Ebook Marinos the little ICU book (2/E): Part 2

Chapter23

Acid-BaseAnalysis
Thischapterdescribeshowtoidentifyacid-basedisordersusingthepH,PCO2andbicarbonate(HCO3)
concentration in blood. Included are: (a) simple rules for the identification of primary, secondary, and
mixedacid-basedisorders,(b)formulasfordeterminingtheexpectedacid-basechangesforeachofthe
primaryacid-basedisorders,and(c)adescriptionofthe“aniongap”andhowitisused.

I.ACID-BASEBALANCE
According to traditional concepts of acid-base physiology, the hydrogen ion (H+) concentration in
extracellularfluidisdeterminedbythebalancebetweenthepartialpressureofcarbondioxide(PCO2)
andthebicarbonate(HCO3)concentration(1):
[H+]=k×(PCO2/HCO3)

(23.1)

(kisaproportionalityconstant).Thismeansthatallacid-basedisordersaredefinedbytwovariables:
PCO2andHCO3.ThisisshowninTable23.1.

A.TypesofAcid-BaseDisorders
1.

Arespiratoryacid-basedisorderisachangein[H+] that is a direct result of a change in PCO2.
AccordingtoEquation23.1,anincreaseinPCO2willincreasethe[H+]andproducearespiratory
acidosis,whileadecreaseinPCO2willdecreasethe[H+]andproducearespiratoryalkalosis.

2.

A metabolic acid-base disorder is a change in [H+] that is a direct result of a change in HCO3.
Equation23.1predictsthatanincreaseinHCO3willdecreasethe[H+]andproduceametabolic
alkalosis,whileadecreaseinHCO3willincreasethe[H+]andproduceametabolicacidosis.

3.

Acid base disorders can be primary (the principal disturbance) or secondary (an additional
disturbance).


B.CompensatoryResponses
1.

Compensatory responses are designed to limit the change in H+ concentration produced by the
primaryacid-basedisorder.Thisisaccomplishedbychangingthesecondaryvariableinthesame
direction as the primary variable (e.g., a primary increase in PCO2 is accompanied by a
compensatoryincreaseinHCO3),asshowninTable23.1.

2.

Compensatory responses do not completely correct the change in [H+] produced by the primary
acid-basedisorder(2).

3.

The specific features of compensatory responses are described next. The equations that describe
theseresponsesareshowninTable23.2.

C.ResponsestoPrimaryMetabolicDisorders
Theresponsetoametabolicacid-basedisorderinvolvesachangeinminuteventilationthatismediated
byperipheralchemoreceptorsinthecarotidbody,locatedatthecarotidbifurcationintheneck.

1. ResponsetoMetabolicAcidosis

Thecompensatoryresponsetometabolicacidosisisanincreaseinminuteventilation(tidalvolume
andrespiratoryrate)andasubsequentdecreaseinarterialPCO2(PaCO2).Thisresponseappears
in30–120minutes,andcantake12to24hourstocomplete(2).Themagnitudeoftheresponseis
definedbytheequationbelow(2).
ΔPaCO2=1.2×ΔHCO3

(23.2)

UsinganormalPaCO2of40mmHgandanormalHCO3of24mEq/L,theaboveequationcanbe
rewrittenasfollows:


ExpectedPaCO2=40–[1.2×(24–HCO3)]
a.

(23.3)

EXAMPLE:ForaprimarymetabolicacidosiswithaplasmaHCO3of14mEq/L,theΔHCO3
is24–14=10mEq/L,theΔPaCO2is1.2×10=12mmHg,andtheexpectedPaCO2is40–
12 = 28 mm Hg. If the measured PaCO2 is >28 mm Hg, there is a secondary respiratory
acidosis,andifthemeasuredPaCO2is<28mmHg,thereisasecondaryrespiratoryalkalosis.

2. ResponsetoMetabolicAlkalosis
The compensatory response to metabolic alkalosis is a decrease in minute ventilation and a
subsequent increase in PaCO2. This response is not as vigorous as the response to metabolic
acidosis(becausethebaselineactivityofperipheralchemoreceptorsislow,sotheyareeasierto
stimulatethaninhibit).Themagnitudeoftheresponseisdefinedbytheequationbelow(2).
ΔPaCO2=0.7×ΔHCO3

(23.4)

UsinganormalPaCO2of40mmHgandanormalHCO3of24mEq/L,theaboveequationcanbe
rewrittenasfollows:
ExpectedPaCO2=40+[0.7×(HCO3–24)]
a.

(23.5)

EXAMPLE:ForametabolicalkalosiswithaplasmaHCO3of40mEq/L,theΔHCO3is40–
24=16mEq/L,theΔPaCO2is0.7×16=11mmHg,andtheexpectedPaCO2is40+11=51
mmHg.


D.ResponsestoPrimaryRespiratoryDisorders
The compensatory response to changes in PaCO2 occurs in the kidneys, where HCO3 absorption in the
proximaltubulesisadjustedtoproducetheappropriatechangeinplasmaHCO3(inthesamedirectionas
the change in PaCO2). This renal response is relatively slow, and can take 2 or 3 days to reach
completion.Asaresult,respiratoryacid-basedisordersareseparatedintoacuteandchronicdisorders.

1. AcuteRespiratoryDisorders
Acute changes in PaCO2 are not accompanied by large changes in the plasma HCO3, as
demonstratedbythefollowingequations(2).
a.

Foracuterespiratoryacidosis:
ΔHCO3=0.1×ΔPaCO2

b.

