REVISTA BRASILEIRA DE ANESTESIOLOGIA … · líquidos em cirurgia não ... 2015 Sociedade Brasileira de Anestesiologia. ... advances in perioperative management, the inci-dence optimization
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Rev Bras Anestesiol. 2016;66(5):513---528
REVISTABRASILEIRA DEANESTESIOLOGIA Publicação Oficial da Sociedade Brasileira de Anestesiologia
Javier Ripollésa,∗, Angel Espinosab, Eugenio Martínez-Hurtadoa,Alfredo Abad-Gurumetac, Rubén Casans-Francésd, Cristina Fernández-Péreze,Francisco López-Timonedaf, José María Calvo-Vecinoa,EAR Group (Evidence Anestesia Review Group)
a Departamento de Anestesia, Hospital Universitario Infanta Leonor, Universidad Complutense de Madrid, Madrid, Spainb Department of Anesthesia, Blekinge County Council Hospital, WämöCenter, Karlskrona, Swedenc Departamento de Anestesia, Hospital Universitario la Paz, Madrid, Spaind Departamento de Anestesia, Hospital Universitario Lozano Blesa, Zaragoza, Spaine Departamento de Medicina Preventiva y Salud Pública, Hospital Clínico San Carlos, Madrid, Spainf Departamento de Anestesia, Hospital Clinico San Carlos, Universidad Complutense de Madrid, Madrid, Spain
Received 23 December 2014; accepted 18 February 2015Available online 14 September 2015
KEYWORDSGoal directed fluidtherapy;Meta-analysis;Hemodynamic goal;Noncardiac surgery
AbstractBackground: The goal directed hemodynamic therapy is an approach focused on the use ofcardiac output and related parameters as end-points for fluids and drugs to optimize tissueperfusion and oxygen delivery. Primary aim: To determine the effects of intraoperative goaldirected hemodynamic therapy on postoperative complications rates.Methods: A meta-analysis was carried out of the effects of goal directed hemodynamic ther-apy in adult noncardiac surgery on postoperative complications and mortality using PreferredReporting Items for Systematic Reviews and Meta-Analyses methodology. A systematic searchwas performed in Medline PubMed, Embase, and the Cochrane Library (last update, October2014). Inclusion criteria were randomized clinical trials in which intraoperative goal directedhemodynamic therapy was compared to conventional fluid management in noncardiac surgery.Exclusion criteria were trauma and pediatric surgery studies and that using pulmonary artery
catheter. End-points were postoperative complications (primary) and mortality (secondary).
he entry criteria were examined in full and subjected to quantifiableup analysis (stratified by type of monitor, therapy, and hemodynamictivity analysis.
Those studies that fulfilled tanalysis, predefined subgrogoal), and predefined sensi
he perioperative management of high-risk surgical patientsontinues to be a challenge for the anesthesiologists.espite advances in perioperative management, the inci-ence of serious complications after major surgery remainsigh.1,2 A decrease in perioperative oxygen transport is
losely related to the development of organ failure andeath.3 It has also been demonstrated that a large high-riskurgical population accounts for 12.5% of surgical proce-ures and for more than 80% of deaths.4 Surgical patients
aoac
an be classified as high risk based on surgical factors oratient-related factors.5 Goal directed hemodynamic ther-py (GDHT) is based on the optimization of preload with these of algorithms based on fluids, inotropes and/or vasopres-ors to achieve a certain goal in stroke volume (SV), cardiacndex (CI), or oxygen delivery (DO2). The ultimate goal of thisptimization is to avoid fluid overload, tissue hypoperfusion,
6
nd hypoxia. All the studies of perioperative hemodynamicptimization had the same starting point, fluid loading,nd the same endpoint, achieving adequate DO2. However,linical heterogeneity between studies of GDHT cannot be
ignored, with regard to type of surgery, patient’s character-istics, therapeutic goals, methods for achieving these goalsand monitoring. The pulmonary arterial catheter (PAC) hasbeen considered to be the ‘‘gold standard’’ for monitor-ing preload, afterload, contractility, and tissue oxygenation.The invasiveness and high rate of complications associatedwith this device render it as unsuitable for routine use inmost cases. The use of minimally invasive monitoring hasgained popularity in the past few years; these devices havebeen validated intraoperatively. Currently PAC is not rec-ommended in most surgeries, and for this reason it was notanalyzed in this meta-analysis. There are no data to supportthe practice of using central venous pressure to guide fluidtherapy,7 therefore, central venous pressure-guided fluidtherapy was not included in analysis.
Yet there are no studies in which different algorithmsor different objectives are compared. The best method forassessing tissue oxygenation and intravascular volume hasnot yet been defined. The present review was designed toupdate the published evidence and determine the effective-ness of intraoperative GDHT with regard to complicationsand mortality with different types of algorithms and moni-tors used.
Material and methods
Selection criteria
The Preferred Reporting Items for Systematic Reviews andMeta-Analyses (PRISMA) methodology8 was used to identifythe studies, based on the following inclusion criteria:
1. Participants: Adult patients (over 18 years) undergoingelective noncardiac surgery were included. The studieswere not limited in terms of surgical risk.
2. Types of intervention: Intraoperative goal directedhemodynamic therapy: defined as hemodynamic moni-toring that allows to perform a hemodynamic optimiza-tion algorithm based on the use of fluids, inotropesand/or vasopressors to achieve normal or supranormalhemodynamic values. GDHT guided by pulmonary arterycatheter, transesophageal echocardiography or centralvenous pressure-guided GDHT were excluded.
