Hemodynamic Evaluation and Monitoring in · 2017-08-14 · Hemodynamic Evaluation and Monitoring in the ICU* Michael R. Pinsky, MD, FCCP Hemodynamic monitoring, a cornerstone in the

DOI 10.1378/chest.07-0073 2007;132;2020-2029Chest

Michael R. Pinsky

*the ICUHemodynamic Evaluation and Monitoring in

http://chestjournal.chestpubs.org/content/132/6/2020.full.htmland services can be found online on the World Wide Web at: The online version of this article, along with updated information

ISSN:0012-3692)http://chestjournal.chestpubs.org/site/misc/reprints.xhtml(

of the copyright holder.may be reproduced or distributed without the prior written permission Northbrook, IL 60062. All rights reserved. No part of this article or PDFby the American College of Chest Physicians, 3300 Dundee Road,

2007Physicians. It has been published monthly since 1935. Copyright CHEST is the official journal of the American College of Chest

© 2007 American College of Chest Physicians by David Quintero on December 2, 2009chestjournal.chestpubs.orgDownloaded from

http://chestjournal.chestpubs.org/site/misc/reprints.xhtml

http://chestjournal.chestpubs.org/

Hemodynamic Evaluation andMonitoring in the ICU*

Michael R. Pinsky, MD, FCCP

Hemodynamic monitoring, a cornerstone in the management of the critically ill patient, is usedto identify cardiovascular insufficiency, its probable cause, and response to therapy. Still it isdifficult to document the efficacy of monitoring because no device improves outcome unlesscoupled to a treatment that improves outcome. Several clinical trials have consistently docu-mented that preoptimization for high-risk surgery patients treated in the operating room andearly (< 12 h) goal-directed resuscitation in septic patients treated in the emergency departmentreduce morbidity, mortality, and resource use (costs) when the end points of resuscitation werefocused on surrogate measures of adequacy of global oxygen delivery (DO2). The closer theresuscitation is to the insult, the greater the benefit. When resuscitation was started after ICUadmission in high-risk surgical patients, reduced length of stay was also seen. The focus of thesemonitoring protocols is to establish a mean arterial pressure > 65 mm Hg and then to increaseDO2 to 600 mL/min/m2 within the first few minutes to hours of presentation. To accomplish thesegoals, hemodynamic monitoring focuses more on measures of cardiac output and mixed venousoxygen saturation to access adequacy of resuscitation efforts than on filling pressures. Althoughthese protocols reduce mortality and morbidity is selected high-risk patient groups, thewidespread use of monitoring-driven treatment protocols has not yet happened, presumablybecause all studies have been single-center trials using a single, proprietary blood flow-monitoring device. Multicenter trials are needed of early goal-directed therapies for all patientspresenting in shock of various etiologies and when the protocol and not the monitoring device isthe primary variable. (CHEST 2007; 132:2020–2029)

Key words: blood-flow monitoring; goal-directed therapy; hemodynamic monitoring; ICU

Abbreviations: Do2 � oxygen delivery; HR � heart rate; LV � left ventricular; MAP � mean arterial pressure;PLR � passive leg raising; Ppao � pulmonary artery occlusion pressure; PPV � pulse pressure variation; Pra � rightatrial pressure; RV � right ventricular; Spo2 � oxygen saturation by pulse oximetry; SPV � systolic pressure variation;Sv�o2 � mixed venous oxygen saturation; SVV � stroke volume variation; V̇o2 � oxygen consumption

H emodynamic monitoring is a cornerstone in thecare of the critically ill patient in the ICU. The

ICU provides a place for monitoring and care of

patients with potentially severe physiologic instabilityrequiring advanced artificial life support. Within thiscontext, hemodynamic monitoring is used to identifyhemodynamic instability and its cause and to monitorthe response to therapy. We have witnessed animpressive number of medical technological ad-vances, allowing monitoring, display, and assessmentof physiologic variables not even imagined before,1yet the utility of most hemodynamic monitoring isunproven. It is the commonly available technologiesfor which clinical studies have demonstrated rele-vance. Physiologic measures available from com-monly available monitoring devices are given inTable 1. Despite the many options available, mostICUs monitor and display only BP, heart rate (HR)

*From the Department of Critical Care Medicine, University ofPittsburgh Medical Center, Pittsburgh, PA.This work was supported in part by National Institutes of Healthgrants HL67181, HL07820, and HL073198.Dr. Pinsky is medical consultant for Edwards LifeSciences,LiDCO Ltd, and Cheetah Medical, and owns stock in LiDCOLtd and Cheetah Medical.Manuscript received January 9, 2007; revision accepted April 30, 2007.Reproduction of this article is prohibited without written permissionfrom the American College of Chest Physicians (www.chestjournal.org/misc/reprints.shtml).Correspondence to: Michael R. Pinsky, MD, FCCP, 606 Scaife Hall,3550 Terrace St, Pittsburgh, PA 15213; e-mail: [email protected]: 10.1378/chest.07-0073

CHEST Postgraduate Education CornerCONTEMPORARY REVIEWS IN CRITICAL CARE MEDICINE

2020 Postgraduate Education Corner



and oxygen saturation by pulse oximetry (Spo2), asthey have done for the last 20 years. Furthermore,with few exceptions, such monitoring does not drivetreatment protocols but rather serves as an auto-mated vital signs record to trigger further attention.It is hard to validate the utility of monitoring when itis used in this fashion because no hemodynamicmonitoring device will improve outcome uselesscoupled to a treatment that itself improves outcome.Thus, the effectiveness of hemodynamic monitoringto improve outcome is limited to specific patientgroups and disease processes for which proven ef-fective treatments exist. Although, like most of med-icine, the utility of hemodynamic monitoring is notwell documented, a primary rationale for the use of

hemodynamic monitoring is to identify cardiovascu-lar instability and its specific etiology, and to guidetherapy.

Interestingly, physicians have developed a psycho-logical dependence on feedback from continuoushemodynamic monitoring tools, independent of theirutility. Spo2 monitoring in low-risk patients is anexample. One would presume that continual mea-sure of Spo2 should improve patient outcomes byidentifying hypoxemia and brady/tachyarrhythmias,thus allowing for effective and rapid correctionbefore the development of global tissue ischemia.However, Moller et al2 examined the benefit ofintraoperative Spo2 monitoring in low-risk surgerypatients. They monitored 20,802 patients: 10,312 pa-tients assigned to an oximetry group, and 10,490 pa-tients assigned to a control group without oximetry.They found no numerical differences in cardiovas-cular, respiratory, neurologic, or infectious compli-cations, duration of hospital stay, or number ofin-hospital deaths between the two groups. Whenshown these results, 80% of the anesthesiologistsreplied by questionnaire that they still felt moresecure in their practice when they used a pulseoximeter in these patients.3 In this article, I shalldiscuss the rationale for commonly available monitor-ing, the usefulness of static measured variables to assessspecific disease states (hemodynamic profile analysis),and interactive monitoring to predict response to ther-apy (applied physiology), and monitoring-driven treat-ment protocols (functional hemodynamic monitoring)that improve outcomes.

