Anaerobic Metabolism During Cardiopulmonary
Bypass: Predictive Value of Carbon Dioxide
Derived Parameters
Marco Ranucci, MD, Giuseppe Isgrò, MD, Federica Romitti, MD, Sara Mele, MD,
Bonizella Biagioli, MD, PhD, and Pierpaolo Giomarelli, MD, PhD
Department of Cardiothoracic Anesthesia, Policlinico San Donato, Milan, and the Thoracic and Cardiovascular Unit, Department
of Surgery and Bioengineering, University of Siena, Siena, Italy
Background. Hyperlactatemia during cardiopulmonary
bypass (CPB) is a common event and is associated to a
high morbidity and mortality after cardiac operations.
The present study is aimed to identify the possible
predictors of hyperlactatemia during CPB among a series
of oxygen and carbon dioxide derived parameters mea-
sured during CPB.
Methods. This is a prospective observational study on
54 patients undergoing cardiac surgery with CPB. Hyper-
lactatemia was defined as an arterial lactate concentration
higher than 3 mMol/L. Serial blood lactate assays have
been performed during CPB, and their association to a
number of oxygen and carbon dioxide derived parame-
ters was explored.
Results. Arterial blood lactate concentration was posi-
tively correlated to the CPB duration, the carbon dioxide
elimination, and the respiratory quotient, and negatively
A t the end of cardiac operations, the finding of ele-
vated blood lactate levels is quite common [1–5].
This pattern is generally attributed to tissue hypoxia
(type A hyperlactatemia) [1–5] but type B hyperlac-
tatemia (in absence of tissue hypoxia) has been advo-
cated in some cases [2].
The presence of hyperlactatemia at the intensive care
unit (ICU) admission after cardiac operations is associ-
ated to a poor outcome [5]; however, the development of
lactic acidosis may occur during the early phases of ICU
recovery, or during cardiopulmonary bypass (CPB) [6].
Even in this last case, it is associated to an increased risk
of morbidity and mortality [6].
Conventional monitoring with arterial and mixed ve-
nous blood gas analysis during CPB may help in detect-
ing the adequacy of tissue perfusion, and the on-line
measurement of mixed venous oxygen saturation (Svo2)
may offer additional information. However, blood lactate
concentration monitoring seems more adequate for de-
tecting the correct matching of oxygen supply and de-
Accepted for publication Jan 3, 2006.
Address correspondence to Dr Ranucci, Policlinico S. Donato, Via Morandi
30, 20097 San Donato Milan, Italy; e-mail: cardioanestesia@virgilio.it.
© 2006 by The Society of Thoracic Surgeons
Published by Elsevier Inc
CARDIOVASCULAR
correlated to the presence of the aortic cross-clamping,
the body surface area, the ratio between the oxygen
delivery and the carbon dioxide production, and the
arterial oxygen saturation. Predictors of hyperlactatemia
during CPB are a carbon dioxide production higher than
60 mL · min-1· m-2, a respiratory quotient higher than 0.9,
and a ratio between oxygen delivery and carbon dioxide
production lower than 5.
Conclusions. Carbon dioxide derived parameters are
representative of hyperlactatemia during CPB, as a result
of the carbon dioxide produced under anaerobic condi-
tions through the buffering of protons by the bicarbonate
system. The carbon dioxide elimination rate measured at
the exhaled site of the oxygenator may be used for an
indirect assessment of the metabolic state of the patient.
(Ann Thorac Surg 2006;81:2189 –95)
© 2006 by The Society of Thoracic Surgeons
mand during CPB [7, 8]. The association of a low oxygen
delivery during CPB with an increased postoperative
morbidity and mortality has been recently hypothesized
in various papers focused on excessive hemodilution
during CPB [9, 10], and is demonstrated as an indepen-
dent risk factor for acute renal failure after cardiac
operations with CPB [11].
Blood lactate concentration is presently not available
as on-line monitoring during CPB or in critically ill
patients. For this reason, various possible predictors of
critical hypoperfusion (defined as hyperlactatemia) have
been tested in critically ill patients and good correlations
have been found for carbon dioxide derived parameters
alone [12] or in association with oxygen derived param-
eters [13, 14]. In spite of the evidence that hyperlac-
tatemia during CPB is associated with a bad outcome [6]
we have found no information about the association
between these parameters and blood lactate levels dur-
ing CPB. The primary endpoint of this study is exploring
a number of oxygen and carbon dioxide derived param-
eters in order to detect their association with hyperlac-
tatemia during CPB; the secondary endpoint is establish-
ing an indirect on-line monitoring system for blood
lactate formation during CPB.
