Stroke Volume Variation as a Predictor of Intravascular Volume ...

Annals of Surgical Oncology
DOI: 10.1245/s10434-008-0139-0
Stroke Volume Variation as a Predictor of Intravascular
Volume Depression and Possible Hypotension During
the Early Postoperative Period After Esophagectomy
Makoto Kobayashi, MD, PhD, 1 Masayoshi Koh, MD, 2 Takashi Irinoda, MD, PhD, 1
Eiji Meguro, MD, PhD, 1 Yoshiro Hayakawa, MD, PhD, 1 and Akinori Takagane, MD, PhD 1
1 Surgical Division, Hakodate Goryoukaku Hospital, 38-3 Goryoukaku-cho, Hakodate City, Hokkaido 040-8611, Japan
2 Anesthetic Division, Hakodate Goryoukaku Hospital, 38-3 Goryoukaku-cho, Hakodate City, Hokkaido 040-8611, Japan
Background: Perioperative hypotension during esophagectomy results from hypovolemia
caused by a shift of extracellular fluid from the intravascular to the extravascular compartment.
Fluid management is often difficult to gauge during major surgery because there are no
reliable indicators of fluid status, and some patients still experience cardiorespiratory instability.
In this retrospective study, we evaluated stroke volume variation (SVV), calculated by
using a new arterial pressure-based cardiac output measurement device, as a predictor for fluid
responsiveness after esophageal surgery.
Methods: Eighteen patients undergoing esophagectomy with extended radical lymphadenectomy
were monitored by the FloTrac sensor/Vigileo monitor system during the perioperative
and immediate postoperative period. Fluid responsiveness was assessed and compared
with concurrent SVV and central venous pressure (CVP) values, and routine hemodynamic
variables.
Results: Eleven of 18 patients needed additional volume loading within the first 10 postoperative
hours as a result of hypotension. The maximum SVV value of fluid resuscitated
patients was >15% in all cases, whereas six of seven patients without postoperative hypotension
had maximum SVV values of 0.05).
Conclusion: We conclude that SVV, as displayed on the Vigileo monitor, is an accurate
predictor of intravascular hypovolemia and is a useful indicator for assessing the appropriateness
and timing of applying fluid for improving circulatory stability during the perioperative
period after esophagectomy.
Over the past decade, morbidity and mortality
associated with radical esophagectomy have both
improved, 1 whereas postoperative management has
remained problematic. 2 In particular, hypotension
often occurs during the perioperative and immediate
Address correspondence and reprint requests to: Makoto
Kobayashi, MD, PhD; E-mail: neo-coba@mub.biglobe.ne.jp
Published by Springer Science+Business Media, LLC Ó 2008 The Society of
Surgical Oncology, Inc.
postoperative periods associated with major surgery,
such as extended radical lymphadenectomy for
esophageal cancer, and it is almost certainly caused
by hypovolemia. While postoperative hemorrhage
needs to be ruled out, most cases of hypovolemic
hypotension seem to be due to a shift of extracellular
fluid from the central to peripheral compartments,
and it has been suggested that this is a direct consequence
of the development of a third space associated
with increased vascular permeability caused by

hypercytokinemia. 3 The destruction of the lymphatic
tract due to interruption of the pulmonary lymph
outflow tract as a result of mediastinal lymphadenectomy
and removal of the thoracic duct also promotes
deterioration of circulatory dynamics. 4 With
the introduction of minimally invasive surgery, 5 corticosteroid
administration to prevent hypercytokinemia,
6,7 and treatment with a specific neutrophil
elastase inhibitor, 8 postoperative management of
esophageal cancer is safer than ever. However, it is
also true that some patients still experience cardiorespiratory
instability; especially those with poor
preoperative nutrition and those receiving neoadjuvant
chemoradiotherapy. 9,10 Even when it is thought
that sufficient fluid has been administered, it is
sometimes difficult to determine whether intravascular
fluid depression has been relieved by monitoring
routine hemodynamic parameters. An added complication
is that although appropriate fluid transfusion
is often crucial to avoid the deleterious effects of
overresuscitation, underresuscitation, or inappropriate
resuscitation, it is also reported that static indicators
of cardiac preload, such as central venous
pressure (CVP), pulmonary artery occlusion pressure,
and cardiac end-diastolic dimensions, may be unreliable
in detecting volume responsiveness in critically
ill patients. 11
The FloTrac sensor in combination with the Vigileo
monitor (Edwards Lifesciences, Tokyo, Japan) is
a recently introduced arterial pressure-based system
for continuously monitoring cardiac output (CO),
which has applicability in the critical care setting. The
FloTrac sensor is a less invasive hemodynamic
monitoring device than those used for thermodilution
assessment, and it can be used to monitor continuously
CO, stroke volume, and stroke volume variation
(SVV) through a peripheral arterial pressure line.
