The Physiologic Implications of Isolated Alpha1 Adrenergic - DePa

13 Pages·2011·2.4 MB·English

Society of Critical Care Anesthesiologists
SectionEditor:MichaelJ.Murray
REVIEW ARTICLE
The Physiologic Implications of Isolated Alpha
1
Adrenergic Stimulation
Robert H. Thiele, MD, Edward C. Nemergut, MD, and Carl Lynch III, MD, PhD
Phenylephrine and methoxamine are direct-acting, predominantly (cid:1) adrenergic receptor (AR)
1
agonists.Tobetterunderstandtheirphysiologiceffects,wescreened463articlesonthebasisof
PubMedsearchesof“methoxamine”and“phenylephrine”(limitedtohuman,randomizedstudies
published in English), as well as citations found therein. Relevant articles, as well as those
discoveredinthepeer-reviewprocess,wereincorporatedintothisreview.Bothmethoxamineand
phenylephrine increase cardiac afterload via several mechanisms, including increased vascular
resistance, decreased vascular compliance, and disadvantageous alterations in the pressure
waveforms produced by the pulsatile heart. Although pure (cid:1) agonists increase arterial blood
pressure, neither animal nor human studies have ever shown1 pure (cid:1)-agonism to produce a
1
favorable change in myocardial energetics because of the resultant increase in myocardial
workload. Furthermore, the cost of increased blood pressure after pure (cid:1)-agonism is almost
1
invariably decreased cardiac output, likely due to increases in venous resistance. The venous
system contains (cid:1) ARs, and though stimulation of (cid:1) ARs decreases capacitance and may
1 1
transiently increase venous return, this gain may be offset by changes in afterload, venous
compliance, and venous resistance. Data on the effects of (cid:1) stimulation in the central nervous
1
systemshowconflictingchanges,whileexperimentalanimaldatasuggestthatrenalbloodflowis
reducedby(cid:1)-agonists,andbothanimalandhumandatasuggestthatgastrointestinalperfusion
maybereduc1edby(cid:1) tone. (AnesthAnalg2011;113:284–96)
1
Phenylephrine is a direct-acting, predominantly (cid:1)- useinmodernclinicalpractice.9Becausetheirmechanismsof
1
adrenergic receptor ((cid:1)-AR) agonist synthetically de- action are similar (predominantly (cid:1) agonism) and because
1 1
rived from epinephrine, structurally different only in some physiologic studies of methoxamine were never re-
itslackofanhydroxylgroupatposition4onitsbenzenering.1 peatedwithphenylephrine,thisreviewofthephysiologyand
Itexertsmildpositiveionotropiceffectswhenadministeredat experimental data of (cid:1)-AR agonism will include data on
1
highconcentrations.2–4 Methoxamineisalong-acting(cid:1)-AR both.
1
agonist,syntheticallyderivedfromepinephrinebutdifferent But why, some 60 years after phenylephrine was intro-
inthenumberandlocationofsidegroups(includingO-CH ducedintoclinicalpractice,10didwechoosetoreviewthe
3
groupsatboththeC andC locationsofthebenzylring,as physiologiceffectsofthesedrugs?
2 5
wellasaCH groupattachedtothe(cid:1)carbon)1(Fig.1). First, practicing physicians are now squarely in the
3
Phenylephrineandmethoxaminehavesimilareffectson midst of a movement towards “goal-directed therapy.”11
vascularresistance,althoughphenylephrineis5to10times Thethoughtprocessunderpinninggoal-directedtherapyis
more potent5,6 with a 3-fold higher maximum attainable that, rather than using an intervention to treat an “abnor-
response.5 Phenylephrine is also shorter acting; a single mal” number, one should think critically about (a) which
dose of phenylephrine generally lasts (cid:1)20 minutes,7 physiologicvariablesaremostimportant(ifthisisknown);
whereas a single dose of IV methoxamine can exert its (b) how a particular intervention affects these variables,
effectsforaslongas60minutes.7,8 even if they cannot be directly measured; and (c) whether
Many early studies of (cid:1)-AR agonism were conducted manipulatingthesevariablescanchangeoutcomes.
1
with methoxamine. Methoxamine’s relatively long duration Favoringlessimportantbutimmediatelymeasurablevari-
ofactionandconsequentlackoftitratability,combinedwith ables,suchasmeanarterialbloodpressure(MAP),overmore
theabilitytovariablyinfusephenylephrine,haveobviatedits important but less measurable variables, such as tissue oxy-
gen delivery (DO ), is the result of “tangible bias,” our
2
From the Department of Anesthesiology, University of Virginia Health tendencytofavorwhatwecanseeandunderstandoverwhat
System,Charlottesville,Virginia. wecannot.Despitethepracticalitiesthatprecludetheroutine
AcceptedforpublicationJanuary14,2011. measurement of regional blood flow, changes in global and
Funding:Departmental. regionalbloodflowshouldbeanticipatedanytimehemody-
Theauthorsdeclarenoconflictofinterest. namics are manipulated, with the goal being adequate DO
2
Reprintswillnotbeavailablefromtheauthors. andnutrientstoorgansofinterest.
AddresscorrespondencetoRobertH.Thiele,MD,DepartmentofAnesthe- Second,andequallyimportant,istheideathatmuchof
siology,UniversityofVirginiaHealthSystem,P.O.Box800710,Charlottes-
medical lore is based on tightly controlled animal experi-
ville,[email protected].
ments that may or may not be applicable to the intact
Copyright©2011InternationalAnesthesiaResearchSociety
DOI:10.1213/ANE.0b013e3182124c0e organism.Althoughitmayneverbepossibletoreproduce
284 www.anesthesia-analgesia.org August2011•Volume113•Number2
ExperimentalAlpha-AgonismData
1
Figure1.Chemicalstructureofepinephrine,norepinephrine,phenylephrine,andmethoxamine.
thesestudiesinhumansinvivo,advancesinmathematical The 6-Compartment Model
modeling and computational biology (as exhibited by This review assumes that Magder et al.’s model of the
Magderetal.12)havemadeitpossibletocriticallyassessthe cardiovascular system closely approximates reality.12
validityoftheselong-heldphysiologictruths. Specifically, Magder et al. considered the cardiovascular
Thus, we critically examined the use of both methox- system to be a “6-compartment” system (right heart, pul-
amineandphenylephrine.Ourinitialsearchwasconducted monary arteries, pulmonary veins, left heart, systemic
in PubMed, using the word methoxamine and limiting our- arteries,systemicveins)thatformsanin-seriesclosedloop.
selvestorandomized,controlled,humantrialspublishedin At any point in time, all 6 compartments have a given
English.Thisresultedin28articles,theabstractsofwhich pressureandvolume.Additionally,the4vascularcompart-
werereviewedforrelevance.Articlesdescribingthehemo- ments can be further described by their compliance
dynamic effects of methoxamine were examined in detail. (dV/dP), whereas the 2 cardiac compartments can be
Wethenrepeatedoursearchusingphenylephrineasourkey further described by the rate at which they move volume
word (same limits), resulting in 435 articles that were (dV/dt). In accordance with the law of conservation of
similarly reviewed, acquired, and if applicable, read. Ar- mass, volume that is added to (or removed from) one
ticles brought to our attention during the review process compartment must be removed from (or added to) an
wereincludedaswell. adjacentcompartment.