Rationale for Hemodynamic Monitoring

Three progressive arguments can be made forusing specific monitoring. At the basic level, onecites its common use. Here, prior experience hasshown that such monitoring can identify diseasestates and/or known complications, even though thelink between the monitoring and disease may not beknown or even postulated. The second level ofdefense rests with an understanding of shock patho-physiology, the etiologies of which are usually cate-gorized into four broad groups: hypovolemic, cardio-genic, obstructive, or distributive, all of which mayhave different causes and treatments.4 Since theprimary goal of the cardiovascular system is to supplyadequate amounts of oxygen to meet the metabolicdemands of the body, calculation of systemic oxygendelivery (Do2) and oxygen consumption (V̇o2), iden-tifying tissue ischemia (usually monitored by mixedvenous oxygen saturation [Sv�o2]) as well as measuresof ventricular performance (stroke work) are oftencalculated from such primary variables. At this level,

Table 1—Hemodynamic Monitoring-Defined PrimaryHemodynamic Variables*

Noninvasive monitoringECG

HR, dysrhythmnias, HR variabilityPulse oximetry

Spo2, HRArterial pressure

SphygmomanometrySystolic and diastolic BP, HR, pulsus paradoxus

Central venous pressureJugular venous distention, hepatojugular reflux, cannon

waves (A-V dissociation), tricuspid regurgitationInvasive monitoring

Arterial catheterizationSystolic BP, diastolic BP, MAP, HR, and pulse pressureArterial blood gas analysis

pH, Pao2, Sao2, Pco2, hemoglobinArterial pressure waveform analysis

Stroke volume, cardiac output, PPV and SVVCentral venous catheterization

Central venous pressure, venous pressure waveform(“v” waves), respiratory variations

Central venous blood gas analysispH, Pcvo2, Scvo2, Pcvco2, hemoglobin

Thermodilution indices (when coupled to an arterial thermalsensor)

Stroke volume, cardiac output, intrathoracic blood volume,global end-diastolic volume, and Do2

Pulmonary artery catheterSystolic BP, diastolic BP, MAP, pressure waveform

(“v” waves), and PpaoMixed venous blood gas analysis

pH, Pv�o2, Sv�o2, Pvco2, hemoglobinThermodilution cardiac output (by thermodilution either

intermittent or continuous)Stroke volume, cardiac output, RV ejection fraction, and RV

end-diastolic volumeEsophageal Doppler echocardiographic monitoring

Stroke volume, cardiac output, and SVV

*Sao2 � arterial oxygen saturation; Pcvo2 � central venous O2 par-tial pressure; Scvo2 � central venous oxygen saturation; Pcvco2 �central venous CO2 partial pressure; Pv�o2 � mixed venous O2

partial pressure; Pv�c02 � mixed venous CO2 pressure.

www.chestjournal.org CHEST / 132 / 6 / DECEMBER, 2007 2021



hemodynamic monitoring is used to define cardio-vascular status, not response to treatments based onassumed pathophysiology, and predict outcome.Most of the rationale for hemodynamic monitoringresides at this pathophysiology level. The impliedassumption here is that restoration of normal hemo-dynamic values will prevent further organ injury andreduce mortality. Unfortunately, this argument maynot be valid, primarily because hemodynamic mon-itoring usually only assesses global circulatory status,not organ function or microcirculation, and does notaddress the mechanisms by which disease occurs.5–7

The highest level of defense comes from documen-tation of improved outcomes based on hemodynamicmonitoring-driven treatments that may alter therapyin otherwise unexpected ways. These treatment pro-tocols often have a mechanistic rationale, but suchscientific rationale is not mandatory. This final de-fense is supported by clinical trials adding weight tothe applied physiologic approach to hemodynamicmonitoring and can form the basis for evidence-based medicine recommendations. Such validationhas only recently been shown for hemodynamicmonitoring-driven protocolized resuscitation in se-lected high-risk patient groups as described below.

Hemodynamic Profile Analysis

Circulatory shock causes tissue hypoperfusion.Cellular dysfunction, organ injury, and death mayoccur proportional to the degree and duration oftissue hypoperfusion as quantified by oxygen debt.8The four pathophysiologic categories of shock areusually characterized by different specific hemody-namic variables, induced by the associated primaryhemodynamic event and the autonomic response toit. These variables can be measured by a variety ofnoninvasive and invasive means (Table 1) and de-rived hemodynamic parameters calculated that re-flect global cardiovascular status (Table 2).

The relation between specific hemodynamic vari-ables is complex is health, and even more complex isdisease. However, a solid understanding of the car-diovascular underpinnings of blood flow homeostasisis required to interpret hemodynamic variables ef-fectively. If disease causes cardiac output and Do2 todecrease, mean arterial pressure (MAP) decreases aswell. Baroreceptors in the aortic arch and carotidbody alter vasomotor tone through modulation ofsympathetic tone to maintain cerebral perfusionpressure (eg, MAP � 65 mm Hg).9 The hemody-namic effects of this increased sympathetic tone aretachycardia and restoration of MAP toward normalvalues by reducing unstressed circulatory blood vol-ume and increased arterial vasomotor tone. Thus,

hypotension reflects failure of the sympathetic ner-vous system to compensate for circulatory shock,while normotension does not insure hemodynamicstability.10 Since regulation of blood flow distributionoccurs by regional vasodilation of arterial resistancevessels, hypotension impairs autoregulated bloodflow distribution.11,12 Except in conditions of severehypoxemia and anemia, the primary means by whichDo2 is varied to match metabolic requirements is byvarying cardiac output and tissue oxygen extraction.Since metabolic demands can vary widely, there is nonormal cardiac output or Do2 value, but ratherminimal thresholds for resting conditions and poten-tially adequate higher levels during stress. Opera-tionally, it is better to access cardiac output as beingeither adequate or inadequate to meet the metabolicdemands of the body. Inadequate Do2 is presumedto occur if tissue oxygen extraction is markedlyincreased, as manifest by a decrease in Sv�o2� 70%.13

Table 2—Derived Hemodynamic Parameters FromHemodynamic Monitoring*

Primary hemodynamic variablesHR, beats/minMAP, mm HgPra, mm HgMPAP, mm HgPpao, mm HgCO, L/minSao2, %Spo2 as an estimate of Sao2, %Sv�o2, %Hb, g/dLHeight and weight needed to calculate BSA, m2

Calculated hemodynamic parametersCI � CO/BSA, L/min/m2

Stroke volume � CO/HR � 1,000, mL/minStroke index � stroke volume/BSA, mL/m2

LV stroke work � stroke volume � (MAP � Ppao), mL � mm HgLV stroke work index � LV stroke work/BSA, mL � mm Hg/m2