In addition, other CO devices require calibration to
correct for the patient’s changing vascular tone,
whereas the FloTrac sensor/Vigileo monitor system
needs no calibration because it continuously adjusts
for the patient’s ever-changing vascular tone by use
of a novel algorithm incorporated within the Vigileo
monitor, which is applied to the digitized arterial
pressure wave. 12 The usefulness of SVV in assessing
fluid responsiveness has previously been reported in
patients with reduced cardiac function. 13 We started
routinely using the FloTrac sensor/Vigileo monitor
system during esophageal surgery, including an
assessment of its advantages for perioperative management
after radical esophagectomy, in May 2006.
This followed early findings that indicated that SVV
was very good at predicting the development of
Ann. Surg. Oncol.
M. KOBAYASHI ET AL.
hypercytokinemia-induced intravascular hypovolemia
in patients undergoing major surgery.
Here we report retrospective results from the first
18 patients undergoing surgery for esophageal cancer
in whom the FloTrac sensor/Vigileo monitor system
was used to assess fluid responsiveness as an integral
part of routine postoperative management follow-up
and care. In addition, we compared SVV with CVP in
terms of reliability in predicting fluid responsiveness
during the perioperative and postsurgical periods.
MATERIALS AND METHODS
Between May 2006 and September 2007, 18 men of
mean ± standard deviation age 66.8 ± 4.8 (range, 61–
73) years underwent perioperative monitoring with
the FloTrac sensor/Vigileo monitor system after
curative esophagectomy for esophageal squamous
cell carcinoma at Hakodate Goryoukaku Hospital.
The tumor, node, metastasis system classification
according to the Guidelines for the Clinical and
Pathologic Studies on Carcinoma of the Esophagus
(Japan Society for Esophageal Diseases, 9th edition)
was as follows: stage I, n = 5; stage II, n = 5; stage
III, n = 4; and stage IVa, n = 4. Tumor location
was upper thoracic esophagus, n = 3; middle thoracic
esophagus, n = 13; and lower thoracic esophagus,
n = 2. Two-field lymph node dissection was
performed in 5 patients and three-field dissection in
13 patients. A combination of general (intravenous
propofol) and epidural (bupivacaine) anesthesia was
used to manage perioperative anesthetic requirements
during surgery. The surgical approach for tumor
resection for all patients was made by intercostal
thoracotomy through a 10- to 12-cm skin incision,
preserving the latissimus dorsi and serratus anterior
muscle. After subtotal esophagectomy and extended
mediastinal lymph node dissection, reconstruction via
stomach and cervical esophagogastrostomy was
performed. All patients were selected for posterior
mediastinal reconstruction via a gastric tube, and a
hand-sutured anastomosis was conducted at the
neck site. During the surgical procedure, fluid was
administered at a rate of 12 to 15 mL/kg/h of crystalloids.
Perioperative dopamine or furosemide was
not permitted.
After surgery, all patients were immediately
transferred to the intensive care unit (ICU) under
tracheal intubation. Mechanical ventilation was
adjusted to supply tidal volumes of 8 to 10 mL/kg of
preoperative body weight with pressure-support
ventilation. Midazolam and morphine were admin-

istrated intravenously for sedation during tracheal
intubation.
Fluid administration in the early postoperative
period was started at a rate of 3.5 mL/kg/h and
continued until an extravascular to intravascular
shift was observed. Patients were considered to be
hypotensive when systolic blood pressure could not
be maintained above 80 mm Hg for longer than
15 minutes, at which time additional volume loading
was provided. If the serum albumin concentration
dropped below the normal range (3.4 to 5.4 g/dL),
250 mL of 5% plasma protein fraction was administered.
