Real time, continuous monitoring of cardiac output and
oxygen delivery
M Jonas, Consultant Intensivist, D Hett,
Consultant Cardiothoracic Anaesthetist, J Morgan, Consultant
Cardiologist, Southampton Hospitals University Trust,
UK
Oxygen delivery is the product of the cardiac output and the
arterial oxygen content, and represents the amount of oxygen delivered to
the tissues in the arterial supply. The clinical rel-evance of these
physiological values has been sub-stantiated by well conducted,
randomised, con-trolled trials 1-4, which have provided
convincing evidence for measurement and manipulation of oxygen delivery in
critically ill patients. These studies conclude that increasing the
cardiac out-put - and therefore oxygen delivery - in selected patients
reduces the risk of them dying. A retrospective analysis of one of the
initial studies 1 determined that survival was strongly
associated with cardiac output, oxygen delivery and oxygen consumption (i.
e., the determinants of oxygen flux). The review analysed the relative
importance of monitored variables in survivors and non-survivors and
concluded that achieving a targeted oxygen delivery index of 600 ml/
minute/ m 2 , reduced mortality and morbidity in high-risk
surgical patients. More recent stud-ies 3- 5 and a consensus conference 6
have support-ed the concept that achieving a target oxygen delivery, as an
'optimising' procedure pre-opera-tively in high-risk surgical patients,
saves lives and reduces morbidity. Even in sepsis, evidence of
cardiovascular manipulation with survival benefits has recently been
shown. 7 If all of the reviewed studies arrive at the same explicit
conclusion in terms of oxygen de-livery and survival in selected patients,
why have these strategies not been adopted clinical-ly?
MEASUREMENT OF CARDIAC OUTPUT The reason for the
slow translation of evidence-based research into clinical practice appears
to be mainly technological. Until recently, there were few devices
available to measure cardiac output without recourse to pulmonary artery
catheterisation with its associated problems. For most clinicians, one of
the major problems in dealing with haemodynamics especially in unstable
patients, is the difficulty in estimating flow without attempting to
measure it. Because of this, given the relative simplicity of measur-ing
pressures, the combination of clinical/ labora-tory assessment and blood
pressure measure-ment are frequently employed as surrogates for flow
measurement. Unfortunately clinical assessment is often wrong 8 and there
is no rela-tionship between changes in arterial pressure and flow
(mathematically the relationship be-tween flow and pressure requires
reference to
the vascular resistance). This means
that clini-cians are relying heavily on pressure measure-ment as an index
to perfusion despite available data actually showing that there is
virtually no correlation between observed changes in pres-sure and changes
in flow. This is the clinician's problem: does a hypotensive patient
require fill-ing or vasoactive drugs, vasodilators or vasocon-strictors?
An unpublished study in which clinicians' esti-mates of cardiac
output were compared to indica-tor dilution clearly illustrates this
(Bruce R, Knight R, Jonas M et al. Personal communica-tion). For 40
patients on a general ICU, the clini-cian responsible for each patient -
using all available data and clinical examination - was asked to estimate
the cardiac output and sys-temic vascular resistance, using all available
data and clinical examination, and to categorise their clinical findings
as high, normal or low. When compared with the results of indicator
dilution, 39% of cardiac output and 51% of sys-temic vascular resistance
estimates were incor-rect. Significantly, the clinician, when informed of
the measured cardiac output, instigated a change of therapy in 42% of the
patients.
This inability to guess the global cardiac out-put
correctly, without measurement has led to the development of a number of
methods to esti-mate cardiac output. Each method has to be evaluated with
reference to the additional clini-cal risk, the potential for measurement
errors and the requirement for technical expertise, all of which will
limit the device's utility. In the cur-rent healthcare environment, the
overwhelming critical issue for any device is the incremental risk to the
patient in obtaining a flow measure-ment. This is the major factor
generating con-cern over pulmonary artery catheterisation, 9-11
and also the motor behind development of alter-native, less invasive
techniques.
