terapia antitrombotica en embarazo

8/11/2019 Terapia Antitrombotica en Embarazo

1/46

CHEST Supplement

www.chestpubs.org CHEST / 141 / 2 / FEBRUARY, 2012 SUPPLEMENT e691S

ANTITHROMBOTIC THERAPY AND PREVENTION OF THROMBOSIS, 9TH ED: ACCP GUIDELINES

Summary of Recommendations

Note on Shaded Text: Throughout this guideline,shading is used within the summary of recommenda-tions sections to indicate recommendations that arenewly added or have been changed since the publica-tion of Antithrombotic and Thrombolytic Therapy:American College of Chest Physicians Evidence-BasedClinical Practice Guidelines (8th Edition). Recom-mendations that remain unchanged are not shaded.

2.2.1. For pregnant patients, we recommendLMWH for the prevention and treatment of

VTE, instead of UFH(Grade 1B).

3.0.1. For women receiving anticoagulation forthe treatment of VTE who become pregnant, werecommend LMWH over vitamin K antagonistsduring the first trimester(Grade 1A), in the sec-ond and third trimesters(Grade 1B), and during latepregnancy when delivery is imminent(Grade 1A).

Background:The use of anticoagulant therapy during pregnancy is challenging because of thepotential for both fetal and maternal complications. This guideline focuses on the managementof VTE and thrombophilia as well as the use of antithrombotic agents during pregnancy.

Methods:The methods of this guideline follow the Methodology for the Development of Anti-thrombotic Therapy and Prevention of Thrombosis Guidelines: Antithrombotic Therapy and Pre-vention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based ClinicalPractice Guidelines in this supplement.

Results:We recommend low-molecular-weight heparin for the prevention and treatment of VTEin pregnant women instead of unfractionated heparin (Grade 1B). For pregnant women withacute VTE, we suggest that anticoagulants be continued for at least 6 weeks postpartum (for aminimum duration of therapy of 3 months) compared with shorter durations of treatment (Grade 2C).For women who fulfill the laboratory criteria for antiphospholipid antibody (APLA) syndromeand meet the clinical APLA criteria based on a history of three or more pregnancy losses,

we recommend antepartum administration of prophylactic or intermediate-dose unfraction-ated heparin or prophylactic low-molecular-weight heparin combined with low-dose aspirin(75-100 mg/d) over no treatment (Grade 1B). For women with inherited thrombophilia and ahistory of pregnancy complications, we suggest not to use antithrombotic prophylaxis (Grade 2C).For women with two or more miscarriages but without APLA or thrombophilia, we recommendagainst antithrombotic prophylaxis (Grade 1B).Conclusions:Most recommendations in this guideline are based on observational studies andextrapolation from other populations. There is an urgent need for appropriately designed studiesin this population. CHEST 2012; 141(2)(Suppl):e691Se736S

Abbreviations: APLA5antiphospholipid antibody; aPPT5activated partial thromboplastin time; HIT5heparin-inducedthrombocytopenia; INR5 international normalized ratio; LMWH5 low-molecular-weight heparin; NNT5numberneeded to treat; PE5pulmonary embolism; RR5 risk ratio; UFH5unfractionated heparin

VTE, Thrombophilia, AntithromboticTherapy, and Pregnancy

Antithrombotic Therapy and Prevention of Thrombosis,

9th ed: American College of Chest Physicians

Evidence-Based Clinical Practice Guidelines

Shannon M. Bates, MDCM; Ian A. Greer, MD, FCCP; Saskia Middeldorp, MD, PhD;David L. Veenstra, PharmD, PhD; Anne-Marie Prabulos, MD;and Per Olav Vandvik, MD, PhD

wnloaded From: http://journal.publications.chestnet.org/ on 08/28/2014
http://www.chestpubs.org/http://www.chestpubs.org/


2/46

e692S VTE, Thrombophilia, Antithrombotic Therapy, and Pregnancy

4.0.2. For lactating women using LMWH, dana-paroid, or r-hirudin who wish to breast-feed,

we recommend continuing the use of LMWH,danaparoid, or r-hirudin(Grade 1B).

4.0.3. For breast-feeding women, we suggest alter-native anticoagulants rather than fondaparinux(Grade 2C).

4.0.4. For breast-feeding women, we recommendalternative anticoagulants rather than oral directthrombin (eg, dabigatran) and factor Xa inhibi-tors (eg, rivaroxaban, apixaban)(Grade 1C).

4.0.5. For lactating women using low-dose aspi-rin for vascular indications who wish to breast-feed, we suggest continuing this medication(Grade 2C).

5.1.1. For women undergoing assisted reproduc-tion, we recommend against the use of routine

thrombosis prophylaxis(Grade 1B).

5.1.2. For women undergoing assisted reproduc-tion who develop severe ovarian hyperstimulationsyndrome, we suggest thrombosis prophylaxis(prophylactic LMWH) for 3 months postresolu-tion of clinical ovarian hyperstimulation syndromerather than no prophylaxis(Grade 2C).

Remarks:Women who are averse to taking medica-tion for very small benefit and those who considerself-injecting a considerable burden will be disin-

clined to use LMWH for extended thrombosis pro-phylaxis. Given that the absolute benefit decreasesas time from the hyperstimulation event increases,such women will be very disinclined to continue pro-phylaxis throughout the entire resultant pregnancy.

6.2.1. For women undergoing cesarean sectionwithout additional thrombosis risk factors, werecommend against the use of thrombosis pro-phylaxis other than early mobilization(Grade 1B).

6.2.2. For women at increased risk of VTE aftercesarean section because of the presence of one

major or at least two minor risk factors, we sug-gest pharmacologic thromboprophylaxis (prophy-lactic LMWH) or mechanical prophylaxis (elasticstockings or intermittent pneumatic compres-sion) in those with contraindications to anti-coagulants while in hospital following deliveryrather than no prophylaxis(Grade 2B).

Remarks: The reduced bleeding risk with mechanicalprophylaxis should be weighed against the inconve-nience of elastic stockings and intermittent pneu-matic compression.

3.0.2. For women requiring long-term vitamin Kantagonists who are attempting pregnancyand are candidates for LMWH substitution, wesuggest performing frequent pregnancy testsand substituting LMWH for vitamin K antago-nists when pregnancy is achieved rather thanswitching to LMWH while attempting pregnancy(Grade 2C).

Remarks:Women who place little value on avoidingthe risks, inconvenience, and costs of LMWH therapyof uncertain duration while awaiting pregnancy and ahigh value on minimizing the risks of early miscarriageassociated with vitamin K antagonist therapy are likelyto choose LMWH while attempting pregnancy.

3.0.3. For pregnant women, we suggest limitingthe use of fondaparinux and parenteral directthrombin inhibitors to those with severe allergicreactions to heparin (eg, HIT) who cannot receivedanaparoid(Grade 2C).

3.0.4. For pregnant women, we recommendavoiding the use of oral direct thrombin (eg,dabigatran) and anti-Xa (eg, rivaroxaban, apixa-ban) inhibitors(Grade 1C).

4.0.1. For lactating women using warfarin,acenocoumarol, or UFH who wish to breast-feed, we recommend continuing the use of war-farin, acenocoumarol, or UFH(Grade 1A).

Revision accepted August 31, 2011.

Affiliations: From the Department of Medicine (Dr Bates),McMaster University and Thrombosis and Atherosclerosis ResearchInstitute, Hamilton, ON, Canada; Faculty of Health and Life Sci-ences (Dr Greer), University of Liverpool, Liverpool, England;Department of Vascular Medicine (Dr Middeldorp), AcademicMedical Center, Amsterdam, The Netherlands; Department ofPharmacy (Dr Veenstra), University of Washington, Seattle, WA;Department of Obstetrics and Gynecology (Dr Prabulos), Universityof Connecticut School of Medicine, Farmington, CT; and MedicalDepartment (Dr Vandvik), Innlandet Hospital Trust and NorwegianKnowledge Centre for the Health Services, Gjvik, Norway.Funding/Support:The Antithrombotic Therapy and Prevention ofThrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines received support from theNational Heart, Lung, and Blood Institute [R13 HL104758] andBayer Schering Pharma AG. Support in the form of educational

grants were also provided by Bristol-Myers Squibb; Pfizer, Inc;Canyon Pharmaceuticals; and sanofi-aventis US.Disclaimer: American College of Chest Physician guidelines areintended for general information only, are not medical advice, anddo not replace professional medical care and physician advice, whichalways should be sought for any medical condition. The completedisclaimer for this guideline can be accessed at http://chestjournal.chestpubs.org/content/141/2_suppl/1S. Correspondence to:Shannon M. Bates, MDCM, Department ofMedicine, HSC 3W11, 1280 Main St W, Hamilton, ON, L8S 4K1,Canada; e-mail: [email protected] 2012 American College of Chest Physicians.Reproductionof this article is prohibited without written permission from theAmerican College of Chest Physicians (http://www.chestpubs.org/site/misc/reprints.xhtml ).DOI: 10.1378/chest.11-2300

http://chestjournal.chestpubs.org/content/141/2_suppl/1Shttp://chestjournal.chestpubs.org/content/141/2_suppl/1Smailto:[email protected]://www.chestpubs.org/site/misc/reprints.xhtmlhttp://www.chestpubs.org/site/misc/reprints.xhtmlhttp://www.chestpubs.org/site/misc/reprints.xhtmlhttp://www.chestpubs.org/site/misc/reprints.xhtmlhttp://www.chestpubs.org/site/misc/reprints.xhtmlhttp://www.chestpubs.org/site/misc/reprints.xhtmlmailto:[email protected]://chestjournal.chestpubs.org/content/141/2_suppl/1Shttp://chestjournal.chestpubs.org/content/141/2_suppl/1S


3/46


dose LMWH rather than clinical vigilance orroutine care(Grade 2C).