(23.6)

Foracuterespiratoryalkalosis:
ΔHCO3=0.2×ΔPaCO2

(23.7)


2. ChronicRespiratoryDisorders
TherenalresponsetoachronicincreaseinPaCO2involvesanincreaseinHCO3absorptioninthe
proximalrenaltubules,whichincreasestheplasmaHCO3.Theresponsetoachronicdecreasein
PaCO2isadecreaseinrenalHCO3absorption,whichlowerstheplasmaHCO3concentration.The
magnitude of this response is similar in chronic respiratory acidosis and chronic respiratory
alkalosis,sothesameformulacanbeusedtodescribetheexpectedchangesinbothconditions.
ΔHCO3=0.4×ΔPaCO2

(23.8)

UsinganormalPaCO2of40mmHgandanormalHCO3of24mEq/L,theaboveequationcanbe
rewrittenasfollows:
a.

Forchronicrespiratoryacidosis:
ExpectedHCO3=24+[0.4×(PaCO2–40)]

b.

(23.9)

Forchronicrespiratoryalkalosis:
ExpectedHCO3=24–[0.4×(40–PaCO2)]

(23.10)

II.ACID-BASEEVALUATION
The following is a structured, rule-based approach to acid-base evaluations using the relationships
between the [H+], PCO2, and HCO3 described in the previous section. The normal range of values for
thesevariablesareshownbelow.
pH=7.36–7.44
PCO2=36–44mmHg
HCO3=22–26mEq/L

Step1:IdentifyPrimaryandMixedDisorders
The first step in the evaluation focuses on the PaCO2 and pH to identify primary and mixed acid-base
disorders.
1.

IfthePaCO2andpHarebothabnormal,comparethedirectionalchange.
a.

If the PaCO2 and pH change in the same direction, there is a primary metabolic acid-base
disorder(andthepHidentifieswhetheritisanacidosisoralkalosis).

b.

IfthePaCO2andpHchangeinoppositedirections,thereisaprimaryrespiratoryacid-base
disorder.


c.

2.

EXAMPLE:ConsideracasewherethearterialpHis7.23andthePaCO2is23mmHg.The
pH and PaCO2 are both decreased (indicating a primary metabolic disorder) and the pH is
low(indicatinganacidosis),sothisrepresentsaprimarymetabolicacidosis.

If only one variable (pH or PaCO2) is abnormal, there is a mixed metabolic and respiratory
disorder(ofequivalentstrength).
a.

If the PaCO2 is abnormal, the directional change in PaCO2 identifies the type of respiratory
disorder,whichthenidentifiestheopposingmetabolicdisorder.

b.

IfthepHisabnormal,thedirectionalchangeinpHidentifiesthetypeofmetabolicdisorder
(e.g.,lowpHindicatesametabolicacidosis)andtheopposingrespiratorydisorder.

c.

EXAMPLE:ConsideracasewherethearterialpHis7.38andthePaCO2is55mmHg.Only
onevariable(thePaCO2)isabnormal,indicatingamixedmetabolicandrespiratorydisorder.
ThePaCO2iselevated,indicatingarespiratoryacidosis,sotheopposingmetabolicdisorder
mustbeametabolicalkalosis.Therefore,thisconditionisamixedrespiratoryacidosisand
metabolicalkalosis.BothdisordersareequalinstrengthbecausethepHisnormal.

Step2:IdentifySecondaryDisorders
Ifthefirststepidentifiedaprimarydisorder(insteadofamixeddisorder),thenextstepistocalculatethe
expectedacid-basechangesusingtheequationsinTable23.2.Theexpectedchangesarethencomparedto
theactualchanges,anddiscrepanciesbetweenthetwoareusedtoidentifysecondaryacid-baseproblems.

Thisprocessisdemonstratedinthefollowingexample.
1.

EXAMPLE:Consideracasewiththefollowingarterialbloodgasresults:pH=7.32,PaCO2=23
mmHg,HCO3=16mEq/L.
a.

ThisrepresentsaprimarymetabolicacidosisbecausethepHandPCO2arebothdecreased.

b.

Equation23.3isthenusedtocalculatetheexpectedPaCO2fromthecompensatoryresponse.
TheexpectedPaCO2is40–[1.2×(24–16)]=30.4mmHg.

c.

TheexpectedandmeasuredPaCO2arethencompared.ThemeasuredPaCO2(23mmHg)is
lowerthantheexpectedPaCO2(30.4mmHg),indicatingasecondaryrespiratoryalkalosis.

d.

Therefore,thiscaseisaprimarymetabolicacidosiswithasecondaryrespiratoryalkalosis.

III.THEANIONGAP
Theaniongapisameasureoftherelativeabundanceofunmeasuredanionsinextracellularfluid,andcan
beusefulintheevaluationofmetabolicacidosis(6,7),asexplainednext.

A.Derivation

Electrochemical balance requires equal concentrations of negatively-charged anions and positivelycharged cations in extracellular fluid. This balance is expressed in the equation below using the
predominantanionsandcationsinplasma(sodium,chloride,andbicarbonate)aswellastheunmeasured
cations(UC)andunmeasuredanions(UA).
Na+UC=CL+HCO3+UA

(23.11)

Rearrangingthetermsintheaboveequationyieldsthefollowing:
Na–(CL+HCO3)=UA–UC
1.

(23.12)

The difference between unmeasured anions and unmeasured cations (UA – UC) is the anion gap
(AG),soEquation23.12canberestatedas:
AG=Na–(CL+HCO3)(mEq/L)

(23.13)

Theaniongapisthusaverysimplecalculationthatinvolvesroutinelymonitoredelectrolytes.

