22) were cal-
culated. The severity of an underlying medical
condition was stratified according to the crite-
ria of McCabe and Jackson (23) as rapidly fatal
(2), ultimately fatal (1) or not fatal (0).
Statistical Analysis. Continuous variables
were expressed as median (1st–3rd quartile).
Between-group comparisons were performed
using the Wilcoxon rank sum test for con-
tinuous variables. The agreement between
the hemodynamic profiles determined at T1
from the TEE interpretation and according
to SSC guidelines was assessed by means of
unweighted Cohen’s κ coefficient (95% con-
fidence interval), interpretation of which is
formalized by the table of Landis and Koch
(24). The concordance of the two approaches
was considered as a disagreement (κ <0), very
weak (κ, 0–0.20), weak (κ, 0.21–0.40), moder-
ate (κ, 0.41–0.60), strong (κ, 0.61–0.80), or as
excellent (κ, 0.81–1.00). A value of p < .05 was
considered significant.
RESULTS
During the study period, 62 mechani-
cally ventilated patients with septic
shock were screened and 46 were finally
included. Sixteen patients were excluded
as follows: seven patients were consid-
ered as moribund, seven patients had
a contraindication to TEE (esophageal
surgery, n = 4; esophageal varices: n = 3),
and two patients were finally diagnosed
with cardiogenic shock complicated by
infection.
Demographic characteristics are pre-
sented in Table 1. In 69% of cases, the
infection was community acquired with
a predominance of pneumonia. At the
time of the initial TEE examination (T1),
which was performed 2.0 (1.4–3.6) hrs
after admission, 39 patients (85%) were
receiving norepinephrine and five patients
(11%) epinephrine. Lactate level was 3.2
(2.5–5.1) mmol/L and base deficit 7.7
(4.2–11.0) mmol/L (Table 2). Five patients
(11%) fulfilled a single therapeutic tar-
get of the SSC, while 25 patients (54%)
and 13 patients (28%) met two and three
SSC criteria, respectively. The remaining
patients (7%) were not in accordance with
the SSC guidelines. The second TEE eval-
uation (T2) was performed 3.7 (2.5–5.6)
hrs later to assess the efficacy of treat-
ment interventions performed at T1. ICU
mortality was 32%.
The main results are illustrated
in Fig. 2. Although TEE and SSC
approaches were concordant to manage
fluid loading in 32 of 46 patients (70%),
echocardiography depicted a ΔSVC
<36% in the remaining 14 patients who
had a CVP <12 mm Hg (Fig. 2A). Accord-
ingly, the agreement for the decision of
fluid loading was weak between echocar-
diography and the SSC, with a κ of 0.37
(0.16–0.59]. With a cut-off value <8 mm
Hg for CVP, κ was 0.33 (−0.03 to 0.69)
(see detais in Supplemental Figure 1;
Supplemental Digital Content 1, http://
links.lww.com/CCM/A513).
were prescribed based on TEE assess-
ment in 14 patients but would have
been considered in only four patients
according to SCC guidelines (Fig. 2B).
As a result, agreement between TEE
and the SCC was weak, with a κ of
0.23  (−0.04 to 0.50). A disagreement
(κ, −0.12 [−0.41 to 0.16]) was also
observed between echocardiography
and SCC for the initiation or increase
of vasopressor support (Fig. 2C) because
the latter was proposed in the absence
of requirement of fluids or inotropes.