3. Types of comparison: The studies that were selectedfor analysis included those that compared GDHT withconventional fluid management (monitoring of bloodpressure, electrocardiogram, heart rate, urine outputand/or central venous pressure).
4. Results: RCTs reporting any of the following outcomes:postoperative complications and/or mortality.
5. Types of studies: RCTs where intraoperative GDHT wasperformed in adult patients scheduled for noncardiacmajor surgery. Only peer-reviewed manuscripts wereincluded.
Sources of information
Following the PRISMA protocol8 different search strategies(last updated in October 2014) were used to iden-tify relevant studies that met inclusion criteria usingEMBASE, MEDLINE and the Cochrane Library. There were no
RlM
515
estrictions on the publication date or language. In addi-ion to electronic searching, industry representatives wereontacted for additional material. All identified review arti-les and evidence-based guidelines were hand-searched fordditional references.
earch itemshe search was conducted using the following key words:urgery, fluid, goal directed, end point, hemodynamic, tar-et, goal and randomized controlled trial.
tudy selection and data extraction
wo independent investigators assessed each title andbstract in order to discard any irrelevant RCTs and identifyhose potentially relevant. These RCTs were analyzed select-ng those that met the inclusion criteria outlined above.CT data extraction was performed by two different inves-igators and any discrepancy required further analysis andonfirmation by a third investigator. The authors reviewedhe data analysis in order to avoid errors in data transcrip-ion.
ssessment of risk of bias in included studies
ias assessment risk was performed using the Cochrane riskf bias tool. From this tool, we used seven domains to assesshe methodological quality of the studies included in thenalysis.
utcome variableshe primary outcome was the overall postoperativeomplications. The results were stratified according to theollowing variables: monitor utilized, therapy used to reach
hemodynamic goal and the hemodynamic goal. For theredefined subgroup analysis, studies were grouped:
1) Monitor: (a) Arterial pulse contour analysis methods(Vigileo/Flotrac®, Edwards Lifesciences Corporation,USA; ProAQT®, Pulsion medical systems SE, Germany;LiDCO Plus®, LiDCO Ltd., UK); (b) oesophageal DopplerMonitoring-ODM (CardioQ®, Deltex Medical, UK); (c)noninvasive methods (Masimo®, Masimo Corporation;CNAP® PPV, CNSystems Medizintechnik AG) and (d)measures of oxygen delivery and extraction methods.
2) Therapy: (a) Fluids; (b) fluids and inotropes; (c)vasopressors and fluids and (4) fluids, inotropes andvasopressors.
eview manager (‘‘Revman’’) 10 for MAC (Cochrane col-aboration, Oxford, UK) was used for statistical analysis.eta-analysis was carried out using the Mantel---Haenszel
516 J. Ripollés et al.
Records identified throughdatabase searching
(n=14 160)
Iden
tific
atio
nS
cree
ning
Elig
ibili
tyIn
clud
edRecords after duplicates removed
(n=10 475)
Records screened(n=1003)
Full-text articles assessedfor eligibility
(n=51)
Studies included inqualitative synthesis
(n=29)
Records excluded(n=9472)
Full-text articles excluded,
Outcomes not reportedn=2
Perioperative (n=6)Cardiac surgery (n=5)
Emergency surgery (n=1)Sepsis (n=2)
Both groups GDT (n=6)
Studies included inquantitative synthesis
(meta-analysis)(n=29)
Additional records identifiedthrough other sources
(n=5)
illu
r(wsuaa
hsc
Fl
Figure 1 Flow diagram
andom-effects model, with results presented as risk ratioRR) with a 95% confidence interval (CI). Forest plotsere then constructed, considering p < 0.050 as statistically
ignificant effect. Statistical heterogeneity was evaluatedsing I2 statistics; I2 values of less than 25% were defineds low heterogeneity, 25---50% as moderate heterogeneitynd greater than 50% as high heterogeneity. A �2 test for
ctwa
Random sequence generation (selection b
Allocation concealment (selection b
Blinding of participants and personnel (performance b
Blinding of outcome assessment (detection b
Incomplete outcome data (attrition b
Selective reporting (reporting b
Other b
Low risk of bias Unclear risk of bia
igure 2 Review authors’ judgements about each risk of bias itemow risk of bias; white, unclear risk of bias; red, high risk of bias.
strating search strategy.
eterogeneity was performed, with p < 0.100 regarded astatistically significant. A priori sensitivity analyses wereonducted on both the primary and the secondary out-
omes by restricting the analysis to high quality trials: tohose studies that showed no allocation bias and thoseithout randomization/allocation bias. Publication bias wasssessed using funnel plot techniques.
ias)
ias)
ias)
ias)
ias)
ias)
ias
s High risk of bias
0% 25% 50% 75% 100%
presented as percentages across all included studies. Green,
517
Ran
dom
seq
uenc
e ge
nera
tion
(sel
ectio
n bi
as)
Inco
mpl
ete
outc
ome
data
(at
triti
on b
ias)
Sel
ectiv
e re
port
ing
(rep
ortin
g bi
as)
Oth
er b
ias
Blin
ding
of o
utco
me
asse
ssm
ent (
dete
ctio
n bi
as)
Blin
ding
of p
artic
ipan
ts a
nd p
erso
nnel
(pe
rfor
man
ce b
ias)
Allo
catio
n co
ncea
lmen
t (se
lect
ion
bias
)
Bartha et al. 2012
Benes et al. 2010
Brandstrup et al. 2012
Buettner et al. 2008
Cecconi et al. 2011
Challand et al. 2012
Conway et al. 2002
Donati et al. 2007
Forget et al. 2010
Forget et al. 2013
Gan et al. 2002
Jammer et al. 2010
Lopes et al. 2007
Mayer et al. 2010
McKenny et al. 2013
Nobblet et al. 2006
Peng et al. 2014
Salzwedel et al. 2013
Scheeren et al. 2013
Senagore et al. 2009
Sinclair et al. 1997
Srinivasa et al. 2013
Van der Linden et al. 2010
Venn et al. 2002
Walkening et al. 2005
Zakhaleva et al. 2013
Zhang et al. 2012
Zhang et al. 2013
Zheng et al. 2013
Figure 3 Review authors’ judgements about each risk of biasitem for each included study.