2. ReferenceRange
The reference range for the AG was originally 12±4 mEq/L (8–16 mEq/L) (7), but advances in
automatedelectrolytemeasurementsledtoareductioninthereferencerangeto7±4mEq/L (3–11
mEq/L)(8).Unfortunate-ly,thischangeisnotuniversallyrecognized.

B.UsingtheAnionGap
In the presence of a metabolic acidosis, an increase in the AG is evidence of an increase in strong
(readilydissociated)acidsinextracellularfluid,whileanormalAGindicatesthatbicarbonatelossisthe
sourceofacidosis.Thecausesofmetabolicacidosiscanthusbeseparatedintotwogroups,basedonthe

AG,asshowninTable23.3.

1. HighAnionGapAcidosis
Frequent sources of high AG metabolic acidosis include lactic acidosis, ketoacidosis, and endstagerenalfailure(duetolossofH+secretioninthedistalrenaltubules).Othernotablesourcesare
toxicingestionsofmethanol(whichproducesformicacid),ethyleneglycol(whichproducesoxalic
acid),andsalicylates(whichproducesalicylicacid)(9).

2. NormalAnionGapAcidosis
Common causes of metabolic acidosis with a normal AG include diarrhea (especially secretory
diarrhea),isotonicsalineinfusion(seeChapter10,SectionI-B-3),andearlyrenalfailure(dueto
lossofHCO3reabsorptionintheproximaltubules).TheHCO3lossintheseconditionsisreplaced
bychlorideforelectricalneutrality,andthetermhyperchloremicmetabolicacidosisisalsoused
for this type of metabolic acidosis. (In high AG metabolic acidoses, the acids dissociate and
generateanionsthatbalancethedecreaseinHCO3,sothereisnoassociatedhyperchloremia.)


C.Reliability
Thereliabilityoftheaniongapfordetectingstrongacidshasbeeninconsistent,andthereareanumberof
reportsshowinganormalAGinpatientswithlacticacidosis(10,11)andketoacidosis(seeReference21
inChapter24).

1. CorrectableFactors
TherearetwocorrectablefactorsthatlimitthesensitivityoftheAG.
a.

One factor is the continued use of the original, higher reference range for the AG, which
substantiallyreducesthesensitivityoftheAGfordetectinglacticacidosiswhencomparedto
thelower,morecurrentreferencerange(12).

b.

The other factor is the ability of hypoalbuminemia to decrease the AG (13), which is
describednext.


2. InfluenceofAlbumin
The unmeasured anions and cations that normally contribute to the anion gap are shown in Table
23.4.Notethatalbuministheprincipalunmeasuredanion,andtheprincipaldeterminantofthe
aniongap.
a.

Albuminisaweakacidthatcontributesabout3mEq/LtotheAGforeach1g/dLofalbuminin
plasma(atanormalpH)(3).

b.

Hypoalbuminemia lowers the AG, and this could hinder or prevent an increase in AG in
metabolic acidoses caused by the accumulation of strong acids. Considering that
hypoalbuminemiaispresentinasmanyas90%ofICUpatients(13),theinfluenceofalbumin
ontheAGcannotbeignored.

c.

The AG can be adjusted for low albumin levels by using the following formula for the
correctedaniongap(AGc).
AGc=AG+[2.5×(4.5–PlasmaAlbumining/dL)](23.14)
(4.5representsthenormalconcentrationofalbumininplasma).ThecorrectedAGhasshown
anim-proveddiagnosticperformanceincriticallyillpatients(14).

REFERENCES

1.
2.
3.
4.

Adrogue HJ, Gennari J, Gala JH, Madias NE. Assessing acid-base disorders. Kidney Int 2009;
76:1239–1247.
AdrogueHJ,MadiasNE.Secondaryresponsestoalteredacid-basestatus:Therulesofengagement.
JAmSocNephrol2010;21:920–923.
KellumJA.Disordersofacid-basebalance.CritCareMed2007;35:2630–2636.
WhittierWL,RuteckiGW.Primeronclinicalacid-baseproblemsolving.DisMon2004;50:117–
162.


5.

FenclV,LeithDE.Stewart’squantitativeacid-basechemistry:applicationsinbiologyandmedicine.
RespirPhysiol1993;91:1–16.

6.

Narins RG, Emmett M. Simple and mixed acid-base disorders: a practical approach. Medicine
1980;59:161–187.
EmmetM,NarinsRG.Clinicaluseoftheaniongap.Medicine1977;56:38–54.
Winter SD, Pearson JR, Gabow PA, et al. The fall of the serum anion gap. Arch Intern Med
1990;150:311–313.
Judge BS. Metabolic acidosis: differentiating the causes in the poisoned patient. Med Clin N Am
2005;89:1107–1124.

7.

8.
9.

10. IbertiTS,LiebowitzAB,PapadakosPJ,etal.Lowsensitivityoftheaniongapasascreentodetect
hyperlactatemiaincriticallyillpatients.CritCareMed1990;18:275–277.
11. Schwartz-Goldstein B, Malik AR, Sarwar A, Brandtsetter RD. Lactic acidosis associated with a
normalaniongap.HeartLung1996;25:79–80.
12. AdamsBD,BonzaniTA,HunterCJ.Theaniongapdoesnotaccuratelyscreenforlacticacidosisin
emergencydepartmentpatients.EmergMedJ2006;23:179–182.
13. Figge J, Jabor A, Kazda A, Fencl V. Anion gap and hypoalbuminemia. Crit Care Med 1998;
26:1807–1810.
14. MallatJ,BarraillerS,LemyzeM,etal.Useofsodiumchloridedifferenceandcorrectedaniongap
assurrogatesofStewartvariablesincriticallyillpatients.PLoSONE2013;8:e56635.