8.2.4. For pregnant women receiving long-termvitamin K antagonists, we suggest adjusted-doseLMWH or 75% of a therapeutic dose of LMWHthroughout pregnancy followed by resumptionof long-term anticoagulants postpartum, rather

than prophylactic-dose LMWH(Grade 2C).9.2.1. For pregnant women with no prior his-tory of VTE who are known to be homozygousfor factor V Leiden or the prothrombin 20210Amutation and have a positive family historyfor VTE, we suggest antepartum prophylaxis

with prophylactic- or intermediate-dose LMWHand postpartum prophylaxis for 6 weeks withprophylactic- or intermediate-dose LMWH orvitamin K antagonists targeted at INR 2.0 to 3.0rather than no prophylaxis(Grade 2B).

9.2.2. For pregnant women with all other throm-bophilias and no prior VTE who have a positivefamily history for VTE, we suggest antepartumclinical vigilance and postpartum prophylaxis

with prophylactic- or intermediate-dose LMWHor, in women who are not protein C or S defi-cient, vitamin K antagonists targeted at INR2.0 to 3.0 rather than routine care(Grade 2C).

9.2.3. For pregnant women with no prior his-tory of VTE who are known to be homozygousfor factor V Leiden or the prothrombin 20210A

mutation and who do not have a positive familyhistory for VTE, we suggest antepartum clin-ical vigilance and postpartum prophylaxis for6 weeks with prophylactic- or intermediate-dose LMWH or vitamin K antagonists tar-geted at INR 2.0 to 3.0 rather than routine care(Grade 2B).

9.2.4. For pregnant women with all other throm-bophilias and no prior VTE who do not have apositive family history for VTE, we suggest antepar-tum and postpartum clinical vigilance ratherthan pharmacologic prophylaxis(Grade 2C).

10.2.1. For women with recurrent early preg-nancy loss (three or more miscarriages before10 weeks of gestation), we recommend screeningfor APLAs(Grade 1B).

10.2.2. For women with a history of pregnancycomplications, we suggest not to screen for inher-ited thrombophilia(Grade 2C).

10.2.3. For women who fulfill the laboratory cri-teria for APLA syndrome and meet the clinical

6.2.3. For women undergoing cesarean sectionwho are considered to be at very high risk forVTE and who have multiple additional risk fac-tors for thromboembolism that persist in thepuerperium, we suggest that prophylactic LMWHbe combined with elastic stockings and/or inter-mittent pneumatic compression over LMWHalone(Grade 2C).

6.2.4. For selected high-risk patients in whomsignificant risk factors persist following delivery,

we suggest extended prophylaxis (up to 6 weeksafter delivery) following discharge from the hos-pital(Grade 2C).

7.1.1. For pregnant women with acute VTE, werecommend therapy with adjusted-dose sub-cutaneous LMWH over adjusted-dose UFH(Grade 1B).

7.1.2. For pregnant women with acute VTE, we

recommend LMWH over vitamin K antagonisttreatment antenatally(Grade 1A).

7.1.3. For pregnant women with acute VTE, wesuggest that anticoagulants should be continuedfor at least 6 weeks postpartum (for a minimumtotal duration of therapy of 3 months) in com-parison with shorter durations of treatment(Grade 2C).

7.1.4. For pregnant women receiving adjusted-dose LMWH therapy and where delivery is

planned, we recommend discontinuation ofLMWH at least 24 h prior to induction of laboror cesarean section (or expected time of neurax-ial anesthesia) rather than continuing LMWHup until the time of delivery(Grade 1B).

8.2.1. For all pregnant women with prior VTE,we suggest postpartum prophylaxis for 6 weekswith prophylactic- or intermediate-dose LMWHor vitamin K antagonists targeted at INR 2.0to 3.0 rather than no prophylaxis(Grade 2B).

8.2.2. For pregnant women at low risk of recur-rent VTE (single episode of VTE associated witha transient risk factor not related to pregnancyor use of estrogen), we suggest clinical vigilanceantepartum rather than antepartum prophylaxis(Grade 2C).

8.2.3. For pregnant women at moderate to highrisk of recurrent VTE (single unprovoked VTE,pregnancy- or estrogen-related VTE, or mul-tiple prior unprovoked VTE not receiving long-term anticoagulation), we suggest antepartumprophylaxis with prophylactic- or intermediate-



4/46


agulants should be resumed postpartum when ade-quate hemostasis is assured.

12.1.2. In women judged to be at very high riskof thromboembolism in whom concerns existabout the efficacy and safety of UFH or LMWHas dosed above (eg, older generation prosthe-sis in the mitral position or history of thrombo-

embolism), we suggest vitamin K antagoniststhroughout pregnancy with replacement by UFHor LMWH (as above) close to delivery ratherthan one of the regimens above(Grade 2C).

Remarks:Women who place a higher value on avoid-ing fetal risk than on avoiding maternal complications(eg, catastrophic valve thrombosis) are likely to chooseLMWH or UFH over vitamin K antagonists.

12.1.3. For pregnant women with prostheticvalves at high risk of thromboembolism, wesuggest the addition of low-dose aspirin, 75 to100 mg/d(Grade 2C).

This article is devoted to the use of antithrombotictherapy in pregnant women. Anticoagulant ther-

apy is indicated during pregnancy for the preventionand treatment of VTE; for the prevention and treat-ment of systemic embolism in patients with mechan-ical heart valves; and, in combination with aspirin,for the prevention of recurrent pregnancy loss in

women with antiphospholipid antibodies (APLAs).The use of anticoagulation for prevention of preg-

nancy complications in women with hereditary throm-bophilia is becoming more frequent. Given theabsence of proven-effective therapy in women withunexplained recurrent pregnancy loss, there is alsogrowing pressure to intervene with antithrombotictherapy in affected women with no known underly-ing thrombophilia. The use of anticoagulant therapyduring pregnancy is challenging because of the poten-tial for fetal and maternal complications.

1.0Methods

Table 1 describes both the question definition (ie, population,intervention, comparator, and outcomes) and the eligibility cri-teria for studies considered in each section of the recommenda-tions that follow. We consider the desirable and undesirable fetaland maternal consequences of antithrombotic therapy in thefollowing populations: (1) breast-feeding women, (2) women usingassisted reproductive technology, (3) women undergoing cesareansection, (4) pregnant women with newly diagnosed VTE, (5) preg-nant women with prior VTE, (6) pregnant women with asymptom-atic thrombophilia, (7) pregnant women with a history of pregnancycomplications (including pregnancy loss, preeclampsia, fetal growthrestriction, and placental abruption), and (8) pregnant women withmechanical heart valves.

APLA criteria based on a history of three or morepregnancy losses, we recommend antepartumadministration of prophylactic- or intermediate-dose UFH or prophylactic LMWH combined

with low-dose aspirin, 75 to 100 mg/d, over notreatment(Grade 1B).

10.2.4. For women with inherited thrombo-

philia and a history of pregnancy complications,we suggest not to use antithrombotic prophy-laxis(Grade 2C).

11.1.1. For women considered at risk for pre-eclampsia, we recommend low-dose aspirinthroughout pregnancy, starting from the secondtrimester, over no treatment(Grade 1B).

11.2.1. For women with two or more miscar-riages but without APLA or thrombophilia, werecommend against antithrombotic prophylaxis(Grade 1B).

12.1.1. For pregnant women with mechanicalheart valves, we recommend one of the followinganticoagulant regimens in preference to no anti-coagulation(all Grade 1A):

(a) Adjusted-dose bid LMWH throughout preg-nancy. We suggest that doses be adjusted toachieve the manufacturers peak anti-Xa LMWH4 h postsubcutaneous-injection or

(b) Adjusted-dose UFH throughout pregnancy

administered subcutaneously every 12 h in dosesadjusted to keep the mid-interval aPTT at leasttwice control or attain an anti-Xa heparin levelof 0.35 to 0.70 units/mL or

(c) UFH or LMWH (as above) until the 13th week,with substitution by vitamin K antagonistsuntil close to delivery when UFH or LMWH isresumed.

Remarks:For pregnant women with mechanical heartvalves, the decision regarding the choice of anticoag-

ulant regimen is so value and preference dependent(risk of thrombosis vs risk of fetal abnormalities) thatwe consider the decision to be completely individ-ualized. Women of childbearing age and pregnant

women with mechanical valves, should be counseledabout potential maternal and fetal risks associated

with various anticoagulant regimens, including con-tinuation of vitamin K antagonists with substitutionby LMWH or UFH close to term, substitution of

vitamin K antagonists by LMWH or UFH until the13th week and then close to term, and use of LMWHor UFH throughout pregnancy. Usual long-term antico-



5/46
http://www.chestpubs.org/


6/46


7/46


PICOQuestion

Section

InformalQuestion

Population

Intervention

Comparator

Outcome

Methodology

Roleofvena

cavalfilters

whenantithrombotic

therapyis

contraind

icated

Pregnantwomen

withprovenacute

VTE

Venalcavalfilter

Novenacavalfilter

Symptomaticrecurrent

DVTorpulmonary

embolism

Fatalpulmonaryembolism

Majorbleeding

Postthromboticsyndrome

Randomizedcontrolled

trials

Observationalstudies

-Caseseries

-Cohortstudies

-Case-controlstudies

Managemen

tof

antithrom

botictherapy

arounddelivery

Pregnantwomen

withprovenacute

VTE

Electivedeliveryewith

discontinuationof

antithrombotic

therapy24to48h

priortodelivery

Noelectivedelivery,e

transitionto

unfractionated

heparin

Noelectivedelivery,e

transitionto

prophylacticdoseof

antithromboticagent

Noelectivedeliverye

withdiscontinuation

ofantithrombotic

therapyassoonas

laborcommences

Otherintervention


DVTorpulmonary

embolism

Majorbleeding:totald

Epiduralhematoma



trials


-Caseseries

-Cohortstudies


Preventionofrecurrent

VTEinpregnant

womenwithprior

VTE(section8.0)