Before proceeding, 3 prerequisite axioms, on which the The implications of the 6-compartment model are that
utility of this review are based, must be established. First, bloodflowthroughthecardiovascularsystemisnotmerely
thisreviewassumesthatDO iscriticalforthesurvivalof a function of just how much pressure the left heart gener-
2
cells, organs, and whole organisms. Second, that for all ates and how much resistance the systemic arterioles
organsandorganisms,thereisanoptimalDO ,whichcan provide, but the result of a much greater number of
2
be organ specific. Third, and perhaps most relevant, that interacting, codependent variables. The details of these
despite a lack of data regarding the optimal DO in most interactions will be discussed in the appropriate sections
2
pathophysiologic states, anesthesiologists and intensivists below.
will develop their own upper and lower limits of accept-
Regional Versus Global Blood Flow
ability and, in general, will attempt to manipulate hemo-
Whentheeffectsofanyinterventiononallcomponentsof
dynamicstomaintainDO inthisrange.
2
thecardiovascularsystemareconsidered,itisimportantto
distinguish global from regional changes. For example,
manydrugstudiesfocusoncardiacoutput(CO),presum-
A COMPREHENSIVE APPROACH TO BLOOD FLOW
ablybecauseCOiseasytomeasureandchangesinglobal
Rationale
flow are thought to result in similar changes in regional
When studying the effects of any vasoactive drug on the
bloodflow;however,drugsmayaffectglobalandregional
cardiovascularsystem,itisnotsufficienttofocus(asmany
blood flow inversely (e.g., increasing CO while decre-
textbooksandarticlesdo)onanarbitrarilychosensubsetof
asing blood flow to the kidneys). Thus, simply measur-
the system (e.g., systemic arteries), and in the process
ing CO changes may not give adequate insight into the
ignoreothercomponents(e.g.,centralveins,rightventricle
utilityofaparticulardrug,andinsomeinstancesmaybe
[RV]) that also affect the system as a whole as well as the
misleading.
region of interest. On the other hand, isolating the indi-
vidual components of the cardiovascular system, all of
THE MATHEMATICS AND PHYSICS OF FLOW
whichinteractwitheachotherinvivo,isusefulfromboth
Ohm’s Law and Mathematical Notation
experimental(suchexperimentsareeasiertoconduct)and
Textbookscommonlycomparebloodflowinthecardiovas-
educational (simplified concepts are easier to understand)
cular system with the flow of electrical charge (dQ/dt)
standpoints.
across a voltage differential (E) in a single resistor (R)
Thus, although many of the experiments cited in this
circuit(Ohm’slaw):
review were either conducted in isolated experimental
modelsorfocusedonasubsetofthecardiovascularsystem, I(cid:2)E/r(cid:2)dQ/dt (1)
theirrelevancedependsheavilyonone’sabilitytointegrate
these findings into a more global model of the cardiovas- Often, when this relationship is applied to blood flow, it is
cularsystem. rearrangedasfollows(commonrearrangementofOhm’slaw):
August2011•Volume113•Number2 www.anesthesia-analgesia.org 285
REVIEWARTICLE
MAP(cid:2)CO(cid:3)systemicvascularresistance(SVR) (2) and are of great importance. But Ohm’s law, which de-
scribesthemovementofelectricalcurrentthroughacircuit
Although “mathematically correct,” this equation can be when a constant voltage is applied, and is often used to
misleading. Convention dictates that the dependent vari- model the human cardiovascular system, is not perfect. It
ables of an equation are always written on the left-hand does not adequately describe the movement of electrical
side of an equals sign and that the independent variables currentincasesinwhichthevoltagedifferentialvarieswith
areontheright.Justastheplacementofasingledigithas time(e.g.,alternatingcurrent).Whentheelectricalpotential
meaning when describing a multidigit number (123 does changes, the resulting current changes as well; how much
not equal 213), the location of variables in a mathematical sodependsonboththeresistanceandthe“capacitance”of
equationisdesignedtoconveyimportantinformation.One thecircuit.
is led to infer from Equation 2 that changes in CO affect
An Introduction to Capacitance
MAP. Blood flow can most easily (and correctly) be de-
(and Compliance)
scribedasfollows:
Capacitance is the ability to store potential energy. In
CO(cid:2)(1/SVR)(cid:3)MAP(cid:2)(1/RVR (cid:4)1/RVR (cid:4)1/RVR electrical engineering, electrical capacitance (CAP,elec) is
1 2 3 defined as the amount of charge (Q) stored given an
(cid:4)…(cid:4)1/RVR )(cid:3)MAP (3) appliedvoltagepotential(E):
n
C (cid:2)Q/E (4)
(analogy of Ohm’s law for global blood flow). Equation 3 AP,elec
introduces2importantpoints–first,becausetheindividual
An electrical circuit with a high capacitance will be more
organs that resist blood flow exist in parallel, SVR can be
resistanttorapidchangesinvoltage.Whenappliedvoltage
related to the inverse of all individual regional vascular
is increasing, some of the current that would normally
resistances(RVR)inaccordancewithKirchoff’scircuitlaws
travel through the resistive elements of the circuit instead
(1/R (cid:2)1/R (cid:3)1/R (cid:3)…(cid:3)1/R ).Second,becausethe
total 1 2 n accumulates in the capacitive elements of the circuit,
body regulates blood pressure (and not CO) primarily
“smoothingout”fluctuationsincurrent.Thisphenomenon
throughalterationsinvascularresistancebroughtaboutby
isoftenreferredtoasdampening.
changesinsympathetictone,MAPcanremainstableover
If one considers electrical capacitance, movement of
a wide variety of hemodynamic states in which CO is
current through an electrical circuit can no longer be
inversely related to vascular resistance. Thus, CO is a
describedinsimplelinearterms.Adifferentialequationis
complexfunctionofglobaleffortstoregulateMAPdespite
required:
multiple regional systems that alter RVR in an attempt to
autoregulate. I(cid:2)dQ/dt(cid:2)[E(t)(cid:5)Q/C ]/R (5)
AP,elec
Still, the common rearrangement of Ohm’s law (Equa-
tion 2) does have practical utility in physiologic states in which is flow of current through an RC series circuit,15
whichCOisfixed,suchasduringcardiopulmonarybypass whereE(t)isvoltageasafunctionoftime.