Chapter24

OrganicAcidoses
This chapter describes two clinical disorders that involve excessive production of organic (carbonbased)acidsbyintermediarymetabolism.Bothofthesedisorders,lacticacidosisandketoacidosis,can
beadaptiveprocessesintherightsetting,butarepathologicalprocessesintheICUsetting.

I.LACTICACIDOSIS
Lacticacidosisisprobablythemostconcerningofallmetabolicacidoses,butthesourceoftheconcernis
nottheacidosis,buttheconditionthatisresponsiblefortheacidosis.

A.ResponsibleConditions
(Note:Becausethepertinentissuesinlacticacidosisareoftenrelatedtothelactatelevelratherthanthe
acidosis,thetermhyperlactatemiawillbeusedinterchangeablywithlacticacidosis.)Severalconditions
canberesponsibleforhyperlactatemia,asshowninTable24.1.Themostprevalentoftheseconditions
aresepsisandtheclinicalshocksyndromes(i.e.,hypovolemic,cardiogenic,andsepticshock).

1. ClinicalShockSyndromes
Hyperlactatemiaisuniversalintheclinicalshocksyndromes(sinceitisrequiredforthediagnosis)
andtheprognosisintheseconditionsisrelatedtotheseverityofthelactateelevation,andthetime
required for the lactate levels to normalize (lactate clearance). These relationships are
demonstratedinFigure6.2(Chapter6).


2. Sepsis
Serumlactatelevelshavethesamediagnosticandprognosticsignificanceinsepsisastheydoinin
theshocksyndromes.Thelacticacidosisinsepsisisnottheresultofinadequatetissueoxygenation
(see Chapter 6, Section III-F), which has important implications for the traditional emphasis on
promotingtissueoxygenationinpatientswithlacticacidosis.

3. ThiamineDeficiency
Thiaminedeficiencyisoftenoverlookedasacauseofelevatedbloodlactatelevels.Thiamineisa
cofactorforpyruvatedehydrogenase(theenzymethatconvertspyruvatetoacetylcoenzymeA,and
limitsconversiontolactate),andthiaminedeficiencycanresultinseverelacticacidosis(2). (See
Chapter36,SectionIII-Aformoreinformationonthiaminedeficiency.)

4. Drugs
Avarietyofdrugscanproducehyperlactatemia,asindicatedinTable24.1.Mostcasesaredueto
an impaired oxidative metabolism, but epinephrine and high-dose β2 agonists promote
hyperlactatemiabyincreasingtheproductionofpyruvate(1).
a.

METFORMIN:Metforminisanoralhypoglycemicagentthatproduceslacticacidosisduring
therapeutic dosing. The mechanism is unclear, and it occurs primarily in patients with renal
insufficiency(3).Thelacticacidosiscanbesevere,withamortalityratethatexceeds45%if
untreated (3,4). Plasma metformin levels are not routinely available, and the diagnosis is

based on excluding other causes of lactic acidosis. The preferred treatment is hemodialysis
(3,4).

5. PropyleneGlycol
Propyleneglycolisusedasasolventinintravenouspreparationsoflorazepam,diazepam,esmolol,
nitroglycerin,andphenytoin.Itismetabolizedprimarilyintheliver,andtheprincipalmetabolites
arelactateandpyruvate(5).


a.

Propyleneglycoltoxicity(i.e.,agitation,coma,seizures,hypotension,andlacticacidosis)has
been reported in 19–66% of patients receiving high-dose IV lorazepam infusions for more
than48hours(5,6).

b.

The diagnosis can be elusive. There is an assay for propylene glycol in blood, but the
acceptablerangehasnotbeendetermined.

c.

Prolonged infusions of lorazepam should be avoided. (In fact, prolonged infusions of any
benzodiazepine should be avoided because these drugs accumulate in the brain and produce
excessiveandprolongedsedation.)

6. OtherNotableConditions
a.

Generalized seizures can produce marked increases in serum lactate levels, but this is a

hypermetaboliceffect,anditresolvesquicklyaftertheseizuressubside(7).

b.

The liver is responsible for 70% of lactate clearance. Hyperlactatemia is common in acute,
fulminantliverfailure(1),butisnotafrequentfindinginchronicliverfailureunlessthereisa
coexistingconditionthatincreaseslactateproduction(e.g.,sepsis)(1).

B.DiagnosticConsiderations
1.

Serumlactatelevelsarereadilyavailable,andscreeningtestsforlacticacidosis,suchastheanion
gap,arenotnecessary(andcanbeunreliable,asdescribedinChapter23,SectionIII-C).

2.

Lactatelevelscanbemeasuredinvenousorarterialblood,withequivalentresults(1).

3.

The upper limit of normal for serum lactate varies from 1.0 to 2.2 mmol/L in individual
laboratories (1), but 2 mmol/L seems to be a common cutoff point. However, lactate levels must
riseabove4mmol/Ltoshowanassociationwithincreasedmortality(8),soacutoffof4mmol/L
maybemoreappropriateforclinicallysignificanthyperlactatemia.

C.AlkaliTherapy
Therapyaimedatcorrectingtheacidosisdoesnothaveamajorroleinthemanagementofpatientswith
lactic acidosis. The following is a brief summary of the relevant issues in alkali therapy for lactic
acidosis.