RiskofrecurrentVTEin

pregnant

womenwith:

-Asingleunprovokedevent

-Asingleeventthatwas

associatedwitha

transient

riskfactor

(all,estro

gen-related

[OCP,pregnancy])

-Multiplepriorevents

-Thrombophiliab

Pregnantwomen

withpriorVTE

Noprophylaxis

Nointervention

SymptomaticDVT,

pulmonaryembolism

Mortality

Majorbleeding:total


Controlarmsof

randomized

controlledtrials


-Caseseries

-Cohortstudies


Choiceand(ifappropriate)

routeand

doseof

antithrom

botic

prophylaxis

Pregnantwomen

withpriorVTE

Noantepartum

prophylaxis,

postpartumonly

-Allrelevantagents

considereda

Antepartumand

postpartumprophylaxis

-Allrelevantagents

considereda

Noprophyla

xis


DVTorpulmonary

embolism

Majorbleeding:total



trials


-Caseseries

-Cohortstudies


Table1Continued

(Continued)



8/46


PICOQuestion

Section

InformalQuestion

Population

Intervention

Comparator

Outcome

Methodology

Preventionof

pregnancy-related

VTEinwomen

withthrombophilia

(section9.0)

Riskofpregnancy-related

VTEinw

omenwith

thrombophiliabQ16

Pregnantwomen

withthrombophiliab

andnopriorVTE

Noprophylaxis

Nointervention

SymptomaticDVT,

pulmonaryembolism

Mortality

Majorbleeding

Controlarmsof

randomized

controlledtrials


-Caseseries

-Cohortstudies


Choice,duration,and(if

appropriate)route/dose

ofprophy

laxis

Pregnantwomen

withthrombophiliab

andnopriorVTE

Noantepartum

prophylaxis,

postpartumonly

-Low-molecular-weight

heparin

-Unfractionatedheparin

-Otherrelevantagentsa

-Graduatedcompression

stockings

-Combinedmechanical

andpharmacologic

prophylaxis

Antepartumand

postpartum

prophylaxis

-Similaragentsasabove

Noprophyla

xisor

Otherinterv

ention

SymptomaticDVT,

pulmonaryembolism

Mortality

Majorbleeding


trials


-Caseseries

-Cohortstudies


Preventionofpregnancy

complicationsin

womenwith

thrombophiliab

(section10.0)

Riskofpregnancy

complicationsin

womenw

ith

thrombophiliab

Pregnantwomenwith

thrombophiliab

andahistory

ofpregnancy

complications

-Recurrentearly

pregnancylossf

-Latepregnancyloss

(single)g

-Latepregnancyloss

(multiple)h

-Pre-eclampsia

-Intrauterinegrowth

restriction

-Placentalabruption

Noprophylaxis

Nointervention

Recurrentpregnancy

complication(asdefined

underpatientpopulation)

SymptomaticDVT,

pulmonaryembolism

Mortality

Majorbleeding

Controlarmof

randomized

controlledtrials


-Caseseries

-Cohortstudies


Table1Continued

(Continued)



9/46


PICOQuestion

Section

InformalQuestion

Population

Intervention

Comparator

Outcome

Methodology


routeand

durationof

antithrom

botic

prophylaxis

Pregnantwomen

withthrombophiliab

(antiphospholipid

antibodies

vscongenital

thrombophiliavs

specificcongenital

thrombophilia)anda

historyofpregnancy

complications

-Recurrentearly

pregnancylossf

-Latepregnancyloss

(single)g

-Latepregnancyloss

(multiple)h

-Preeclampsia

-Intrauterinegrowth

restriction

-Placentalabruption

Aspirin

Unfractionated

heparin(aspirin)

Low-molecular-weight

heparin(aspirin)

Noprophyla

xisor

Otherantith

rombotic

strategy

Recurrentpregnancy

complication(asdefined

underpatientpopulation)

SymptomaticDVT,

pulmonaryembolism

Mortality

Majorbleeding


trials


-Caseseries

-Cohortstudies


Preventionofrecurrent

preeclampsiaor

recurrent

pregnancylossin

womenwithout

knownthrombophiliab

(section11.0)


routeand

durationof

antithrom

botic

prophylaxis

Pregnantwomen

withnoknown

thrombophiliaband

priorpreeclampsia

Pregnantwomen

withnoknown

thrombophiliaand

atleasttwoprior

pregnancylosses

Aspirin

Unfractionated

heparin(aspirin)

Low-molecular-weight

heparin(aspirin)

Noprophyla

xis

Recurrentpreeclampsia

Recurrentpregnancyloss


trials


-Caseseries

-Cohortstudies


Preventionof

thromboembolismin

pregnantwomenwith

mechanicalheart

valves(section12.0)

Riskofthromboembolism

inpregna

ntwomen

withmechanicalheart

valves

Pregnantwomen

withmechanical

heartvalves

Noantithrombotic

therapy

Nointervention

Maternalthromboembolism

Majorbleeding:total

Majorbleeding:maternal

death

Congenitalmalformations

Fetal/neonatalhemorrhage

Pregnancyloss

Controlarmof

randomized

controlledtrials


-Caseseries

-Cohortstudies


Table1Continued

(Continued)



10/46


11/46


Adjusted-dose LMWH: weight-adjusted or full-treatmentdoses of LMWH given once daily or bid (eg, dalteparin200 units/kg or tinzaparin 175 units/kg once daily or dalteparin100 units/kg every 12 h or enoxaparin 1 mg/kg every 12 h)

Postpartum anticoagulation refers to vitamin K antagonistsfor 6 weeks with a target INR of 2.0 to 3.0, with initial UFHor LMWH overlap until the INR is 2.0 or prophylactic- orintermediate-dose LMWH for 6 weeks. The term clinical vigi-lance refers to patient and physician alertness to the signs and

symptoms of VTE and awareness of the need for timely andappropriate objective investigation of women with symptoms sus-picious of DVT or PE. A family history of VTE refers to DVT orPE in a first-degree relative.

1.1 The Implications of Womens Preferencesand Values During Pregnancy

In considering womens choices regarding risks and benefits ofantithrombotic therapy in pregnancy, two considerations are ofparticular importance. First, treatment decisions during preg-nancy and breast-feeding have implications not only for the healthand life of the mother but also for the health and life of the fetusor child. Second, many women prefer to see pregnancy as a nor-

mal part of a healthy womans life course rather than as a medicalcondition. On the background of these considerations, many fac-tors, including the frequency and type of medication adminis-tration; pain, discomfort, and possible side effects; and the need,frequency, and type of testing associated with a given regimen,

will affect womens choices.The weight given to harmful effects (eg, maternal bleeding

events, congenital malformations) and burden of treatment (eg,self-injecting with LMWH for 9 months) compared with benefi-cial effects (eg, avoiding VTE or pregnancy loss) affects trade-offsbetween benefits and harms of antithrombotic treatment in preg-nancy. A systematic review of patient preferences for antithrom-botic treatment did not identify any studies of pregnant women.5The findings of this systematic review, and the value and prefer-ence rating exercise described in Guyatt et al4 suggest that one

VTE should be viewed as more or less equivalent to one majorextracranial bleed. Our clinical experience and preliminary resultsfrom a cross-sectional interview study (S. M. Bates, MDCM, per-sonal communication, March 27, 2011) to determine the willingnessof women with prior VTE who are either pregnant, actively plan-ning a pregnancy, or who may in the future consider pregnancyto receive LMWH prophylaxis during pregnancy for preventionof recurrent VTE suggest that many, but not all women will chooselong-term prophylaxis when confronted with the burden of self-injecting with LMWH over several months. Therefore, in general,

we only make weak recommendations for long-term prophylaxiswith LMWH.

In addition, the burden of long-term prophylaxis or treatmentwith LMWH or warfarin throughout pregnancy will have an

impact on the choice of antithrombotic therapy. Clinical experi-ence suggests that many, but not all women give higher priority tothe impact of any treatment on the health of their unborn babythan to effects on themselves, placing a low value on avoiding thepain, cost, and inconvenience of heparin therapy in order to avoidthe small risk of even a minor abnormality in their child. Attemptsto balance the burden of long-term prophylaxis against the dis-utility associated with VTE or major bleeding events are furthercomplicated by the fact that all pregnant women will experiencethe disutility of long-term prophylaxis, whereas only a minority

will avoid VTE with treatment (because the baseline risk of suchevents generally is low).

Recommendations in this article, therefore, reflect our beliefthat although average women considering antithrombotic therapy

In addition to considering fetal outcomes (eg, pregnancy loss,congenital malformations) and maternal outcomes (eg, mortality,

VTE, major maternal hemorrhage), we also consider burden oftreatment as an important outcome for pregnant women takinglong-term low-molecular-weight heparin (LMWH) or warfarin.

When considered relevant, we report deaths (preferably as dis-ease and treatment-specific mortality). Maternal thromboem-bolism includes DVT and pulmonary embolism (PE) in sectionsdiscussing the treatment and prevention of VTE and systemicembolization and valve thrombosis in sections discussing the

management of pregnant women with mechanical heart valves.Major nonfatal maternal hemorrhage is defined as a symptomaticbleeding complication noted during pregnancy or within 6 weekspostpartum that involves bleeding into a critical site (intracranial,intraspinal, intraocular resulting in vision changes, retroperitoneal,pericardial, intramuscular with compartment syndrome, or pla-cental abruption), causing a fall in hemoglobin level of 20 g/L,and bleeding leading to transfusion of two or more units of wholeblood or red cells. This definition is in part based on the definitionrecommended by the International Society on Thrombosis andHaemostasis.1Where major bleeding was not explicitly definedin primary research articles, the authors definition was accepted.Fetal loss refers to loss at any time after confirmation of a viableintrauterine pregnancy, not including elective termination.