A variety of new methods for measuring car-diac output
have been developed. All of these methods attempt to reduce clinical risk
to the patient by being less invasive in clinical use, this representing
one of the first attributes of the ideal cardiac output monitor (Table 1).
Few de-vices have addressed these concepts, especially that of real time,
continuous monitoring, which would logically deliver considerable benefits
(Table 2).
LIDCOTM/ PULSECO
TM The LiDCOTM/ PulseCOTM
device represents a combination of a simple indicator dilution
tech-
INTERNATIONAL JOURNAL OF INTENSIVE CARE SPRING
2002
Real
time continuous monitoring
Table 1. Properties of the ideal
monitor
Minimally invasive and therefore widely applicable
Accurate
Real
time: beat-to-beat CO
Real
time: preload + afterload
Real
time: oxygen delivery
Nurse
driven
Data
interpretation
Bedside
information management
Neonates to aduts
Table 2. Benefits of continuous cardiac output
True
monitor = early warning of deterioration
Weight
of scientific evidence for improved outcome
Optimum
fluid management
Rational drug administration (e. g., inotropes)
Optimising patient- ventilator interaction
Patient
'condition' communication to clinical staff
Reduced
work of health care staff
Decreased procedural complications (e. g., bolus
injections)
nique used to calibrate an arterial waveform
analysis algorithm. Theoretically this combina-tion should provide
beat-by-beat measurement of cardiac output, with little clinical
increment of risk, assuming that in these critically ill patients
arterial and venous lines would already be in situ.
LiDCOTM indicator dilution A bolus
indicator dilution technique for measur-ing cardiac output was
originally described by Henriques and developed by Hamilton et al.,
12 based on the concept of the dilution of a known amount of
indicator. The original technique, using indocyanine green, is
technically difficult, time consuming and requires frequent blood
sampling and analysis. The use of lithium as an alternative
indicator for the estimation of cardiac output was first described
in 1993 13 and has now been exten-sively validated. 14 In brief,
isotonic lithium chlo-ride (150 mM) is injected as a bolus (0.002-
0.004 mmol/ kg) via the central, or peripheral, venous route and a
concentration- time curve generated by an ion-selective electrode
attached to the arte-rial line manometer system. The cardiac output
is calculated from the lithium dose and the area under the
concentration-time curve prior to recir-culation 15 using equation
1:
Cardiac output
=
Lithium dose (mmol) x 60
Area x (1 - PCV) (mmol/
second)
where the
area is the integral of the primary curve, and PCV is packed cell
volume (Hb (g/ dl)/ 34). (A correction for PCV is necessary because
lithium is distributed in the plasma.) The voltage response of the
lithium ion-sensi-tive electrode is to percentage change of ion
con-centration, and as lithium is not normally pre-sent in the
plasma, extremely small doses can be used. These doses are too small
to exert pharma-cological effects. Multiple dosing with lithium has
been extensively investigated and the phar-
macokinetics of
intravenous lithium chloride in man and in animals has been
described 16 . The safety profile is well established,
with the recom-mended maximum total dose having to be exce-eded many
times before toxic levels are reached. For indicator dilution
theory, a critical bound-ary condition is that there is no first
pass loss of marker from the circulation, as this would lead to
overestimation of cardiac output. To investi-gate this for lithium
chloride, a study in patients comparing measurements of LiDCO using
right or left atrial injection of lithium was undertak-en, which
clearly showed that there was no sig-nificant loss of lithium during
its passage through the pulmonary circulation. 15 The
technical innovation of the LiDCO system is the design and
application of the ion-selective electrode (Figure 1), which
comprises a lithium sensitive electrode situated in a flow-through
cell. The electrode is disposable and packaged sterile. Set-up is
simple and swift, the electrode is primed and is attached to the
arterial manometer line via a three-way tap (Figure 2). When set up
and the tap opened, blood flows into the sensor assembly at a rate
that is controlled by a peristaltic, bat-tery- powered pump to
remain at 4 ml/ minute.