Total peripheral resistance � (MAP/CO) � 80, dyne � s/cm5

Systemic vascular resistance � (�MAP � Pra�)/CO � 80,dyne � s/cm5

RV stroke work � stroke volume � (MPAP � Pra), mL � mm HgRV stroke work index � RV stroke work/BSA, mL � mm Hg/m2

Pulmonary vascular resistance � (�MPAP – Ppao�/CO) � 80,dyne � s/cm5

Global Do2† � CO � (Sao2 � Sv�o2) � Hb � 1.36 � 1,000, mLoxygen/min

Global Do2 index† � CI � (Sao2 � Sv�o2) Hb � 1.36, mLoxygen/min

Global V̇o2† � CO � Sao2 � Hb � 1.36 � 1,000, mL oxygen/min

Global V̇o2 index† � CI � Sao2 � Hb � 1.36 � 1,000, mLoxygen/min

*CO � cardiac output; CI � cardiac index; BSA � body surfacearea; Sao2 � arterial oxygen saturation; MPAP � mean pulmonaryartery pressure; Hb � hemoglobin.

†Spo2 can be substituted for arterial oxygen saturation in thesecalculations.




Of the four categories of shock, only distributiveshock states following intravascular volume resusci-tation are associated with an increased cardiac out-put but decreased vasomotor tone.4 Thus, cardiacoutput, stroke work, Do2, and Sv�o2 are decreased incardiogenic, hypovolemic, and obstructive shock butmay be normal or even increased in distributiveshock. However, in all conditions, HR increasesassociated with an increased sympathetic tone. Car-diogenic shock represents primary cardiac failure. Itcan be due to impaired contractility (myocardialischemia/infarction, electrolyte imbalance, hypox-emia, hypothermia, endocrinologic diseases, meta-bolic poisoning, �-blockers), pump function (valvu-lopathy, ventriculoseptal defect, dyssrythmias), ordiastolic compliance (fibrosis, infiltrative cardiomy-opathies, hypertrophy). The specific cardinal find-ings of cardiogenic shock are increased back pres-sure to cardiac filling (right atrial pressure [Pra] andpulmonary artery occlusion pressure [Ppao]) andupstream edema (peripheral and pulmonary). Hypo-volemic shock represents a decrease in effectivecirculating blood volume and venous return. It canbe due to primary intravascular volume loss (hemor-rhage, capillary leak), secondary intravascular vol-ume loss (third-space loss, insensible loss throughskin with burns, diarrhea, vomiting), and increasedunstressed vascular volume (loss of sympathetictone, spinal cord injury, vasodilating drugs). Thespecific findings of hypovolemic shock are decreasedfilling pressures. Obstructive shock represents ablockage of blood flow. It may be due to rightventricular (RV) outflow obstruction (pulmonaryembolism, hyperinflation), tamponade (pericardialeffusion, hyperinflation), or left ventricular (LV)outflow obstruction (aortic stenosis, dissecting aorticaneurysm). The specific findings of obstructive shockare often more subtle but include decreased LVdiastolic compliance (small LV volume with in-creased Ppao) and signs of cor pulmonale (Pragreater than Ppao, tricuspid regurgitation). Distrib-utive shock represents loss of normal sympatheticresponsiveness resulting in decreased vasomotortone. In the nonresuscitated subject, this presents ashypovolemic shock,14 but with fluid resuscitation BPdoes not increase despite an increase in cardiacoutput. It can be due to loss of vascular responsive-ness (sepsis, spinal shock, vasodilating drugs, meta-bolic poisons). The specific findings of distributiveshock are an increased cardiac output, Do2, andSv�o2 despite persistent hypotension. Hemodynamicmonitoring can aid in determining circulatory shocketiology.

Since most forms of circulatory shock reflectinadequate tissue Do2, a primary goal of resuscita-tion is to increase Do2. Three important functional

questions are usually asked of the hemodynamicallyunstable patient. First, will cardiac output increasewith fluid resuscitation and, if so, by how much?Physiologically speaking, this equates to preloadresponsiveness. Second, in the hypotensive patient isarterial vasomotor tone increased, decreased, ornormal? Finally, is the heart capable of sustaining aneffective cardiac output once arterial pressure isrestored without going into failure? Clearly, patient-specific hemodynamic questions are also asked but,in general, these are the fundamental questionsaddressed by most effective treatment algorithms.Unfortunately, although specific patterns of hemo-dynamic values, as described above, reflect specifictypes of disease, they do not predict individualpatient response to therapy.

Applied Physiology Applied toHemodynamic Monitoring

To address the question of preload responsiveness,physicians usually attempt to measure intravascularvolume status, either by indirect measures (skinturgor, mucus membrane wetness and venous con-gestion, or vascular pedicle diameter of the chestradiograph)15 or by attempting to estimate RV andLV end-diastolic volumes. Importantly, the pub-lished clinical literature does not support the use ofdirect or indirect measures of end-diastolic volumeas a means to predict preload responsiveness. Read-ers are referred to the metaanalysis by Michard andTeboul16 published in CHEST for further discussion.Although general trends in filling pressures andvolumes define patient populations, their use inclinical decision making is poor. Specifically neitherabsolute values for Pra, Ppao, RV end-diastolic vol-ume, or LV end-diastolic area predict preload re-sponsiveness. Furthermore, the changes in eitherPra or Ppao do not reflect changes in either cardiacoutput or stroke volume in hemodynamically unsta-ble patients.17 Although increases in either RV or LVend-diastolic volumes do increase stroke volume,knowing ventricular pressures or volumes at a singlepoint in time is not useful in making this prediction.Although the reasons for such inaccuracies of usingPra or Ppao to estimate preload may reflect inaccu-racies in measures,18 or in understanding what Ppaoreflects even when these values are measured accu-rately,19 even when measured accurately they do notpredict preload responsiveness.20 Thus, the lack ofthe ability of measures of Pra or Ppao to predictpreload responsiveness may explain the lack of dif-ference in outcome from Pra- vs Ppao-guided fluidresuscitation therapies in the two published articleson the ARDS Clinical Trials Network Fluid and