A comprehensive English-language literature search (January 2005-January 2010) was conducted to update our existing literature base.We followed the approach articulated by Grades of Recommen-dations, Assessment, Development, and Evaluation for formula-tion of recommendations.2-4 In making recommendations, we haveplaced the burden of proof with those who would claim a benefit oftreatment. Therefore, when there is uncertain benefit and a prob-ability of important harm associated with therapy, we generallyrecommend against intervention.

There is a paucity of high-quality studies addressing risk factorsfor the outcomes discussed in this article as well as for the risksand benefits of antithrombotic therapy during pregnancy. Mostrecommendations, therefore, are based on low- to moderate-quality evidence and mirror our limited confidence in relativeeffect estimates from studies of antithrombotic treatment dur-ing pregnancy. To obtain baseline risk estimates for pregnancycomplications, we summarize available observational studiesof pregnant women, including case reports and case series ofpregnant women in the absence of studies with a cohort design.

We then apply the baseline risk estimates to the relative risk esti-mates to establish anticipated absolute benefits and harms ofintervention. In the absence of direct evidence from random-ized trials of reasonable quality, indirect evidence from random-ized trials in nonpregnant patients is considered applicable tothe present patient population (eg, we extrapolate the effectof thromboprophylaxis with LMWH on the incidence of VTEin patients undergoing general surgery to women undergoingcesarean section).

When describing the various regimens of unfractionated hep-

arin (UFH) and LMWH, we use the following short forms:

Adjusted-dose UFH: UFH subcutaneously every 12 h indoses adjusted to target a midinterval activated partial thrombo-plastin time (aPTT) into the therapeutic range

Prophylactic LMWH: for example, dalteparin 5,000 unitssubcutaneously every 24 h, tinzaparin 4,500 units subcutane-ously every 24 h, nadroparin 2,850 units subcutaneouslyevery 24 h, or enoxaparin 40 mg subcutaneously every 24 h(although at extremes of body weight, modification of dosemay be required)

Intermediate-dose LMWH: for example, dalteparin 5,000 unitssubcutaneously every 12 h or enoxaparin 40 mg subcutane-ously every 12 h



12/46


prolongation of the aPTT persisted for up to 28 hafter the last injection of adjusted-dose subcutaneousheparin.11The mechanism for this prolonged effectis unclear. A similar effect has not been noted with IVUFH.

Thrombocytopenia during pregnancy is notuncommon, and pregnancy-specific causes12 shouldbe differentiated from IgG-mediated thrombocyto-

penia or HIT, which occurs in 3% of nonpregnantpatients receiving UFH.13The diagnosis, prevention,and treatment of HIT are described in Linkins et al14in these guidelines. In pregnant women who developHIT and require ongoing anticoagulant therapy,use of the heparinoid danaparoid sodium is recom-mended because it is an effective antithromboticagent15 that does not cross the placenta16-18 and haslow cross-reactivity with UFH19; therefore, it rarelyproduces HIT (danaparoid was withdrawn from theUS market in 2002). Although there are reports offondaparinux20,21being used for this indication in

pregnancy, experience with this agent during preg-nancy is too limited to recommend fondaparinux overdanaparoid.

Long-term treatment with UFH has been reportedto cause osteoporosis in both laboratory animals andhumans.22-30A number of studies have attempted toquantify the risk of osteoporosis during prolongedtreatment (.1 month) with UFH. Symptomatic ver-tebral fractures have been reported to occur in 2%to 3% of the patient population, and significantreductions of bone density have been reported in upto 30%.22,23A small study (n540) reported an even

higher percentage of fractures (15%) when oldernonpregnant patients were treated with bid subcu-taneous injections of 10,000 units UFH for a periodof 3 to 6 months.26

Adverse skin reactions to UFH include bruising,urticarial rashes, erythematous well-circumscribedlesions (because of a delayed type 4 hypersensitivityreaction), skin necrosis (often due to vasculitis), andHIT. The true incidence of skin reactions caused byUFH is unknown.31

2.2 LMWH Therapy

Despite a paucity of supportive data from con-trolled trials or even large prospective observationalstudies, LMWH is now commonly used for prophy-laxis and treatment of maternal thromboembolism.This change in practice is based largely on the resultsof large trials in nonpregnant patients, showing thatLMWHs are at least as safe and effective as UFHfor the treatment of VTE32,33and acute coronary syn-dromes34as well as for prophylaxis in high-risk patients.35

Retrospective analyses and systematic reviewssuggest that the incidence of bleeding in pregnant

will also want to avoid medicalizing their pregnancy, they will putan extremely high value on avoiding fetal risk. For women whodo not share these values, some of the recommendations in thisarticle may not apply. For most recommendations, optimal decision-making will require that physicians educate patients about theirtreatment options, including their relative effectiveness, the con-sequences for both mother and baby, the method of administra-tion and monitoring, the likely side effects, and the uncertaintyassociated with the estimates of all these effects. Once educated,

women can participate in the selection of the treatment regimen

that best matches their preferences and values.

2.0Maternal Complications ofAnticoagulant Therapy

Maternal complications of anticoagulant therapyare similar to those seen in nonpregnant patients andinclude bleeding (for all anticoagulants) as well asheparin-induced thrombocytopenia (HIT), heparin-associated osteoporosis, bruising, local allergic reac-tions, and pain at injection sites for heparin-related

compounds.

2.1 UFH Therapy

During pregnancy, UFH can be used for both pre-vention and treatment of thromboembolism. Prophy-lactic UFH is typically administered subcutaneouslytwo to three times per day either in fixed doses ordoses adjusted to a target a specific anti-Xa UFHlevel (prophylactic- or intermediate-dose UFH). Whenused in therapeutic doses, UFH is administeredeither intravenously by continuous infusion with dose

adjustment to achieve a target therapeutic aPTT orsubcutaneously by bid injections in doses sufficientto achieve a therapeutic aPTT 6 h after injection.

During pregnancy, the aPTT response to UFHoften is attenuated likely because of increased levelsof heparin-binding proteins, factor VIII, and fibrin-ogen.6Consequently, the use of an aPTT range thatcorresponds to therapeutic heparin levels in non-pregnant patients might result in higher dosing (andheparin levels) in pregnant women than in non-pregnant patients. However, it is not clear whetherthis translates into excessive bleeding because the

reported rates of bleeding using the standard aPTTrange appear to be low. In a retrospective cohortstudy of 100 pregnancies in 77 women,7the rate ofmajor antepartum bleeding in pregnant womentreated with UFH was 1% (95% CI, 0.2%-5.4%),

which is consistent with reported rates of bleedingassociated with heparin therapy in nonpregnantpatients8and with warfarin therapy9,10when used forthe treatment of DVT.

Therapeutic doses of subcutaneous UFH can causea persistent anticoagulant effect, which can compli-cate its use prior to delivery. In a small cohort study,



13/46


3.0Fetal Complications of AntithromboticTherapy During Pregnancy

3.1 Vitamin K Antagonist Exposure In Utero

Vitamin K antagonists cross the placenta and havethe potential to cause fetal wastage, bleeding in thefetus, and teratogenicity.47-58 In a systematic review ofthe literature published between 1966 and 1997 that

examined fetal and maternal outcomes of pregnantwomen with prosthetic valves, Chan and colleagues49provided pooled estimates of risks associated with thefollowing approaches: (1) use of vitamin K antago-nists throughout pregnancy, (2) replacement of vita-min K antagonists with UFH from 6 to 12 weeks, and(3) UFH use throughout pregnancy (Tables S1, S2)(tables that contain an S before the number denotesupplementary tables not contained in the body ofthe article and available instead in an online data sup-plement; see the Acknowledgments for more infor-mation). The authors found that the use of vitamin K

antagonists throughout pregnancy was associatedwith congenital anomalies in 35 of 549 live births(6.4%; 95% CI, 4.6%-8.9%). A subsequent systematicreview covering the years 2000 to 2009 (Tables S1, S2)reported a slightly lower risk estimate (21/559 [3.7%];95% CI, 1.9%-4.8%).50

In both reviews, the most common fetal anomalywas coumarin or warfarin embryopathy consistingof midfacial hypoplasia and stippled epiphyses. Limbhypoplasia has been reported in up to one-third ofcases of embryopathy.51 Embryopathy typically occursafter in utero exposure to vitamin K antagonists

during the first trimester of pregnancy.48The resultsof a recently published multicenter European studynot included in the systematic reviews, in which thepregnancies of 666 consenting women who contactedone of 12 Teratology Information Services between1988 and 2004 seeking advice about gestationalexposure to vitamin K antagonists were prospectivelyfollowed, also suggests that the risk of coumarinembryopathy is not high.58Although the frequency ofmajor birth defects after any first trimester exposureto vitamin K antagonists was increased compared

with that seen in a control group of 1,094 women

counseled during pregnancy about exposures knownto be nonteratogenic (4.8% vs 1.4%, respectively;OR, 3.86; 95% CI, 1.86-8.00), there were only twocases of embryopathy among 356 live births (0.6%).Both cases involved exposure to phenprocoumon untilat least the end of the first trimester.