or in an autoregulated, healthy cardiovascular system. In Similarly,thedrivingforceofthehumancardiovascular
both instances, pressure can be considered dependent on system is pressure generated by a pulsatile heart, and the
bloodflowif,andonlyif,thepressuregeneratoriscapable vesselsthemselvesactascapacitors.Inthearterialsystems,
of increasing pressure (with attendant increases in energy compliant vessels store mass (blood) and potential energy
consumption)inresponsetoincreasedvascularresistance. (pressure (cid:4) volume) during systole and deliver mass
Indeed, human studies have shown that with massive (blood) and energy (pressure (cid:4) volume) to the human
blood loss, the healthy, intact cardiovascular system will “circuit”duringdiastole.Theendresultisthatthecapillary
regulateMAPbymanipulatingafterloadattheexpenseof beds receive a more constant stream of blood, despite the
CO.13Morerecentstudiesofthehemodynamicresponseto pulsatile nature of the heart. This is referred to as the
trachealintubationhaveconfirmedthatfromawhole-body windkesseleffect,16anditsconceptualdevelopmentisattrib-
perspective, preservation of MAP takes precedence over utedtoOttoFrank.17
CO.14 That said, if pressure generation is fixed or limited, The major systemic arterial capacitance vessels include
blood flow will further diminish as resistance to flow is theaortaandlargearteries,whichexistinparallelwiththe
increased.Itmustbekeptinmindthatevenintheseidealized resistive elements of the vasculature, and because the
situations,ifpressuregenerationweretocease,sotoowould upper and lower body contain capacitance vessels of
theflowofbloodthroughthecardiovascularsystem. different lengths, the systemic vascular system is most
When thinking about the cardiovascular system as a appropriately modeled as a 2-capacitor circuit.18 A major
whole,particularlyinnonidealizedsituations(e.g.,cardio- differencebetweenthecardiovascularsystemandananalo-
vascularfailure,lossofautoregulatoryreflexes)orregional gous electrical circuit is that the cardiovascular system
bloodflow,onemustappreciatethatresistancetoflowand storesmass(volume),notcharge.
pressure generated by ventricular contraction make equal Volume exists in 2 states: hemodynamically inactive
contributions to the determination of both regional and “unstressed” volume (defined as the amount of blood
globalbloodflow(andthus,DO ). present in the venous system where venous transmural
2
Thusmathematicalequations—whichdescribephysical pressure is 0 [approximately 70% of total venous blood
reality in terms that can be quantified, understood, and volume]), and hemodynamically active “stressed” volume
applied—profoundly influence our conception of reality, (defined as the difference between total venous blood
286 www.anesthesia-analgesia.org ANESTHESIA&ANALGESIA
ExperimentalAlpha-AgonismData
1
volumeandunstressedvolume).19Fromthestandpointof compliancechangesonventricularefficiencyinadultdogs,
measuring flow as a function of changes in pressure, by altering the compliance of the abdominal aorta with a
volume, compliance, and resistance, it is only the stressed rigid plastic graft. The plastic conduit reduced arterial
volume that matters. Importantly, hemodynamic changes complianceby60%–80%,resultinginan11.9%increasein
(e.g.,vasoconstriction)canconvert“unstressed”volumeto MAP despite a 10.5% decrease in CO (calculated SVR
“stressed” volume as a compensatory means,20 without increasedby20%).Despitetheroughlyequaloppositional
necessarily changing vascular compliance.19 There is no changesinpressureandvolume,implantationoftherigid
electricalequivalentfor“unstressedvolume.” graft resulted in a 32% increase in mVO and a 32%
2
Thus,itisusefultothinkofthevasculaturenotonlyin decrease in ventricular efficiency, reflecting the relatively
terms of the amount of volume stored at a given pressure substantial contribution of pressure work (energy inten-
(defined as vascular capacitance, Equation 6), but also in sive, in comparison with volume work) in the determina-
termsofvascularcompliance,definedasachangeinvolume tionofmyocardialoxygenneeds.Interestingly,theincrease
that results from a change in pressure (Equation 7).19 in myocardial consumption was directly proportional to
Vascularcomplianceisinverselyrelatedtovesselstiffness the increase in pressure volume area (PVA; see PUMP
((cid:6)).Vascularcapacitanceisdefinedas WORK AND VENTRICULAR EFFICIENCY section, be-
low);however,becausetheauthorsdidnotincreasevascu-
C (cid:2)V/P (6) larresistanceindependentlyofcompliance,itisimpossible
AP,vasc
to know for certain whether an isolated decrease in com-
andvascularcomplianceas pliancealsoworsensventricularefficiency.
The venous systems, by contrast, are much more com-
COM,vasc(cid:2)(cid:1)V/(cid:1)P(cid:2)1/(cid:6). (7) pliant than are the arterial systems, and, by comparison,
storemorevolume(70%oftotalblood25)andlesspotential
Notethatwhileconventiondictatesthattheabilitytostore
energy. Thus, although arterial compliance primarily af-
electrical energy in the form of charge is referred to as
fects the arterial waveform and afterload, venous compli-
electricalcapacitance,theabilitytostoreenergyintheformof
ance impacts the cardiovascular system through several
pressureismoreappropriatelydescribedbyvascularcom-
differentmechanisms,allofwhicharebasedonanunder-
pliance(becauseitappropriatelyneglectsthe“unstressed”
standing of the venous function curve and the concept of
volumethatisenergeticallyinactive),andforthepurposes
meancirculatoryfillingpressure(MCFP).
of this analogy, electrical capacitance (Q/E) and vascular
compliance ((cid:5)V/(cid:5)P) can be considered interchangeable,
Venous Function Curve
althoughtheywillbeabbreviatedasC andC
(or1/(cid:6)),respectively. AP,elec OM,vasc The venous function curve, as originally described by
Guyton,26 describes the effect of changes in right atrial
As with the electrical circuit driven by an oscillating
pressure (RAP) on venous return: as RAP is increased,
voltagepotential,bloodflowthroughacompliantvesselis
venousreturn(andCO)decreases,andultimatelybecomes
not simply a function of “resistance,” but must also con-
zero as RAP approaches MCFP. Similarly, as RAP is
sider arterial compliance and the volume of blood con-
decreased,venousreturnincreases,reachingamaximumat
tainedinthevesselatthatmoment:
thepointatwhichveinscollapse(atmosphericpressure,or
Q(cid:2)dV/dt(cid:2)P/R(cid:4)dP/dt(cid:3)(C )(cid:2)P/R(cid:4)dP/dt higherinthesettingofpositiveend-expiratorypressure).