Goal directed hemodynamic therapy
Results
Study selection
There were 14,160 references in electronic databases, ofwhich 1003 were screened. Of those, 55 RCTs were analyzedand 24 of them were included for systematic review andmeta-analysis, excluding those who did not meet the inclu-sion criteria. Finally a total of 29 RCTs were included9---37;5 RCTs were added by manual search. 2654 patients wereincluded. Fig. 1 shows the flowchart used for item selection.
Assessment of risk of bias in individual studies
Bias risk was analyzed with the Cochrane tool. This was per-formed by two authors independently and we resolved anydisparity by discussion and the involvement of a third per-son. We present the methodological quality in a summarytable and a graph (Figs. 2 and 3).
Characteristics of the studies included in theanalysis
The selected articles describe the results of RCTs that eval-uated the use of intraoperative GDHT in noncardiac electivesurgery, and that included postoperative complicationsand/or mortality as outcome. The characteristics of theincluded RCTs are shown in Table 1.
Primary results
1. Total complicationsAnalyzing the 29 RCTs, 26 describe the total associated
complications10---37 GDHT was associated with a significantreduction in overall complication compared with patientstreated in the control group (RR: 0.70, 95% CI: 0.62---0.79,p < 0.001) (Fig. 4).
2. Complications by monitorA significant decrease in complications was found in
subgroup based on pulse contour analysis (RR: 0.78, 95%CI: 0.59---0.99, p = 0.04) and subgroup ODM (RR: 0.67, 95%CI: 0.53---0.85, p < 0.001). However, it was not shown inthe subgroup of noninvasive monitoring (RR: 0.57, 95%CI: 0.28---1.15, p = 0.12) and was based on measures ofoxygen delivery and extraction methods (RR: 0.59, 95%CI: 0.20---1.80, p = 0.36) (Fig. 5).
3. Complications by hemodynamic therapyA significant decrease in complications was observed
in the fluids as monotherapy subgroup (RR: 0.69, 95%CI: 0.57---0.84, p < 0.001), fluids and vasopressors sub-group (RR: 0.76, 95% CI: 0.68---0.85, p < 0.001), and fluids,vasopressors and inotropes subgroup (RR: 0.54, 95% CI:0.32---0.89, p = 0.02). However, the use of fluids andinotropes showed no decrease in complications (RR. 0.66,95% CI: 0.34---1.28, p = 0.22) (Fig. 6).
4. Complications by hemodynamic goal
A decrease in complications was associated with the
use of GDHT in the following subgroups: Svmaximization(RR: 0.73, 95% CI: 0.61---0.89, p < 0.001), in the subgrouppreload responsiveness (RR: 0.73, 95% CI: 0.59---0.95),
518 J. Ripollés et al.
Table 1 PICO characteristic of included studies.
Study Year Patients Intervention Comparator Outcomes Studydesign
Sinclairet al.
1997 Patients over 55years undergoinghip replacementsurgery
CardioQ-guided GDHTby optimizing SV andcFT with fluidsn 20
Standard caren 20
Length of stay,hemodynamicparameters
RCT
Conwayet al.
2002 ASA I---III patientsundergoingcolorectal surgery
p < 0.001), and with a CI target, CI > 2.5 mL/min/m2 (RR:0.58, 95% CI: 0.44---0.76, p < 0.001), whereas in the sub-group that utilized measures of oxygen delivery andextraction methods no significant decrease was observed(Fig. 7).
ortality
o significant differences were found with regard to mortal-ty (RR: 0.76, 95% CI: 0.45---1.28, p = 0.30) (Fig. 8).
hwpp
n delivery index; SVI, indexed stroke volume; PP, pulse pressure;
ensitivity analysis, assessment risk of bias acrosstudies and publication biaso changes in the results with regard to complications[RR: (CI 95%) 0.71 (0.61---0.82), p < 0.01]) or mortality ([RR:CI 95%) 0.77 (0.42---1.40), p = 0.39]) were observed whenestricting the analysis to those studies that had no allo-ation bias; or when restricting the analysis to those that
ad no allocation and/or randomization bias in the resultsith regard to complications [RR (CI 95%) 0.69 (0.59---0.81),
< 0.01] and mortality ([RR (CI 95%) 0.95 (0.45---1.85), = 0.87]) On the other hand, a prespecified group analysis
Goal directed hemodynamic therapy 521
Gan et al. 2002Venn et al. 2002Conway et al. 2002Walkening et al. 2005Nobblet et al. 2006Donati et al. 2007Lopes et al. 2007Senagore et al. 2009Mayer et al. 2010Forget et al. 2010Jammer et al. 2010Van der Linden et al. 2010Benes et al. 2010Cecconi et al. 2011Zhang et al. 2012Brandstrup et al. 2012Challand et al. 2012Bartha et al. 2012Zhang et al. 2013McKenny et al. 2013Scheeren et al. 2013Salzwedel et al. 2013Forget et al. 2013Srinivasa et al. 2013Zakhaleva et al. 2013Zheng et al. 2013Peng et al. 2014
Total (95% CI)Total eventsHeterogeneity: Tau2=0.04 ; Chi2=54.70, df=26 ( P=.0008); I2=52%Test for overall effect: Z=5.71 (P<.00001)
1275
StudyGDHT
Events Total Weight YearRisk Ratio
M-H, Random, 95% CIRisk Ratio
M-H, Random, 95% CIControl
Events Total
1259 100.0% 0.70 [0.62, 0.79]460 645
0.01Favours GDHT Favours control 0.1 1 10 100
roup
ruwavsascawwtftbAaasdcOar
Figure 4 Effect of GDHT in the protocol g
has been conducted and described above. A funnel plot wasdrawn for the primary outcome comparison to explore thepossibility of publication bias. The symmetry of the funnelplot was assessed visually and did not suggest publicationbias (Fig. 9).