Figure
1.
The lithium selective elec-trodein the flow-through cell.
Figure 2.
The
LiDCOTM/ PulseCOTMsystem. Blood is sampled
from the arterial line via the three-way tap in the manometer
line.
INTERNATIONAL JOURNAL OF INTENSIVE CARE SPRING
2002
Real
time continuous monitoring
patients had considerable scatter and higher readings than
LiDCO (Figure 3). The lithium technique has also been success-fully used
in paediatrics. Children were studied comparing lithium dilution
measurements with transpulmonary thermodilution as many of these patients
were too small for pulmonary artery catheterisation. The transpulmonary
technique (Pulsion COLD) uses cold dilution with a ther-mistor placed in
the aorta via the femoral artery, rather than the pulmonary artery. Again
agree-ment was very good between lithium-and ther-modilution.
18
Larger animals have also been studied, 19 to ensure
that the technique remained valid des-pite large differences in body
weight and shape - the weights studied ranged from a 2 kg baby up to a 558
kg horse (Figure 4).
These initial studies of the LiDCO system
measured cardiac output by central venous boluses of the marker lithium.
Although many critically ill patients have central venous access as well
as arterial access, there would be a con-siderable advantage to being able
to measure cardiac output in those patients without central venous access
using a peripheral vein. A study of patients on the ICU, with both central
and peripheral lines, showed that the correlation be-tween peripherally
and centrally injected lithi-um dilution was excellent 20 (r =
0.997), and this and subsequent studies have validated peripher-al bolus
LiDCO. In essence, cardiac output mea-surements can be made in patients
with just arterial and peripheral venous access.
Limitations
of LiDCO Because the concentration change of lithium is used
to calculate the cardiac output, this tech-nique cannot be used in
patients receiving lithi-um therapy, since the increased background
lithium concentration causes an overestimation of cardiac output. The
electrode may also drift in the presence of certain muscle relaxant
infusions and so cause inaccurate measurements. If mus-cle paralysis is
used, bolus techniques of admin-istration have to be adopted.
A
right- left shunt will cause obvious distor-tion of the initial part of
the dilution curve and a left- right shunt will result in the right
ventricu-lar output being higher than the flow into the aorta. In the
paediatric study, analysis of the concentration-time curves clearly
demonstrated anatomical shunts which were otherwise unsus-pected (Figure
5).
The initial upstroke is slurred due to the early
appearance of some lithium through the shunt. As mixing is incomplete, the
LiDCO cur-ves cannot be used to quantify the shunt in this situation
because the amount passing through the shunt is not in proportion to the
flow. PulseCOTM ThePulseCO monitor calculates continous car-diac output following
LiDCO calibration, by an-
Figure
3.X-Y plot
and Bland-Altman plot of 40 patients comparing LiDCOTM(average
of 5 measurements) with bolus thermodilution (BTD) in 40 patients. X-Y
plot: LiDCO = 0.31 + 0.89 BTD (L/ minute) r 2 = 0.94. Bland-Altman plot:
Mean differences BTD -LiDCO 0.25 L/ minute, SD of the dif-ference was
0.46 L/ minute.
Figure 4. Comparisons
(n = 318) of LiDCOTM vs bolus thermodilution in adults,
paediatrics and horses.
The flow-through cell is designed with an eccen-tric inlet
so that blood swirls past the tip of the electrode, which contains a
membrane that is selectively permeable to lithium. The voltage across the
membrane is related by the Nernst equation to the plasma lithium
concentration. A correction is applied for plasma sodium concen-tration
because, in the absence of lithium, the baseline voltage is determined by
the sodium con-centration. The electrode is made of polyurethane with a
central lumen. A wick, which is soaked in heparinised saline when the cell
is first primed, makes the electrical connection between the blood at the
tip of the electrode and the remote refer-ence. The voltage is measured
using an isolated amplifier, digitalised and analysed on-line.