0003-4975/06/$32.00
doi:10.1016/j.athoracsur.2006.01.025
 2190 RANUCCI ET AL
HYPERLACTATEMIA DURING CPB
Patients and Methods
Study Design
This is a prospective observational study conducted dur-
ing one month of activity in the Cardiac Surgery Depart-
CARDIOVASCULAR
ments of the two participating institutions. All the pa-
tients gave a written consent to the scientific treatment of
their data. The Local Ethical Committee waived the need
for the approval.
Patient Population
Exclusion criteria were age less than 18 years and cardiac
transplant operation. Fifty-four patients undergoing car-
diac surgery with CPB were enrolled in the study; iso-
lated coronary artery bypass graft operations were 29
(54%), isolated valve procedures were 7 (13%), and com-
bined coronary artery ⫹ valve or double-triple valve
operations were 18 (33%). Three patients reached the
operating theater under emergency conditions due to
failed percutaneous coronary angioplasty or congestive
heart failure, while the remaining 51 were elective pa-
tients. Thirty-three patients (61%) were male; the mean
age was 67.6 ⫾ 6.2 years and the CPB duration was 89 ⫾
38 minutes.
Anesthesia, Surgery, and CPB Management
The patients were treated with a totally intravenous
anesthesia with remifentanil and midazolam plus cisa-
tracurium for muscle relaxation, or with a combined
intravenous-inhalatory anesthesia according to the anes-
thesiologist’s preference. Cardiopulmonary bypass was
established after a standard median sternotomy, aortic
root cannulation, and single or double atrial cannulation
for venous return. Lowest core body temperature during
CPB varied from 27°C to 37°C as requested by the
surgeon. Body temperature was measured at the naso-
pharingeal site and at the rectal site. This last tempera-
ture was considered for correcting the values of blood gas
analyses. The perfusate temperature was measured at
the oxygenator site and used for correcting the values of
exhaled carbon dioxide. Antegrade intermittent cold
crystalloid or cold blood cardioplegia was used according
to the surgeon’s preference. The circuit was primed with
700 mL of a gelatin solution (Eufusin; Bieffe Medical,
Modena, Italy), and 200 mL of trihydroxymethylamin-
omethane solution. Roller (Stockert, Munich, Germany)
or centrifugal pumps (Medtronic Bio-Medicus, Eden
Prairie, MN) were used according to the availability; a
biocompatible treatment (phosphorylcholine coating)
and a closed circuit with separation of the blood suctions
were used in 24% of the patients. The oxygenator was a
hollow fiber D 905 Avant (Dideco, Mirandola, Italy). The
pump flow was targeted between 2.0 and 2.4 L · min · m2,
and the target mean arterial pressure was settled at 60
mm Hg. The gas flow was initially settled at 50% oxygen
to air ratio and a 1:2 flow ratio with the pump flow
indexed, and subsequently arranged in order to maintain
an arterial oxygen tension greater than 150 mm Hg and
an arterial carbon dioxide tension between 33 and 38 mm
Hg.
Ann Thorac Surg
2006;81:2189 –95
Anticoagulation was established with an initial dose of
300 IU per kilogram of body weight of porcine intestinal
heparin injected into a central venous line 10 minutes
before the initiation of CPB, and a target activated clot-
ting time of 480 seconds; patients receiving closed and
biocompatible circuits received a reduced dose of hepa-
rin with a target activated clotting time settled at 300
seconds. At the end of CPB, heparin was reversed by
protamine chloride at a 1:1 ratio of the loading dose,
regardless of the total heparin dosage.
Immediately after establishing CPB, and every 20 min-
utes, a standard arterial and mixed venous blood gas
analysis was performed on the arterial and venous blood
of the CPB circuit. Additional blood gas analyses were
done according to the perfusionist’s needs and in case of
hyperlactatemia.