Discussion
Numerous studies have reported differences betweentechnologies, especially in their response to typical sur-gical interventions such as fluid and vasoactive drugadministration.38 The results obtained with one type ofmonitoring cannot be extrapolated to those obtained withother monitors.39 RCTs using noninvasive monitors werelimited both in number of patients studied and meth-odological quality. How minimal invasive cardiac outputmonitoring techniques can be used to guide individ-ualized fluid management40 needs to be sustained byvalidation studies that adhere to the proposed methodologi-cal considerations41 as well as large-scale clinical outcomestudies.
There is some evidence that SV maximization strategiescould be harmful in aerobically fit patients by lead-ing to volume overload,25 and recent evidence suggests
that this goal does not provide the benefits previouslydescribed.42
However, the results of our meta-analysis show thatthis hemodynamic goal remains valid. The use of dynamic
tp
s
vs control group on overall complications.
esponse parameters to volume may decrease the risk of vol-me overload. A CI > 2.5 mL/min/m2 as hemodynamic targetithin algorithms in which fluids, vasopressors and inotropesre used avoid the risk of hypotension due to decreasedasomotor tone. The use of inotropes increases the CO inituations where the patient is nonresponsive to the volumend does not present a reduced vasomotor tone. Inotropicupport with dobutamine can result in changes in microvas-ular flow related to direct effects on the microcirculations well as global CO.43 With the exception of ScvO2, thatas only evaluated in one RCT,19 and was not associatedith better outcomes, this meta-analysis has not been able
o detect significant differences between subgroups. There-ore it seems reasonable to adapt GDHT to risk patient,ype of surgery as well as its duration44 as recommendedy recent European Society of Anaesthesiology guidelines.45
multicenter observational trial in patients with intra-bdominal surgery found that low ScvO2 was associated withn increased risk of postoperative complications in high-riskurgery. In this trial, the optimal value of mean ScvO2 toiscriminate between patients who did or did not developomplications was 73% (sensitivity 72%, specificity 61%)46
ne of the major limitations of venous oximetry is that,s a global marker of demand-supply balance, it does noteflect organ-specific malperfusion. Whether ScvO2 moni-
oring improves outcomes in surgical patients remains to beroven in large RCTs.
Unlike our results, a recent meta-analysis has shown aignificant benefit of GDHT in patients receiving fluids and
522 J. Ripollés et al.
SubgroupPulse contour devices
Oesophageal Doppler Monitoring
Measures of Oxygen Delivery and Extraction
Non invasive
Total (95% CI)
GDHTEvents Total Weight
Risk RatioM-H, Random, 95% CI
Risk RatioM-H, Random, 95% CI
ControlEvents Total
0.01Favours GDHT Favours control 0.1 1 10 100
Bartha et al. 2012Benes et al. 2010Cecconi et al. 2011Mayer et al. 2010Peng et al. 2014Salzwedel et al. 2013Scheeren et al. 2013Van der Linden et al. 2010Zhang et al. 2012Zhang et al. 2013Zheng et al. 2013Subtotal (95% CI)
Brandstrup et al. 2012Challand et al. 2012Conway et al. 2002Gan et al. 2002McKenny et al. 2013Nobblet et al. 2006Senagore et al. 2009Srinivasa et al. 2013Venn et al. 2002Walkening et al. 2005Zakhaleva et al. 2013Subtotal (95% CI)
Donati et al. 2007Jammer et al. 2010Subtotal (95% CI)
Forget et al. 2010Forget et al. 2013Lopes et al. 2007Subtotal (95% CI)
Total eventsHeterogeneity: Tau2=0.11 ; Chi2=42.20, df=10 ( P=.00001); I2=76%Test for overall effect: Z=2.02 (P<.04)
Total eventsHeterogeneity: Tau2=0.05 ; Chi2=17.18, df=10 ( P=.07); I2=42%Test for overall effect: Z=3.39 (P<.0007)
Total eventsHeterogeneity: Tau2=0.57 ; Chi2=9.05, df=1 (P=.003); I2=89%Test for overall effect: Z=0.92 (P<.36)
Total eventsHeterogeneity: Tau2=0.24 ; Chi2=6.73, df=2 (P=.03); I2=70%Test for overall effect: Z=1.57 (P<.12)
Total eventsHeterogeneity: Tau2=0.07 ; Chi2=74.18, df=26 ( P=.00001); I2=65%Test for overall effect: Z=4.50 (P<.00001)Test for subgroup differences: Chi2=1.21, df=3 ( P=.75); I2=0%
Figure 5 Effect of GDHT in the protocol group vs control group on overall complications grouped by monitor. Pulse contourd : CN
io
cetm
lfi
evices: Vigileo/Flotrac®, ProAQT, and LiDCO Plus®. No invasive
notropes in order to achieve supraphysiological targets forxygen delivery in high-risk patients.47
This meta-analysis was unable to demonstrate a signifi-
ant reduction in mortality. There are a number of reasons toxplain why the control mortality may have decreased overime. These include: (1) better overall care thus decreasingortality for similar patients; (2) clinicians’ awareness,
atte
AP® PPV and Masimo®.