Validation: comparison with other
techniques The LiDCO system has been extensively validat-ed
in adults, paediatric patients and animals. Validation comparing LiDCO
with bolus pul-monary artery catheter thermodilution 17 demon-strated a
good overall agreement between the two methods (r2 = 0.94).
LiDCO was at least as accurate as bolus thermodilution, with
signifi-cantly greater precision. The coefficient of deter-mination
improved to r2 = 0.96 if the two patients in whom the pulmonary
artery catheter was probably malpositioned were excluded. The bolus
thermodilution measurements in these
INTERNATIONAL JOURNAL OF INTENSIVE CARE SPRING
2002
Real
time continuous monitoring
alysis of the arterial blood pressure trace. The
arterial blood pressure trace undergoes a three-step transformation.
Step 1 - arterial pressure transformation into
a volume-time waveform
An accurate way of
determining the change in blood volume in the arterial tree from
maximum to minimum dilatation would allow an estimate of the volume
of blood flowing out of the arterial tree during a period slightly
longer than diastole. Since the whole period of the cardiac cycle
usually bears a fixed relation to diastole, simple scaling would
give the stroke volume. The relationship between the capacity of the
arterial side of the cir-culation and the intravascular pressure can
be expressed as the compliance (i. e., pressure change per unit
volume change). This relationship would be straightforward if the
compliance were con-stant. However, arterial compliance changes as
arterial pressure changes. A stiffening of the vas-culature occurs
as pressure and volume increase such that, at higher pressures, a
given increase in pressure expands the arterial tree by a smaller
volume. Nevertheless, the form of this curvilinear relationship,
though differing in its scaling, appears to be very similar in
different subjects.
Using a lookup table, the pressure
waveform can be used as the basis for calculating a contin-uous
curve describing the general form of the arterial volume changes
with every cardiac cycle for which arterial blood pressure is
available.
Step 2 - deriving nominal stroke volume and
the heartbeat duration In order to obtain cardiac output
as volume per unit time, the algorithm needs to calculate the
duration of the cardiac cycle and the stroke vol-ume, or a value
proportional to it (the nominal stroke volume). The mathematical
technique of autocorrelation can be used to give both these values.
Nominal stroke volume: initially the soft-ware
subtracts the mean value of the derived arterial blood volume
record, giv-ing a description of how much the arterial blood
volume changes around it. This is periodic like a sine wave but
with differ-ently shaped areas above and below zero. Figure 6
shows how the method works by first using a pure sine wave and
then sub-jecting it to the algorithm. Initially an estimate of the
mean deviation from zero is obtained by multiplying all values of
the waveform by themselves. This gives posi-tive waves for both
the positive and nega-tive parts of the original sine wave,
creat-ing a double waveform. The mean of the values of this new
waveform is the mean square and the square root of this value is a
constant proportion of the amplitude of the original waveform -
known as the root mean square value). This value is approxi
Figure 5. Lithium dilution curve in a paediatric patient aged
5 days and weighing 4 kg. The lithium dose was 0.015 mmol. The
purple line delineates the primary curve and the green line is the
result of subtracting that primary curve from the original curve.
The discrepancy between the two curves at the take off was due to a
right-to-left shunt through a patent foramen ovale. The yellow line
is the recorded curve.
mately 0.7 of the waveform
amplitude and is linearly related to the stroke volume. Figure
6 shows the original sine wave and the squared (double)
waveform is shown for three cycles.
Estimation of the heart beat duration: hav-ing
determined the nominal stroke volume, the precise period of the
cycle can be obtained by moving one version of the vol-ume
waveform relative to another. Again, for autocorrelation, cross
multiplication and addition of the values deliver values that are
both positive and negative. The sum for a given displacement, or
the tau shift, becomes less with maximum opposi-tion of the two
derived versions of the waveform and increases as the waveforms
reinforce each other. Continuation of the step-by-step movement of
one version of the waveform relative to the other gener-ates an
autocorrelogram with a series of maxima and minima at tau shifts,
which represent the duration of the cardiac cycle.