Data Collection and Definitions
The following demographic and operative variables were
collected for each patient: age (years); gender; weight
(kgs); body surface area (BSA, m2); and CPB duration
(minutes). At each sampling time, the following variables
were recorded: pump flow indexed (L/min-1/m-2); arterial
oxygen tension (mm Hg); arterial oxygen saturation (%);
arterial carbon dioxide tension (mm Hg); arterial hemo-
globin (Hb) concentration (mg/dL); arterial lactate con-
centration (mMol/L); mixed venous oxygen tension (mm
Hg); mixed venous oxygen saturation (SVo2) (%); and
mixed venous carbon dioxide tension (mm Hg).
Blood gas analyses were performed using a blood gas
analyzer Nova Stat Profile (Nova Biomedical, Waltham,
MA). All blood gas data were corrected for temperature
according to standard equations.
Simultaneously, the carbon dioxide exhaled from the
oxygenator (eco2, mm Hg), and the gas flow into the
oxygenator (Ve) were recorded. Exhaled carbon dioxide
was measured with a mainstream capnograph Capnostat
(Novametrix Medical Systems Inc, Wallingford, CT).
Arterial and mixed venous oxygen content was calcu-
lated according to the following equation:
oxygen content (mL) ⫽ Hb (mg/dL) · 1.34 · Hb satura-
tion (%) ⫹ 0.003 · oxygen tension (mm Hg).
Carbon dioxide production (Vco2) was calculated ac-
cording to the following equation [14]:
VCO2 indexed 共mL · min-1 · m-2兲:
eCO2共mm Hg兲 · Ve共L/min兲 · 1,000
. (1)
760 · BSA共m2兲
Gas volumes and flows are expressed in standard tem-
perature 0° degrees, pressure 760 mm Hg, and dry
(STPD). Since gas pressures are expressed in body tem-
perature, ambient pressure, and saturated with water
vapor (BTPS), and considering that the body temperature
may change during CPB, the following relationship has
been applied:
VolumeSTPD⫽
共Barometric pressure-H2O vapor pressure兲 · 273
. (2)
760 · 共273⫹Body temperature兲
Ann Thorac Surg RANUCCI ET AL 2191
2006;81:2189 –95 HYPERLACTATEMIA DURING CPB

On the basis of the above data, the following oxygen Table 1. Correlation Between Oxygen-Carbon Dioxide
and carbon dioxide derived variables have been calcu- Derived Parameters, Other Intraoperative Variables and
lated: Arterial Lactate Concentration

(a) Arteriovenous oxygen content difference (mL); Correlation

CARDIOVASCULAR
(b) Oxygen consumption indexed (Vo2i): (mL · min-1 · Parameter Coefficient p
m-2): 10 · pump flow indexed (L · min-1 · m-2) · Arterial oxygen saturation (%) ⫺0.18 0.03
arteriovenous oxygen content difference (mL/100 Mixed venous oxygen saturation (%) ⫺0.19 0.06
mL); Arteriovenous oxygen content ⫺0.10 0.35
(c) Oxygen delivery indexed (Do2i): (mL · min-1 · m-2): difference (mL)
10 · pump flow indexed (L · min-1 · m-2) · arterial Oxygen consumption indexed (mL 0.12 0.26
oxygen content (mL/100 mL); · min⫺1 · m⫺2)
(d) Oxygen extraction ratio (O2ER) : Vo2i/ Do2i; Oxygen delivery indexed (mL · 0.03 0.68
(e) Venoarterial carbon dioxide tension difference min⫺1 · m⫺2)
(mm Hg): mixed venous carbon dioxide tension— Oxygen extraction ratio 0.19 0.06
arterial carbon dioxide tension; Arterial CO2 tension (mm Hg) ⫺0.02 0.90
(f) ⌬Pco2/C(a-v)O2: venoarterial carbon dioxide ten- Mixed venous CO2 tension (mm Hg) ⫺0.04 0.77
sion difference (mm Hg)/arteriovenous oxygen Venoarterial CO2 tension difference ⫺0.07 0.58
content difference (mL/100 mL); (mm Hg)
(g) Do2i/ Vco2i; ⌬Pco2/C(a-v)O2 0.02 0.88
(h) Respiratory quotient (RQ): Vco2i / Vo2i. Vco2 indexed (mL · min⫺1 · m⫺2) 0.72 ⬍ 0.001
Do2/Vco2 ⫺0.54 ⬍ 0.001
Hyperlactatemia was defined as an arterial blood lac-
Respiratory quotient 0.74 ⬍ 0.001
tate level greater than 3 mMol/L [5].