Catheter Treatment Trial comparing central venouscatheters (Pra guided) to pulmonary arterial cathe-ters (Ppao guided)21 and liberal vs restricted fluidresuscitation (high Pra or Ppao vs low Pra or Ppao),other than length of stay being slightly shorter in theconservative fluids arm22 because neither measurecorrelates with Do2, although both tend to parallelchanges in effective circulating blood volume. Fur-thermore, the Surviving Sepsis Campaign recom-mendations23 for targeted values of Pra and Ppao arenot supported by the existing evidence. Fluid re-sponsiveness was documented to be unrelated to therecommended Pra and Ppao values.24 Such negativefindings based on a treatment protocol targetingspecific Pra or Ppao values are not surprising. Inessence, preload is not preload responsiveness.Clearly, as numerous previous studies25–28 have un-derscored, just inserting a catheter to make measure-ments without a defined and effective treatmentprotocol requiring such information is unlikely toresult in improved patient outcomes. What cliniciansneed to know is the latter, and what static measurersestimate is the former. Clearly, as intravascularvolume increases, Pra may also increase, especially inpatients with impaired RV function. Still, one canhave an expanded intravascular volume and a lowPra, as is the case in hyperdynamic hepatic cirrhosispatients. Similarly, Ppao also tends to be higher withhypervolemia and tends to tract intrathoracic bloodvolume especially in heart failure patients. However,as was shown previously by Michard and Teboul,16

absolute Pra or Ppao values are no better than arandom chance at predicting preload responsiveness.There are few relative truths in the assessment ofsingle, fixed hemodynamic variables, but Table 3 liststhose I have come to realize when considering acuteresuscitation of the critically ill.

In the assessment of preload responsiveness, oneneeds to measure other parameters than filling pres-sures and ventricular volumes. The time-honoredmethod of assessing preload responsiveness is theintravascular fluid challenge, wherein a bolus of fluidis rapidly infused and the subsequent changes inspecific flow-dependent variables (cardiac output,MAP, HR, Sv�o2, Pra, Ppao) are measured. Theproblems with performing a fluid challenge forclinical decision making are multiple. First, only half

the hemodynamically unstable patients administereda volume challenge will have increased cardiac out-put.16 Thus, the correct management may have beendelayed in half the patients. Second, in the half ofthose patients who are not preload responsive, vol-ume loading may be directly injurious. For example,both acute cor pulmonale (pulmonary embolism,COPD) or LV failure may deteriorate further withvolume loading. Two alternative methods of per-forming a reversible fluid challenge have recentlygained interest in the acute care setting. Theseinclude the use of positive pressure ventilation-induced changes in arterial pressure and LV strokevolume to cyclically varying venous return, and byperforming a passive leg raising (PLR) maneuver totransiently increase venous return and noting thechange in mean blood flow.

Positive pressure ventilation when applied to a pa-tient at rest and with no spontaneous respiratory effortis associated with a cyclic increase in Pra in phase withinspiration. Since Pra is the back-pressure to venousreturn, if upstream venous pressures do not simulta-neously increase29 then RV filling will also decrease ina cyclic fashion. This cyclic variation in RV filling willinduce a cyclic variation in LV filling if both RV and LVare preload responsive.30 This cyclic variation in LVfilling will induce a cyclic variation in LV strokevolume and arterial pulse pressure if the patient ispreload responsive. Several studies31–33 have docu-mented that the associated variations in LV strokevolume, referred to as stroke volume variation (SVV)and quantified as the maximal to minimal strokevolume values over their mean over three breaths ora defined time interval (eg, 20 to 30 s), is highlypredictive of preload responsiveness. For a tidalvolume of 6 mL/kg, a SVV � 10% predicts a � 15%increase in cardiac output for a 500-mL fluid bo-lus.34–36 Since the primary determinant of arterialpulse pressure is stroke volume, pulse pressurevariation (PPV), calculated in the same manner asSVV, has also been shown to predict preload respon-siveness well. Here however, a � 13% PPV predictsa � 15% increase in cardiac output for a 500-mLvolume bolus. Presently, PPV is easier to measurethan SVV because it only requires inspection of thearterial pressure waveform over time,37 whereas SVVcan be assessed by either esophageal Doppler echo-cardiography38 or echocardiographic measures ofaortic velocity.39 Several commercially availabletechnologies have evolved based on arterial wave-form analysis that can estimate stroke volume fromthe pulse pressure waveform. Furthermore, onlyquantifying systolic pressure variation (SPV) over theventilatory cycle has been proposed.40 This mea-sured, also known as pulse paradoxus, has the advan-tage of being easier to monitor but also has de-

Table 3—Hemodynamic Monitoring Truths

Tachycardia is never a good thing.Hypotension is always pathologic.There is no such thing as normal cardiac output.Central venous pressure is only elevated in disease.Peripheral edema is of cosmetic concern.




creased sensitivity because it does not quantify thevarying diastolic arterial pressure component of thePPV.41 Finally, studies suggest that the Spo2 plethy-somgraphic waveform amplitude co-varies with arte-rial pulse pressure.42, and this plethysmographicsignal variation predicts fluid responsiveness in hy-potensive patients.43 If validated to predict preloadresponsiveness in the broader group of hemodynam-ically unstable patients, then such noninvasive tech-niques could expand the application of this appliedphysiologic approach at the bedside.

Like all hemodynamic monitoring approaches, theuse of SVV, PPV, or SPV to assess preload respon-siveness requires an understanding of its physiologicunderpinnings. SVV, PPV, and SPV are created bytidal volume-induced changes in venous return.They presume a constant R-R interval and aremeasured from diastole to systole, not vice versa,such that SVV, PPV, and SPV reflect only changes invenous return and not diastolic filling time. Thus,these parameters will lose their predictive valueunder conditions of varying R-R intervals (atrialfibrillation), and they may also lose accuracy if tidalvolume varies from breath to breath as may occurwith assisted and spontaneous ventilation.44–46 Thus,these approaches are limited to only a small percent-age of critically ill patients. Furthermore, since theratio of PPV to SVV reflects central arterial compli-ance, if arterial tone varies, PPV and SVV may vary indisproportional ways. However, potentially one canmonitor the PPV/SVV ratio to identify changingcentral arterial vasomotor tone. Finally, preload re-sponsiveness does not mean that the patient requiresvolume resuscitation because normal subjects arealso preload responsive.47

More advanced monitoring using transthoracic48,49

and transesophageal50,51 ultrasound (echo) imag-ing of the vena caval collapse during positivepressure ventilation has also been shown to predictPra � 10 mm Hg. If venal caval diameter is de-creased below a threshold value, the Pra is � 10 mmHg; otherwise, it is � 10 mm Hg. This Pra thresholdvalue is important in a limited way because patientswith a Pra � 10 mm Hg invariably have decreasedcardiac output if additional positive end-expiratorypressure is applied during positive pressure ventila-tion.52 However, if Pra is � 10 mm Hg, no predic-tions can be made as to the change in cardiac outputin response to increasing levels of positive end-expiratory pressure.