earning from previous early published studies and there-ore drifting their practice toward lower risk groups; (3)mprovement in technology, that has become less invasive
48
nd therefore, gaining more credibility. Another reason forhis may be that the most recent studies are not poweredo assess mortality; in earlier studies, mortality was consid-red the most relevant endpoint, but this has changed to
Goal directed hemodynamic therapy 523
Fluids
Fluids and inotropes
Fluids and vasopressor
Fluids, inotropes and vasopressor
Total (95% CI)
0.01Favours GDHT Favours control 0.1 1 10 100
Total eventsHeterogeneity: Tau2= 0.06 ; Chi2=27.25, df=15 (P=.02); I2=46%Test for overall effect: Z=3.65 (P=.0003)
Total eventsHeterogeneity: Tau2=0.31 ; Chi2=12.69, df=3 (P=.005); I2=76%Test for overall effect: Z=1.22 (P=.22)
Total eventsHeterogeneity: Tau2=0.00 ; Chi2=1.98, df=3 (P=.58); I2=0%Test for overall effect: Z=4.77 (P<.00001)
Total eventsHeterogeneity: Tau2=0.14 ; Chi2=8.64, df=2 (P=.01); I2=77%Test for overall effect: Z=2.41 (P=.02)
Total eventsHeterogeneity: Tau2=0.04 ; Chi2=53.21, df=26 (P=.001); I2=51%Test for overall effect: Z=5.82 (P<.00001)Test for subgroup differences: Chi2=2.31, df=3 (P=.51); I2=0%
Gan et al. 2002Venn et al. 2002Conway et al . 2002Walkening et al. 2005Nobblet et al . 2006Lopes et al . 2007Senagore et al. 2009Jammer et al. 2010Brandstrup et al. 2012Challand et al . 2012Zhang et al . 2012Scheeren et al . 2013Zakhaleva et al . 2013Forget et al. 2013McKenny et al . 2013Srinivasa et al. 2013Subtotal (95% CI)
Donati et al . 2007Van der Linden et al. 2010Bartha et al . 2012Zhang et al . 2013Subtotal (95% CI)
Forget et al. 2010Cecconi et al . 2011Zheng et al . 2013Peng et al. 2014Subtotal (95% CI)
Benes et al . 2010Mayer et al. 2010Salzwedel et al . 2013Subtotal (95% CI)
Figure 6 Effect of GDHT in the protocol group vs control group on overall complications grouped by therapy.
wrsb
length of stay and morbidity endpoints with less high-riskpatient group, and as a result, have very low or no mortal-ity. However, a reduction in mortality associated with GDHTwas demonstrated in groups of extremely high-risk patients
49
(baseline mortality rate of >20%) as well as with long-termfollow-up.50
Unlike previous meta-analysis, we have not includedthose studies in which PAC was used, since these studies
opic
ere published over 10 years, and do not reflect cur-ent practice. Grocott et al. meta-analysis51 included 31tudies with 5292 participants. The results are dominatedy a single large RCT with a weight of more than 60%
52
f the overall population in which PAC was used. Theresent meta-analysis confirms that the use of minimallynvasive monitoring is effective and reduces postoperativeomplications. Postsurgical complications have a dramatic
524 J. Ripollés et al.
Conway et al. 2002Gan et al. 2002Venn et al. 2002Walkening et al. 2005Nobblet et al. 2006Senagore et al. 2009Cecconi et al. 2011Challand et al. 2012Brandstrup et al. 2012Bartha et al. 2012Srinivasa et al. 2013McKenny et al. 2013Zakhaleva et al. 2013Subtotal (95% CI)Total events
Lopes et al. 2007Forget et al. 2010Zhang et al. 2012Scheeren et al. 2013Forget et al. 2013Peng et al. 2014Subtotal (95% CI)Total events
Donati et al. 2007Van der linden et al. 2010Mayer et al. 2010Benes et al. 2010Zheng et al. 2013Zhang et al. 2013Salzwedel et al. 2013Subtotal (95% CI)Total events
936
1819
952
110
68203060303079
287
272
1535301672
182
67173050303081
275
2.4%0.5%
4.4%6.6%2.6%8.2%
24.6%
Jammer et al. 2010Subtotal (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z=0.06 (P=.96)
Heterogeneity: Tau2=0.04; Chi2=50.42, df=25 (P=.002); I2=50%Test for overall effect: Z=5.66 (P<.00001)Test for subgroup differences: Chi2=6.94, df=3 (P=0.07), I2=56.8%
Figure 7 Effect of GDHT in the protocol group vs control group on overall complications grouped by hemodynamic goal. (SV,stroke volume; CI, cardiac index, SVV, stroke volume variation; PPV, pulse pressure variation, PVI®, Pleth Variability Index).