Inotropes
In the subset of patients who received
fluid loading based on the presence of a
ΔSVC ≥36% (43 [39–45] %) at T1, CVP was
8 (7–8) mm Hg and ScVO2 81 (74–87) %
(Table 3). After fluid loading (T2, 3000
table 1. Demographic and clinical characteristics
on intensive care unit admission
Age (yrs)
Sex ratio (male/female)
Body mass index (kg/m²)
McCabe score (n)
0
1
2
Simplified Acute Physiology
Score II
Sequential Organ Failure
Assessment score
Community-acquired infection
Pneumonia
Peritonitis
Mesenteric infarction
Bacteremia
Urosepsis
Meningitis
Cholangitis
Other
Heart rate (beats/min)
Systolic blood pressure
(mm Hg)
Diastolic blood pressure
(mm Hg)
Mean arterial pressure
(mm Hg)
Temperature (°C)
pH
Lactate (mmol/L)
Base deficit (mmol/L)
63 (54–76)
30/16
26 (19–37)
28
13
5
51 (37–59)
8 (4–12)
32 (69%)
19 (42%)
11 (24%)
5 (11%)
2 (4%)
2 (4%)
2 (4%)
2 (4%)
3 (7%)
107 (93–128)
95 (84–119)
54 (41–62)
61 (47–68)
36.7 (36.1–37.4)
7.24 (7.12–7.42)
4.8 (3.6–6.3)
8.1 (5.2–11.9)
table 2. Hemodynamic evaluation at T1 in the
overall population
Variables at T1
Heart rate (beats/min)
Systolic blood pressure
(mm Hg)
Diastolic blood pressure
(mm Hg)
Mean arterial pressure
(mm Hg)
Central venous pressure
(mm Hg)
Central venous oxygen
saturation (%)
Left ventricular
fractional area
contraction (%)
Superior vena cava
collapsibility index
(%)
Cardiac index (L/min/
m2)
Lactate (mmol/L)
Base deficit (mmol/L)
Cumulative volume
expansion (mL)
Norepinephrine, n
(µg/kg/min)
Dobutamine, n
(µg/kg/min)
Epinephrine, n
(µg/kg/min)
105 (91–127)
99 (87–117)
55 (44–62)
71 (61–79)
12 (9–16)
79 (76–86)
51 (32–57)
14 (4–24)
2.18 (1.83–2.65)
3.2 (2.5–5.1)
7.7 (4.2–11.0)
1125 (750–2000)
39 (0.306 [0.123–0.575])
1 (5.0)
5 (0.256 [0.133–0.357])
figure 2. Contingency tables for the decision of
fluid loading, inotropic, and vasopressor supports
according to transesophageal echocardiography
(TEE) and the Surviving Sepsis Campaign (SSC)
guidelines.

Page 4

2824 Crit Care Med 2012 Vol. 40, No. 10
[1875–3125] mL), ΔSVC decreased to 13
(6–19) % and CVP increased to 13 (9–14)
mm Hg, while cardiav index (CI) signifi-
cantly increased from 2.1 (1.9–2.5) L/min/
m2 at T1 to 2.5 (2.4–2.8) L/min/m2 at T2
(Table 3). Catecholamine infusions were
substantially tapered. In the remaining
38 patients, ΔSVC was <36% (9 [3–19]
%), while mean CVP was 14 (10–17) mm
Hg and ScVO2 79 (76–85) % (Table 3). All
patients with a CVP≥ 12 mm Hg had a
ΔSVC <36% (Fig. 3).
No RV dysfunction was observed using
TEE assessment at T1. Inotropic support
was prescribed or increased based on the
presence of LV systolic dysfunction (LVFAC,
24 [18–37] %) at T1 in 14 patients (30%).
CVP was 15 (11–19) mm Hg, ScVO2 75
(73–80) %, and CI 1.7 (1.4–2.1) L/min/m2
(Table 4). No patient had anemia and only
three patients had a ScVO2 <70% (Fig. 4).
At T2, LVFAC significantly increased to 48
(39–61) %, CVP decreased to 9 (6–12) mm
Hg, and CI rose to 2.4 (2.1–2.9) L/min/m2
(Table 4). In half of the patients, base defi-
cit decreased (7.6 [4.5–8.7] mmol/L at T1
vs. 4.2 (2.6–6.8) mmol/L at T2), whereas
it tended to increase in the remaining
patients (8.5[7.1–10.9] mmol/L at T1 vs.
10.3 [7.4–12.6] mmol/L at T2). The three
patients with a ScVO2 <70% at T1 had a
partially or totally corrected ScVO2 at T2.