1. TheBicarbonateExperience
Clinical studies have consistently shown that sodium bicarbonate infusions are without
hemodynamic benefit or survival benefit in lactic acidosis (9-11). Furthermore, bicarbonate
infusionsareaccompaniedbyseveralundesirableeffects(seeTable24.2),includinganincreasein
arterialPCO2andaparadoxicaldecreaseinintracellularpH(attributedtotranscellularmovement
ofthegeneratedCO2)(9,12).


2. CurrentRecommendations
Given the lack of benefit, and the associated risk, bicarbonate therapy is not recommended as a
treatment modality in lactic acidosis (9,13). Furthermore, bicarbonate therapy has been removed
from the ACLS guidelines on cardiac arrest (14). Nevertheless, there continue to be
recommendationsforbicarbonatetherapyincasesofsevereacidosis,whenthepHfallsbelow7.0
(15). The current use of bicarbonate is predominantly as a “desperation measure” to restore
vasopressorresponsivenessinpatientswhoarerapidlydeteriorating.

3. ReplacementRegimen
Thepopularfluidforbicarbonatereplacementisa7.5%sodiumbicarbonatesolution,andTable
24.2 shows the composition of this fluid. Note the hyperosmolality (which mandates infusion
throughalargevein)andtheextremelyhighPCO2 (which explains the increase in arterial PCO2
associatedwithbicarbonateinfusions).
a.

The bicarbonate dose is determined by estimating the HCO3 deficit with the following
equation(15,16).
HCO3deficit=0.6×wt(kg)×(15–plasmaHCO3)

(24.1)

wherewtisidealbodyweight,and15mEq/ListhedesiredplasmaHCO3.(Foranadultwith

anidealbodyweightof70kgandaplasmaHCO3of10mEq/L,theHCO3deficitis0.6×70
×(15–10)=210mEq.)
b.

TheHCO3deficitcanbereplacedatarateof1mEq/kgperhour(11).ThePaCO2shouldbe
monitored during bicarbonate infusions, and increases in the PaCO2 should be corrected by
adjustingtheventilatorsettingstoprovideanincreasedminuteventilation.

c.

Ifthereisnohemodynamicorclinicalimprovementafterafewhours,thebicarbonateinfusion
shouldbediscontinued.

II.DIABETICKETOACIDOSIS


Whenglucosemovementintocellsisimpaired,adiposetissuereleasesfreefattyacidsthataretakenupin
theliverandmetabolizedtoketonesthatcanbeusedasoxidativefuels.Theseketonesincludeacetone,
acetoacetate(AcAc)andβ-hydroxybutyrate(β-OHB).

A.Ketoacids
AcAc and β-OHB are strong acids (ketoacids), and plasma concentrations above 3 mmol/L produce a
metabolicacidosis(17).β-OHBisthepredominantketoacid(seeFigure24.1),andisaboutthreetimes
more abundant than AcAc. Acetone is not a ketoacid, but is responsible for the characteristic “fruity”
odorofthebreathinpatientswithketoacidosis.

1. TheNitroprussideReaction
The nitroprusside reaction is a popular, colorimetric method for detecting ketones in blood and
urine. The test can be performed with tablets (Acetest) or reagent strips (Ketostix, Labstix,
Multistix).

a.

THEPROBLEM:Thenitroprussidereactionhasonemajorshortcoming;i.e.,itdetectsonly
acetoneandAcAc,anddoesnotdetectβ-OHB(17),thepredominantketoacidinblood.This
limitation is illustrated in Figure 24.1. Note that, in alcoholic ketoacidosis, the total
concentration of ketoacids in blood is 13 mmol/L (about 4 times the concentration that
producesanacidosis),buttheketoacidswillnotbedetectedbecausetheAcAclevelisbelow
thethresholdfordetection(3mmol/L).

2. β-hydroxybutyrateMonitoring
Portable “ketone meters” are now available that provide reliable measurements of β-OHB in
fingerstick (capillary) blood in about 10 seconds (18). The American Diabetes Association
considersthisthepreferredmethodformonitoringketoacidosis(19).


FIGURE24.1Theconcentrationsofacetoacetateandβ-hydroxybutyrateinthebloodindiabetic
ketoacidosis(DKA)andalcoholicketoacidosis(AKA).Thehorizontalhatchedlinerepresentsthe
thresholdforapositivenitroprussidereaction.

B.ClinicalFeatures
1.

2.

3.

According to the American Diabetes Association (ADA), diabetic ketoacidosis (DKA) has the
followingcharacteristics(19):
a.

Bloodglucose>250mg/dL.

b.

PlasmaHCO3<18mEq/LandplasmapH≤7.3.

c.

Increasedaniongap.

d.

Evidenceofketonesinbloodorurine.

ThefollowingareexceptionstotheADAcriteria:
a.

Thebloodglucoseis<250mg/dLinabout20%ofcasesofDKA(20).

b.

TheaniongapcanbenormalinDKA(21).Therenalexcretionofketonesisaccompaniedby
an increase in chloride reabsorption, and the resulting hyperchloremia limits the increase in
theaniongap.(SeeChapter23,SectionIIIforadescriptionoftheaniongap.)

OtherclinicalfeaturesofDKAthatdeservementionaresummarizedbelow.
a.

Leukocytosis is not a reliable marker of infection in DKA because ketonemia produces a
leukocytosis (19), but an increase in immature neutrophils (band forms) can be a reliable

markerofinfectioninDKA(22).

b.

Elevated troponin I levels without an acute coronary event has been reported in 27% of
patientswithDKA(23).


c.

HyperamylasemiaiscommoninDKA,buttheamylaseisextrapancreatic(19).

d.