Step 3 - nominal stroke volume and
calibration The algorithm derived stroke volume and
there-fore cardiac output are initially uncalibrated.
Figure 6.
Autocorrelation demonstra-ted with a pure sine wave (see
text).
INTERNATIONAL JOURNAL OF INTENSIVE CARE SPRING
2002
Real
time continuous monitoring
Figure 7. Comparison of the continu-ous
PulseCOTM readings in 10 patients with intermittent
LiDCOTM readings. The mean continuous CO was 8.2 L/
minute (range 5.3- 17.1 L/ minute) and mean ID was 7.9 L/ min (range
4.6- 14.8 L/ minute) The mean bias was -0.3 L/ minute and precision
was 0.85 L/ minute.
They are converted to actual values by multiply-ing
the nominal stroke volume by a calibration factor. This is a
patient-specific correction factor generated by the
PulseCOTM algorithm when the nominal data are corrected
to actual data by a LiDCO calibration. In summary, raw haemody-namic
data from the patient bedside monitor are converted to volume using
the pressure-volume transform and autocorrelation. The lithium
dilu-tion cardiac is performed and the result is en-tered into the
calibration screen to derive actual cardiac output from
PulseCOTM.
VALIDATION
Experience with this system is rapidly develop-ing. Several
studies have recently been complet-ed and suggest good accuracy and
precision. The data from a recent study in Southampton 21 are shown
in Figure 7. This study compared CO measurements in ITU patients
over a 9-hour period using PulseCOTM as compared with the
LiDCOTM intermittent indicator dilution tech-nique. The
objective was to analyse drift of the CCO over time from initial
calibration in ITU patients. This is a problem frequently found in
Figure 9.
PulseCOTM data screen showing continuous cardiac
output and oxygen delivery.
morphology dependent algorithms. Ten ITU patients
with a variety of diagnoses were stud-ied; eight patients had radial
arterial lines and two had femoral arterial lines. The
PulseCOTM device was connected to the patient monitor and
initially calibrated with two consecutive indica-tor dilutions. At
2.5 hours, 5 hours and 7.5 hours indicator dilutions were performed
and the results were compared to those of the autocorre-lation
waveform analysis. The values from PulseCOTM and the
LiDCOTM measurements agreed closely over the 9-hour study
period. Similar accuracy and utility has also been reported from
studies in cardiothoracic and pae-diatric patients. 22,23
ARTERIAL
PRESSURE VARIATIONS AND PRELOAD
An
adequate preload is an essential component of the resuscitation and
management of the criti-cally ill. The use of central venous or
pulmonary artery catheter estimates of vascular filling is
controversial and at best a coarse guide to intravascular filling.
Several studies 24,25 have documented the poor correlation of the
pul-monary artery occlusion pressure (PAOP) with left ventricular
end diastolic volume (LVEDV). The implication is that PAOP (and
central ven-ous pressure) only provide qualitative assess-ment of
over or under-filling, i. e., high or low values; intermediate
values are difficult to in-terpret.
The PulseCOTM algorithm continuously
mea-sures arterial pressure variations. These pres-sure variations
have to be measured real time and are represented as stroke volume
variation, systolic pressure variation and pulse pressure variation.