Among the remaining 32 patients with an
LVFAC >45%, ScVO2 was 75 (71–82) %,
CVP 9 (7–13) mm Hg, and CI 2.4 (2.1–2.9)
L/min/m2 (Table 4).
DISCUSSION
The SSC implemented guidelines to
improve the management of severe sep-
sis and septic shock, hence its prognosis
(2). Initial resuscitation relies on a step-
by-step hemodynamic assessment lead-
ing to repeated goal-targeted therapeutic
interventions. Fluid loading is recom-
mended to obtain a CVP ≥12 mm Hg in
mechanically ventilated patients, vaso-
pressors should be titrated to obtain a
mean arterial pressure ≥65 mm Hg, and
a dobutamine infusion should be initiated
if ScVO2 is <70% despite the correction
of an underlying anemia (2). Substantial
limitations of these guidelines, which are
mainly based on a positive trial conducted
in emergency department settings in
patients who were seriously ill, as reflected
by a mean ScVO2 <50% and mean lactate
>7 mmol/L (3), have been emphasized (4).
In addition, ScVO2 is usually in the normal
to high range in septic patients because
of an impaired oxygen extraction (7), as
reported in recent clinical series (25).
The present study showed that the
agreement of therapeutic interventions
derived from early echocardiographic
assessment and SSC guidelines was weak
regarding the prescription of fluid load-
ing and positive inotropes in patients pre-
senting to the ICU with severe sepsis or
septic shock. According to our standard
of care (10), we used a threshold value
table 3. Hemodynamic evaluation at T1 and T2 according to the superior vena cava collapsibility index measured at T1
ΔSVC <36% (n = 38) ΔSVC ≥36% (n = 8)
T1 T2 T1T2
Heart rate (bpm)
Systolic blood pressure (mm Hg)
Diastolic blood pressure (mm Hg)
Mean arterial pressure (mm Hg)
Central venous pressure (mm Hg)
Central venous oxygen saturation (%)
Left ventricular fractional area contraction
(%)
ΔSVC (%)
Cardiac index (L/min/m2)
Lactate (mmol/L)
Base deficit (mmol/L)
Cumulative volume expansion (mL)
Norepinephrine, n
(µg/kg/min)
Dobutamine, n
µg/kg/min
Epinephrine, n
µg/kg/min
101 (91–124)
103 (89–116)
58 (48–65)
72 (64–80)
14 (10–17)
79 (76–85)
51 (28–56)
108 (93–118)
120 (106–135)
61 (54–69)
83 (71–90)
11 (9–15)
79 (73–84)
50 (45–58)
117 (92–133)
97 (84–105)
43 (41–46)a
59 (53–68)
8 (7–8)
81 (74–87)
53 (51–65)
117 (114–120)
105 (103–111)
51 (48–56)
70 (68–82)
13 (9–14)b
79 (77–83)
64 (56–66)
9 (3–19)
2.3 (1.8–2.8)
3.5 (2.2–5.5)
8.1 (4.0–11.0)
1000 (750–2000)
32
0.354 (0.12–0.569)
1
5.0
5
0.256 (0.133–0.357)
9 (3–17)
2.7 (1.9–3.4)b
3.0 (1.4–4.9)
5 (1.8–9.5)
1350 (0–2300)
32
0.301 (0.138–0.668)
8
5.0 (5.0–5.63)
8
0.401 (0.126–0.8)
43 (39–45)a
2.1 (1.9–2.5)
3.0 (2.9–3.1)
6.9 (5.7–9.5)
1750 (875–2000)
7
0.201 (0.143–0.388)
0
Not applicable
0
Not applicable
13 (6–19)c
2.5 (2.4–2.8)b
2.2 (1.9–2.6)
5.7 (3.7–8.5)
3000 (1875–3125)
7
0.129b (0.093–0.228)
0
Not applicable
0
Not applicable
ΔSVC, superior vena cava collapsibility index.
ap< .05 T1 ΔSVC < 36% vs. T1 ΔSVC > 36%; bp < .05 T2 vs. T1; cp < .01 T2 vs.T1.