Pump flow indexed (L · min⫺1 · 0.12 0.11
m⫺2)
Statistical Analysis
Hemoglobin concentration (mg/dL) 0.05 0.55
Data are expressed as mean ⫾ standard deviation (con- Aortic cross-clamping on ⫺0.29 ⬍0.001
tinuous variables), or as frequency and percentage (cat-
Body surface area (m2) ⫺0.50 ⬍0.001
egoric variables). Operative and demographics variables,
CPB time (min) 0.18 0.02
and oxygen-carbon dioxide derived variables during CPB
Temperature (°C) 0.13 0.11
have been tested for association with arterial blood
lactate value, first using a bivariate linear regression CPB ⫽ cardiopulmonary bypass; ⌬Pco2/C(a-v)O2 ⫽ venoarterial car-
analysis and subsequently testing different regression bon dioxide tension difference/arteriovenous oxygen content difference;
Do2/Vco2 ⫽ oxygen delivery/carbon dioxide elimination; eco2 ⫽
analyses (linear, quadratic, cubic, exponential, logarith- exhaled carbon dioxide tension; Vco2 ⫽ carbon dioxide elimination.
mic, potential) for defining the best approximating equa-
tion. Factors being significantly associated to arterial
blood lactate value were subsequently tested for associ-
ation with hyperlactatemia, using an unpaired t test or a
Pearson’s ␹2 test when appropriate. for each factor. The Vco2i relationship with arterial blood
The predictive value of the variables associated to lactate concentration follows a cubic equation with a p
hyperlactatemia was tested using receiver operating value less than 0.001; the Do2/Vco2 ratio relationship
characteristics (ROC) curves. The area under the ROC with arterial blood lactate concentration follows a cubic
curve was used to define the best predictive variables; equation with a p value less than 0.001; and the Vco2/Vo2
adequate cutoff values have been searched based on the ratio relationship with arterial blood lactate concentra-
best coupling between sensitivity and specificity. For all tion follows a quadratic equation with a p value less than
the statistical tests, a p value less than 0.05 was consid- 0.001 (Fig 1). In all three cases, curvilinear equations
ered significant. demonstrated a higher correlation coefficient than simple
linear relationships (r2 values, respectively: 0.59 vs 0.52;
0.45 vs 0.29; 0.61 vs 0.54). In particular, the Vco2i and
Results Do2/Vco2 ratio relationships with arterial blood lactates
The various intraoperative factors considered in the tend to reach an asymptotic value for the higher levels of
study were tested for association with the arterial blood blood lactates, therefore reflecting the common clinical
lactate concentration (Table 1). Seven factors were signif- practice, where arterial blood lactates very rarely reach
icantly associated to arterial blood lactate concentration: values higher than 18 to 20 mMol/L.
a positive correlation was found for Vco2i, Vco2/Vo2 According to previous published papers [5] and to the
ratio, and CPB time; a negative correlation was found for usually accepted higher value for normal arterial lactates
Do2/Vco2 ratio, arterial oxygen saturation, aortic cross- concentration (2 mMol/L), hyperlactatemia was defined
clamping on, and BSA. A borderline (p ⫽ 0.06) correlation as an arterial blood lactate concentration greater than 3
was found for SVo2 (negative) and O2ER (positive). mMol/L. According to this cutoff value, the abovemen-
The univariate relationship for Vco2i, Do2/Vco2 ratio, tioned variables were tested with respect to the presence
and Vco2/Vo2 ratio was explored with a best-fit equation of hyperlactatemia (Table 2). Six factors were signifi-
 2192 RANUCCI ET AL Ann Thorac Surg
HYPERLACTATEMIA DURING CPB 2006;81:2189 –95
CARDIOVASCULAR

Fig 1. Relationships between arterial blood lactate concentration and (A) carbon dioxide production (Vco2i), (B) oxygen delivery (Do2i) to car-
bon dioxide production (Vco2i) ratio, and (C) respiratory quotient (Vco2i/Vo2i).

cantly different in normal versus hyperlactatemia condi-
tions: Vco2i, Do2i/Vco2i, Vco2i/Vo2i, aortic cross-clamp
on, BSA, and CPB time.
A receiver operating characteristic (ROC) analysis was
applied to each of the above variables (except the binary
variable aortic cross-clamp on) in order to assess their
predictive value for hyperlactatemia and the adequate
best cutoff values according to sensitivity and specificity.