Stroke volume variation is the reduc-tion in left ventricular stroke
volume associated with reduced venous return during positive
pres-sure ventilation. Pulse pressure variation is defined as the
maximum pulse pressure minus the minimum, divided by the average of
these two pressures. The systolic pressure variation is the
difference between the maximum and mini-mum systolic pressure
following inspiration. Physiologically these variations are thought
to be due to cyclical changes in intrathoracic pres-sure with
positive pressure ventilation resulting in transient changes in
ventricular preload and hence cardiac output. Several recent studies
have suggested that these variations, in mech-anically ventilated
patients, are sensitive predic-tors of the response of the left
ventricle to vol-ume administration and can be used as a guide for
fluid therapy. 24-29
CONCLUSION
This is a new system, which integrates a novel indicator
dilution technique with an arterial waveform analysis program. The
system is fully configurable with touch screen technology and expert
software for data interpretation. It is not
INTERNATIONAL JOURNAL OF INTENSIVE CARE SPRING
2002
Real
time continuous monitoring
artery specific or waveform shape dependent and appears to
be accurate in adults and paediatrics. Further studies are shortly to be
published. If old technology produced the inertia for the slow translation
of scientific evidence into patient care, then the arrival of reliable,
mini-mally invasive and continuous technology should encourage an
accelerating change in clinical practice.
REFERENCES
1.
Shoemaker WC. Monitoring and management
of the high risk surgical patient. Care Crit Ill 1990; 6: 39- 47.
2.
Shoemaker WC, Appel PL, Kram HB et al.
Prospective trial of supranormal values of survivors as therapeutic
goals in high risk surgical patients. Chest 1988; 94: 1176-
1186.
3.
Boyd O, Grounds RM, Bennet D et al. A
randomized clini-cal trial of the effect of deliberate perioperative
increase of oxygen delivery on mortality in high-risk surgical
patients. JAMA 1993; 270: 2699- 2707.
4.
Wilson J, Woods I, Fawcett J et al.
Reducing the risk of major elective surgery: randomised controlled
trial of pre-operative optimisation of oxygen delivery. BMJ 1999;
318: 1099- 1103.
5.
Grocott M, Hamilton M, Rowan K.
Targeted increases in perioperative global blood flow to explicit
goals and surgi-cal mortality. 8th World Congress Intensive Care
Sydney, Australia, November 2001 (abstract).
6.
Grocott MPW, Ball JAS. Consensus
meeting: management of the high risk surgical patient. Int J Crit
Care Med 2000; 11( Special report): 1- 19.
7.
Rivers E, Nguyen B, Hallstad S et al.
Early goal-directed therapy in the treatment of severe sepsis and
septic shock. N Engl J Med 2001; 345: 1368- 1367.
8.
Tibby SM, Hatherill M, Marsh MJ et al.
Clinicians' abili-ties to estimate cardiac index in ventilated
children and infants. Arch Dis Child 1997; 77: 516- 518.
9.
Connors AF Jr, Speroff T, Dawson NV et
al. The effective-ness of right heart catheterisation in the initial
care of critically ill patients. JAMA 1996; 276: 889- 897.
10.
Boyd KD, Thomas SJ, Gold J et al. A
prospective study of complications of pulmonary artery
catheterisations in 500 consecutive patients. Chest 1983; 84:
245- 249.
11.
Dalen JE, Bone RC. Is it time
to pull the pulmonary artery catheter? JAMA 1996; 276: 916-
918.
12.
Hamilton WF, Moore JW, Kinsman J et al.
Studies on the circulation IV. Further analysis of the injection
methods and of changes in hemodynamics under physiological and
pathological conditions. Am J Physiol 1932; 99: 534- 551.
13.
Linton R, Band DM, Haire KM. A new
method of measur-ing cardiac output in man using lithium dilution.
Br J Anaesth 1993; 71: 262- 266.
14.
Linton RAF, Band DM, O'Brien T et al.
Lithium dilution cardiac output measurement - a brief review. In:
Ikeda K, Doi M, Kazama T, (eds.), State-of-the-Art Technology in
Anaesthesia and Intensive Care. Amsterdam: Elsevier Science BV,
1998: 61- 66.
15.
Band DM, Linton RA, Jonas MM et al. The
shape of indi-cator dilution curves used for cardiac output
measurement in man. J Physiol 1997; 498: 225- 229.
16.
Jonas MM, Linton RA, O'Brien TK et al.
The pharmacoki-netics of intravenous lithium chloride in patients
and nor-mal volunteers. J Trace Elements Microprobe Techniques 2001;
19: 313- 320.