Patients with a ΔSVC ≥36% were considered as requiring fluids.
figure 3. Correlation between the central venous pressure (CVP) and the superior vena cava collaps-
ibility index (ΔSVC) in the study population at the time of the initial hemodynamic evaluation.

Page 5

Crit Care Med 2012 Vol. 40, No. 10 2825
of ΔSVC ≥36% to perform blood volume
expansion in our patients. This dynamic
index has previously been shown to predict
fluid responsiveness accurately in venti-
lated septic patients, with 100% specificity
and 90% sensitivity (18). Importantly, the
second TEE assessment depicted a 23%
increase of CI and a concomitant decrease
in norepinephrine doses, thereby confirm-
ing the efficacy of fluid loading in this sub-
set of patients. All our patients with a ΔSVC
≥36% who received fluids had an initial
CVP <12 mm Hg, as previously suggested
(5), but 14 of 38 patients with a ΔSVC <36%
also had a CVP <12 mm Hg, suggesting the
need for further blood volume expansion
according to SSC guidelines (2). Although
CVP is routinely used on clinical grounds
by a majority of intensivists, this variable
has been widely shown to be inaccurate
for the prediction of fluid responsiveness
(5, 18, 26). In addition, the cut-off value of
12 mm Hg targeted in the SSC guidelines
is mainly supported by a 30-yr-old study
performed in 15 patients with hypovole-
mic or septic shock (27). The rationale
for using CVP to guide fluid resuscitation
is based on the hypothesis that the right
atrial pressure accurately reflects right
atrial blood volume. Nevertheless, in ven-
tilated patients CVP is not altered only by
the distending right atrial pressure (i.e.,
preload) but also by the transmitted posi-
tive intrathoracic pressure.
Early echocardiography disclosed LV
systolic dysfunction in one third of the
patients, a proportion similar to that
reported in previous studies (6, 18, 28).
On the basis of initial TEE findings, the
administration of positive inotropes to
these patients resulted in an 18% increase
in CI, without any change in heart rate,
and a significant decrease in CVP. The
detection of LV dysfunction is based on
the conjunction of elevated CVP (>12
mm Hg) and low ScVO2 (<70%) in SSC
guidelines (2). Interestingly, CVP was <12
mm Hg in four of our 14 patients with
echocardiographically documented LV
systolic dysfunction, and a low ScVO2 was
observed in only three of them. These dis-
crepancies are presumably related to the
absence of marked elevation of cardiac
filling pressures in the presence of septic
cardiomyopathy and to impaired oxygen
extraction in septic patients. In the land-
mark study of Parker et al (6), pulmonary
capillary occlusion pressure was similar
in patients with or without sepsis-induced
severe LV systolic dysfunction (14 ± 2 mm
Hg vs. 13 ± 2 mm Hg). In 1990, Jardin
et al (29) reported a mean CVP of 11 ± 3
mm Hg and a mean pulmonary capillary
occlusion pressure of 11 ± 4 mm Hg in
patients with echocardiographically doc-
umented septic cardiomyopathy, similar
to those of patients with preserved LV sys-
tolic function. Recently, Bouhemad et al
(30) reported similar values of pulmonary
capillary occlusion pressure in patients
with and without septic cardiomyopathy
(10 ± 4 mm Hg and 11 ± 5 mm Hg, respec-
tively). The absence of elevated cardiac
filling pressures despite documented LV
systolic dysfunction in septic patients is
figure 4. Correlation between the central venous oxygen saturation (ScVO2) and the left ventricular
fractional area change (LVFAC) in the study population at the time of the initial hemodynamic evalua-
tion. Base deficit is noted in brackets for patients with a LVFAC <45%.