In lieu of corticosteroid treatment to treat
hypercytokinemia, sivelestat sodium hydrate, a specific
neutrophil elastase inhibitor (Elaspol, Ono
pharmacy, Tokyo, Japan), was administrated intravenously
at a rate of 0.2 mg/kg/h on completion of
surgery. To avoid the possibility of intravascular
hypoperfusion, our policy is not to use low-dose
dopamine or loop diuretics to protect against oliguria
until after adjustments have been made for
volume depression. After confirming circulatory
stability and a shift toward diuresis, the rate of
fluid administration was immediately decreased to
1.5 mL/kg/h to avoid congestive heart failure
developing as a consequence of overhydration.
Patients were then weaned from assisted mechanical
ventilation.
Assessment of Hemodynamic Parameters
Before surgery, hemodynamic monitoring was initiated
via a 22-gauge elastic catheter that was inserted
into the radial artery and connected to a FloTrac
sensor. CO and SVV were measured every 20 seconds
according to the algorithm housed within the Vigileo
monitor. Other parameters routinely monitored
included, continuous electrocardiography, pulse
oximetry, end-tidal CO2 and arterial blood pressure
(ABP). A central venous catheter was inserted into
the internal jugular vein, and CVP was measured
continuously during the perioperative period with a
low-pressure transducer. All raw data from the Vigileo
monitor were recorded directly onto a computer
and were subsequently reviewed and analyzed statistically.
To assess hemodynamic variability associated
with volume loading mean SVV, CVP and CO were
determined 30 minutes before and 30 minutes after
fluid administration. To investigate the relation
between changes in preload and postload variables,
changes (D) in SVV, CVP, and CO were calculated by
the following formulas:
SVV AFTER ESOPHAGECTOMY
DSVVð%Þ ¼preload SVV postload SVV
DCVP ðmm HgÞ ¼postload CVP preload CVP
DCO ¼ðpostload CO preload COÞ=preload CO.
Statistical Analysis
All data are expressed as mean ± SD, unless stated
otherwise. The v 2 test for independence was used to
assess the relationship between SVV and the development
of postoperative hypotension. To determine
whether changes in hemodynamic variables (DSVV,
DCVP) were related to the increased ratio in CO
(DCO) after additional volume loading, both linear
regression and Pearson’s correlation coefficient were
calculated by StatView software (Abacus Concepts,
Berkeley, CA). Values of P < 0.05 were considered
statistically significant.
RESULTS
The mean duration of surgery was 303 ± 58 minutes,
and mean blood loss was 280 ± 320 mL. Postoperatively,
11 of 18 patients required additional
volume loading within the first 10 hours due to
hypotension, and 3 of these received blood transfusions.
The mean duration under mechanical assist
ventilation after surgery was 1.7 ± 0.7 days. The
mean lengths of ICU and hospital stay after surgery
were 3.6 ± 1.4 days and 16.0 ± 3.2 days, respectively.
There was no significant difference in cardiac function,
as assessed by echocardiographically measured
ejection fraction before surgery, in patients who needed
fluid resuscitation after hypotension (69.1 ± 7.2%)
compared with those who did not (66.1 ± 8.8%;
P = 0.390, Mann-Whitney U-test). Furthermore,
there was no statistically significant difference in the
volume of fluid administered (14.1 ± 3.8 vs. 16.1 ± 3.5
mL/kg/h; P = 0.2204, Mann-Whitney U-test), blood
loss (241 ± 165 vs. 427 ± 332 mL; P = 0.092, Mann-
Whitney U-test), or operating time (293 ± 39 vs.
294 ± 70 minutes; P = 0.9025, Mann-Whitney Utest)
between the group requiring fluid resuscitation
versus the group that did not.
Importantly, in this series of patients who had
undergone esophagectomy with extended radical lymphadenectomy
and who developed intravascular volume
depression, ICU management with appropriate
fluid replacement for critical hypotension (as predicted
by SVV changes on the Vigileo monitor) resulted in
resuscitation and recovery in all cases and no clinically
Ann. Surg. Oncol.