iGwnCdQ
R
M
mpact on costs. Potential costs savings resulting fromDHT are substantial53 and seem to be cost effective evenith moderate clinical effect.54 Particularly, ODM tech-
ology has been considered favorably by both the NHSenter for evidence-based purchasing in the United King-om and United States Agency for Healthcare Research anduality.55,56
daat
esearch implications
ore studies in which different types of monitoring and
ifferent types of algorithms and hemodynamic therapiesre compared in patients with different risk in order tochieve optimal hemodynamic goals are needed. In addi-ion, outcome report should be standardized. In this regard,
Goal directed hemodynamic therapy 525
Study
Sinclair et al. 1997Venn et al. 2002Conway et al. 2002Walkening et al. 2005Nobblet et al. 2006Donati et al. 2007Lopes et al. 2007Buettner et al. 2008Senagore et al. 2009Mayer et al. 2010Benes et al. 2010Forget et al. 2010Van der Linden et al. 2010Jammer et al. 2010Cecconi et al. 2011Challand et al. 2012Brandstrup et al. 2012Zhang et al. 2012Bartha et al. 2012McKenny et al. 2013Srinivasa et al. 2013Scheeren et al. 2013Forget et al. 2013Zheng et al. 2013Zakhaleva et al. 2013Zhang et al. 2013Peng et al. 2014
recommendations for the evaluation and standardization ofperioperative complications have been recently published.57
In summary, more studies are needed to demonstrate a sig-nificant reduction in mortality associated with GDHT.
Weaknesses in study
The study by Mayer et al.22 has been under investigationfor ethical reasons, the manuscript has not been withdrawn
0
0.5
1.5
1
20.01 0.1 1 10 100
RR
SE(log[RR])
Figure 9 Funnel plot of the published studies in relation tothe primary outcome. The measure of precision used is thestandard error (SE) of the log RR.
iisoa
ctiderra
dhntt
p
group vs control group on mortality.
nd remains part of the scientific record at the time weearched the literature. To verify potential biases in ouresults, both the primary and the secondary outcome weree-examined without including the Mayer et al.22 manuscriptnd no differences were found.
Many trials were single center trials and only one hasnvestigated more than 100 patients per group.19 Differencesn methodological quality may cause heterogeneity. Smallertudies tend to be conducted and analyzed with less method-logical rigor than larger studies, and trials of lower qualitylso tend to show larger intervention effects.
The major limitation of our analysis is that overallomplications were analyzed, regardless of the severity ofhese and their impact on length of stay and/or mortal-ty. Furthermore, the use of different surgical interventions,ifferent monitoring systems and algorithms adds more het-rogeneity to the analysis. Thus, study heterogeneity mayeduce the precision of treatment effect estimates andeduce the generalizability of the results of this meta-nalysis.
The present meta-analysis is based on studies thatescribe the incidence of postoperative complications. Itas to be recognized that the reporting of complications isot consistent and that the definitions used can differ inype, definition and importance between studies, limiting
he applicability of some of our findings.
Furthermore, and unlike previous meta-analysis, theresent meta-analysis conducted a global analysis of total
5
csrwq
C
Toisdt
C
T
F
Tp
A
TfM
Gb
R
1
1
1
1
1
1
1
1
1
1
2
2
2
26
omplications, without conducting an organ-specific47,58 ortratified by risk49 analysis. Despite these limitations, theesults are consistent in most subgroups analyzed and evenhen the analysis is restricted to those studies with higheruality.
onclusions
he results of this meta-analysis show that the usef intraoperative GDHT with minimally invasive monitor-ng decreases postoperative complications in noncardiacurgery, although it was not possible to show a significantecrease in mortality rate. ScvO2 monitoring was not ableo decrease the frequency of complications.
onflicts of interest
he authors declare no conflicts of interest.
unding
his study had no funding. EAR group is a collaborative inde-endent nonprofit group, endless external funding.
cknowledgements
he authors wish to thank the professionals of the Pro-essional Library of the University Hospital Infanta Leonor,adrid.
The authors wish to thank Drs. Angela Camps, Susanaonzález and Miriam de Nadal for their invaluable contri-ution in training and interest in this area.
eferences
1. Turrentine FE, Wang H, Simpson VB, et al. Surgical risk factors,morbidity, and mortality in elderly patients. J Am Coll Surg.2006;203:865---77.
2. Story DA, Leslie K, Myles PS, et al. Complications and mortal-ity in older surgical patients in Australia and New Zealand (theREASON study): a multicentre, prospective, observational study.Anaesthesia. 2010;65:1022---30.
3. Mythen MG, Webb AR. Intra-operative gut mucosal hypoper-fusion is associated with increased post-operative complicationsand cost. Intensive Care Med. 1994;20:99---104.
4. Bland RD, Shoemaker WC. Probability of survival as a prognosticand severity of illness score in critically ill surgical patients. CritCare Med. 1985;13:91---5.
5. Pearse RM, Harrison DA, James P, et al. Identification of thehigh risk surgical population in the United Kingdom. Crit Care.2006;10:R81.
6. Older P, Hall A. Clinical review: how to identify high-risk surgicalpatients. Crit Care. 2004;8:369---72.
7. Lobo SM, Salgado PF, Castillo VG, et al. Effects of maximizingoxygen delivery on morbidity and mortality in high-risk surgicalpatients. Crit Care Med. 2000;28:3396---404.
2
J. Ripollés et al.
8. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting itemsfor systematic reviews and meta-analyses: the PRISMA state-ment. BMJ. 2009;339:b2535.
9. Sinclair S, James S, Singer M. Intraoperative intravascular vol-ume optimization and length of hospital stay after repair ofproximal femoral fracture: randomised controlled trial. BMJ.1997;315:909---12.
0. Conway DH, Mayall R, Abdul-Latif MS, et al. Randomised con-trolled trial investigating the influence of intravenous fluidtitration using oesophageal Doppler monitoring during bowelsurgery. Anaesthesia. 2002;57:845---9.
1. Gan TJ, Soppitt A, Maroof M, et al. Goal-directed intraoperativefluid administration reduces length of hospital stay after majorsurgery. Anesthesiology. 2002;97:820---6.