To simplify these approaches, the clinically vali-dated PLR method can be used as a transient andreversible increase in venous return.53 The PLRmethod requires that the legs be raised 30° abovethe chest and held there for 1 min. PLR causes anapproximate 300-mL blood bolus in a 70-kg man that

persists for about 2 to 3 min before resulting inintravascular volume redistribution. The immediatehemodynamic response from before to during thePLR is taken to reflect preload response.54 Tominimize the need for a constant HR and tidalvolume, measures of mean aortic flow averaged over20 to 30 s can be measured and are actually superiorto SVV and PPV measures in the same subjects.46

There are two important implications of these findings.First, since measures of changing mean blood flowduring PLR accurately predict preload responsivenessduring both spontaneous and positive pressure ventila-tion and with or without arrhythmias, this approach canbe applied in all hemodynamically unstable patients.Second, since measures of mean blood flow can beascertained at the bedside using many commerciallyavailable devices, including esophageal Doppler flow-measuring devices55–57 and arterial pressure waveformestimates of flow,35,36,58,59 most ICUs are capable ofmaking these measures today.

Unfortunately, although SVV, PPV, and SPV havebeen described for several years, and recently thechange in mean blood flow with PLR, none of thesetechniques has been used to drive treatment proto-cols. Clearly, this application of these simple moni-toring approaches is the next step in the evolution offunctional hemodynamic monitoring.

Functional Hemodynamic Monitoring:Goal-Directed Therapy

Numerous clinical trials have attempted to docu-ment improved patient outcome when resuscitationstrategies were driven by measured hemodynamicvariables. Early on, the results were either mixed ornegative. However, with increased understanding ofthe pathophysiology of shock and a heightenedawareness of the need to prevent tissue ischemia,clinical trials in the emergency department by Riverset al60 and the operating room61–63 have clearlydocumented improved outcome. Clearly, intraoper-ative volume expansion improves organ perfusionand reduces gastric ischemia64 as assessed by tono-metric measures of gastric Pco2.65 Importantly, inthe study by Rivers et al,60 the total amount ofresuscitation fluids administered was similar in thetreatment and control groups, but the treatmentgroup received more additional early fluid when thecontrol group protocol did not require it becausetraditional hemodynamic measures such as Pra andMAP were at their target levels. The benefits real-ized from these studies have recently filtered into theICU environment, where two prospective clinicaltrials66,67 have shown that goal-directed therapy im-proves outcome. All these studies follow the same




theme: the earlier treatment is begun and tissueischemia resolved, the better the outcome.

For example, the greatest outcome benefit ofgoal-directed therapy appears to exist in the field ofhigh-risk surgery or, speaking from a physiologicperspective, scheduled trauma. This form of resus-citation has been termed preoptimization becausethe resuscitation starts prior to the cardiovascularstress and surgical trauma. In essence, resuscitationoccurs before tissue injury. Shoemaker et al68 docu-mented improved outcome and reduced cost whenhigh-risk surgery patients were resuscitated to highDo2 values (� 600 mL/min/m2) prior to surgery.These findings were duplicated by Boyd et al69 andLobo et al.70 Importantly, Lobo et al70 showed thatthe improved patient outcomes were realized acrossthe entire treatment group of elderly patients, evenin those patients who did not achieve the target Do2levels. One need not target Do2 to see improvedoutcome. Goepfert et al71 measured the sum end-diastolic volume of the heart (eg, right and left atrialand ventricular volumes at end diastole) and targetedglobal end-diastolic volume in cardiac surgery patientsand documented reduced need for catecholamines andless time on the ventilator but an increase in net fluidbalance of approximately 500 mL.

Studies72 in critically ill ICU patients using goal-directed therapy presumed that if Do2 were in-creased to supranormal levels (as references to rest-ing Do2 values), as was done in the preoptimizationstudies above, patients would have improved sur-vival. This approach is referred to as postoptimiza-tion in distinction to the intraoperative preoptimiza-tion protocols because it is started after the patientpresents in shock. However, neither Tuchschmidt etal73 nor Gattinoni et al74 were able to document anyimproved outcome when critically ill patients wereenrolled 12 to 36 h after presenting with shock.75 Infact, Hayes et al76 saw increased mortality in theirtreatment group presumably because of overly ag-gressive attempts to reach target Do2 values. How-ever, Rivers et al60 underscored the importance ofimmediate (emergency department at presentation)and appropriate (adequate level of Do2 as defined bythe central venous oxygen saturation as a surrogate ofSv�o2)77 resuscitation of critically ill patients to im-prove outcome.78 These authors79 also reported thatproinflammatory cytokine levels were reducedin treatment patients, suggesting that early goal-directed therapy also reduces the systemic inflam-matory response. Their study79 validated the prin-cipal of immediate restoration of cardiovascularstability as the primary treatment for circulatoryshock and focused the issue on rapid triage andmanagement. Clearly, allowing patients to remain inshock for hours before starting aggressive resuscita-

tion is a major cause of increased morbidity andmortality. One prior study80 documented improvedoutcome and reduced cost when resuscitation wastargeted to a minimal Sv�o2; however, these studieswere not followed up using defined treatment protocolsuntil recently. If one delays resuscitation further, thebenefits of that activity diminish. For example,McKendry et al66 used esophageal Doppler monitor-ing of mean blood flow to maximize preload in theimmediate postoperative resuscitation of cardiac sur-gery patients. Their nurse-driven protocol targetedDo2 values for only the first 6 postoperative hours.They observed a reduced length of hospital stay anda markedly reduced incidence of complications, mostnotably postoperative wound infections. Similarly,Pearse et al67 followed that up with a similar studydesign in postoperative high-risk patients. They tar-geted a postoperative Do2 of 600 mL/kg/min usingarterial pressure-derived estimates of cardiac output.Importantly, the treatment group received morecolloid and dopexamine infusions but had similarstroke volumes, Pra, and blood lactate levels with thecontrol groups. They found similar reductions inhospital length of stay primarily because of a reducedincidence of postoperative complications.

These data demonstrate two important things.First, that in high-risk surgical populations, preopti-mization applied prior to surgery and postoptimiza-tion therapies applied in the ICU in a protocolizedfashion improves outcomes and reduce cost. Second,the longer one delays aggressive metabolic targetedresuscitation, the less the observed benefit. It is notclear how long the therapeutic window remains openbefore such aggressive therapies worsen outcome, asexemplified by Hayes et al.76 Furthermore, it is notclear that similar metabolically targeted therapieswill also be beneficial in other ICU patient popula-tions, such as those with septic shock or single-organfailures such as primary ARDS and trauma. Further-more, none of the above-mentioned clinical trialsused the newly established SVV, PPV, or SPV meth-ods of assessing preload responsiveness for clinicaldecision making. Clearly, prospective clinical trials ofthese proven treatment strategies and novel robustdecision-support parameters used in different pa-tient populations are needed. Still, the results ofstudies that have been completed using functionalmeasures in targeted high-risk populations early intheir disease have all been positive, whereas thosestudies using more traditional measures or usingsimilar functional measures but applied later in thecourse of shock have been unsuccessful. The themetherefore appears to be clear: target patients at risk fortissue ischemia prior to severe organ injury usingtitrated therapies that monitoring circulatory suffi-ciency, and administer that therapy as soon as possible.