DehydrationisalmostuniversalinDKA,butthismaynotbereflectedintheplasmasodium
concentration because hyperglycemia draws water from intracellular fluid, which causes a
dilutional decrease in the serum sodium concentration, and this masks free-water loss
(dehydration).

e.

The dilutional effect of hyperglycemia results in decrease in serum sodium of 1.6 to 2.4
mEq/Lforevery100mg/dLincreaseintheserumglucoseconcentration(24,25).

C.Management
ThemanagementofDKAdescribedhereisbasedontheADAguidelines(19).

1. IntravenousFluids
AprotocolforintravenousfluidtherapyinDKAisshowninTable24.3.Thefollowingaresome
highlightsofthisprotocol.

a.

VolumedeficitsinDKAaverageabout50to100mL/kg,andfluidtherapybeginswithisotonic
(0.9%) saline at a rate of 15–20 mL/kg per hour (or 1–1.5 L/hr) until the patient is
hemodynamicallystable.

b.

Note in Table 24.3 that the “corrected” serum sodium concentration is used to select the
appropriate IV fluid after hemodynamic stability is achieved. This correction refers to the
previously-described dilutional effect of hyperglycemia on the serum sodium concentration,
whichis1.6to2.4mEq/Lforevery100mg/dLincreaseintheserumglucose.

c.

EXAMPLE:Using2mEq/Lasthecorrectionfactor,ifthesodiumconcentrationis140mEq/L
and the plasma glucose is 600 mg/dL, the dilutional effect is 2 × 5 = 10 mEq/L, so the
correctedsodiumconcentrationis140+10=150mEq/L.

d.

Note also in Table 24.3 that 5% dextrose is added to the IV fluids when the serum glucose


fallsto250mg/dL.Thisreducestheriskofhypoglycemiauntilfoodintakebegins.

2. Insulin
AprotocolforinsulintherapyinDKAisshowninTable24.4.Thefollowingaresomehighlights.
a.

Notethatinsulinshouldnotbestartedifthepatientishypokalemic(whichisuncommonwhen
DKAfirstpresents).

b.

Regular insulin is started with an IV bolus dose of 0.15 units/kg (some consider this
unnecessary)followedbyacontinuousinfusionat0.1units/kg/hr.

c.

Insulin infusions are continued until the ketoacidosis has resolved (see later for how this is
determined)andoralnutrientintakeispossible.Thereafter,subcutaneousinsulinisstartedas
directedinTable24.4.

d.

AchievingeuglycemiaisneveradvisedintheICUsettingbecauseoftheriskofhypoglycemia,
andthegoalofglycemiccontrolisaserumglucoseof150–200mg/dL(26).

3. Potassium
a.

PotassiumdepletionisuniversalinDKA,withanaveragedeficitof3–5mEq/kg(20),butthe


serumpotassiumisoftennormal(74%ofpatients)orelevated(22%ofpatients)whenDKA
presents(20).
b.

The serum potassium can fall precipitously during insulin therapy (transcellular shift), so
potassium replacement should be started as soon as possible, and serum potassium levels
shouldbemonitoredevery1–2hoursuntillevelsstabilize.AddingpotassiumtotheIVfluids,
asindicatedinTable24.5,isusuallyeffectiveinmaintainingnormokalemia(19).

4. Phosphate
Thesituationwithphosphateisverysimilartopotassium(i.e.,depletioncommonbutserumlevels
rarelylowatpresentation,andserumlevelsdeclineduringinsulininfusions)withoneexception;
i.e., routine phosphate replacement has no documented benefit in DKA, and is not recommended
unlessphosphatelevelsare<1mg/dL(19,26).

5. AlkaliTherapy
TherecommendationsforbicarbonatereplacementinDKAarethesameasthosedescribedearlier
for lactic acidosis; i.e., bicarbonate therapy has no documented benefit in DKA, even when the
acidosisissevere(pH6.9–7.1)(27),anditisnotrecommendeduntilthepHfallsbelow7.0(19).

D.Acid-BaseMonitoring
1.

ResolutionofDKAhasbeendefinedasaplasmaglucose<200mg/dL,plasmaHCO3≥18mEq/L,
andvenouspH>7.3(19).

2.

TheHCO3andpHwillbeunreliablewhenisotonic(0.9%)salineisthepredominantresuscitation
fluid because the high chloride concentration in isotonic saline produces a hyperchloremic
metabolicacidosis(seeChapter10,SectionI-B-3)thatwillcounteracttheincreaseinserumHCO3
fromtheresolvingketoacidosis.

3.

TheaniongapshouldbeamorereliablemeasureformonitoringtheresolutionofDKA.


III.ALCOHOLICKETOACIDOSIS
Alcoholic ketoacidosis (AKA) is a sporadic disorder that occurs primarily in chronic alcoholics who
bingedrink(17,28).

A.ClinicalFeatures
Clinicalmanifestationsappear1–3daysafteraperiodofheavy,intensedrinking.
1.

Common manifestations include abdominal pain, vomiting, dehydration, and multiple electrolyte
abnormalities(e.g.,hypokalemia,hypomagnesemia,hypoglycemia,andhypophosphatemia).

2.

Theelectrolyteabnormalitiesmightexplainwhyasmanyas10%ofpatientswithAKAexperience
anunexpectedcardiacarrest(17).

B.Diagnosis
1.

The diagnosis of AKA can be elusive because the nitroprusside reaction for detecting ketones
canbenegativeinAKA.ThisisillustratedinFigure24.1.

2.