An area under the curve (AUC) greater than 0.75 was
considered acceptable for predictivity [15]; the BSA and
CPB time failed to reach this value. The Do2i/Vco2i ratio
had an AUC of 0.852, the Vco2i had an AUC of 0.838, and
the Vco2i/Vo2i ratio had an AUC of 0.803 (Fig 2). The
complete analysis, with the best cutoff values identified
for the three variables is reported in Table 3. The best
predictive values for hyperlactatemia are a Do2i/Vco2i

Table 2. Univariate Analysis of Oxygen-Carbon Dioxide Derived Parameters and Other Intraoperative Variables at Arterial
Lactate Determinations Below or Above the Threshold Value (3 mmol/L).
Arterial Lactates ⱕ 3 mmol/L Arterial Lactates ⬎ 3 mmol/L
Parameter (n ⫽ 130) (n ⫽ 37) p

Pao2 (mmHg) 225 ⫾ 56 228 ⫾ 44 0.7

Svo2 0.78 ⫾ 0.76 0.76 ⫾ 0.8 0.15
VCO2i (mL · min⫺1 · m⫺2) 51.4 ⫾ 15.2 82.1 ⫾ 38.4 ⬍ 0.001
DO2i/VCO2i 6.35 ⫾ 1.7 4.14 ⫾ 1.2 ⬍ 0.001
VCO2i/VO2i 0.77 ⫾ 0.22 1.35 ⫾ 0.68 ⬍ 0.001
Aortic cross-clamp on 72% 46% 0.003
BSA (m2) 1.85 ⫾ 0.2 1.62 ⫾ 0.45 0.005
CPB time (min) 44.7 ⫾ 36.3 68.9 ⫾ 47.7 0.006

BSA ⫽ body surface area; CPB ⫽ cardiopulmonary bypass; Do2i/Vco2i ⫽ oxygen delivery indexed/carbon dioxide elimination indexed; Pao2
⫽ arterial oxygen tension; Svo2 ⫽ mixed venous oxygen saturation; Vco2i ⫽ carbon dioxide elimination indexed; Vco2i/Vo2i ⫽ respiratory
quotient.
Ann Thorac Surg
2006;81:2189 –95
Fig 2. Receiver operating characteristic curves for carbon dioxide
production (Vco2i), oxygen delivery to carbon dioxide production
ratio (Do2i/Vco2i), and respiratory quotient (Vco2i/Vo2i), as predic-
tors of hyperlactatemia. (— ⫽ Vco2i; - - - ⫽ Vco2i/Vo2i; ···· ⫽
Do2i/Vco2i.)
ratio lower than 5, a Vco2i higher than 60 mL · minute-1
· m-2, and a Vco2i/Vo2i ratio higher than 0.9.
Comment
Under normal resting conditions, the oxygen delivery
matches the overall metabolic demands of the organs, the
Vo2 is about 25% of the Do2, and energy is produced
basically through the aerobic mechanism (oxidative
phosphorylation). When the Do2 starts decreasing (due
to a decreased cardiac output, extreme hemodilution, or
both), the Vo2 is maintained until a “critical level” is
reached [16 –18]. Below this critical point the oxygen
consumption starts decreasing, becoming dependent on
the oxygen delivery, and the failing aerobic energy pro-
duction is progressively replaced by anaerobic adenosine
triphosphate production (pyruvate conversion to lactate).
As a result, blood lactate concentration starts rising, and
numerous studies have established the use of lactates as
a marker of global tissue hypoxia in circulatory shock
[19 –21]. Under these circumstances, the anaerobic me-
tabolism results in an excess of proton production and
tissue acidosis; buffering of the protons by bicarbonate
Table 3. Receiver Operating Characteristic Analysis and Relative Cutoff Values
Factor AUC p Cutoff Value
Do2i/Vco2i 0.852 ⬍ 0.001
Vco2i 0.838 ⬍ 0.001
Vco2i/Vo2i 0.803 ⬍ 0.001
AUC ⫽ area under the curve; Do2i/Vco2i ⫽ oxygen delivery indexed/ carbon dioxide elimination indexed;
indexed; Vco2i/Vo2i ⫽ respiratory quotient.