2. Venn R, Steele A, Richardson P, et al. Randomized controlledtrial to investigate influence of the fluid challenge on durationof hospital stay and perioperative morbidity in patients with hipfractures. Br J Anaesth. 2002;88:65---71.
3. Wakeling HG, McFall MR, Jenkins CS, et al. Intraoperativeoesophageal Doppler guided fluid management shortens post-operative hospital stay after major bowel surgery. Br J Anaesth.2005;95:634---42.
4. Noblett SE, Snowden CP, Shenton BK, et al. Randomized clinicaltrial assessing the effect of Doppler-optimized fluid manage-ment on outcome after elective colorectal resection. Br J Surg.2006;93:1069---76.
5. Donati A, Loggi S, Preiser JC, et al. Goal-directed intra-operative therapy reduces morbidity and length of hos-pital stay in high-risk surgical patients. Chest. 2007;132:1817---24.
6. Lopes MR, Oliveira MA, Pereira VOS, et al. Goal-directed fluidmanagement based on pulse pressure variation monitoring dur-ing high-risk surgery: a pilot randomized controlled trial. CritCare. 2007;11:R100.
7. Buettner M, Schummer W, Huettemann E, et al. Influence ofsystolic-pressure-variation guided intraoperative fluid manage-ment on organ function and oxygen transport. Br J Anaesth.2008;101:194---9.
8. Senagore A, Emery T, Luchtefeld M, et al. Fluid managementfor laparoscopic colectomy: a prospective randomized assess-ment of goal directed administration of balanced salt solutionor hetastarch coupled with an enhanced recovery program. DisColon Rectum. 2009;52:1935---40.
9. Jammer I, Ulvik A, Erichsen C, et al. Does centralvenous oxygen saturation-directed fluid therapy affect post-operative morbidity after colorectal surgery? A randomizedassessor-blinded controlled trial. Anesthesiology. 2011;113:1072---80.
0. Van Der Linden PJ, Dierick A, Wilmin S, et al. A random-ized controlled trial comparing an intraoperative goal-directedstrategy with routine clinical practice in patients undergo-ing peripheral arterial surgery. Eur J Anaesthesiol. 2010;27:788---93.
1. Forget P, Lois F, De KM. Goal-directed fluid management basedon the pulse oximeter-derived pleth variability index reduceslactate levels and improves fluid management. Anesth Analg.2010;111:910---4.
2. Mayer J, Boldt J, Mengistu AM, et al. Goal-directed intra-operative therapy based on autocalibrated arterial pressurewaveform analysis reduces hospital stay in high-risk surgicalpatients: a randomized, controlled trial. Crit Care.2010;14:R18.
3. Benes J, Chytra I, Altmann P, et al. Intraoperative fluid opti-mization using stroke volume variation in high risk surgical
patients: results of prospective randomized study. Crit Care.2010;14:R118.
patients. Health Technol Assess. 2009;13, iii---iv, ix---xii, 1---95.56. Agency for Healthcare Research and Quality. Technology Assess-
Goal directed hemodynamic therapy
24. Cecconi M, Fasano N, Langiano N, et al. Goal-directedhaemodynamic therapy during elective total hip arthroplastyunder regional anaesthesia. Crit Care. 2011;15:R132.
25. Challand C, Struthers R, Sneyd JR, et al. Randomized controlledtrial of intraoperative goal-directed fluid therapy in aerobi-cally fit and unfit patients having major colorectal surgery. Br JAnaesth. 2012;108:53---62.
26. Brandstrup B, Svendsen PE, Rasmussen M, et al. Which goal forfluid therapy during colorectal surgery is followed by the bestoutcome: near-maximal stroke volume or zero fluid balance? BrJ Anaesth. 2012;109:191---9.
27. Bartha E, Arfwedson C, Imnell A, et al. Randomized controlledtrial of goal-directed haemodynamic treatment in patientswith proximal femoral fracture. Br J Anaesth. 2013;110:545---53.
28. Zhang J, Qiao H, He Z, et al. Intraoperative fluid management inopen gastrointestinal surgery: goal-directed versus restrictive.Clinics (Sao Paulo). 2012;67:1149---55.
29. Salzwedel C, Puig J, Carstens A, et al. Perioperative goal-directed hemodynamic therapy based on radial arterial pulsepressure variation and continuous cardiac index trendingreduces postoperative complications after major abdominalsurgery: a multi-center, prospective, randomized study. CritCare. 2013;17:R191.
30. Scheeren TWL, Wiesenack C, Gerlach H, et al. Goal-directedintraoperative fluid therapy guided by stroke volume and itsvariation in high-risk surgical patients: a prospective ran-domized multicentre study. J Clin Monit Comput. 2013;27:249---57.
31. Zhang J, Chen CQ, Lei XZ, et al. Goal-directed fluid optimiza-tion based on stroke volume variation and cardiac index duringone-lung ventilation in patients undergoing thoracoscopy lobec-tomy operations: a pilot study. Clinics (Sao Paulo). 2013;68:1065---70.
32. Forget P, Lois F, Kartheuser A, et al. The concept of titrationcan be transposed to fluid management, but does is changethe volumes? Randomised trial on pleth variability index dur-ing fast-track colonic surgery. Curr Clin Pharmacol. 2013;8:110---4.
33. Zakhaleva J, Tam J, Denoya PI, et al. The impact of intravenousfluid administration on complication rates in bowel surgerywithin an enhanced recovery protocol: a randomized controlledtrial. Colorectal Dis. 2013;15:892---9.