FIG. 1. Postoperative SVV, CO, and CVP in a patient with circulatory instability after esophagectomy in a 69-year-old man. Duration of
surgery was 237 minutes; blood loss was 127 mL. Pathological classification according Japanese guidelines (see Materials and Methods) was
pT3 pN0 M0 pIM0; pStage II. SVV, stroke volume variation; CO, cardiac output; CVP, central venous pressure; PPF, plasma protein
fraction.
important medical problems such as renal dysfunction,
respiratory failure, cardiac insufficiency, or death
occurred. Furthermore, there were no complications
associated with the use of the FloTrac sensor/Vigileo
monitoring system in this cohort of patients.
Figure 1 shows a typical graphical presentation for
SVV, CO, and CVP in a patient who underwent
esophagectomy with three-field lymph node dissection.
This patient experienced postoperative hypotension
and required fluid resuscitation in the ICU.
The initial SVV of the patient at the time of entering
the ICU was 8%, but it gradually increased to 15% to
20% by postoperative hour 4. Shortly thereafter,
systolic ABP decreased to 80 mm Hg, but SVV still increased to 26%. ABP
once again dropped to 20%, and hypotension reoccurred.
After additional fluid resuscitation, the circulation
stabilized, and SVV was finally maintained near 10%.
During these events, CVP values showed almost no
response before or after volume loading, whereas the
change in SVV observed graphically on the Vigileo
monitor clearly predicted hypotension resulting from
intravascular hypovolemia. In contrast, Fig. 2 shows
the typical graphical presentation for SVV, CVP, and
CO in a patient with stable circulation without any
hypotensive events. CO and ABP remained stable,
Ann. Surg. Oncol.
M. KOBAYASHI ET AL.
and SVV remained 15% at any stage during the 12 hours after surgery.
In contrast, maximum SVV values in the patient
group with hypotension were >15% in all cases
(n = 11), even though the initial SVV value was
15% is statistically significantly higher
than in patients with maximum SVV of

P = 0.049). In this small cohort of patients, the
correlation between SVV and CO only just achieved
statistical significance, but it did show promise as a
predictor of circulatory instability induced by intravascular
hypovolemia after esophagectomy. In this
regard it, was clearly superior to CVP.
DISCUSSION
Hypovolemic hypotension induced by hypercytokinemia
is often observed in the early postoperative
SVV AFTER ESOPHAGECTOMY
FIG. 2. Postoperative SVV, CO, and CVP in a patient with circulatory stability after esophagectomy in a 61-year-old man. Duration of the
operation was 335 minutes; blood loss was 82 mL. Pathological classification was pT2 pN3 M0 IM1, pStage II. SVV, stroke volume variation;
CO, cardiac output; CVP, central venous pressure.
FIG. 3. Initial and maximum intensive care unit (ICU) stroke volume variation (SVV) values. Patients are divided into 2 groups according to
need of fluid resuscitation after postoperative hypotension caused by intravascular hypovolemia. The initial value of SVV on entering the ICU
(j) and the maximum value before fluid resuscitation figures (•) are presented.
period of esophagectomy for esophageal cancer, and
it is reported that >60% of patients develop hypotension
on the operative day. 14 Postoperative hypotension
is often associated with a further reduction in
intravascular volume caused by unusual shift of
extracellular fluid into the third space. 3 Even though
an effective strategy for hypercytokinemia can be
adopted that uses corticosteroids or a specific neutrophil
elastase inhibitor such as sivelestat, and even
though early weaning from mechanical ventilation is
possible, 8 some patients still experience severe circulatory
instability. The usual protocol in our institute
Ann. Surg. Oncol.

FIG. 4. Responsiveness of CO according to changes in CVP and SVV after fluid loading. DSVV (%) = preload SVV - postload SVV; DCVP
(mm Hg) = postload CVP - preload CVP; DCO = (postload CO - preload CO)/preload CO; SVV, stroke volume variation; CO, cardiac
output; CVP, central venous pressure.
for the perioperative management of patients undergoing
esophagectomy is to administer fluid during
anesthesia at a rate of 15 mL/kg/h of crystalloids.