RANUCCI ET AL 2193
HYPERLACTATEMIA DURING CPB
ions results, in turn, in an anaerobic carbon dioxide
production [22]. Therefore, below the critical Do2, there is
a linear decrease of both Vo2 and Vco2, but due to the
anaerobic CO2 production, the RQ increases.
When the critical Do2 is reached due to a decrease in
CARDIOVASCULAR
cardiac output (cardiogenic shock), the above relation-
ship becomes more complex. Due to the reduced pulmo-
nary flow and to ventilation-perfusion mismatch the
ability of the lung to eliminate carbon dioxide is im-
paired, and carbon dioxide elimination and end-tidal
carbon dioxide tension are decreased [12]. Consequently,
carbon dioxide starts accumulating in the venous com-
partment, and venoarterial carbon dioxide gradient is
increased. In other terms, the Vco2 (intended as carbon
dioxide production by the tissues) becomes progressively
higher than carbon dioxide elimination.
Under CPB conditions the above pattern changes
again. The artificial lung is much more efficient than the
natural lung in terms of carbon dioxide clearance, and is
maintained even for a very low pump flow. Not by
chance, under specific circumstances like deep hypother-
mia and according to the pH strategy, it is clinically
needed to add carbon dioxide to the gas flow in order to
avoid dramatic and dangerous patterns of hypocapnia. In
this setting, the Vco2 is strictly correlated to the carbon
dioxide elimination. Therefore, while in a normal setting
the venous carbon dioxide tension (Pvco2) is inversely
correlated to the carbon dioxide elimination [12], during
CPB the two parameters are positively correlated, as we
could check through a linear regression analysis in our
patient population (Vco2i ⫽ ⫺6.7 ⫹ 1.67Pvco2; r2 ⫽ 0.11,
p ⫽ 0.005).
On the basis of the above pathophysiological consid-
erations, our results may be interpreted in the following
ways.
(1) At the lactate threshold of 3 mMol/L, there is an
increase of Vco2i and RQ above their respective
cutoff values of 60 mL -1 · m2 and 0.9. This behavior
reflects the increased anaerobic carbon dioxide
production with concomitant normal or slightly
decreased Vo2.
(2) The best predictor of lactate threshold is the Do2i/
Vco2i ratio, with a cutoff value at 5. Actually, the
normal Do2i/Vco2i is 5, being the Do2 about 1,000
mL/minute and the Vco2 about 200 mL/minute.
This ratio is maintained until the critical Do2 is
reached, because above this limit the Vo2 does not
change and the aerobic-derived Vco2 is un-
changed as well. Below the critical Do2 the Vo2
Sensitivity Specificity
4.99 78.4% 74%
60.7 75.7% 70.7%
0.90 75.7% 77.2%
Vco2i ⫽ carbon dioxide elimination
 2194 RANUCCI ET AL
HYPERLACTATEMIA DURING CPB
decreases, the aerobic-derived Vco2 decreases in a
linear fashion with Vo2, but the total Vco2 de-
creases less than the Vo2 due to the contribution of
the anaerobic-derived Vco2. Therefore, the Do2i/
Vco2i decreases below 5.
CARDIOVASCULAR
(3) The venoarterial carbon dioxide tension gradient,
and its ratio with the arteriovenous oxygen satu-
ration, which in previous papers had a clear cor-
relation with the arterial blood lactate concentra-
tion in patients not under CPB [12, 13], failed to
demonstrate this association during CPB. Again,
we must consider that the artificial lung is much
more efficient than the natural lung in terms of
carbon dioxide elimination; therefore, the effect of
venous blood carbon dioxide accumulation in case
of critical Do2 is blunted by the artificial lung
carbon dioxide removal, and the excess carbon
dioxide anaerobically produced is found at the gas
exhaled site of the oxygenator (eco2) rather than in
the venous compartment.
(4) The value of oxygen derived parameters (namely,
the SVo2) is poor in terms of predictivity for the
lactate threshold during CPB. Our data are in
agreement with other observations [15], demon-
strating that SVo2 and other oxygen derived pa-
rameters are not predictive for hyperlactatemia. It
has been demonstrated that during CPB systemic
microvascular control may become disordered,
inducing peripheral arteriovenous shunting that is
associated to a rise in lactate levels despite an
apparently adequate oxygen supply [6]. Moreover,
selective splanchnic hypoperfusion has been con-
sidered responsible for the production of lactate
during CPB [1, 8]. Even considering that under
these circumstances the venous blood from the
splanchnic district has probably a low oxygen
content, the mixing of this blood with highly
oxygen saturated blood from many other organs at
metabolic rest during anesthesia may result in a
normal oxygen content of the mixed venous blood.