The substitution of heparin at or prior to 6 weeksappears to eliminate the risk of embryopathy, raisingthe possibility that vitamin K antagonists are safe

with regard to embryopathy during the first 6 weeksof gestation. In the systematic review by Chan andcolleagues,49none of the 125 live births (95% CI,

women receiving LMWH is low.36-38 A systematicreview of 64 studies that included 2,777 pregnanciesin which LMWH was used reported that the fre-quencies of significant bleeding were 0.43% (95% CI,0.22%-0.75%) for antepartum hemorrhage, 0.94%(95% CI, 0.61%-1.37%) for postpartum hemorrhage,and 0.61% (95% CI, 0.36%-0.98%) for wound hema-toma, giving an overall frequency of 1.98% (95% CI,

1.50%-2.57%).38The risk of HIT appears much lowerwith LMWH than with UFH.13,37,38

Evidence suggests that LMWHs carry a lower riskof osteoporosis than UFH. In a study by Monreal andcolleagues26 in which 80 patients (men and women;mean age, 68 years) with DVT were randomized toeither subcutaneous dalteparin 5,000 units bid (inter-mediate dose) or subcutaneous UFH 10,000 unitsbid for a period of 3 to 6 months, the risk of vertebralfractures with UFH (six of 40 [15%] patients; 95% CI,6%-30%) was higher than with dalteparin (one of40 [3%] patients; 95% CI, 0%-11%). In another ran-

domized trial of 44 pregnant women allocated toprophylactic doses of dalteparin (n5 21) or UFH(n523),27bone density did not differ between womenreceiving dalteparin and those in a concurrent non-randomized cohort of healthy pregnant women but

was significantly lower in those receiving UFH. Aprospective observational study in which 55 pregnant

women treated with prophylactic LMWH and aspirinand 20 pregnant untreated volunteers reported sim-ilar results.39Finally, in an a priori substudy of anongoing randomized comparison of prophylacticLMWH (subcutaneous dalteparin 5,000 units/d) with

placebo for prevention of pregnancy complications,there was no difference between the two groups withrespect to mean bone mineral density.40

Despite these reassuring data, there have beencase reports41-44 of symptomatic osteoporosis occur-ring with LMWH. Osteoporosis may be due to indi-

vidual susceptibility, reflecting the presence of riskfactors for osteoporosis, a variable effect of differentLMWH preparations or doses on bone density, or acombination of both. Risk factors that make womensusceptible to this complication when exposed toLMWH in pregnancy remain to be identified.

Adverse skin reactions similar to those seen withUFH can also occur with LMWH, although the fre-quency appears reduced in patients receiving thelatter. The reported incidence ranges from 1.8% to29%.38,45 Most LMWH-induced skin lesions arebenign; however, HIT should be excluded.46

Recommendation

2.2.1. For pregnant patients, we recommendLMWH for the prevention and treatment of

VTE, instead of UFH(Grade 1B).



14/46


vitamin K antagonist therapy before pregnancy occurs.If pregnancy is still desired, the following two optionscan reduce the risk of warfarin embryopathy:

1. Performance of frequent pregnancy tests andsubstitution of adjusted-dose LMWH or UFHfor warfarin when pregnancy is achieved or

2. Replacement of vitamin K antagonists with

LMWH or UFH before conception is attemptedBoth approaches have limitations. The first assumes

that vitamin K antagonists are safe during the first 4 to6 weeks of gestation. Although the second approachminimizes the risks of early miscarriage associated with

vitamin K antagonist therapy, it lengthens the dura-tion of exposure to heparin and, therefore, is costly andexposes the patient to a greater burden of treatmentassociated with the use of parenteral anticoagulants.

3.2 UFH Exposure In Utero

UFH does not cross the placenta63

and, therefore,does not have the potential to cause fetal bleeding orteratogenicity; although bleeding at the uteroplacental

junction is possible. Several studies provide high-qualityevidence that UFH therapy is safe for the fetus.7,47,64

3.3 LMWH Exposure In Utero

As determined by measurement of anti-Xa activityin fetal blood, LMWH also does not cross the pla-centa.65,66 There is no evidence that LMWH causesteratogenicity or increases the risk of fetal bleeding.36

3.4 Danaparoid Exposure In UteroAnimal experiments and human case reports sug-

gest negligible transport of danaparoid across theplacenta16-18,67 Thus, there is no demonstrable fetaltoxicity with maternal danaparoid use. However, thequality of evidence available to support that claimis low. (Note: Danaparoid was withdrawn from theUS market in 2002.)

3.5 Pentasaccharide Exposure In Utero

Although no placental passage of fondaparinux was

demonstrated in a human cotyledon (small lobeon the uterine or maternal surface of the placenta)model,68anti-Xa activity (at approximately one-tenththe concentration of maternal plasma) was found inthe umbilical cord plasma of five newborns of motherstreated with fondaparinux.69 Although there havebeen a small number of reports of the successful useof this agent in pregnant woman,70-76most of theseinvolve second trimester or later exposure. Thus,the quality of evidence regarding supporting use offondaparinux in pregnancy is very low. Potential del-eterious effects on the fetus cannot be excluded.

0%-3.0%) in which vitamin K antagonists werereplaced with UFH at or before 6 weeks gestation orUFH used throughout pregnancy was associated withcongenital fetal anomalies. In the European multicenterTeratology Information Services study, there were nocases of embryopathy among 235 live births when

vitamin K antagonists were discontinued before week 8after the first day of the last menstrual period.58

Vitamin K antagonists have also been associatedwith CNS abnormalities after exposure during anytrimester.48Two patterns of CNS damage have beendescribed: dorsal midline dysplasia (agenesis of thecorpus callosum, Dandy-Walker malformation, andmidline cerebellar atrophy) and ventral-midline dys-plasia leading to optic atrophy.48These complicationsare uncommon.48,49

Although one cohort study reported that the use ofcoumarins during the second and third trimester

was not associated with major risks for abnormalitiesin growth and long-term development of offspring,

the authors noted an increased risk of minor neuro-developmental problems (OR, 1.7; 95% CI, 1.0-3.0)in children exposed to coumarins in the second andthird trimester of pregnancy compared with age-matched nonexposed children (14% vs 8%, respec-tively).59However, these minor neurodevelopmentalproblems are likely of minor importance becausethere were no differences in mean IQ or performanceon tests for reading, spelling, and arithmetic betweenexposed and nonexposed children.60

Vitamin K antagonists have been linked to anincreased risk of pregnancy loss49,50,58,61and can cause

fetal hemorrhagic complications likely because thefetal liver is immature and fetal levels of vitaminK-dependent coagulation factors are low. Fetal coag-ulopathy is of particular concern at the time ofdelivery when the combination of the anticoagulanteffect and trauma of delivery can lead to bleeding inthe neonate. The risk of delivering an anticoagulatedinfant can be reduced by substituting UFH or LMWHfor vitamin K antagonists approximately 3 weeks priorto planned delivery61and discontinuing these medi-cations shortly before delivery. Others have advocatedthe use of planned cesarean section at 38 weeks with

only a brief (2 to 3 day) interruption of anticoagulanttherapy.62This approach resulted in good neonatal andmaternal outcomes in a study of 30 babies. Cesareansection is not without risk and is not recommendedfor other conditions associated with an increased riskof neonatal intracranial hemorrhage at the time ofdelivery (eg, immune thrombocytopenia purpura).

3.1.1 Thromboprophylaxis in Women Using Vitamin KAntagonists and Planning Pregnancy: Physicians shouldcounsel women receiving vitamin K antagonist ther-apy and contemplating pregnancy about the risks of



15/46


3.9 Thrombolysis During Pregnancy

Although investigations with 131I-labeled streptoki-nase or tissue plasminogen activator showed minimaltransplacental passage,91concerns remain about theuse of thrombolytic therapy during pregnancy due tomaternal and placental effects. Although there havebeen reports of successful thrombolysis in pregnancy(most involving streptokinase),91-94 the number ofcases is small. Given this and limitations of availabledata regarding the safety of this intervention in preg-nancy, the use of thrombolytic therapy is best reservedfor life-threatening maternal thromboembolism.95

Recommendations

3.0.1. For women receiving anticoagulation forthe treatment of VTE who become pregnant,

we recommend that LMWH over vitamin Kantagonists during the first trimester (Grade 1A),in the second and third trimesters(Grade 1B),

and during late pregnancy when delivery isimminent(Grade 1A).

3.0.2. For women requiring long-term vitaminK antagonists who are attempting pregnancyand are candidates for LMWH substitution, wesuggest performing frequent pregnancy testsand substituting LMWH for vitamin K antago-nists when pregnancy is achieved rather thanswitching to LMWH while attempting pregnancy(Grade 2C).

Remarks:Women who place little value on avoid-ing the risks, inconvenience, and costs of LMWHtherapy of uncertain duration while awaiting pregnancyand a high value on minimizing the risks of earlymiscarriage associated with vitamin K antagonist ther-apy will probably choose LMWH while attemptingpregnancy.

3.0.3. For pregnant women, we suggest limitingthe use of fondaparinux and parenteral directthrombin inhibitors to those with severe allergicreactions to heparin (including HIT) who cannotreceive danaparoid(Grade 2C).

3.0.4. For pregnant women, we recommendavoiding the use of oral direct thrombin (eg,dabigatran) and anti-Xa (eg, rivaroxaban, apixa-ban) inhibitors(Grade 1C).

4.0Use of Anticoagulants inBreast-feeding Women

In order for a drug to pose a risk to the breast-fedinfant, not only must it be transferred and excretedinto breast milk but also it must be absorbed from

3.6 Parenteral Direct Thrombin InhibitorExposure In Utero

Investigations have documented placental transferof r-hirudin in rabbits and rats.77,78Although smallnumbers of case reports have documented successfuloutcomes with r-hirudin use in pregnancy,77,79,80thereare insufficient data to evaluate its safety. Three casereports have been published describing the use ofargatroban late in pregnancy.81-83There are no pub-lished reports on the use of bivalirudin.

3.7 New Oral Direct Thrombin and Anti-XaInhibitor Exposure In Utero

Pregnant women were excluded from participatingin clinical trials evaluating these new agents. Thereare no published reports describing the use of neworal direct thrombin inhibitors (eg, dabigatran) oranti-Xa inhibitors (rivaroxaban, apixaban) in preg-nancy. The Summaries of Product Characteristics

for dabigatran and rivaroxaban describe animal repro-ductive toxicity.84,85The human reproductive risks ofthese medications are unknown.