table 4. Hemodynamic evaluation at T1 and T2 according to left ventricular fractional area change at T1
LVFAC >45% (n = 32)LVFAC <45% (n = 14)
T1 T2T3T4
Heart rate (bpm)
Systolic blood pressure (mm Hg)
Diastolic blood pressure (mm Hg)
Mean arterial pressure (mm Hg)
Central venous pressure (mm Hg)
Central venous oxygen saturation (%)
LVFAC (%)
Superior vena cava collapsibility index (%)
Cardiac index (L/min/m2)
Lactate (mmol/L)
Base deficit (mmol/L)
Cumulative volume expansion (mL)
Norepinephrine, n
µg/kg/min
Dobutamine, n
µg/kg/min
Epinephrine, n
µg/kg/min
103 (99–111)
111 (89–119)
51 (38–61)
62 (53–79)
9 (7–13)
75 (71–82)
58 (50–62)
17 (11–21)
2.4 (2.1–2.9)
5.1 (3.2–7.3)
6.0 (4.1–9.2)
2000 (1750–2750)
28
0.375 (0.22–0.651)
109 (101–121)
129 (98–135)
69 (49–78)
70 (55–76)
13 (10–16)
81 (73–83)
55 (48–64)
10 (3–25)
3.0 (2.1–3.8)b
3.2 (1.9–5.8)
4.6 (3.2–5.9)
22523 (1985–2750)
28
0.755 (0.54–1.01)b
0
Not applicable
1
0.241
119 (110–135)a
100 (86–113)
58 (48–69)
66 (57–76)
15 (11–19)a
75 (73–80)
24 (18–37)a
10 (0–16)a
1.7 (1.4–2.1)a
4.8 (3.6–5.8)
8 (4.3–11.9)
1921 (1563–2239)
11
0.543 (0.40–0.876)a
1
5
2
0.340 (0.33–0.34)
102 (94–113)b
128 (118–141)b
58 (51–73)
85 (70–91)b
9 (6–12)b
76 (70–79)
48 (39–61)c
12 (6–16)
2.4 (2.1–2.9)b
5.0 (3.1–6.5)
7.0 (4.5–9.0)
2125 (1750–2645)
10
0.387 (0.28–0.54)
8
6.3 (5–7.5)
6
0.475 (0.32–0.55)
0
Not applicable
3
0.310 (0.212–0.324)
LVFAC, left ventricular fractional area change.
ap< .05 T1 LVFAC <45% vs. T1 LVFAC >45%; bp < .05 T2 vs. T1; cp < .01 T2 vs. T1.
Patients with a LVFAC <45% were considered as having a septic cardiomyopathy.

Page 6

2826 Crit Care Med 2012 Vol. 40, No. 10
presumably explained by the transient
increase in LV compliance because of the
septic process (31, 32). ScVO2 is related to
oxygen transport but also to tissue oxygen
extraction. The latter is deeply impaired
in severe sepsis (7) especially because of
mitochondrial dysfunction (33). In the
randomized study by Gattinoni et al (34)
performed in septic patients, mean SvO2
was in the normal range in the entire
population and remained unaffected by
a strategy aimed at optimizing CI, when
compared with the control group.