However, these consistent findings across studies,although promising, still reflect single-center trialsusing one proprietary blood flow-monitoring device(eg, esophageal Doppler, arterial pulse contour) inhighly selected high-risk patient groups. What isneeded is a large multicenter clinical trial aimed atearly goal-directed treatment of all patients in shockfrom various etiologies for which the goals of therapyand the rapidity of treatment rather than the meansto access treatment are the primary operative vari-ables, while any within study comparison of moni-toring device differences would be of secondaryimportance. If such studies documented improvedpatient outcomes, then the choice of which monitor-ing devices one uses would be subject more to issuesof cost, convenience, and complications.

Future Monitoring Approaches

The future of hemodynamic monitoring is alreadyhere and can be summarized as focusing on measur-ing tissue wellness using continuous, noninvasive,and metabolic markers. Examples of these devicesinclude sublingual Pco2,81,82 tissue oxygen satura-tion,83 and capillary blood flow measured under thetongue.84 The above continuous noninvasive measuresdescribe metabolic effects of circulatory function. Po-tentially, they may be used to identify compensatedshock and to define functional end points of resuscita-tion. When applied using the above-mentioned titra-tion of resuscitation to restore and sustain tissue bloodflow, such novel monitoring devices may add an extradimension to our monitoring options by allowing real-time assessment of response to therapy and potentiallywhen to stop. Since no prospective outcomes clinicaltrials have been done, the use of these novel monitor-ing approaches is speculative.

Conclusion

Enough clinical data have accumulated over thepast 30 years to defend abolishing the use of statichemodynamic values, such as Pra and Ppao, asmarkers of preload responsiveness. Dynamic re-sponses, to either a volume challenge or a physiologicreversible volume challenge using either positive pres-sure ventilation or PLR are highly sensitive and specificfor preload responsiveness. Numerous prospective clin-ical trials60–71 have documented improved outcomeand reduced cost when early goal-directed therapiesare applied in a protocolized fashion in high-risk pa-tients, whereas no benefit and even harm may occurwhen aggressive resuscitation is applied late (� 12 h) inthe course of circulatory shock.

References1 Pinsky MR, Payen D. Functional hemodynamic monitoring.

Crit Care 2005; 9:566–5722 Moller JT, Pedersen T, Rasmussen LS, et al. Randomized

evaluation of pulse oximetry in 20,802 patients: i. Design,demography, pulse oximetry failure rate, and overall compli-cation rate. Anesthesiology 1993; 78:436–444

3 Moller JT, Johannessen NW, Espersen K, et al. Randomizedevaluation of pulse oximetry in 20,802 patients: ii: perioper-ative events and postoperative complications. Anesthesiology1993; 78:445–453

4 Weil MH, Shubin H. Shock following acute myocardiuminfarction: current understanding of hemodynamic mecha-nisms. Prog Cardiovasc Dis 1968; 11:1–17

5 Schmidt-Nielsen K. Circulation. In: Animal physiology.4th ed. Cambridge, UK: Cambridge University Press, 1983;97–133

6 Lush CW, Kvietys PR. Microvascular dysfunction in sepsis.Microcirculation 2000; 7:83–101

7 Buwalda M, Ince C. Opening the microcirculation: canvasodilators be useful in sepsis? Intensive Care Med 2002;28:1208–1217

8 Shoemaker WC, Appel PL, Kram HR. Tissue oxygen debt asa determinant of lethal and non-lethal postoperative organfailure. Crit Care Med 1988; 16:1117–1120

9 LeDoux D, Astix ME, Carpati C, et al. Effects of perfusionpressure on tissue perfusion in septic shock. Crit Care Med2000; 28:2729–2732

10 Partrick DA, Bensard DD, Janik JS, et al. Is hypotension areliable indicator of blood loss from traumatic injury inchildren? Am J Surg 2002; 184:555–559

11 Ruokonen E, Takala J, Uusaro A. Effect of vasoactive treat-ment on the relationship between mixed venous and regionaloxygen saturation. Crit Care Med 1991; 19:1365–1369

12 Bellomo R, Kellum JA, Wisniewski SR, et al. Effects ofnorepinephrine on the renal vasculature in normal andendotoxemic dogs. Am J Respir Crit Care Med 1999; 159:1186–1192

13 Pinsky MR, Vincent JL. Let us use the pulmonary arterycatheter correctly and only when we need it. Crit Care Med2005; 33:1119–1122

14 Carroll GC, Snyder JV. Hyperdynamic severe intravascularsepsis depends on fluid administration in cynomolgus mon-key. Am J Physiol 1982; 243:R131–R141

15 Passos RH, Ricardo JS, Pereira AJ, et al. Radiologic assess-ment of volemic status: vascular pedicle width. Crit Care2005; 9(suppl 2):P110–P112

16 Michard F, Teboul JL. Predicting fluid responsiveness in ICUpatients: a critical analysis of the evidence. Chest 2002;121:2000–2008

17 Kumar A, Anel R, Bunnell E, et al. Pulmonary arteryocclusion pressure and central venous pressure fail to predictventricular filling volume, cardiac performance, or the re-sponse to volume infusion in normal subjects. Crit Care Med2004; 32:691–699

18 Papadakos PJ, Vender JS. Training requirements for pulmo-nary artery catheter utilization in adult patients. New Horiz1997; 5:287–291

19 Pinsky MR. Clinical significance of pulmonary artery occlu-sion pressure. Intensive Care Med 2003; 29:175–178

20 Michard F, Chemla D, Richard C, et al. Clinical use ofrespiratory changes in arterial pulse pressure to monitor thehemodynamic effects of PEEP. Am J Respir Crit Care Med1999; 159:935–939

21 The National Heart, Lung, and Blood Institute Acute Respi-ratory Distress Syndrome (ARDS) Clinical Trials Network.




Pulmonary artery versus central venous catheter to guidetreatment of acute lung injury. N Engl J Med 2006; 354:2213–2224

22 The National Heart, Lung, and Blood Institute Acute Respi-ratory Distress Syndrome (ARDS) Clinical Trials Network.Comparison of two fluid-management strategies in acute lunginjury. N Engl J Med 2006; 354:2564–2575

23 Dellinger RP, Carlet JM, Masur H, et al. Surviving SepsisCampaign guidelines for management of severe sepsis andseptic shock. Crit Care Med 2004; 32:858–873

24 Osman D, Ridel C, Ray P, et al. Cardiac filling pressures arenot appropriate to predict hemodynamic response to volumechallenge. Crit Care Med 2007; 35:64–68

25 Connors AF, Speroff T, Dawson NV, et al. The effectivenessof right heart catheterization in the initial care of critically illpatients. JAMA 1996; 276:889–897