3.8 Aspirin Exposure In Utero

Aspirin crosses the placenta, and animal studieshave shown that aspirin may increase the risk of con-genital anomalies. Several systematic reviews haveexamined the safety of aspirin use during pregnancy(Tables S1-S3).86-88 A meta-analysis of 31 randomizedstudies comparing antiplatelet agents with either pla-cebo or no antiplatelet agents in 32,217 pregnant

women at risk for developing preeclampsia86reportedthat aspirin therapy was not associated with anincrease in the risk of pregnancy loss, neonatal hem-orrhage, or growth restriction. However, in a meta-analysis of eight studies that evaluated the risk ofcongenital anomalies with aspirin exposure duringthe first trimester, aspirin use was associated witha twofold increase in the risk for gastroschisis (OR,2.37; 95% CI, 1.44-3.88).87 The validity of this riskestimate is questionable because of a significant riskof bias in the contributing studies.

One population-based study did note an increased

risk of miscarriage with aspirin use that was greatestwhen aspirin was taken around the time of concep-tion89; however, the number of aspirin users was small,aspirin doses were unknown, and users may have hadconditions associated with an increased risk of preg-nancy loss.90 A meta-analysis of seven randomizedtrials in which women started aspirin later in preg-nancy (Tables S1, S3) failed to establish or refute anincrease in risk of miscarriage with aspirin compared

with placebo (risk ratio [RR], 0.92; 95% CI, 0.71-1.19for first or second trimester exposure; RR, 1.3;95% CI, 0.63-2.69 for first trimester exposure only).88



16/46


on the excretion of fondaparinux into human milk,and the effects on the nursing infant are unknown.As a negatively charged oligosaccharide, only minoramounts of fondaparinux are expected to pass theintestinal epithelial barrier after oral administration,and significant absorption by the nursing infant isunlikely.105However, the manufacturer recommendsthat caution be used when administering fondaparinux

to breast-feeding women.

4.5 Use of Parenteral Direct ThrombinInhibitors in Breast-feeding Women

In a single-case report, no r-hirudin was detectedin the breast milk of a nursing mother with a thera-peutic plasma hirudin level.106Enteral absorption ofr-hirudin appears to be low.78Therefore, it is unlikelythat exposed infants would experience a significantanticoagulant effect, even if small amounts of r-hirudinappear in breast milk.

4.6 Use of New Oral Direct Thrombin andFactor Xa Inhibitors in Breast-feeding Women

Breast-feeding women were excluded from trialsevaluating new oral direct thrombin and anti-Xainhibitors, and there are no clinical data on the effectof these agents on breast-fed infants. The Summaryof Product Characteristics for rivaroxaban notes thatanimal data indicate that this agent is secreted intobreast milk.85The manufacturers of dabigatran andrivaroxaban both recommend against using thesemedications in breast-feeding women.84,85

4.7 Use of Aspirin in Breast-feeding Women

Although aspirin is a polar, acidic drug that is poorlylipid soluble and highly bound to plasma proteins,maternal aspirin ingestion is associated with excre-tion of salicylates into breast milk.107There are, there-fore, potential risks of platelet dysfunction and GIbleeding in nursing infants of mothers using highdoses of this drug.107,108Metabolic acidosis has beenreported in breast-fed infants of mothers takingseveral grams of aspirin per day.109,110Theoretically,nursing infants of mothers taking aspirin could be at

risk for developing Reye syndrome.107The use of low-dose aspirin (,100 mg/d) late in pregnancy was notassociated with significant effects on neonatal plateletfunction.111,112In a prospective study of 15 breast-feeding mothers taking aspirin therapy, no negativeeffects were noted (Tables S4, S5).113

Recommendations

4.0.1. For lactating women using warfarin, aceno-coumarol, or UFH who wish to breast-feed,

we recommend continuing the use of warfa-rin, acenocoumarol, or UFH(Grade 1A).

the infants gut. Drugs that are poorly absorbed areunlikely to affect the neonate. Lipid-soluble drugs

with a low molecular weight that are not highlyprotein bound are more likely to be transferred intobreast milk.96

4.1 Use of Vitamin K Antagonistsin Breast-feeding Women

Despite a lack of data suggesting any harmfuleffect to breast-feeding infants, many obstetriciansremain reluctant to prescribe warfarin to lactating

women. This might reflect concerns that less polar,more lipophilic vitamin K antagonists rarely usedin North America (eg, phenindione, anisindione, andphenprocoumon) might be excreted into breast milk.97

Warfarin, the oral anticoagulant prescribed for mostpatients in North America, is polar, nonlipophilic,and highly protein bound. There have been twoconvincing reports demonstrating that warfarin is not

detected in breast milk and does not induce an anti-coagulant effect in the breast-fed infant when nurs-ing mothers consume the drug.98,99Acenocoumarol,

which is commonly used in Europe, has similar prop-erties (Tables S4, S5).100,101 Therefore, the use of

warfarin and acenocoumarol in women who requirepostpartum anticoagulant therapy is safe.

4.2 Use of UFH and LMWH inBreast-feeding Women

Because of its high molecular weight and strong

negative charge, UFH does not pass into breast milkand can be safely given to nursing mothers.102In acase series of 15 women receiving 2,500 InternationalUnits of LMWH after cesarean section, there wasevidence of excretion of small amounts of LMWHinto the breast milk in 11 patients (Tables S4, S5).103However, given the very low bioavailability of oralheparin, there is unlikely to be any clinically relevanteffect on the nursing infant.

4.3 Use of Danaparoid in Breast-feeding Women

Very little is known about the passage of dana-

paroid into breast milk. A small number of case reportshave reported no or very low anti-Xa activity in thebreast milk of danaparoid-treated women.77Becausedanaparoid is not absorbed by the GI tract after oralintake, it is unlikely that any anticoagulant effect

would appear in breast-fed infants.

4.4 Use of Fondaparinux in Breast-feeding Women

According to the manufacturers prescribing infor-mation, fondaparinux was found to be excreted in themilk of lactating rats.104There are no published data



17/46


5.1 Prevention of VTE in Patients UndergoingAssisted Reproductive Technology

The bleeding risk most relevant to this popula-tion is intraabdominal and vaginal bleeding. The esti-mates of normal blood loss during uncomplicatedoocyte retrieval vary, ranging from approximately230 mL in one prospective cohort of 220 women123to13 mL (range, 0-98 mL) in a study of 83 women.124Although patient-important vaginal bleeding appearsto occur in up to 2% to 3% of patients, significantintraabdominal bleeding is much less common(0.5% procedures) (Table S7).120,125-132 Whetherthese risks are increased with antithrombotic prophy-laxis is uncertain.

All studies that address the impact of prophylaxisin in vitro fertilization have important limitations, andthe number of patients who have received anticoagu-lants is too small to draw any conclusions about safetyand efficacy (Table S8).133-135Therefore, we used indi-rect evidence from a meta-analysis of thrombopro-phylaxis in patients undergoing hip arthroplasty136toestimate the relative effects LMWH prophylaxis inassisted reproductive technology. Table 2 and Table S9summarize the quality of evidence and anticipatedabsolute effects of thrombosis prophylaxis in women

with and without ovarian hyperstimulation syndrome.We rate the quality of evidence as low due to indi-rectness of the population and intervention and due toconsiderable imprecision in risk estimates for majorbleeding events and VTE. In women with severeovarian hyperstimulation syndrome, thromboprophy-laxis may result in 26 fewer VTE per 1,000 womentreated (number needed to treat [NNT] of 39 [givenan estimated baseline risk of VTE of 4%]), with noincreased risk of significant bleeding. However, in

women without ovarian hyperstimulation syndromein whom the baseline risk of VTE is estimated tobe 0.2%, the use of LMWH prophylaxis is of verylimited value (NNT, 781).

Data regarding the risk of VTE in women withthrombophilia or prior VTE who undergo assistedreproduction are lacking. Given the low baseline riskof VTE associated with assisted reproduction, if themagnitude of relative risk increases is similar to thatreported with pregnancy-related VTE (sections 8and 9), women with low-risk thrombophilias or prior

VTE associated with major transient risk factors willreceive only very small benefit from prophylaxis.

Dosage and duration of thromboprophylaxis afterassisted reproductive therapy has not been wellstudied. If LMWH is used in women who developovarian hyperstimulation, extension of prophylaxisfor 4 to 8 weeks postresolution of hyperstimulation114or throughout any resultant pregnancy and into thepostpartum period115has been suggested given that

4.0.2. For lactating women using LMWH, dana-paroid, or r-hirudin who wish to breast-feed,

we recommend continuing the use of LMWH,danaparoid, or r-hirudin(Grade 1B).

4.0.3. For breast-feeding women, we suggestalternative anticoagulants rather than fonda-parinux (Grade 2C).

4.0.4. For breast-feeding women, we recom-mend alternative anticoagulants rather thanoral direct thrombin and factor Xa inhibitors(Grade 1C).

4.0.5. For lactating women using low-dose aspi-rin for vascular indications who wish to breast-feed, we suggest continuing this medication(Grade 2C).

5.0VTE in Patients Using AssistedReproductive Technology

Assisted reproductive technology, which refers toall treatments or procedures involving in vitro han-dling of human oocytes and sperm or embryos forthe purpose of achieving pregnancy,114,115 may beassociated with VTE. Data regarding the frequencyof VTE, however, comprise predominantly of casereports, case series, and relatively small cohort studies(Table S6).116-121 In two large retrospective series ofpatients undergoing in vitro fertilization, thrombosiscomplicated 0.1% (95% CI, 0%-0.3%)116 and 0.3%(95% CI, 0%-0.8%)117 of cycles. A hospital-basedcase-control study demonstrated a fourfold increase inantenatal VTE with assisted reproductive technologyfor singleton pregnancies and a sixfold incidencein twin pregnancies but no statistically significantassociation with postpartum VTE.121Thus, althoughin vitro fertilization appears to be a risk factor forantepartum thromboembolism, the overall absoluteincidence of symptomatic thrombosis appears to below.