Recently, the implementation of
the SSC guidelines has been shown to
be associated with a 20% reduction in
hospital mortality in the participating
centers (35). Similar results were also
reported by Ferrer et al (36). Neverthe-
less, SSC guidelines contain >80 rec-
ommendations that are not limited to
hemodynamic resuscitation but rather
encompass all aspects of the manage-
ment of septic patients (2). In the above
mentioned studies, the implementation
of the SSC hemodynamic guidelines
was not specifically associated with bet-
ter survival in contrast to the guidelines
concerning antibiotic
administration of activated protein C
in persistent shock, and tight glucose
control (35, 36). In the present study, the
application of SSC guidelines would have
led to fluid administration in 16 addi-
tional patients who had no TEE findings
of preload dependency. ΔSVC is a fairly
sensitive index of preload dependence
because we reported only two false nega-
tives out of 66 septic shock patients in a
similar population of patients with septic
cardiomyopathy (18). Increasing evidence
supports the deleterious effect of exces-
sive fluid loading on the prognosis of ICU
patients (37, 38). Specifically, Boyd et al
(38) showed that a positive fluid balance
during the first 4 days of ICU hospitaliza-
tion and the presence of an elevated CVP
were independent factors of death, the
highest mortality rate being observed in
the subset of patients with a CVP >12 mm
Hg at hour 12 after admission. In the
present study, 30% of patients were diag-
nosed with LV systolic dysfunction dur-
ing the initial TEE assessment, which
prompted the administration of positive
inotropes. Small studies have suggested
that in the presence of LV systolic dys-
function, patients who have a preserved
cardiovascular response to inotropic
stimulation, as reflected by a positive
response to low-dose dobutamine, have a
better survival (39–41). Accordingly, the
management,
accurate detection of septic cardiomyopa-
thy using early TEE in septic patients with
organ hypoperfusion could be valuable in
guiding the intensivist to initiate inotro-
pic support, whereas no study has clearly
demonstrated the value of dobutamine. In
addition, systematic optimization of car-
diac output has long been known to have
deleterious effects (34). According to SSC
guidelines, inotropic support is highly
recommended in the presence of a CVP
>12 mm Hg and an ScVO2 <70%, which
are suggestive of LV systolic dysfunction.
The present data suggest that relevant LV
systolic dysfunction may be clearly dis-
closed by TEE in patients who fail to meet
SSC criteria to receive inotropic support.
This study has several limitations.
First, the relatively small size of our
population precludes definitive conclu-
sions and it remains to be seen how our
results can be applied to spontaneously
breathing patients. In the study of Rivers
et al (3), only half of the septic patients
were mechanically ventilated. Second,
our study shows that treatment deci-
sions are heavily influenced by the type of
monitoring and guidelines used but fails
to provide information on their potential
impact on outcome. Third, our definition
of LV systolic dysfunction is somewhat
arbitrary but is based on our previous
experience (10, 19). Importantly, a consis-
tent improvement in hemodynamics after
inotropic support was observed in this
subset of patients. Finally, as opposed to
the SSC guidelines, which were designed
to be applicable anywhere, echocardiog-
raphy is not yet readily accessible in all
ICUs, and intensivists must be adequately
trained to use our approach in the early
management of patients presenting with
severe sepsis or septic shock (42, 43).
CONCLUSIONS
We found that initial hemodynamic
monitoring with echocardiography might
result in less fluid loading and more use
of inotropes than SSC guidelines in ven-
tilated patients presenting to the ICU
with septic shock. Our findings should be
confirmed in a larger, more statistically
robust population and further studies are
required to evaluate the potential impact
of these strategies on outcome.
REFERENCES
1. Parrillo JE: Pathogenetic mechanisms of sep-
tic shock. N Engl J Med 1993; 328:1471–1477
2. Dellinger RP, Levy MM, Carlet JM, et al:
Surviving Sepsis Campaign: International
guidelines for management of severe sepsis
and septic shock: 2008. Crit Care Med 2008;
36:296–327
3. Rivers E, Nguyen B, Havstad S, et al: Early
goal-directed therapy in the treatment of
severe sepsis and septic shock. N Engl J Med
2001; 345:1368–1377
4. Perel A: Bench-to-bedside review: The ini-
tial hemodynamic resuscitation of the septic
patient according to Surviving Sepsis Cam-
paign guidelines–does one size fit all? Crit
Care 2008; 12:223
5. Osman D, Ridel C, Ray P, et al: Cardiac fill-
ing pressures are not appropriate to predict
hemodynamic response to volume challenge.
Crit Care Med 2007; 35:64–68
6. Parker MM, Shelhamer JH, Bacharach SL,
et al: Profound but reversible myocardial
depression in patients with septic shock. Ann
Intern Med 1984; 100:483–490
7. Krafft P, Steltzer H, Hiesmayr M, et al: Mixed
venous oxygen saturation in critically ill sep-
tic shock patients. The role of defined events.