OM,vasc
Guyton’svenousfunctioncurvescanalsobeunderstood
(cid:3)(cid:6)1/(cid:6)(cid:7) working backwards, i.e., starting from the position of no
CO.19 When CO is zero, blood pressure in the pulmonary
Equation 8 describes blood flow through a compliant andsystemicarterialandvenoussystemswillbeequal;this
vessel,orwhatisknownasthe“TwoElementWindkessel isreferredtoastheMCFP,andisafunctionoftotalblood
Model.”Modelsincorporatingupto4elementshavebeen volume as well as arterial and venous compliance in both
developed, and increasingly approximate experimental the pulmonary and systemic vasculatures. As the left
observations.21,22 ventricle (LV) and right ventrical (RV) begin to pump
Therefore, although hemodynamic data can be used to blood,arterialpressureswillincreaseaboveMCFP.Venous
calculate systemic vascular “resistance,” this value is a pressureswilldecreasebelowMCFP,therebyestablishinga
combination of resistance to blood flow when a constant pressure gradient (required for blood flow) across the
driving force is applied, the instantaneous directional pulmonaryandsystemicvasculatures,andshiftingvolume
changeinpressure(dP(t)/dt),andcompliance,allofwhich from the venous compartments to the arterial compart-
affect blood flow—their relative contribution changes de- ments. As CO increases further, arterial pressures will
pendingonthehemodynamicstate.Totrulyappreciatethe necessarily continue to increase, venous pressures will
oscillatory component of afterload, one must decompose continuetodecrease,andincreasingvolumewillbeshifted
boththepressureandflowwaveformsintotheirharmonic towardsthearterialcompartments.
components,theendresultofwhichisthebipartiteconcept Foragivenbloodvolumeandvascularcompliance,the
of vascular impedance (abbreviated Z, comprising modu- cardiovascularsystemcanexistatanystatethatliesonits
lusandphase),thedetailsofwhicharebeyondthescopeof venous function curve. Traditional teaching espouses that
thisreviewbutarethoroughlydetailedelsewhere.23 this state depends on where the venous function curve
Arterial compliance may also affect myocardial oxygen intersects the CO curve (because, at steady state, total
consumption (mVO ). Kelly et al.24 studied the effects of venousbloodreturnmustequalCO).TheCOandvenous
2
August2011•Volume113•Number2 www.anesthesia-analgesia.org 287
REVIEWARTICLE
function curves are often plotted together to make this venous return and P .29 Thus, Guyton’s initial experi-
RA
calculation, which can be misleading because although ments, which established that decreased resistance to ve-
RAP is closely related to right ventricular preload, the nousreturn(andtheresultantincreaseinvenouspressure
ability of RAP to reflect LV preload is dependent on the gradients)leadtoincreasedvenousreturninanexperimen-
physiologicstateofthepulmonaryvasculartreeaswellas talsystem,wereunabletoattributethesechangesinvenous
onthatoftheleftventricle. returntochangesinRAP.
ThisaddedcomplexitycomplicatesGuyton’sclassicteach- To better understand the impact of changes in venous
ingandsuggeststhatGuyton’svenousfunctioncurves,which resistance on venous return and CO, Guyton compared
were derived in experimental animal models, may not be selectiveincreasesineitherarterialorvenousresistancein
directly applicable in vivo.27–29 Regardless, Guyton’s major anesthetized dogs (right-heart bypass preparation).36 Vas-
premise, that venous return is just as important as cardiac cular reflexes were abolished using spinal anesthesia. Ar-
outflow in determining steady-state blood flow, still holds terialresistancewasthenincreasedbyinjectingglassbeads
true,andhasmajorimplications,asnotedbelow. into the aorta, and venous resistance was increased by
Despite the controversy surrounding the shape of the tighteninginflatablecuffsimplantedaroundthevenacava.
venous function curve in vivo, it is generally thought that Interestingly, doubling SVR via the injection of glass
venous compliance impacts CO through 3 mechanisms. beads led to a 15% decrease in CO and a 75% increase in
First, because venous pressures decrease with increasing blood pressure, whereas doubling SVR via constriction of
CObutcannotdecreasebelowthepointofvenouscollapse, thevenacavareducedCOby65%,presumablybyseques-
the maximum attainable CO is impacted by both blood teringbloodinthevenoussystem,andthusdeprivingthe
volumeandvenouscompliance(or,aswasmoreprecisely ventricles of the preload needed to maintain CO (postu-
described by Levy in 1979, the ratio of venous to arterial lated, but not proven, by Guyton in this article36). Indeed,
compliance30).Second,steady-stateCOisatleastpartially no amount of isolated arterial resistance (even a 500%
dependent on the configuration of the venous function increaseinSVR)coulddecreaseCOtotheextentachieved
curve,whichisafunctionofvenouscompliance.Third,by by a relatively modest increase in venous resistance (50%
redistributing blood volume towards (or away from) the increaseinSVRviaconstrictionofvenacava).
central vessels, atria, and ventricles,31 changes in venous Clearly, the canine left ventricles in this experiment
resistance and compliance can profoundly affect ventricu- were better able to maintain stable CO despite increased
larenddiastolicvolume,andthusCO.32 arterial resistance, in comparison with increased venous
resistance. This was likely due to differences in vascular
Relative Importance of Arterial and Venous distensibility.Arteries,whicharerelativelynondistensible,
Vascular Resistance areunabletoremovesignificantvolumefromthecirculation
BecausethemajorityofSVRisprovidedbythearterioles,it despite increased resistance to flow. Increased arterial resis-
may seem counterintuitive that venous resistance and tance does not markedly decrease cardiac filling unless the
compliancecouldsignificantlyimpactCO.Indeed,51years heart cannot maintain constant CO despite increased after-
afterGuytonpublishedhisvenousfunctiondata,theutility load.Bycontrast,veins,whicharehighlydistensible(compli-
ofhismodelswasstillbeingdebatedintheliterature.33,34 ant)andstoreapproximately70%oftotalbloodvolume,can
The debate about whether Guyton’s models are appli- almost immediately sequester relatively large amounts of
cableinvivoisamisunderstandingofhisexperimentsand blooddespiteincreasedresistancetoflow,essentiallyrobbing
his conclusions. Guyton’s venous function curves were boththeleftandrightventriclesofpreload.
developed by cannulating the right atrium and aorta,26 Thus, in 1958, Guyton’s experimental data began to
bypassing both ventricles and the lungs. A mechanical provide proof that the all-encompassing concept of “sys-
pump (connected in series with a piece of collapsible temicvascularresistance”andtacitassumptionthatSVRis
tubing, connected proximally [i.e., a “Starling resistor”]) due to arterial tone may not be physiologically relevant,
was placed in between.33,35 As the height of the Starling becausethelocationofvascularresistanceiscriticalandis
“resistor” was changed, inflow to the pump was variably notaccountedforbysimplydividingpressuregradientsby
throttled, resulting in changes in both RAP and venous CO. This idea was further refined by separating the regu-
return. Technically, the independent variable in this ar- lation of venous flow into changes in compliance and
rangementwasflow33,35(determinedbytheStarlingresis- resistance,whichcanoccurindependentlyofoneanother.19
tor), not RAP (although if one accepts that blood moves
down a pressure gradient, the origin of the pressure Pressure Wave Reflections
gradient is not relevant). Maximal output was limited by Further complicating hemodynamic predictions is the
the point at which the tubing collapsed (0 mmHg). This 3-dimensionalshapeofthecardiovascularsystem.Whereas
arrangement is not necessarily what happens in live ani- electricalcircuitsaremadeofwirethatrarelyvaryinsize,
mals with a closed chest and interacting pulmonary and shape, or composition, the “wires” of the human cardio-
systemiccirculations,pumpsthatdependonpreload,vary- vascularsystemvarygreatly,bothintermsoftheirstiffness
ing thoracic pressures, and central venous systems that andshapeaswellastheirbranchpoints.