The risk of thrombosis may be higher in women withovarian hyperstimulation syndrome, with an incidenceof thrombosis of up to 4.1% (95% CI, 1.1%-13.7%) insevere cases.116In a review of thrombosis associated

with assisted reproductive technology, Chan and col-leagues122identified 61 reports of venous thrombosis(49 cases involving the veins of the neck and arm) and35 arterial events. Ovarian hyperstimulation syndrome

was reported in 90% of arterial cases and 78% ofvenous events. In 98% of cases, thrombosis occurredafter ovulation induction. Venous events were delayedcompared with those involving the arterial circulation(42.4 days after embryo transfer and 10.7 days post-transfer, respectively).122



18/46


Table 2[Section 5.1.1, 5.1.2] Summary of Findings: Prophylactic-Dose LMWH vs No Thromboprophylaxis forWomen Who Undergo Assisted Reproductive Therapy

Outcomes

Participants(Studies),Follow-up

Quality of theEvidence (GRADE)

Relative Effect(95% CI)a

Anticipated Absolute Effects During Pregnancy

Risk WithoutProphylaxis

Risk Difference With LMWH(95% CI)

SymptomaticVTE, DVT,and pulmonaryembolism

1,953 (6 RCTs),27-35 dpostoperative

Low due toindirectnessbandimprecisiona

RR 0.36(0.20-0.67)

Without severe ovarian hyperstimulation syndrome2 VTE per 1,000c 1 fewer VTE per 1,000 (from

2 fewer to 0 fewer)With severe ovarian hyperstimulation syndrome

40 VTE per 1,000c 26 fewer per 1,000 (from32 fewer to 13 fewer)

Major bleed 5,456 (7 RCTs),3 wks-9 mo

Low due toindirectnessbandimprecisiona

RR 0.43(0.11-1.65)

30 bleeding eventsper 1,000c

No significant difference 17 fewerbleeding events per 1,000 (from27 fewer to 20 more)

GRADE5Grades of Recommendations, Assessment, Development, and Evaluation; LMWH5 low-molecular-weight heparin; RCT5 randomizedcontrolled trial; RR5 risk ratio.aRated down for imprecision due to imprecise control group risk estimates for bleeding events and for VTE in the subset of women with ovarianhyperstimulation (Tables S6-S8).bThe population did not include pregnant women. Different doses of LMWH were used, treatment was initiated variably before or after surgery

with a duration of7 d (in hospital). Outcomes were variably reported; meta-analysis also provides other outcomes such as mortality, asymptomaticDVT, and specific bleed outcomes (wound hematoma, transfusion). Follow-up varied between trials from 3 wk to 9 mo.cControl group risk for VTE and major bleed come from observational studies of women undergoing assisted reproductive technology, with manystudies following women until delivery (Tables S6-S8).

tional studies have assessed the risk of VTE aftercesarean section, with absolute risk estimates rang-ing from ,1 in 1,000 up to 18 of 1,000 cesareandeliveries.121,139-148However, studies based on hospitalrecords and disease coding may result in an underes-timation of the true incidence of symptomatic VTE.149A Norwegian study of 59 low-risk women undergoingelective cesarean section who underwent screeningfor DVT using triplex ultrasonography (compression

ultrasonography, color Doppler echocardiography,and spectral Doppler echocardiography) 2 to 5 daysafter delivery and followed up for 6 weeks reportedthat none had symptomatic or asymptomatic VTE(95% CI, 0%-6.1%).144 A small prospective studyin which patients after cesarean section underwentscreening ultrasounds at hospital discharge and 2 weekspostpartum and were followed for 3 months reporteda symptomatic event rate of five of 1,000 (95% CI,0.1%-2.8%).147 This is consistent with estimates basedon hospital discharge data antedating the use ofthromboprophylaxis.138,140

Observational studies provide evidence concerningrisk factors for VTE in pregnant women (Tables S10,S11)121,146,148,150-152; these are likely to be relevant in

women undergoing cesarean section. In assessingrisk in this setting, the number of risk factors, themagnitude of risk associated with these factors, andtheir impact when occurring together are all rele-

vant. Table 3 provides an overview of major andminor risk factors we suggest clinicians use to identify

women at increased risk of VTE after cesarean sec-tion. The presence of one major or at least two minorrisk factors will indicate whether patients qualify for

most reported events have developed days to weeks(range, 2 days-11 weeks) after resolution of ovarianhyperstimulation.115However, given the lack of a clearassociation between assisted reproductive technologyand postpartum events,117,121 continuing anticoagulantprophylaxis after delivery is less likely to be of benefit.

Recommendations

5.1.1. For women undergoing assisted reproduc-tion, we recommend against the use of routinethrombosis prophylaxis(Grade 1B).

5.1.2. For women undergoing assisted reproduc-tion who develop severe ovarian hyperstimulationsyndrome, we suggest thrombosis prophylaxis(prophylactic LMWH) for 3 months postresolu-tion of clinical ovarian hyperstimulation syn-drome rather than no prophylaxis(Grade 2C).

Remarks:Women who are averse to taking medica-tion for very small benefit and those who consider

self-injecting a considerable burden will be disin-clined to use LMWH for extended thrombosis pro-phylaxis. Given that the absolute benefit decreases astime from the hyperstimulation event increases, such

women will be very disinclined to continue prophy-laxis throughout the entire resultant pregnancy.

6.0VTE Following Cesarean Section

6.1 Risk of VTE Following Cesarean Section

The puerperium is the time of maximal daily riskof pregnancy-associated VTE.137,138Several observa-



19/46


20/46


Table 4[ Section 6.2.1-6.2.4] Summary of Findings: LMWH vs No Thromboprophylaxis for Prevention of VTEin Women Undergoing Cesarean Section

OutcomesParticipants (Studies),

Follow-upQuality of the

Evidence (GRADE)Relative Effect

(95% CI)a

Anticipated Absolute Effects Over 6 wk Postpartum

Risk WithoutProphylaxis

Risk Difference With LMWH(95% CI)

SymptomaticVTE, DVT,and pulmonaryembolism

4,890 (3 RCTs),3 wk-9 mo

Moderate due toindirectnessb

RR 0.29(0.11-0.73)

Low risk (see Table 13)5 VTE per 1,000a 3 fewer VTE per 1,000 (from 4

fewer to 1 fewer)High risk (see Table 13)

40 VTE per 1,000a 21 fewer per 1,000 (from 27fewer to 9 fewer)

Major bleedc 5,456 (7 RCTs),3 wk-9 mo

Moderate due toindirectnessc

RR 2.03(1.37-3.01)

20 bleeding eventsper 1,000d

20 more bleeding events per 1,000(from 8 more to 40 more)

See Table 2 legend for expansion of abbreviations.aControl group risk estimates come from studies providing risk factors for VTE after cesarean section (Tables S10 and S11).bRated down for indirect study population (general surgery patients). We did not rate down for risk of bias, although only five of eight RCTs ofLMWH vs placebo/no treatment reported mortality.cRated down for indirectness due to variable bleeding definitions in trials: bleeding leading to death, transfusion, reoperation, or discontinuation oftherapy. Measured at end of therapy.dControl group risk estimate comes from a decision analysis by Blondon et al.141

Remarks:The reduced bleeding risk with mechanicalprophylaxis should be weighed against the inconve-nience of elastic stockings and intermittent pneu-maric compression.

6.2.3. For women undergoing cesarean sectionwho are considered to be at very high riskfor VTE and who have multiple additional riskfactors for thromboembolism that persist inthe puerperium, we suggest that prophylacticLMWH be combined with elastic stockings and/orintermittent pneumatic compression over LMWHalone(Grade 2C).

6.2.4. For selected high-risk patients in whomsignificant risk factors persist following delivery,

we suggest extended prophylaxis (up to 6 weeksafter delivery) following discharge from thehospital(Grade 2C).

7.0 Treatment of Proven Acute

VTE During Pregnancy

PE remains a leading cause of maternal mortalityin the western world,168,169 and VTE in pregnancyis an important cause of maternal morbidity.168,170,171Results from studies in which either all or mostpatients underwent accurate diagnostic testing for

VTE report that the incidence of VTE ranges from0.6 to 1.7 episodes per 1,000 deliveries.138,139,146,148,152,172A meta-analysis showed that two-thirds of DVT occurantepartum, with these events distributed through-out all three trimesters.173In contrast, 43% to 60%

The optimal duration of prophylaxis after cesareansection is not established. If we extrapolate from gen-eral surgery,156,163-166 treatment until discharge fromthe hospital, with extended prophylaxis for those withsignificant ongoing risk factors, may be appropriate. Weexpress a preference for LMWH over UFH becauseof its favorable safety profile (see section 4.0).

There are no relevant cost-effectiveness data inthis setting using UFH or LMWH; however, in onestudy modeling the cost-effectiveness of intermittentpneumatic compression, this intervention was con-sidered cost-effective when the incidence of post-cesarean section DVT was at least 6.8 of 1,000167(Tables S13, S14). However, these devices are notreadily available at all sites, and patients and nursesoften find them to be inconvenient and cumbersometo use.

Recommendations

6.2.1. For women undergoing cesarean sectionwithout additional thrombosis risk factors, werecommend against the use of thrombosis pro-

phylaxis other than early mobilization(Grade 1B).6.2.2. For women at increased risk of VTE aftercesarean section because of the presence ofone major or at least two minor risk factors(Table 3), we suggest pharmacologic thrombo-prophylaxis (prophylactic LMWH), or mechan-ical prophylaxis (elastic stockings or intermittentpneumatic compression) in those with contrain-dications to anticoagulants while in the hospitalfollowing delivery rather than no prophylaxis(Grade 2B).