After transfer to the ICU, fluid administration starts
at a rate of 3.5 mL/kg/h. Our strategy for minimizing
effects related to the adverse release of the neutrophil
elastase is to administer sivelestat rather than a corticosteroid,
and we also avoid the use of dopamine
and furosemide. By means of this protocol, in 18
patients with esophageal cancer undergoing radical
esophagectomy, 15 patients (83%) were successfully
extubated within 2 days after surgery without complication,
and the mean period was 1.7 days after
surgery. Seven of 18 patients avoided hypotension
and escaped additional volume loading, but 11 patients
(61%) needed fluid resuscitation to treat
intravascular hypovolemia.
Precise control of fluid balance is a primary goal of
postoperative management after surgery, but traditional
hemodynamic monitoring parameters (heart
rate, mean arterial pressure, and CVP) are often
insensitive and sometimes misleading in the assessment
of circulating blood volume. 15 From May 2006, our
institute introduced the FloTrac sensor/Vigileo monitor
system for tracking SVV during the perioperative
period of esophagectomy. It was reported that SVV
calculated from stroke volume changes within the
respiratory cycle under mechanical ventilation could
be used to assess the volume status and cardiac preload
of critically ill 16 and cardiac surgery patients. 17 Our
early clinical experience of applying the FloTrac sensor/Vigileo
monitor system to patients with esophageal
cancer, also demonstrated that monitoring changes in
Ann. Surg. Oncol.
M. KOBAYASHI ET AL.
SVV accurately predicted the development of hypotension
in patients undergoing esophageal surgery.
In our series, the initial value of SVV on entering to
the ICU was15% in all cases, and
the occurrence rate of hypotension was statistically
significantly higher (P = 0.0012) in these patients
(Fig. 3). These data indicate that an increase in SVV
above 15% might usefully be used to predict the
development of hypotension and the need for additional
fluid during the early postoperative period,
even when traditional parameters (e.g., mean arterial
pressure, CVP, heart rate) may not highlight such
changes. In the literature concerning the accuracy of
SVV for estimating fluid responsiveness after cardiac
surgery, it has been reported that real-time monitoring
of SVV is a more sensitive and specific predictor
than CVP and other hemodynamic parameters. 18 Our
SVV-based data is the first pertaining to esophagectomy
patients and confirms that SVV also has excellent
predictive qualities in this group of patients
undergoing esophageal surgery. Comparing CVP and
SVV for their predictability in assessing fluid responsiveness
(Fig. 4) indicated that a decrease in SVV
values was significantly correlated to CO improvement
(r = 0.638, P = 0.049), but no such correlation
between CVP and CO value existed. CVP, which is a
commonly used parameter for the evaluation of
intravascular volume status, 11,17 demonstrated no
predictive value for cardiorespiratory instability during
the perioperative period after esophagectomy.

Although the usefulness of SVV is clearly demonstrated
in our data, the clinical use of this hemodynamic
parameter has certain limitations. First, this
monitoring method can only be used in mechanically
ventilated patients without arrhythmias. Moreover,
severe peripheral constriction and aortic regurgitation
may affect absolute values. Nevertheless, despite
these limitations, we suggest that the FloTrac sensor/
Vigileo monitor system provides marked advantages
over more traditional vital sign monitoring systems
alone in postoperative fluid resuscitation after
esophageal surgery. Because individual responses to
surgical stress from this form of surgery vary and are
difficult to predict, perioperative cardiorespiratory
instability is more unpredictable than after cardiac
surgery. 19 Even though relatively large amounts of
fluid were transfused during the surgical procedures
reported, postoperative hypotension resulting from
intravascular hypovolemia still occurred unexpectedly.
This is a predicament for the physician, who has
difficult decisions to make with regards fluid resuscitation.
Persistent hypotension can lead to serious
tissue hypoperfusion and organ distress, while
excessive fluid replacement may lead to congestive
heart failure and pulmonary edema during volume
resuscitation. To maintain low mortality rates in
esophagectomy, a safer and more exact fluid management
method is required.
On the basis of our experience to date, we conclude
that SVV, as displayed on the Vigileo monitor, is an
accurate predictor of intravascular hypovolemia and
is a useful indicator for assessing the appropriateness
and timing of applying fluid for improving circulatory
stability during the perioperative period after
esophagectomy. A larger, prospective trial is needed
to help ascertain the overall effectiveness of SVV/
Vigileo monitoring.