26 Sandham JD, Hull RD, Brant RF, et al. A randomized,controlled trial of the use of pulmonary-artery catheters inhigh-risk surgical patients. N Engl J Med 2003; 348:5–14

27 Richard C, Warszawski J, Anguel N, et al. Early use of thepulmonary artery catheter and outcomes in patients withshock and acute respiratory distress syndrome: a randomizedcontrolled trial. JAMA 2003; 290:2713–2720

28 Yu DT, Platt R, Lanken PN, et al. Relationship of pulmonaryartery catheter use to mortality and resource utilization inpatients with severe sepsis. Crit Care Med 2003; 31:2734–2741

29 Van den Berg P, Jansen JRC, Pinsky MR. The effect ofpositive-pressure inspiration on venous return in volumeloaded post-operative cardiac surgical patients. J Appl Physiol2002; 92:1223–1231

30 Vieillard-Baron A, Chergui K, Augarde R, et al. Cyclicchanges in arterial pulse during respiratory support revisitedby Doppler echocardiography. Am J Respir Crit Care Med2003; 168:671–676

31 Feissel M, Michard F, Mangin I, et al. Respiratory changes inaortic blood velocity as an indicator of fluid responsiveness inventilated patients with septic shock. Chest 2001; 119:867–873

32 Berkenstadt H, Margalit N, Hadani M, et al. Stroke volumevariation as a predictor of fluid responsiveness in patientsundergoing brain surgery. Anesth Analg 2001; 92:984–989

33 Slama M, Masson H, Teboul JL, et al. Respiratory variationsof aortic VTI: a new index of hypovolemia and fluid respon-siveness. Am J Physiol Heart Circ Physiol 2002; 283:H1729–H1733

34 Monnet X, Rienzo M, Osman D, et al. Esophageal Dopplermonitoring predicts fluid responsiveness in critically ill ven-tilated patients. Intensive Care Med 2005; 31:1195–1201

35 Reuter D, Felbinger TW, Schmidt C, et al. Stroke volumevariation for assessment of cardiac responsiveness to volumeloading in mechanically ventilated patients after cardiacsurgery. Intensive Care Med 2002; 28:392–398

36 Berkenstadt H, Margalit N, Hadani M, et al. Stroke volumevariation as a predictor of fluid responsiveness in patientsundergoing brain surgery. Anesth Analg 2001; 92:984–989

37 Wesseling KH, Jansen JRC, Settels JJ, et al. Computation ofaortic flow from pressure in humans using a nonlinear,three-element model. J Appl Physiol 1993; 74:2566–2573

38 Slama M, Masson H, Teboul JL, et al. Monitoring of respi-ratory variations of aortic blood flow velocity using esophagealDoppler. Intensive Care Med 2004; 30:1182–1187

39 Feissel M, Michard F, Mangin I, et al. Respiratory changes inaortic blood flow velocity as an indicator of fluid responsive-ness in ventilated patients with septic shock. Chest 2001;119:867–873

40 Perel A. Assessing fluid responsiveness by the systolic pres-

sure variation in mechanically ventilated patients: systolicpressure variation as a guide to fluid therapy in patients withsepsis-induced hypotension. Anesthesiology 1998; 89:1309–1310

41 Michard F, Boussat S, Chemla D, et al. Relation betweenrespiratory changes in arterial pulse pressure and fluid re-sponsiveness in septic patients with acute circulatory failure.Am J Respir Crit Care Med 2000; 162:134–138

42 Cassesson M, Besnard C, Durand PG, et al. Relation betweenrespiratory variations in pulse oximetry plethysmographicwaveform amplitude and arterial pulse pressure in ventilatedpatients. Crit Care 2005; 9:R562–R568

43 Natalini G, Rosano A, Taranto M, et al. Arterial versusplethysmographic dynamic indices to test responsiveness fortesting fluid administration in hypotensive patients: a clinicaltrial. Anesth Analg 2006; 103:1478–1484

44 Reuter DA, Bayerlein J, Goepfert MS, et al. Influence of tidalvolume on left ventricular stroke volume variation measuredby pulse contour analysis in mechanically ventilated patients.Intensive Care Med 2003; 29:476–480

45 De Backer D, Heenen S, Piagnerelli M, et al. Pulse pressurevariations to predict fluid responsiveness: influence of tidalvolume. Intensive Care Med 2005; 31:517–523

46 Monnet X, Rienzo M, Osman D, et al. Response to leg raisingpredicts fluid responsiveness during spontaneous breathingor with arrhythmia. Crit Care Med 2006; 34:1402–1407

47 Pinsky MR. Using ventilation-induced aortic pressure andflow variation to diagnose preload responsiveness. IntensiveCare Med 2004; 30:1008–1010

48 Barbier C, Loubieres Y, Schmit C, et al. Respiratory changesin inferior vena cava diameter are helpful in predicting fluidresponsiveness in ventilated septic patients. Intensive CareMed 2004; 30:1740–1746

49 Feissel M, Michard F, Faller JP, et al. The respiratoryvariation in the inferior vena cava diameter as a guide to fluidtherapy. Intensive Care Med 2004; 30:1834–1837

50 Vieillard-Baron A, Augarde R, Prin S, et al. Influence ofsuperior vena caval zone condition on cyclic changes in rightventricular outflow during respiratory support. Anesthesiol-ogy 2001; 95:1083–1088

51 Vieillard-Baron A, Chergui K, Rabiller A, et al. Superior venacaval collapsibility as a gauge of volume status in ventilatedseptic patients. Intensive Care Med 2004; 2004; 30:1734–1739

52 Jellinek H, Krafft P, Fitzgerald RD, et al. Right atrial pressurepredicts hemodynamic response to apneic positive airwaypressure. Crit Care Med 2000; 28:672–678

53 Thomas M, Shillingford J. The circulatory response to astandard postural change in ischaemic heart disease. BrHeart J 1965; 27:17–27

54 Boulain T, Achard JM, Teboul JL, et al. Changes in bloodpressure induced by passive leg raising predict response to fluidloading in critically ill patients. Chest 2002; 121:1245–1252

55 Singer M, Clarke J, Bennett D. Continuous hemodynamicmonitoring by esophageal Doppler. Crit Care Med 1989;17:447–452

56 Valtier B, Cholley BP, Belot JP, et al. Noninvasive monitoringof cardiac output in critically ill patients using transesopha-geal Doppler. Am J Respir Crit Care Med 1998; 158:77–83

57 Cariou A, Monchi M, Joly LM, et al. Noninvasive cardiacoutput monitoring by aortic blood flow determination: eval-uation of the Sometec Dynemo-3000 system. Crit Care Med1998; 1998; 26:2066–2072

58 Linton R, Band D, O’Brien T, et al. Lithium dilution cardiacoutput measurement: a comparison with thermodilution. CritCare Med 1997; 25:1796–1800