21/46


of major bleeding events.175-178Table 5 and Table S15summarize the quality of evidence and main findingsfrom a systematic review of nonpregnant patientsdeemed applicable to the present population of preg-nant women with acute VTE. Given these results,

we consider the burden of self-injecting with LMWHfor several months and possibility of skin reactions oflesser importance.

If LMWH is used for treatment of acute VTE inpregnancy, a weight-adjusted dosing regimen shouldbe used. LMWH requirements may alter as preg-nancy progresses because the volume of distributionof LMWH changes and glomerular filtration rateincreases in the second trimester. The latter has ledsome to recommend a bid LMWH dosing schedule.However, many clinicians use a once-daily regimento simplify administration and enhance compliance.Observational studies have not demonstrated anyincrease in the risk of recurrence with the once-dailyregimen over the bid regimen.150,174

The need for dose adjustments over the course ofpregnancy remains controversial. Some suggest thatdose should be increased in proportion to the changein weight.181On the basis of small studies showingthe need for dose-escalation to maintain therapeuticanti-Xa LMWH levels,182,183some advocate the per-formance of periodic (eg, every 1-3 months) antifac-tor Xa LMWH levels 4 to 6 h after injection with doseadjustment to maintain a therapeutic anti-Xa level(0.6-1.0 units/mL if a bid regimen is used and higher

of pregnancy-related episodes of PE appear to occurin the 4 to 6 weeks after delivery.139,148Because theantepartum period is substantially longer than the6-week postpartum period, the daily risk of PE, as

well as DVT, is considerably higher following deliverythan antepartum.

7.1 Treatment of VTE During Pregnancy

Based on safety data for the fetus, heparin com-pounds are preferred over vitamin K antagonists forthe treatment of VTE in pregnancy (see section 3.0).LMWH is the preferred option for most patientsbecause of its better bioavailability, longer plasma half-life, more predictable dose response, and improvedmaternal safety profile with respect to osteoporosisand thrombocytopenia (see section 2.0).35-38Further,LMWH is a more convenient option because it canbe given once daily, and unlike UFH, LMWH doesnot require aPTT monitoring.6

A systematic review of LMWH use in pregnancy38and subsequent observational studies36,150,174confirmthe safety and efficacy of LMWH in this patient popu-lation when used for treatment of VTE. Our strongrecommendation for LMWH over vitamin K antago-nists in the treatment of VTE in pregnancy is furthersupported by evidence showing that in the nonpreg-nant population, LMWH is more effective than vita-min K antagonists in preventing recurrent VTE andpostthrombotic syndrome without increasing the risk

Table 5[Section 7.1.2] Summary of Findings: Should LMWH Rather Than VKA Be Used for Long-term

Treatment of VTE in Pregnant Women?

OutcomesParticipants

(Studies), Follow-upQuality of the Evidence

(GRADE)Relative Effect

(95% CI)a

Anticipated Absolute Effects During Pregnancyb

Risk With VKARisk Difference With LMWH

(95% CI)

Recurrentsymptomatic

VTE, DVT,and pulmonaryembolism

2496 (7 RCTs);median, 6 mo

Moderate due to riskof biasa

RR 0.62(0.46-0.84)

30 VTE per 1,000b 11 fewer VTE per 1,000 (from16 fewer to 5 fewer)

Major bleeding 2727 (8 RCTs);median, 6 mo

Moderate due toimprecisionc

RR 0.81(0.55-1.2)

20 bleeding eventsper 1,000d

4 fewer bleeding events per 1,000(from 9 fewer to 4 more)

PTS self-reported

leg symptomsand signs

100 (1 RCT);

median, 3 mo

Low due to risk of biasa

and indirectnesse

RR 0.85

(0.77-0.94)

480 PTS per 1,000f 38 fewer bleeding events per 1,000

(from 110 fewer to 29 fewer)

Limited to LMWH regimens that used50% of the acute treatment dose during the extended phase of treatment. Meta-analysis is based on RCTsas referenced in Kearon et al178in this guideline. PTS5postthrombotic syndrome; VKA5vitamin K antagonist. See Table 2 legend for expansionof other abbreviations.aRisk of bias due to lack of blinding.bControl group risk estimate for VTE with VKAs comes from cohort study by Prandoni et al, 179adjusted to the 6-mo time frame consideredapplicable to the pregnancy period.cRated down for imprecision because CI includes both benefit and harm. Borderline decision not to rate down for risk of bias (considered thisoutcome less subjective, so lack of blinding not serious threat to validity).dControl group risk estimate for major bleeding events comes from cohort studies by Prandoni et al 179and Beyth et al,180adjusted to a 6-mo timeframe considered applicable to the pregnancy period.ePredictive value from 3 mo (follow-up in study) to long term is uncertain.fControl group risk estimate for PTS comes from observational study of pregnant women (most mild).171



22/46


nant population), sufficient to recommend treatmentthroughout pregnancy and the postpartum period fora minimum total duration of 3 months.

The delivery options in women using anticoagu-lants are best considered by a multidisciplinary team.Several options are possible, including spontaneouslabor and delivery, induction of labor, and electivecesarean section. The plan for delivery should take

into account obstetric, hematologic, and anestheticissues. In order to avoid an unwanted anticoagulanteffect during delivery (especially with neuraxial anes-thesia) in women receiving adjusted-dose subcuta-neous UFH11or LMWH who have a planned delivery;twice-daily subcutaneous UFH or LMWH shouldbe discontinued 24 h before induction of labor orcesarean section, whereas patients taking once-dailyLMWH should take only 50% of their dose on themorning of the day prior to delivery (see Kunz et al189in this guideline). If spontaneous labor occurs in womenreceiving anticoagulation, neuraxial anesthesia should

not be used. Where the level of anticoagulation isuncertain and where laboratory support allows forrapid assessment of heparin levels, then testing canbe considered to guide anesthetic and surgical man-agement. In women receiving subcutaneous UFH,careful monitoring of the aPTT is required and, if itis markedly prolonged, protamine sulfate190may berequired to reduce the risk of bleeding. If bleedingoccurs that is considered secondary to LMWH ratherthan to an obstetric cause, protamine sulfate mayprovide partial neutralization.191

Women with a very high risk for recurrent VTE (eg,

proximal DVT or PE close to the expected date ofdelivery) may benefit by having a planned deliveryby induction or cesarean section, as appropriate, sothat the duration of time without anticoagulation canbe minimized. Those at the highest risk of recurrence(eg, proximal DVT or PE within 2 weeks) can beswitched to therapeutic IV UFH, which is then dis-continued 4 to 6 h prior to the expected time ofdelivery or epidural insertion. Alternatively, a tempo-rary inferior vena caval filter can be inserted andremoved postpartum. However, the latter may bebest restricted to women with proven DVT who

have recurrent PE despite adequate anticoagulationbecause experience with these devices during preg-nancy is limited,192-194and the risk of filter migrationand inferior vena cava perforation may be increasedduring pregnancy.193,194

Recommendations

7.1.1. For pregnant women with acute VTE,we recommend therapy with adjusted-dosesubcutaneous LMWH over adjusted-dose UFH(Grade 1B).

if a once-daily regimen is chosen). However, otherresearchers have demonstrated that few womenrequire dose adjustment when therapeutic doses ofLMWH are used.184-186Given the absence of largestudies using clinical end points that demonstratean optimal therapeutic anti-Xa LMWH range or thatdose adjustments increase the safety or efficacy oftherapy, the lack of accuracy and reliability of the

measurement,187the lack of correlation with risk ofbleeding and recurrence,188 and the cost of the assay,routine monitoring with anti-Xa levels is difficult to

justify.Where LMWH cannot be used or when UFH is

preferred (eg, in patients with renal dysfunction), UFHcan be used through one of two alternatives: (1) initialIV therapy followed by adjusted-dose subcutaneousUFH given every 12 h or (2) bid adjusted-dose sub-cutaneous UFH. With subcutaneous therapy, UFHdoses should be adjusted to prolong a midinterval(6 h postinjection) aPTT into therapeutic range,

although it is recognized that aPTT monitoring isless reliable in pregnancy.6As previously discussed,the use of fondaparinux is inadvisable in pregnancy(see section 3.5). In this guideline, Linkins et al14andKearon et al178present evidence regarding plateletcount monitoring for the detection of HIT and therole of compression stockings in the acute manage-ment of DVT.

It remains unclear whether the dose of LMWH(or UFH) can be reduced after an initial phase oftherapeutic anticoagulation. Some suggest that full-dose anticoagulation should be maintained through-

out pregnancy and the puerperium because of theongoing risk of recurrent VTE. However, regimensin which the intensity of LMWH is reduced laterduring the course of therapy to an intermediate-doseregimen26or 75% of a full-treatment dose177havebeen used successfully in the nonpregnant popula-tion, including in cancer patients who have a muchhigher risk of recurrence. A similar approach whenusing LMWH in pregnancy may reduce the smallrisks of anticoagulant-related bleeding and heparin-induced osteoporosis. Although there have beenno studies directly comparing full-dose LMWH with

one of these modified dosing strategies in pregnantwomen, a modified dosing regimen may be useful inpregnant women at increased risk of either of thesetwo complications.

No studies have assessed optimal duration of anti-coagulant therapy for treatment of pregnancy-related

VTE. In nonpregnant patients with VTE, evidencesupports a minimum duration of 3 months treatment(see Kearon et al178in this guideline). We considerthe additional fivefold to 10-fold increase in risk for

VTE in pregnant women, coupled with the high rateof proximal thrombi (compared with the nonpreg-



23/46


24/46


estimated the recurrence rate because objective diag-nostic testing was not used.199

Given the low quality of the direct evidence, we useindirect evidence about the relative effects of thrombo-prophylaxis from

terapia antitrombotica en embarazo

Documents