Figure24.1alsoshowsthatbetahydroxybutyrate(β-OHB)ispresentinhighconcentrationsinAKA
(higherthanDKA),someasuringtheβ-OHBlevelinbloodshouldbeasensitivemethodforthe

detectionofketonesinAKA.

C.Management
1.

The management of AKA is notable for its simplicity; i.e., infusion of dextrose-containing saline
solutions is usually all that is required. The glucose infusion slows hepatic ketone production,
whiletheinfusedvolumepromotestherenalclearanceofketones.

2.

Thiamine supplementation is recommended because glucose infusions can deplete marginal
thiaminereserves.

3.

Theketoacidosisusuallyresolveswithin24hours.

REFERENCES
1.
2.
3.

KrautJA,MadiasNE.Lacticacidosis.NEnglJMed2014;371:2309–2319.
Campbell CH. The severe lactic acidosis of thiamine deficiency: acute, pernicious or fulminating
beriberi.Lancet1984;1:446–449.
Seidowsky A, Nseir S, Houdret N, Fourrier F. Metformin-associated lactic acidosis: a prognostic
andtherapeuticstudy.CritCareMed2009;37:2191–2196.

4.

Perrone J, Phillips C, Gaieski D. Occult metformin toxicity in three patients with profound lactic
acidosis.JEmergMed2011;40:271–275.

5.

Wilson KC, Reardon C, Theodore AC, Farber HW. Propylene glycol toxicity: a severe iatrogenic
illnessinICUpatientsreceivingIVbenzodiazepines.Chest2005;128:1674–1681.
ArrogliaA,ShehabN,McCarthyK,GonzalesJP.Relationshipofcontinuousinfusionlorazepamto
serumpropyleneglycolconcentrationincriticallyilladults.CritCareMed2004;32:1709–1714.
OrringerCE,EusaceJC,WunschCD,GardnerLB.Naturalhistoryoflacticacidosisaftergrand-mal
seizures.Amodelforthestudyofanion-gapacidosesnotassociatedwithhyperkalemia.NEnglJ
Med1977;297:796–781.
Okorie ON, Dellinger P. Lactate: biomarker and potential therapeutic target. Crit Care Clin 2011;
27:299–326.

6.
7.

8.
9.
10.

11.

12.
13.
14.

15.
16.
17.
18.
19.
20.
21.

Forsythe SM, Schmidt GA. Sodium bicarbonate for the treatment of lactic acidosis. Chest 2000;
117:260–267
CooperDJ,WalleyKR,WiggsRR,etal.Bicarbonatedoesnotimprovehemodynamicsincritically
illpatientswhohavelacticacidosis:aprospective,controlledclinicalstudy.AnnInternMed1990;
112:492–498.
MathieuD,NeviereR,BillardV,etal.Effectsofbicarbonatetherapyonhemodynamicsandtissue
oxygenationinpatientswithlacticacidosis:Aprospective,controlledclinicalstudy.CritCareMed
1991;19:1352–1356.
KimmounA,NovyE,AuchetT,etal.Hemodynamicconsequencesofseverelacticacidosisinshock
states:frombenchtobedside.CritCare2015;19:175.
DellingerRP,LevyMM,RhodesA,etal.SurvivingSepsisCampaign:Internationalguidelinesfor
managementofseveresepsisandsepticshock,2012.IntensiveCareMed2013;39:165–228.
Link MS, Berkow LC, Kudenchuk PJ, et al. Part 7: Adult advanced cardiovascular life support:
2015 American Heart Association Guidelines Update for Cardiopulmonary Resuscitation and
EmergencyCardiovascularCare.Circulation.2015;132(Suppl2):S444–S464.
Sabatini S, Kurtzman NA. Bicarbonate therapy in severe metabolic acidosis. J Am Soc Nephrol
2009;20:692–695.
RoseBD,PostTW.Clinicalphysiologyofacid-baseandelectrolytedisorders.5thed.NewYork:
McGraw-Hill,2001:630–632.
Cartwright MM, Hajja W, Al-Khatib S, et al. Toxigenic and metabolic causes of ketosis and
ketoacidoticsyndromes.CritCareClin2012;601–631.
Plüdderman A, Hemeghan C, Price C, et al. Point-of-care blood test for ketones in patients with

diabetes:primarycarediagnostictechnologyupdate.BrJClinPract2011;61:530–531.
American Diabetes Association. Hyperglycemic crisis in diabetes. Diabetes Care 2004;
27(Suppl):S94–S102.
Charfen MA, Fernandez-Frackelton M. Diabetic ketoacidosis. Emerg Med Clin N Am 2005;
23:609–628.
GamblinGT,AshburnRW,KempDG,BeuttelSC.Diabeticketoacidosispresentingwithanormal
aniongap.AmJMed1986;80:758–760.


22. SlovisCM,MorkVG,SlovisRJ,BrainRP.Diabeticketoacidosisandinfection:leukocytecountand
differentialasearlypredictorsofseriousinfection.AmJEmergMed1987;5:1–5.
23. AlMallah M, Zuberi O, Arida M, Kim HE. Positive troponin in diabetic ketoacidosis without
evidentacutecoronarysyndromepredictsadversecardiacevents.ClinCardiol2008;31:67–71.
24. RoseBD,PostTW.Hyperosmolalstates:hyperglycemia.In:Clinicalphysiologyofacid-baseand
electrolytedisorders.5thed.NewYork,NY:McGraw-Hill,2001;794–821.
25. MoranSM,JamisonRL.Thevariablehyponatremicresponsetohyperglycemia.WestJMed1985;
142:49–53.
26. WesterbergDP.Diabeticketoacidosis:evaluationandtreatment.AmFamPhysician2013;87:337–
346.
27. Morris LR, Murphy MB, Kitabchi AE. Bicarbonate therapy in severe diabetic ketoacidosis. Ann
InternMed1986;105:836–840.
28. McGuireLC,CruickshankAM,MunroPT.Alcoholicketoacidosis.EmergMedJ2006;23:417–420.