Chest 1993; 103:900–906
8. Vieillard-Baron A, Slama M, Cholley B, et al:
Echocardiography in the intensive care unit:
From evolution to revolution? Intensive Care
Med 2008; 34:243–249
9. Vignon P, Slama M: Diagnosing the mecha-
nisms of circulatory failure. In: Hemody-
namic Monitoring Using Echocardiography
in the Critically Ill. De Backer D, Cholley
BP, Slama M, et al (Eds). Berlin, Heidelberg,
Springer-Verlag, 2011, pp 99–107
10. Vieillard-Baron A, Prin S, Chergui K, et al:
Hemodynamic instability in sepsis: Bedside
assessment by Doppler echocardiography. Am
J Respir Crit Care Med 2003; 168:1270–1276
11. Bone RC, Balk RA, Cerra FB, et al: Definitions
for sepsis and organ failure and guidelines for
the use of innovative therapies in sepsis. The
ACCP/SCCM Consensus Conference Commit-
tee. American College of Chest Physicians/
Society of Critical Care Medicine. Chest 1992;
101:1644–1655
12. Levy MM, Fink MP, Marshall JC, et al: SCCM/
ESICM/ACCP/ATS/SIS: 2001 SCCM/ESICM/
ACCP/ATS/SIS International Sepsis Defi-
nitions Conference. Crit Care Med 2003;
31:1250–1256
13. Antonelli M, Levy M, Andrews PJ, et al: Hemo-
dynamic monitoring in shock and implications
for management. International Consensus
Conference, Paris, France, 27-28 April 2006.
Intensive Care Med 2007; 33:575–590
14. Schuster M, Nave H, Piepenbrock S, et al: The
carina as a landmark in central venous cathe-
ter placement. Br J Anaesth 2000; 85:192–194
15. Pottecher T, Calvat S, Dupont H, et al: SFAR/
SRLF workgroup: Haemodynamic manage-
ment of severe sepsis: Recommendations of
the French Intensive Care Societies (SFAR/
SRLF) Consensus Conference, 13 October
2005, Paris, France. Crit Care 2006; 10:311
16. Perrino AC Jr, Harris SN, Luther MA:
Intraoperative determination of cardiac
output using multiplane transesophageal

Page 7

Crit Care Med 2012 Vol. 40, No. 10 2827
echocardiography: A comparison to thermo-
dilution. Anesthesiology 1998; 89:350–357
17. Zoghbi WA, Quinones MA: Determina-
tion of cardiac output by Doppler echocar-
diography: A critical appraisal. Herz 1986;
11:258–268
18. Vieillard-Baron A, Chergui K, Rabiller A, et al:
Superior vena caval collapsibility as a gauge
of volume status in ventilated septic patients.
Intensive Care Med 2004; 30:1734–1739
19. Vieillard-Baron A, Caille V, Charron C, et al:
Actual incidence of global left ventricular
hypokinesia in adult septic shock. Crit Care
Med 2008; 36:1701–1706
20. Jardin F, Dubourg O, Bourdarias JP: Echocar-
diographic pattern of acute cor pulmonale.
Chest 1997; 111:209–217
21. Le Gall JR, Lemeshow S, Saulnier F: A new
Simplified Acute Physiology Score (SAPS II)
based on a European/North American mul-
ticenter study. JAMA 1993; 270:2957–2963
22. Vincent JL, Moreno R, Takala J, et al: The
SOFA (Sepsis-related Organ Failure Assess-
ment) score to describe organ dysfunction/
failure. On behalf of the Working Group on
Sepsis-Related Problems of the European
Society of Intensive Care Medicine. Intensive
Care Med 1996; 22:707–710
23. McCabe WR, Jackson GG: Gram-negative bac-
teremia: I. Etiology and ecology. Arch Intern
Med 1962; 110:847–855
24. Bakeman R, Gottman JM: Observing Interac-
tion: An Introduction to Sequential Analysis.