(5) Other determinants of lactate production, albeit
nonpredictive factors for hyperlactatemia, are the
CPB duration, a small BSA, and the release of the
aortic cross-clamp. Cardiopulmonary bypass du-
ration already has been mentioned as a determi-
nant of lactate concentration [6]. A small BSA is
often associated to a poor venous blood return to
the heart-lung machine, with consequent reduced
pump flow, and to a higher hemodilution. Both
these factors may determine a reduced oxygen
delivery during CPB. Finally, the release of the
aortic cross-clamp admits to the systemic circula-
tion the previously ischemic heart, with a release
of lactates from the coronary circulation that has
been already demonstrated in studies dealing with
myocardial ischemia-reperfusion during cardiac
operations [23].
Body temperature during CPB was not significantly
correlated to arterial blood lactate values, but only dem-
Ann Thorac Surg
2006;81:2189 –95
onstrated a trend (p ⫽ 0.11). This apparently could be
difficult to explain, as it is reasonable to hypothesize that
with increasing metabolic needs the likelihood of having
an inadequate oxygen supply may be higher. However,
some of our patients developed an anaerobic status
before the operation (emergency procedures) or during
the operation, before going on CPB (due to overt or
subtle low cardiac output), and therefore the relationship
is probably biased by this condition.
We are aware that interpretation of lactate measure-
ment requires caution. The lactate concentration de-
pends on a balance between production and clearance;
while the first is very rapid, the second depends on
metabolic elimination and requires a prolonged (hours)
time in critical patients [24]. Therefore, the presence of an
elevated lactate concentration in blood does not neces-
sarily mean that the anaerobic metabolism is activated at
that time, often being associated to lactate production
which occurred maybe hours before. For this reason, in
our series we have considered only the serial measure-
ments until the highest lactate concentration was
reached, not considering the relationship between oxy-
gen and carbon dioxide derived parameters and lactate
concentration when (and if) the lactate concentration
started decreasing.
This study is not intended to address the complex topic
of the origin of hyperlactatemia during CPB, but to
identify predictive parameters being clinically measur-
able in a continuous way. To this respect, we believe that
the online monitoring of carbon dioxide derived param-
eters, together with the oxygen delivery, may be of
considerable aid during CPB in order to optimize the
pump flow, the arterial oxygen content, and therefore the
oxygen delivery, to finally avoid the establishment of a
critical hyperlactatemia that has a well-defined role in
determining postoperative morbidity and mortality.
References
1. Landow L. Splanchnic lactate production in cardiac surgery
patients. Crit Care Med 1993;21(suppl):S84 –S91.
2. Raper RF, Cameron G, Walker D, Bowey CJ. Type B lactic
acidosis following cardiopulmonary bypass. Crit Care Med
1997;25:46 –51.
3. Boldt J, Piper S, Murray P, Lehmann A. Severe lactic acidosis
after cardiac surgery: sign of perfusion deficits. J Cardiotho-
rac Vasc Anesth 1999;13:220 – 4.
4. Totaro R, Raper RF. Epinephrine induced lactic acidosis
following cardiopulmonary bypass. Crit Care Med 1997;25:
1693–9.
5. Maillet J-M, Le Besnerais P, Cantoni M, et al. Frequency, risk
factors, and outcome of hyperlactatemia after cardiac sur-
gery. Chest 2003;123:1361– 6.
6. Demers P, Elkouri S, Martineau R, Couturier A, Cartier R.
Outcome with high blood lactate levels during cardiopulmo-
nary bypass in adult cardiac operation. Ann Thorac Surg
2000;70:2082– 6.
7. Fiaccadori E, Vezzani A, Coffrini E, et al. Cell metabolism in
patients undergoing major valvular heart surgery: relation-
ship with intra- and postoperative hemodynamics, oxygen
transport and oxygen utilization patterns. Crit Care Med
1989;17:1286 –92.
Ann Thorac Surg
2006;81:2189 –95
8. Landow L, Phillips DA, Heard SO, et al. Gastric tonometry
and venous oximetry in cardiac surgery patients. Crit Care
Med 1991;19:1226 –33.