Chapter25

MetabolicAlkalosis
Metabolicacidosisgetsalltheheadlines,butmetabolicalkalosisisthemostcommonacid-basedisorder
inhospitalizedpatients(1-3).Theprevalenceofmetabolicalkalosiscanbeattributedtothreefactors:(a)
common etiologies (e.g., diuretic therapy), (b) a tendency for the alkalosis to be sustained (thanks to

chloride),and(c)failuretoidentifyandcorrectthefactorsthatmaintainthealkalosis.

I.ORIGINS
Metabolic alkalosis is defined as an increase in the bicarbonate (HCO3) concentration in extracellular
fluid(plasma)thatisnotanadaptiveresponsetohypercapnia.ThenormalrangefortheplasmaHCO3is
22–26mEq/L.

A.Pathogenesis
1.

2.

Metabolicalkalosisisusuallytheresultofoneofthefollowingconditions(3):
a.

Lossofgastricacidfromvomitingornasogastricsuction.

b.

Enhanced secretion of hydrogen ions (H+) in the distal renal tubules (e.g., from diuretics or
mineralocorticoidexcess).

c.

TranscellularshiftofH+asaresultofhypokalemia.

d.

LossoffluidsthatcontainlittleornoHCO3(contractionalkalosis).

The normal response to metabolic alkalosis is an increase in the renal excretion of HCO3. This
response is reversed by chloride depletion and hypokalemia (3,4), and this helps to maintain a
metabolicalkalosis.
a.

Chloride depletion promotes the renal retention of HCO3 by increasing HCO3 reabsorption,
and inhibiting HCO3 secretion, in the distal renal tubules. Both effects are mediated by a
decrease in the luminal chloride concentration. The renal actions of chloride depletion are
consideredtheprincipalcauseofsustainedcasesofmetabolicalkalosis(3,4).

b.

Hypokalemiahasthesameeffectsaschloridedepletion(thoughthemechanismsdiffer).

B.Etiologies


Thecommonconditionsthatprecipitateand/ormaintainametabolicalkalosisareshowninTable 25.1,
alongwiththemechanismsinvolvedineachcondition.

1. VolumeLoss
LossoffluidsthatcontainlittleornoHCO3isawell-knowncauseofmetabolicalkalosis,andhas
been called contraction alkalosis because the assumed mechanism was a simple concentrating
effectontheplasmaHCO3.However,therealculpritischloridedepletion,becausethealkalosis
isnotcorrectedbyreplacingthefluiddeficitunlessthechloridedeficitisalsoreplaced(4).

2. LossofGastricSecretions
GastricsecretionsarerichinH+(50–100mEq/L),CL-(120–160mEq/L),and,toalesserextent,
K+ (10–15 mEq/L) (5). As a result, loss of gastric secretions (e.g., from nasogastric suction)
createsmultiplerisksformetabolicalkalosis(i.e.,lossofH+,CL-,K+,andvolumeloss).

3. Diuretics
Thiazide diuretics and “loop” diuretics like furosemide promote metabolic alkalosis via urinary
losses of H+, CL-, K+, and volume (1-3). Urinary chloride losses (chloruresis) match the sodium


losses(natriuresis),andmustbereplacedtocorrectthealkalosis.

4. Hypokalemia
Hypokalemiacanprecipitateametabolicalkalosis(viatranscellularshiftofH+)andalsohelpsto
maintainthealkalosis(bydecreasingtherenalexcretionofHCO3)(1-3).

5. ChlorideDepletion
As mentioned, chloride depletion helps to maintain the metabolic alkalosis by promoting renal
HCO3retention.

6. PosthypercapnicAlkalosis
Chronic CO2 retention is associated with an increase in plasma HCO3 (from enhanced HCO3
reabsorptioninthekidneys),andpatientswithchronicCO2retentionwhoareplacedonmechanical
ventilation can experience an abrupt decrease in arterial PCO2 from overventilation. In this
situation,theplasmaHCO3remainselevatedandresemblesametabolicalkalosis.Thiscondition
isoftensustainedbecauseofcoexistingchloridedepletion(3).

7. MassiveTransfusion
Each unit of packed red blood cells (PRBCs) contains about 17 mEq of citrate (used as an
anticoagulant), which generates HCO3 when metabolized. Transfusion of more than 8 units of
PRBCscanproduceametabolicalkalosis(3).

8. Others
Other causes of metabolic alkalosis include mineralocorticoid excess (primary

hyperaldosteronism), hypercalcemia and milk-alkali syndrome (chronic ingestion of calcium
carbonate-containingantacidsthatpromotehypercalcemia),andlaxativeabuse.

II.CLINICALMANIFESTATIONS
Metabolicalkalosishasremarkablyfewadverseeffects.

A.Hypoventilation
1.

Metabolic alkalosis produces respiratory depression and a subsequent increase in the arterial
PCO2 (PaCO2). However, this is not a vigorous response (unlike the respiratory stimulation
producedbyametabolicacidosis)(6).Themagnitudeoftheresponseisdefinedbythefollowing
equation(7):
ΔPaCO2=0.7×ΔHCO3

(25.1)