Cambridge, UK: Cambridge University Press,
1997
25. van Beest PA, Hofstra JJ, Schultz MJ, et al:
The incidence of low venous oxygen satu-
ration on admission to the intensive care
unit: A multi-center observational study in
The Netherlands. Crit Care 2008; 12:R33
26. Michard F, Teboul JL: Predicting fluid
responsiveness in ICU patients: A critical
analysis of the evidence. Chest 2002; 121:
2000–2008
27. Packman MI, Rackow EC: Optimum left heart
filling pressure during fluid resuscitation of
patients with hypovolemic and septic shock.
Crit Care Med 1983; 11:165–169
28. Etchecopar-Chevreuil C, François B, Clavel
M, et al: Cardiac morphological and func-
tional changes during early septic shock: A
transesophageal echocardiographic study.
Intensive Care Med 2008; 34:250–256
29. Jardin F, Brun-Ney D, Auvert B, et al: Sepsis-
related cardiogenic shock. Crit Care Med
1990; 18:1055–1060
30. Bouhemad B, Nicolas-Robin A, Arbelot C, et al:
Acute left ventricular dilatation and shock-
induced myocardial dysfunction. Crit Care
Med 2009; 37:441–447
31. Ognibene FP, Parker MM, Natanson C, et al:
Depressed left ventricular
Response to volume infusion in patients with
sepsis and septic shock. Chest 1988; 93:903–910
32. Suffredini AF, Fromm RE, Parker MM, et al:
The cardiovascular response of normal
humans to the administration of endotoxin.
N Engl J Med 1989; 321:280–287
33. Levy RJ, Piel DA, Acton PD, et al: Evidence
of myocardial hibernation in the septic heart.
Crit Care Med 2005; 33:2752–2756
34. Gattinoni L, Brazzi L, Pelosi P, et al: A trial of
goal-oriented hemodynamic therapy in criti-
cally ill patients. SvO2 Collaborative Group. N
Engl J Med 1995; 333:1025–1032
35. Levy MM, Dellinger RP, Townsend SR, et al:
The Surviving Sepsis Campaign: Results of an
international guideline-based performance
improvement program targeting severe sep-
sis. Intensive Care Med 2010; 36:222–231
performance.
36. Ferrer R, Artigas A, Suarez D, et al: Edusep-
sis Study Group: Effectiveness of treatments
for severe sepsis: A prospective, multicenter,
observational study. Am J Respir Crit Care
Med 2009; 180:861–866
37. Vincent JL, Sakr Y, Sprung CL, et al: Sepsis
Occurrence in Acutely Ill Patients Investiga-
tors: Sepsis in European intensive care units:
Results of the SOAP study. Crit Care Med
2006; 34:344–353
38. Boyd JH, Forbes J, Nakada TA, et al: Fluid
resuscitation in septic shock: A positive fluid
balance and elevated central venous pressure
are associated with increased mortality. Crit
Care Med 2011; 39:259–265
39. Vallet B, Chopin C, Curtis SE, et al: Prognostic
value of the dobutamine test in patients with
sepsis syndrome and normal lactate values: A
prospective, multicenter study. Crit Care Med
1993; 21:1868–1875
40. Rhodes A, Lamb FJ, Malagon I, et al: A pro-
spective study of the use of a dobutamine
stress test to identify outcome in patients
with sepsis, severe sepsis, or septic shock. Crit
Care Med 1999; 27:2361–2366
41. Kumar A, Schupp E, Bunnell E, et al: Car-
diovascular response to dobutamine stress
predicts outcome in severe sepsis and septic
shock. Crit Care 2008; 12:R35
42. Mayo PH, Beaulieu Y, Doelken P, et al: Ameri-
can College of Chest Physicians/La Société de
Réanimation de Langue Française statement
on competence in critical care ultrasonogra-
phy. Chest 2009; 135:1050–1060
43. Expert round table on ultrasound in ICU:
International expert statement on train-
ing standards for critical care ultraso-
nography. Intensive Care Med 2011; 37:
1077–1083