17.
Linton R, Band D, O'Brien T et al.
Lithium dilution car-diac output measurement: a comparison with
thermodilu-tion. Crit Care Med 1997; 25: 1796- 1800.
18.
Linton RA, Jonas MM, Tibby SM et al.
Cardiac output measured by lithium dilution and transpulmonary
ther-modilution in a paediatric intensive care unit. Intensive Care
Med 2000; 26: 1507- 1511.
19.
Linton RA, Young LE, Marlin DJ et al.
Cardiac output measured by lithium dilution, thermodilution and
tran-soesophageal Doppler echocardiography in anaesthetised horses.
Am J Vet Res 2000; 61: 731- 737.
20.
Jonas MM, Kelly FE, Linton RA et al. A
comparison of lithium dilution cardiac output measurement made using
central and antecubital venous injection of lithium chlo-ride. J
Clin Monit 1999; 15: 525- 528.
21.
Jonas M, Bruce R, Kelly F et al.
Comparison of cardiac output measurements using a continuous
arterial wave-form analysis monitor (PulseCO) with an indicator
dilu-tion technique (LiDCO) in patients in intensive care. 8th World
Congress of Intensive Care Medicine, Sydney, October 2001, Abstract.
22.
Mappes A, Gruendel M, Kuppe M.
Perioperative hemody-namic beat-to-beat monitoring of cardiac
output: the new PulseCO System. American Society of Anesthesiology,
New Orleans, October 2001, Abstract.
23.
Hamilton T. Jessen M. PulseCO accurate
monitoring of cardiac output from the arterial waveform. 8th Annual
CTT Meeting 2002: Current Trends in Thoracic Surgery VIII, Miami,
USA, January 2002, Abstract.
24.
Godje O, Peyerl M, Seebauer T et al.
Central venous pres-sure, pulmonary capillary wedge pressure, and
intratho-racic blood volumes as preload indicators in cardiac
surgi-cal patients. Eur J Cardiothorac Surg 1998; 13: 533-
539.
25.
Ellis RJ, Mangano DT, VanDyke DC.
Relationship of wedge pressure to end-diastolic volume in patients
under-going myocardial revascularization. J Thorac Cardiovasc Surg
1979; 78: 605- 613.
26.
Tavernier B, Makhotine O, Lebuffe G et
al. Systolic pres-sure variation as a guide to fluid therapy in
patients with sepsis-induced hypotension. Anesthesiology 1998;
89: 1313- 1321.
27.
Michard F, Boussat S, Chemla D et al.
Relation between respiratory changes in arterial pulse pressure and
fluid responsiveness in septic patients with acute circulatory
failure. Am J Respir Crit Care Med 2000; 162: 134-
138.
28.
Michard F, Chemla D, Richard C.
Clinical use of respira-tory changes in arterial pulse pressure to
monitor the hemodynamic effects of PEEP. Respir Crit Care Med 1999;
159: 935- 939.
29.
Reuter D, Felbinger T, Kilger
E et al. Optimising fluid therapy in mechanically ventilated
patients after cardiac surgery by on-line monitoring of left
ventricular stroke volume variations. Comparison with aortic
systolic pres-sure variations. Br J Anaesth 2002; 88: 124-
126.
Correspondence
to: Dr M Jonas Consultant Intensivist Southampton Hospitals
University Trust Tremona Road Southampton SO16 7BQ, UK
e-mail: max. jonas@suht.swest.nhs.uk
INTERNATIONAL JOURNAL OF INTENSIVE CARE SPRING
2002
Continuous, Real-time Cardiovascular
Monitoring
UK Contact: +44 (0) 1223 893081
US Contact: 972 462 7316
LiDCO Ltd Sales and Marketing, Unit M South Cambridge Business Park, Babraham Road, Sawston, Cambridge CB22 3JH, UK Telephone: +44 (0) 1223 830666 Fax: +44 (0) 1223 837241 VAT number 672 4757 08