may, in some instances, remain patent even at subatmo- As a pressure wave travels down the vascular tree, it
spheric pressures. Studies of closed-chest humans after meets additional resistance at places where the vascular
cardiac surgery have produced mixed results. Some au- treebranchesorwherevascularimpedance(acombination
thors have suggested that Guyton’s relationship holds,27 of resistance and compliance) changes quickly.37 At these
and others have failed to elucidate a relationship between branchpointsandchangesinimpedance,partofthepressure
288 www.anesthesia-analgesia.org ANESTHESIA&ANALGESIA
ExperimentalAlpha-AgonismData
1
waveisreflectedbacktowardstheheart(muchasultrasound mVO by 77%. The arterial–coronary O saturation differ-
2 2
waves emitted by an echocardiography probe are partially ence was unaffected (64% in both instances), however.
reflectedbytissueinterfaces),thusreducingthedrivingforce Fifteen percent of patients receiving methoxamine dis-
forforwardbloodflow.Normally,thesepressurewavesreach playedSTsegmentchanges.
the left ventricle during diastole, where they can either Antonopoulos et al. administered phenylephrine (80
contribute to coronary perfusion38 or be absorbed by the (cid:7)g/min)to41hemodynamicallystablepatientswithdocu-
closedaorticvalve.Incasesinwhichvascularcomplianceis mented coronary artery disease, increasing MAP by 30%
significantly reduced (e.g., with atherosclerosis or aging), abovebaseline.Sixtysecondsafterachievinganincreasein
these reflected pressure waves travel more quickly, and can blood pressure, 2 mCurie (mCi) of thallium (Tl) were
arrive back at the left ventricle before aortic valve closure, injected and Tl scintigraphy was performed 2 and 240
decreasingthespeedofmyocyteshortening.39 minutesafterTlinjection.Scintigraphyafterphenylephrine
Therelativecontributionsofresistance,compliance,and infusion revealed 152 defects (average 14% of evaluated
pressurewavestoventricularafterloadaredifficulttotease segments),andthesizeofthedefectwasdirectlyrelatedto
apart, because most interventions (e.g., vasoconstrictors) thenumberofdiseasedvessels.Theauthorsconcludedthat
affect all 3 variables simultaneously.37 In addition to in- bloodpressureincreaseaccompanyingphenylephrinepro-
creasing arteriolar resistance, vasoconstrictors (including ducedasignificantimpairmentofmyocardialperfusion.47
phenylephrine6) appear to augment these arterial tree Unfortunately, the authors did not include data on myo-
reflectionsbydecreasingcomplianceandacceleratingpres- cardialperfusionbeforeadministrationofphenylephrine.
surewaveconduction.40 Taken together, these studies suggest that although
increasedpressurecanmaintainglobalperfusion,depend-
SUPPLY AND DEMAND—THE ECONOMICS OF ingonhowachieved,itmaystillleadtoamaldistribution
CARDIAC PERFORMANCE ofregionalmyocardialbloodflow.
Oxygen delivery (DO ) is preeminently important in the
2
contextofoxygenconsumption(VO ).Thus,whenconsid- Demand (Afterload)
2
ering the hemodynamic effects of drugs on the heart, one Afterload ((cid:8)) is a measure of the forces against which the
musthaveanappreciationforbothcoronaryarteryperfu- heartmustworktogenerateagivenCO.Initspurestform,
sion(supply)andmVO (demand). it is defined as the forces opposed to LV fiber shortening
2
(i.e.,LVwallstress).48SVRisconsideredbymostpractitio-
Supply (Coronary Artery Perfusion) nersasequivalenttoafterload.48However,asnotedabove,
Heyndrickx et al. studied the effects of methoxamine on SVRisanoversimplifiedquantificationofhemodynamics.
coronarybloodflowinhealthy,consciousdogs,withboth Unfortunately, true afterload cannot be readily mea-
pacedandspontaneouslybeatinghearts.Inthepacedhearts, sured except in the experimental setting. It can be best
methoxamineincreasedcoronarybloodflow,whereasinthe approximatedbycalculatingcircumferentialwallstress(S),
spontaneouslybeatinghearts,coronarybloodflowdecreased inavariationofLaplace’slawknownasLame’sequation.37
8% (despite increasing MAP by 55%, and decreasing heart AlthoughmoredifficulttocalculatethanisSVR,itgivesa
rateby13beatsperminute[bpm]).41WoodmanandVatner42 more accurate indication of cardiac energy expenditure,
gavephenylephrine(0.5and1(cid:7)g/kg/min)tounanesthetized andismorespecificallyproportionaltomVO .49
2
dogs autonomically blocked with hexamethonium,
propranolol, and atropine and found that phenylephrine S(cid:2)Pr/h (8)
increased MAP but had no effect on coronary blood flow.
Crystaletal.43administeredphenylephrine(2.8(cid:7)g/kg/min) where S (cid:2) wall stress, P (cid:2) pressure, r (cid:2) radius, and h (cid:2)
to anesthetized dogs and found that although myocardial thickness.AnimportantimplicationofLame’sequationisthat
blood flow increased by 60%, mVO increased by 61%. afterload is not simply a function of resistance, compliance,
2
Neither arterial–coronary sinus oxygen difference, coronary andwavereflections,butisalsodependentonthegeometry
sinusPo ,orcoronarysinussaturationchangedsignificantly. oftheleftventricleitself.Duringsystoletheventricularwall
2
Milleretal.studiedtheeffectsofphenylephrineoncoro- thickensasitsradiuscontracts,whichreducesLVwallstress.
narybloodflowinhumans,byadministeringnitroglycerinto In the latter half of systole, afterload is reduced simply
17pacedpatientsundergoingcardiacdiagnosticcatheteriza- becauseLVradiusissmallerandwallthicknessisgreater.All
tion(MAPdecreasedbyanaverageof10.5mmHg),andthen otherthingsbeingequal,aheartoperatingwithmoreshort-
randomizingthemtophenylephrine(50to90(cid:7)g/min)versus enedmyocytes(e.g.,asmayoccurafter (cid:9)adrenergicstimu-
no intervention. At 10 minutes postnitroglycerin, coronary lation)willworkagainstless“afterload,”becauseitwillspend
sinus blood flow was significantly higher in the phenyleph- proportionately more time in a favorable (smaller radius)
rinegroup.44Indeed,studiesof(cid:1) receptordensityinhumans geometricalconfiguration.