9. Habib RH, Zacharias A, Schwann TA, Riordan CJ, Durham
SJ, Shah A. Adverse effects of low hematocrit during cardio-
pulmonary bypass in the adult: should current practice be
changed? J Thorac Cardiovasc Surg 2003;125:1438 –50.
10. Karkouti K, Beattie WS, Wijeysundera DN, et al. Hemodilu-
tion during cardiopulmonary bypass is an independent risk
factor for acute renal failure in adult cardiac surgery. J Tho-
rac Cardiovasc Surg 2005;129:391– 400.
11. Ranucci M, Romitti F, Isgrò G, et al. Oxygen delivery during
cardiopulmonary bypass and acute renal failure following
coronary operations. Ann Thorac Surg 2005;80:2213–20.
12. Zhang H, Vincent JL. Arteriovenous differences in Pco2 and
pH are good indicators of critical hypoperfusion. Am Rev
Respir Dis 1993;148:867–71.
13. Mekontso-Dessap A, Castelain V, Anguel N, et al. Combi-
nation of venoarterial PCO2 difference with arteriovenous
O2 content difference to detect anaerobic metabolism in
patients. Intensive Care Med 2002;28:272–7.
14. Jaffe MB, Orr J. Combining flow and carbon dioxide. In:
Gravenstein JS, Jaffe MB, Paulus DA, eds. Capnography:
clinical aspects. Cambridge, UK: Cambridge University
Press; 2004:423-6.
15. Myles PS, Gin T. Statistical methods for anesthesia and
intensive care. Oxford, UK: Butterworth-Heinemann; 2000.
The Society of Thoracic Surgeons Policy Action Center
The Society of Thoracic Surgeons (STS) is pleased to
announce a new member benefit—the STS Policy Action
Center, a website that allows STS members to participate
in change in Washington, DC. This easy, interactive,
hassle-free site allows members to:
● Personally contact legislators with one’s input on key
issues relevant to cardiothoracic surgery
● Write and send an editorial opinion to one’s local media
© 2006 by The Society of Thoracic Surgeons
Published by Elsevier Inc
RANUCCI ET AL 2195
HYPERLACTATEMIA DURING CPB
16. Soni N, Fawcett WJ, Halliday FC. Beyond the lung: oxygen
delivery and tissue oxygenation. Anaesthesia 1993;48:704 –11.
17. Pinsky MR. Beyond global oxygen supply-demand relations:
in search of measures of dysoxia. Intensive Care Med
1994;20:1–3.
18. Steltzer H, Hiesmayr M, Mayer N, Krafft P, Hammerle AF.
CARDIOVASCULAR
The relationship between oxygen delivery and uptake in the
critically ill: is there a critical optimal therapeutic value?
Anaesthesia 1994;49:229 –36.
19. Bakker J, Coffernils M, Leon M, Gris P, Vincent JL. Blood
lactate levels are superior to oxygen derived variables in
predicting outcome in human septic shock. Chest 1991;99:
956 – 62.
20. Weil MH, Afifi AA. Experimental and clinical studies on
lactate and pyruvate as indicators of the severity of acute
circulatory failure (shock). Circulation 1970;41:989 –1001.
21. Vincent JL, Dufaye P, Berre J, Leeman M, Degaute JP, Kahn
RJ. Serial determinations during circulatory shock. Crit Care
Med 1983;11:449 –51.
22. Randall HM Jr, Cohen JJ. Anaerobic CO2 production by dog
kidney in vitro. Am J Physiol 1966;211:493–505.
23. Elvenes OP, Korvald C, Myklebust R, Sorlie D. Warm
retrograde blood cardioplegia saves more ischemic myocar-
dium but may cause a functional impairment compared to
cold crystalloid. Eur J Cardiothorac Surg 2002;22:402–9.
24. Nguyen HB, Rivers EP, Knoblich BP, et al. Early lactate
clearance is associated with improved outcome in severe
sepsis and shock. Crit Care Med 2004;32:1637– 42.
● E-mail senators and representatives about upcoming
medical liability reform legislation
● Track congressional campaigns in one’s district—and
become involved
● Research the proposed policies that help— or hurt—
one’s practice
● Take action on behalf of cardiothoracic surgery
This website is now available at www.sts.org/takeaction.
Ann Thorac Surg 2006;81:2195 • 0003-4975/06/$32.00