34. Srinivasa S, Taylor MH, Singh PP, et al. Randomized clini-cal trial of goal-directed fluid therapy within an enhancedrecovery protocol for elective colectomy. Br J Surg. 2013;100:66---74.
35. McKenny M, Conroy P, Wong A, et al. A randomised prospec-tive trial of intra-operative oesophageal Doppler-guided fluidadministration in major gynaecological surgery. Anaesthesia.2013;68:1224---31.
36. Zheng H, Guo H, Ye JR, et al. Goal-directed fluid ther-apy in gastrointestinal surgery in older coronary heartdisease patients: randomized trial. World J Surg. 2013;37:2820---9.
37. Peng K, Li J, Cheng H, et al. Goal-directed fluid therapy based onstroke volume variations improves fluid management and gas-trointestinal perfusion in patients undergoing major orthopedicsurgery. Med Princ Pract. 2014 [Epub ahead of print].
38. Meng L, Tran NP, Alexander BS, et al. The impact of phenyle-phrine, ephedrine, and increased preload on third-generationVigileo---FloTrac and esophageal Doppler cardiac output mea-surements. Anesth Analg. 2011;113:751---7.
39. Feldheiser A, Hunsicker O, Krebbel H, et al. OesophagealDoppler and calibrated pulse contour analysis are not inter-changeable within a goal-directed haemodynamic algorithm in
527
major gynaecological surgery. Br J Anaesth. 2014 [Epub aheadof print].
0. Chikhani M, Moppett IK. Minimally invasive cardiac out-put monitoring: what evidence do we need. Br J Anaesth.2011;106:451---3.
1. Critchley LA, Lee A, Ho AM. A critical review of the ability ofcontinuous cardiac output monitors to measure trends in car-diac output. Anesth Analg. 2010;111:1180---92.
3. Jhanji S, Vivian-Smith A, Lucena-Amaro S, et al. Haemodynamicoptimization improves tissue microvascular flow and oxygen-ation after major surgery: a randomised controlled trial. CritCare. 2010;14:R151.
4. Della Rocca G, Ventrugno L, Tripi G, et al. Liberal orrestricted fluid administration: are we ready for a proposal of arestricted intraoperative approach? BMC Anesthesiol. 2014;14:62.
5. Kristensen SD, Knuuti J, Saraste A, et al. 2014 ESC/ESA Guide-lines on non-cardiac surgery: cardiovascular assessment andmanagement: The Joint Task Force on non-cardiac surgery:cardiovascular assessment and management of the EuropeanSociety of Cardiology (ESC) and the European Society of Anaes-thesiology (ESA). Eur Heart J. 2014;35:2383---431.
6. Collaborative Study Group on Perioperative ScvO2 Monitoring.Multicentre study on peri- and postoperative central venousoxygen saturation in high-risk surgical patients. Crit Care.2006;10:R158.
7. Arulkumaran N, Corredor C, Hamilton MA, et al. Car-diac complications associated with goal-directed therapy inhigh-risk surgical patients: a meta-analysis. Br J Anaesth.2014;112:648---59.
8. Hamilton MA, Cecconi M, Rhodes A. A systematic reviewand meta-analysis on the use of preemptive hemodynamicintervention to improve postoperative outcomes in moder-ate and high-risk surgical patients. Anesth Analg. 2011;112:1392---402.
9. Cecconi M, Corredor C, Arulkumaran N, et al. Clinical review:goal-directed therapy --- what is the evidence in surgicalpatients? The effect on different risk groups. Crit Care.2013;17:209---23.
0. Rhodes A, Cecconi M, Hamilton M, et al. Grounds goal-directedtherapy in high-risk surgical patients: a 15-year follow-up study.Intensive Care Med. 2010;36:1327---32.
1. Grocott MPW, Dushianthan A, Hamilton MA, et al. Periopera-tive increase in global blood flow to explicit defined goals andoutcomes after surgery: a Cochrane systematic review. Br JAnaesth. 2013;111:535---48.
2. Sandham JD, Hull RD, Brant RF, et al. A randomized, controlledtrial of the use of pulmonary---artery catheters in high-risk sur-gical patients. N Engl J Med. 2003;348:5---14.
3. Manecke G, Asemota A, Michard F. Tackling the economicburden of postsurgical complications: would perioperative goal-directed fluid therapy help? Crit Care. 2014;18:566.
4. Bartha E, Davidson T, Hommel A, et al. Cost-effectiveness anal-ysis of goal-directed hemodynamic treatment of elderly hipfracture patients: before clinical research starts. Anesthesiol-ogy. 2012;117.
5. Mowatt G, Houston G, Hernandez R, et al. Systematic review ofthe clinical effectiveness and cost-effectiveness of oesophagealDoppler monitoring in critically ill and high-risk surgical
ment Program: esophageal Doppler ultrasound-based cardiacoutput monitoring for real-time therapeutic management
of hospitalized patients; 2007. Available from: http://www.cms.gov/medicare-coveragedatabase/details/technology-
assessments-details.aspx?TAId=45
7. Jammer I, Wickboldt N, Sander M, et al. Standards for defini-tions and use of outcome measures for clinical effectivenessresearch in perioperative medicine: European Perioperative
5
J. Ripollés et al.
Clinical Outcome (EPCO) definitions: a statement from the ESA-ESICM joint taskforce on perioperative outcome measures. Eur
J Anaesthesiol. 2014 [Epub ahead of print].
8. Corcoran T, Rhodes JE, Clarke S, et al. Perioperative fluid man-agement strategies in major surgery: a stratified meta-analysis.Anesth Analg. 2012;114:640---51.