1
have confirmed that the coronary arteries contain (cid:1) recep- Lang et al. studied the effects of methoxamine, nitro-
1
tors, although the amount (2.1 fmol/mg protein) is signifi- prusside, norepinephrine, and dobutamine on 8 anesthe-
cantly less than that found in the other, similarly sized tized,intubated,andcatheterized(leftandrightheart)but
arteries,suchasthemammaries(6.0fmol/mgprotein).45 otherwise healthy dogs.48 Using Grossman’s previously
Loeb et al.46 studied the effects of methoxamine (2 validatedmethod49tocalculateafterload,Langetal.found
mg/min)in20patientswithstable,ischemicheartdisease thatSVRisanalmost2-foldunderestimateofLVwallstress
(meanMAP90mmHgbeforeintervention),findingthatit after administration of methoxamine (SVR increased 48%,
increased coronary sinus flow by 82% but also increased whereasLVwallstressincreased86%).Norepinephrineled
August2011•Volume113•Number2 www.anesthesia-analgesia.org 289
REVIEWARTICLE
Figure2.Effectofpharmacologicagentsonbloodpressure,cardiac
output, and SVR. Effect of nitroprusside, methoxamine (pure
(cid:1)-agonist), dobutamine, and norepinephrine on aortic pressure,
cardiac output, and systemic vascular resistance in healthy dogs Figure4.Effectofpharmacologicagentsonleftventricularfunction.
(redrawn from Lang et al.,48 with written permission from Wolters Effectsofnitroprusside,methoxamine(pure(cid:1)-agonist),dobutamine,
KluwerHealth).MCFP(cid:2)meancirculatoryfillingpressure;LVEDV(cid:2) andnorepinephrineonleft-ventricularperformance(asmeasuredby
left-ventricularend-diastolicvolume. velocity of left-ventricular fiber shortening, Vcf) in healthy dogs
c
(redrawn from Lang et al.,48 with written permission from Wolters
KluwerHealth).SVR(cid:2)systemicvascularresistance.
produce both pressure work (increasing pressure) and
volumework(movementofvolume),bothofwhichrequire
the expenditure of energy. Mathematically, the amount of
mechanical work done by a pump is represented by the
areainsidethecurveofapressure–volume(PV)loop:
(cid:1)
V2
W(cid:2) (cid:5) Pdv (9)
V1
Figure 3. Effect of pharmacologic agents on SVR and end-systolic Notethatalthoughpressureworkandvolumeworkcanbe
wall stress. Comparative effects of nitroprusside, methoxamine
thought of separately, neither can occur in the absence of
(pure(cid:1)-agonist),dobutamine,andnorepinephrineonsystemicvas-
theother.Increasedpressurewithoutmovingvolumedoes
cular resistance (SVR) and afterload (defined as end-systolic wall
stress, (cid:8) ) in healthy dogs. Note that for norepinephrine, SVR is notresultinwork,nordoesmovingvolumeifpressureis
es
positive, whereas (cid:8) is negative, thus invalidating SVR as an notaccordinglyincreased.
es
indicator of afterload (redrawn from Lang et al.,48 with written In situations in which CO (volume pumped) is desired,
permissionfromWoltersKluwerHealth).CO(cid:2)cardiacoutput.
pressure work is inefficient because increased ventricular
pressuresnecessitatetheformationandreleaseofadditional
myosin–actin cross-bridges (which consumes adenosine
toa21%increaseincalculatedSVR,butmeasuredLVwall triphosphate),butdoesnotincreaseflow.TheshapeofthePV
stress actually decreased by 9% because of increased con- loopcanprovideavisualestimateofvolume-basedventric-
tractility and a subsequent decrease in ventricular size. ularefficiency.Multiplestudiesinhumanshaveshownthat
Thus, in some instances, SVR is not simply off by a phenylephrinereducesCOandincreasesmeasuredvascular
percentage, it changes in the opposite direction of wall resistance,50–52 which heightens the resultant PV loop, and
stress48(Figs.2to4). usuallydecreasesstrokevolume(Fig.5).53,54
Guyton et al.’s experiments on selective vascular resis-
tance further support this view, because changes in SVR Internal Work
bore almost no relationship to changes in either CO or Unlikeaninanimate,mechanicalpump(e.g.,internalcom-
afterload. Indeed, isolated increases in arterial resistance bustionengine)thatrequiresnoenergywhennotfunction-
led to significant increases in aortic blood pressures, ing, the heart requires energy to maintain its cellular
whereas equivalent increases in venous resistance led to integrityevenwhennotcontracting.
almostnochangeinaorticbloodpressure.36 In addition, the heart requires energy expenditure to
generatepressureevenwhennovolumeismoved.Thisis
PUMP WORK AND VENTRICULAR EFFICIENCY referredtoasinternalwork(alsotermedpotentialenergy)and
Pump (Mechanical) Work can be attributed to electrical activation and excitation–
Afterload, which measures force, is related to but distinct contractioncoupling(calciumcycling).55
from work, the application of force over distance. Pumps Sugaetal.notedthatdifferingcombinationsofpressure
work by applying pressure to a displaced volume, and and volume work (which produce an identical amount of
290 www.anesthesia-analgesia.org ANESTHESIA&ANALGESIA
ExperimentalAlpha-AgonismData
1
ofbloodflowismoreimportantthanthepressurerequired
todeliverit.Fromthisvolume-basedventricularefficiency
standpoint,increasedpressureworkandinternalworkare
wastefulandshouldbeminimized.
Phenylephrinedecreasesvolume-basedventriculareffi-
ciencybyshiftingmyocardialmechanicalworkfromvolume
to pressure work, and presumably by increasing internal
work (although PVA has not been specifically studied in
humans,thiscanbesurmisedbyanalyzingtheresultingPV
loops, Fig. 5). A study comparing phenylephrine with epi-
nephrine, norepinephrine, dopamine, dobutamine, and iso-
proterenolinpigletsundergeneralanesthesiasuggestedthat
ofthese6vasoactivedrugs,phenylephrinewastheonlyone
thatdidnotincreasetheCO/PVAratio.62
In instances in which perfusion pressure is deemed
more important than global blood flow (e.g., optimizing
cerebral perfusion pressure), volume work is wasteful,
because it does not necessarily lead to increased blood
Figure5.Effectofphenylephrineonthepressure–volumeloopinthe pressurebutstillrequiresadditionalmechanicalwork(and
setting of human heart failure (adapted from Asanoi et al.,53 with
expenditure of energy). Thus, from a practical standpoint,
writtenpermissionfromWoltersKluwerHealth).
the “efficiency” of the heart depends not only on work
doneperenergyconsumed,butalsoonwhattypeofwork
mechanical work) require different amounts of oxygen
isneededmost(volumeorpressure).