59 Pittman J, Bar-Yosef S, SumPing J, et al. Continuous cardiac




output monitoring with pulse contour analysis: a comparisonwith lithium indicator dilution cardiac output measurement.Crit Care Med 2005; 33:2015–2021

60 Rivers E, Nguyen B, Havstad S, et al. Early goal-directedtherapy in the treatment of severe sepsis and septic shock.N Engl J Med 2001; 345:1368–1377

61 Gan TJ, Soppitt A, Maroof M, et al. Goal-directed intraoper-ative fluid administration reduces length of hospital stay aftermajor surgery. Anesthesiology 2002; 97:820–826

62 Fenwick E, Wilson J, Sculpher M, et al. Pre-operativeoptimisation employing dopexamine or adrenaline for pa-tients undergoing major elective surgery: a cost-effectivenessanalysis. Intensive Care Med 2002; 28:599–608

63 Venn R, Steele A, Richardson P, et al. Randomised controlledtrial to investigate influence of the fluid challenge on durationof hospital stay and perioperative morbidity in patients withhip fractures. Br J Anaesth 2002; 88:65–71

64 Mythen MG, Webb AR. Perioperative plasma volume expan-sion reduces the incidence of gut mucosal hypoperfusionduring cardiac surgery. Arch Surg 1995; 130:423–429

65 Tang W, Weil MH, Sun S, et al. Gastric intramural Pco2 as amonitor of perfusion failure during hemorrhagic and anaphy-lactic shock. J Appl Physiol 1994; 76:572–577

66 McKendry M, McGloin H, Saberi D, et al. Randomisedcontrolled trial assessing the impact of a nurse delivered, flowmonitored protocol for optimisation of circulatory status aftercardiac surgery. BMJ 2004; 329:438–444

67 Pearse R, Dawson D, Fawcett J, et al. Early goal-directedtherapy after major surgery reduces complications and dura-tion of hospital stay: a randomised, controlled trial. Crit Care2005; 9:R687–R693

68 Shoemaker WC, Appel PL, Kram HB, et al. Prospective trialof supranormal values of survivors as therapeutic goals inhigh-risk surgical patients. Chest 1988; 94:1176–1186

69 Boyd O, Grounds M, Bennett ED. A randomized clinical trialof the effect of deliberate perioperative increase of oxygendelivery on mortality in high-risk surgical patients. JAMA1993; 270:2699–2707

70 Lobo S, Salgado P, Castillo VGT, et al. Effects of maximizingoxygen delivery on morbidity and mortality in high-risksurgical patients. Crit Care Med 2000; 28:3396–3404

71 Goepfert MS, Reuter DA, Akyol D, et al. Goal-directed fluidmanagement reduces vasopressor and catecholamine use in

cardiac surgery patients. Intensive Care Med 2007; 33:96–103

72 Bland RD, Shoemaker WC, Abraham E, et al. Hemodynamicand oxygen transport patterns in surviving and nonsurvivingpostoperative patients. Crit Care Med 1985; 13:85–90

73 Tuchschmidt J, Fired J, Astiz M, et al. Elevation of cardiacoutput and oxygen delivery improves outcome in septic shock.Chest 1992; 102:216–220

74 Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-orientedhemodynamic therapy in critically ill patients: Svo2 Collabo-rative Group. N Engl J Med 1995; 333:1025–1032

75 Heyland DK, Cook DJ, King D, et al. Maximizing oxygendelivery in critically ill patients: a methodologic appraisal ofthe evidence. Crit Care Med 1996; 1996; 24:517–524

76 Hayes MA, Timmins AC, Yau EH, et al. Evaluation ofsystemic oxygen delivery in the treatment of the critically ill.N Engl J Med 1994; 330:1717–1722

77 Reinhart K, Kuhn HJ, Hartog C, et al. Continuous centralvenous and pulmonary artery oxygen saturation monitoring inthe critically ill. Intensive Care Med 2004; 30:1572–1578

78 Kern JW, Shoemaker WC. Meta-analysis of hemodynamicoptimization in high-risk patients. Crit Care Med 2002;30:1686–1692

79 Vincent JL, de Carvalho FB, DeBacker D. Management ofseptic shock. Ann Med 2002; 34:606–613

80 Jastremski MS, Chelluri L, Beney KM. Analysis of the effectsof continuous on-line monitoring of mixed venous oxygensaturation on patient outcome and cost-effectiveness. CritCare Med 1989; 17:148–153

81 Nakagawa Y, Weil MH, Tang W, et al. Sublingual capnometryfor diagnosis and quantitation of circulatory shock. Am JRespir Crit Care Med 1998; 157:1838–1843

82 Marik PE, Bankov A. Sublingual capnometry versus tradi-tional markers of tissue oxygenation in critically ill patients.Crit Care Med 2003; 31:818–822

83 Clavijo-Alvarez JA, Sims CA, Soller B, et al. Monitoringskeletal muscle and subcutaneous tissue acid-base status andoxygenation during hemorrhagic shock and resuscitation.Shock 2005; 24:270–275

84 DeBacker D, Creteur J, Preiser JC, et al. Microcirculatoryblood flow is altered in patients with sepsis. Am J Respir CritCare Med 2002; 166:98–104




DOI 10.1378/chest.07-0073 2007;132; 2020-2029Chest

Michael R. Pinsky*Hemodynamic Evaluation and Monitoring in the ICU

December 2, 2009This information is current as of

& ServicesUpdated Information

l.htmlhttp://chestjournal.chestpubs.org/content/132/6/2020.fulhigh-resolution figures, can be found at:Updated Information and services, including

References

0.full.html#ref-list-1http://chestjournal.chestpubs.org/content/132/6/202accessed free at:This article cites 83 articles, 33 of which can be

Open AccessoptionFreely available online through CHEST open access

Permissions & Licensing

http://www.chestjournal.org/site/misc/reprints.xhtml(figures, tables) or in its entirety can be found online at: Information about reproducing this article in parts

Reprints http://www.chestjournal.org/site/misc/reprints.xhtml

Information about ordering reprints can be found online:

Email alerting service

online article.article. Sign up in the box at the top right corner of the Receive free email alerts when new articles cite this

formatImages in PowerPoint

format. See any online article figure for directions downloaded for teaching purposes in PowerPoint slide Figures that appear in CHEST articles can be


http://chestjournal.chestpubs.org/content/132/6/2020.full.html

http://chestjournal.chestpubs.org/content/132/6/2020.full.html#ref-list-1

http://www.chestjournal.org/site/misc/reprints.xhtml

http://www.chestjournal.org/site/misc/reprints.xhtml


Documents

Hemodynamic Evaluation and Monitoring in · 2017-08-14 · Hemodynamic Evaluation and Monitoring in the ICU* Michael R. Pinsky, MD, FCCP Hemodynamic monitoring, a cornerstone in the