ACKNOWLEDGMENT
We thank Steve Clissold, PhD (Content Ed Net),
who provided assistance with English language and
whose work was funded by Edwards Lifesciences,
Japan.
REFERENCES
1. Ando N, Ozawa S, Kitagawa Y, et al. Improvement in the
results of surgical treatment of advanced squamous esophageal
SVV AFTER ESOPHAGECTOMY
carcinoma during 15 consecutive years. Ann Surg 2000;
232:225–32.
2. Fujita H, Kakegawa T, Yamana H, et al. Mortality and
morbidity rates, postoperative course, quality of life, and
prognosis after extended radical lymphadenectomy for esophageal
cancer. Comparison of three-field lymphadenectomy with
two-field lymphadenectomy. Ann Surg 1995; 222:654–62.
3. Ishihara H, Nakamura H, Okawa H, et al. Comparison of
initial distribution volume of glucose and intrathoracic blood
volume during hemodynamically unstable states early after
esophagectomy. Chest 2005; 128:1713–9.
4. Imamura M, Shimada Y, Kanda T, et al. Hemodynamic
changes after resection of thoracic duct for en-bloc resection of
esophageal cancer. Surg Today 1992; 22:226–32.
5. Luketich JD, Alvelo-Rivera M, Buenaventura PO, et al.
Minimally invasive esophagectomy: outcomes in 222 patients.
Ann Surg 2003; 238:486–94.
6. Sayama J, Shineha R, Yokota K, et al. [Control of the excessive
reaction after surgery for esophageal carcinoma with
preoperative administration of the cortico-steriods]. Jpn J
Gasroenterol Surg 1994; 81:49–51.
7. Shimada H, Ochiai T, Okazumi S, et al. Clinical benefits of
steroid therapy on surgical stress in patients with esophageal
cancer. Surgery 2000; 128:791–8.
8. Kobayashi M, Harada O, Shimaya S, et al. [Experiences using
neutrophil elastase inhibitors perioperatively after radical
esophagectomy]. Shinyaku Rinsho 2004; 53:762–5.
9. Bartels H, Stein HJ, Siewert JR. Preoperative risk analysis and
postoperative mortality of oesophagectomy for resectable
oesophageal cancer. Br J Surg 1998; 85:840–4.
10. Reynolds JV, Ravi N, Hollywood D, et al. Neoadjuvant chemoradiation
may increase the risk of respiratory complications
and sepsis after transthoracic esophagectomy. J Thorac Cardiovasc
Surg 2006; 132:549–55.
11. Michard F, Teeboul JL. Predicting fluid responsiveness in ICU
patients: a critical analysis of the evidence. Chest 2002;
121:2000–8.
12. Manecke GR Jr, Auger WR. Cardiac output determination
from the arterial pressure wave: clinical testing of a novel
algorithm that does not require calibration. J Cardiothorac
Vasc Anesth 2007; 21:3–7.
13. Reuter DA, Felbinger TW, Schmidt C, et al. Stroke volume
variations for assessment of cardiac responsiveness to volume
loading in mechanically ventilated patients after cardiac surgery.
Intensive Care Med 2002; 28:392–8.
14. Suzuki A, Ishihara H, Okawa H, et al. Can initial distribution
volume of glucose predict hypovolemic hypotension after
radical surgery for esophageal cancer? Anesth Analg 2001;
92:1146–51.
15. Shippy CR, Appel PL, Shoemaker WL. Reliability of clinical
monitoring to assess blood volume in critically ill patients. Crit
Care Med 1984; 12:107–12.
16. Michard F. Changes in arterial pressure during mechanical
ventilation. Anesthesiology 2005; 103:419–28.
17. Hofer CK, Muller SM, Furrer L, et al. Stroke volume and
pulse pressure variation for prediction of fluid responsiveness
in patients undergoing off-pump coronary artery bypass
grafting. Chest 2005; 128:848–54.
18. Rex S, Brose S, Metzelder S, et al. Prediction of fluid responsiveness
in patients during cardiac surgery. Br J Anaesth 2004;
93:782–8.
19. Tandon S, Batchelor A, Bullock R, et al. Peri-operative risk
factors for acute lung injury after elective oesophagectomy. Br
J Anaesth 2001; 86:633–8.
Ann. Surg. Oncol.