consumption,55,56andthatthesedifferencescouldbeattrib-
uted to increases in internal work required to function at
EFFECT OF ALPHA-ADRENERGIC AGONISTS ON
different physiologic states (Fig. 6A). Unlike the PV loop
INDIVIDUAL ORGAN SYSTEMS
area,totalpressure–volumearea(PVA,Fig.6,B–C)reliably
estimates myocardial VO in an isolated canine heart Cardiovascular System
2
model, as was shown by Suga,57–59 Khalafbeigui,60 and Left heart and CO. Smith et al.63 randomized 60 carotid
Burkhoff61 in a series of experiments conducted over a endarterectomypatientstoanesthesia(MAC1.04)withno
periodof10years. ionotropicdrugsversusdeeperanesthesia(MAC1.43)with
administrationofphenylephrinetomaintainsystolicblood
Ventricular Efficiency pressure within 20% of baseline. Intergroup differences in
Thetermventricularefficiencyhasseveraldefinitions,butis bloodpressurewereinsignificant;however,thephenyleph-
usually thought of as CO per milliliter of oxygen con- rine group was burdened with a 40% increase in LV
sumed.Notethatthisdefinitionassumesthatthequantity end-systolic wall stress as estimated by transesophageal
Figure 6. A,Relationshipbetweenmyocar-
dialoxygenconsumptionandpressure–vol-
umearea[PVA,definedasinternalwork(cid:3)
externalwork]ataconstantlevelofexternal
work. B, Myocardial energetics at state 2
[high-pressure work, low-volume work]. C,
Myocardial energetics at state 3 [high-
volumework,low-pressurework].Allfigures
derivedfromexperimentsusingtheisolated
canineheartmodel.
August2011•Volume113•Number2 www.anesthesia-analgesia.org 291
REVIEWARTICLE
echocardiography, a 160% increase in the incidence of
segmental wall motion or wall thickening abnormalities,
and a 32% reduction in the rate-corrected velocity of
circumferentialfibershortening.
Goertz et al.64,65 compared phenylephrine to norepi-
nephrine in both volunteers and cardiac surgery patients,
bothundergeneralanesthesia.In16volunteers(2(cid:7)g/kgof
phenylephrine vs. 0.1 (cid:7)g/kg of norepinephrine) and 38
cardiacsurgerypatients(2(cid:7)g/kgofphenylephrinevs.0.05
(cid:7)g/kg of norepinephrine), both drugs produced identical
changesinMAP,butphenylephrineproducedsignificantly
higher wall stress (estimated from transesophageal echo-
cardiographicmeasurements)andsignificantlylowerfrac-
tional area change (and presumably, CO) in comparison Figure 7. Effect of escalating doses of methoxamine on cardiac
withnorepinephrine. output and peripheral resistance. Escalating dose response of
methoxamine(long-actingpure(cid:1)agonist)onthehemodynamicstate
Sharrock et al.50 compared phenylephrine (2 to 20
of dogs after administration of spinal anesthesia (dose started at
(cid:7)g/min) with epinephrine (1 to 5 (cid:7)g/min) in 30 patients 0.3mg/kg/h,increasedby0.3mg/kg/hevery15minutes;redrawn
undergoingepiduralanesthesiaandfoundthatdespiteno fromZandbergetal.,73withwrittenpermissionfromWoltersKluwer
differenceinMAP,phenylephrineledtoslowerheartrate Health).
and CO.52 Brooker et al. conducted a similar study in 13
patientsundergoingspinalanesthesiaandfoundthatphen- stationary bikes, increasing workloads from rest (VO
2
ylephrine significantly reduced CO, in comparison with averaged 5.8 mL/kg/min) to VO (average 39
2max
epinephrine.51Langsesaeteretal.alsostudiedtheeffectsof mL/kg/min). CO increased by 170% at maximal values,
phenylephrine (0.25 (cid:7)g/kg/min) in patients undergoing but organ-specific blood volumes (measured with techne-
spinalanesthesia,comparingitwithplaceboin80women tium 99-m scanning) decreased by 46%, 24%, and 18% in
undergoingcesareandelivery.Whilephenylephrineclearly the spleen, kidneys, and liver, respectively. By contrast,
increased systolic blood pressure, it also led to significant organ-specific blood volumes increased by 50% and 24%,
reductions in CO in comparison with that in controls. A respectively,inthelungsandheart.32
morerecentcomparisonofphenylephrineinfusionratesin Particularlyappealingaboutthesplanchnicvenousres-
the setting of spinal anesthesia for elective cesarean deliv- ervoir concept is the fact that the splanchnic vasculature
ery found that the maximal decrease in CO was linearly exists in parallel (1/R (cid:2) 1/R (cid:3) 1/R (cid:3) … (cid:3) 1/R )
Total 1 2 n
relatedtothedose.66 withtheremainderofthesystemiccirculation,thusattenu-
Systemic venous return. Multiple animal studies have ating the increase in venous resistance that would other-
shown that (cid:1)-agonists are venoconstrictors, and thus wise accompany venoconstriction, while at the same time
1
capable of increasing MCFP.2,67–69 Appleton et al., for allowing the body to mobilize a significant amount of
instance, found that 8 to 20 (cid:7)g/kg/min of phenylephrine volume.19
increasedMCFPby49%inlightlysedateddogs.2 Thus, it appears that (cid:1)-agonists’ effects on the venous
1
Although an isolated increase in venous tone may circulation have the ability to both increase (by reducing
transiently increase CO, whether or not a pharmacologi- venous compliance, thus converting unstressed volume to
cally mediated increase in venous tone leads to increased stressed volume and increasing preload) and decrease
venous return in the intact organism has been debated.70 (primarilythroughincreasesinvenousresistance)CO.The
Unliketheincreasedskeletalmuscletonethataccompanies end result is likely related to dose of the drug and the
exercise,ortheincreaseinstressedvolumethatoccursafter sensitivityoftheindividualorganismandtissues.Thisidea
fluid administration, all of the known drugs that increase is supported by Zandberg et al.’s dose-response study of
venous tone have the potential to cause accompanying methoxamineindogswhohadundergonespinalanesthe-
increases in arterial and venous resistance, potentially sia. Initially, methoxamine increased both CO (maximally
negatinganyimprovementsinvenousreturn.71 at 0.6 mg/kg/h of methoxamine) and blood pressure,
Inanilluminatingreview,GelmanandMushlinpointed althoughatincreasingdoses,CObegantodecrease73(Fig.
out that 25% of total body blood volume is present in the 7).Interestingly,calculatedSVRwasincreasedevenatlow
splanchnic organs, making them an important hemody- doses (0.3 to 0.6 mg/kg/h), suggesting that the increased
namic reservoir. Furthermore, they suggested that (cid:1)- COwasduetoincreasesinstressedvolumeandpreload.
1
agonists have a dose-dependent effect on venous return; A limited number of human studies also support the
the initial response being an increase in venous return as ideathat(cid:1)-agonistsreducevenouscomplianceandpoten-
1
the splanchnic vasculature is “unloaded,” followed by a tially increase venous return. In 1975, Marino et al.
decrease in venous return at higher doses as the effects of compared the effects of phenylephrine, isoproteronol, do-
vasoconstriction and venoconstriction (decreased organ pamine, and phentolamine on perfusion pressures and
outflow)predominate.72 reservoirvolumesin73patientsundergoingcardiopulmo-
Theideathatthesplanchniccirculationisanimportant nary bypass. Marino et al.’s experiments showed signifi-
reservoirforvenousbloodisstronglysupportedbyFlamm cant increases in cardiopulmonary bypass reservoir
et al.’s study of blood volume distribution in exercising volumes after administration of both phenylephrine (0.3
humans.Flamm’sgroupexercised14healthyvolunteerson mg)andlow-(0.2mg)andhigh-(4mg)dosedopamine(by
292 www.anesthesia-analgesia.org ANESTHESIA&ANALGESIA
ExperimentalAlpha-AgonismData
1
the intracranial vessels of animals with experimental neu-
rologic injuries (and presumably a disrupted cerebral au-
toregulatory curve).82 However, increased velocity does
notnecessarilyimplyincreasedflow,especiallyifachieved
throughtheactionsofavasoconstrictivedrug.
Studiesofphenylephrineoncerebralbloodflow(CBF),
asopposedtovelocity,arerare.Kitaguchietal.studiedthe
effects of methoxamine on 10 patients with ischemic cere-
brovascular disease undergoing extraintracranial artery
bypass and, using the Kety–Schmidt inert gas saturation
technique, found no relationship between CBF and
methoxamine-induced increases in MAP.83 Joseph et al.
studied the effects of phenylephrine on 5 vasospastic
Figure 8. Effects of phenylephrine and dopamine on vascular subarachnoid hemorrhage patients and found that mean
resistance and preload as measured by changes in perfusion
CBF in the right frontal cortex increased by 75% in the
pressure and venous reservoir volume on the basis of measure-
mentsinpatientsoncardiopulmonarybypass(adaptedfromMarino vasospastic cortical regions, but did not report CBF in the
etal.,74withwrittenpermissionfromWoltersKluwerHealth). nonvasospasticcorticalregions.84
247, 444, and 687 mL, respectively), suggesting that both Kidneys and Other Organs
drugs decrease venous compliance through enhanced tone. Studiesof(cid:1)1receptordensityinhumanshaveshownthat
Perfusion pressures changed by 23, 0.13, and 15 mmHg, the renal arteries contain a relatively high density of (cid:1)1
respectively, suggesting that phenylephrine and high-dose receptors(24fmol/mgprotein),incomparisonwithother,
dopamineincreasedvascularresistance(andthus,afterload), similarly sized vessels, such as the mesenteric arteries (13
whereaslow-dosedopaminehadnoeffectonvascularresis- fmol/mgprotein).45Humanphysiologicstudiesarelargely
tance,despiteitsabilitytoincreasevenousreturn74(Fig.8). absent; however, Grangsjo and Persson85 studied the ad-
Rightheartandpulmonarycirculation. Becausesystemic ministration of vasoactive drugs on canine renal blood
complianceisapproximately7timesthatofthepulmonary flow. Three normotensive dogs under general anesthesia
circulation,67,69 the pulmonary vascular system cannot received methoxamine (doses ranged from 0.17 to 0.6
store as much latent “preload” as does its systemic coun- mg/kg), which immediately resulted in significant reduc-
terpart.Thatsaid,becausetheleftheartreliesontheright tions in urine output (in 1 animal, measured renal blood
heart for preload, and the right heart traditionally faces flowdecreasedfrom110to10mL/min)andtotalcessation
significantly less afterload than does the left, changes in of urine output within minutes. Three hypotensive dogs
pulmonary vascular resistance can profoundly impact the (bled through a femoral artery catheter to blood pressure
cardiovascularsystemasawhole. [systolic or diastolic not distinguished] (cid:1)50 mmHg) also
Studies of (cid:1) receptor density in humans have shown received methoxamine (0.3 to 1 mg/kg), which similarly
1
that the pulmonary arteries contain a higher density of (cid:1) reducedrenalbloodflowby70%–80%,despitesignificant
1
receptors than does any nonsplanchnic organ.45 Tuman et increases in blood pressure and renal perfusion pressure
al.examinedtheeffectsofphenylephrineonrightventric- (Fig. 9). Norepinephrine, by contrast, increased medullary
ular function in patients undergoing coronary artery sur- blood flow and resulted in an increase in urine output
gery,andfoundthatpostinductionphenylephrine(titrated when administered as a continuous infusion (0.03 to 0.6
to increase systolic blood pressure to 20% above baseline) (cid:7)g/kg/min)inGrangsjoandPersson’sexperiments.85 At
significantly increased right ventricular end-diastolic vol- least1caseofovertrenalfailureinducedbyphenyleph-
ume index (RVEDVi) (from 86.3 to 97.5 mL/m2, P (cid:2) rine administration in humans has been reported.86
0.0001)75 without significantly impacting cardiac index. Hoffbrand et al. administered vasopressors to unanes-
Pulmonary vascular resistance, however, increased from thetizedrhesusmonkeys,andfoundthatalthoughnorepi-
62.0to157.2dyne(cid:1)s(cid:1)cm(cid:8)5,suggestingthattheincreasein nephrine (0.5 to 3 (cid:7)g/kg/min) redistributed CO towards
RVEDVi was due to an increase in pulmonary vascular the heart and skeletal muscles, methoxamine increased
resistance,andnotvenousreturn. vascularresistanceuniformlyandreducedCOatbothlow
(20 to 100 (cid:7)g/kg/min) and high (150 to 500 (cid:7)g/kg/min)
Brain doses (by 20% and 43%, respectively). Both doses of me-
It is generally thought that the cerebral vasculature lacks thoxaminesignificantlyreducedbloodflowtothekidneys,
significant (cid:1) receptors, mostly on the basis of animal spleen, pancreas, and lungs, and high-dose methoxamine
1
experiments (such as Harik et al.’s data from rat and pig significantly reduced blood flow to the brain (regional
models76), although this assertion is refuted by other data resistance increased by 54%), heart (regional resistance
from bovine,77 rat,78 and gerbil79 models, as well as some increased79%),gastrointestinaltract,andskeletalmuscles,
humandata.80 andofalltheorgansmeasured,sparedonlytheadrenals87
Nevertheless, phenylephrine has been used to increase (Fig.10).
cerebral perfusion pressure. Although Doppler studies Heyndrickx et al. found that when escalating methox-
have been used to suggest that cerebral autoregulation is aminedosesfrom5to50(cid:7)g/kg/mininhealthy,conscious
intactduringgeneralanesthesia,81theyhavealsobeenused dogs, MAP values increased to 25, 35, and 55% above
toshowthatphenylephrineincreasesbloodflowvelocityin baseline,butCOdecreasedby9%atlowerdosesandbyas
August2011•Volume113•Number2 www.anesthesia-analgesia.org 293