Chapter 55. Thromboembolic Disorders. Will Obs

 Thromboembolic Disorders

BS. Nguyễn Hồng Anh

During the past century, the requency o venous thromboembolism (VE) uring the puerperium ecrease remarkably

as early ambulation became routine practice. Despite this an

other avances in prevention an treatment, thromboembolism

remains a leaing cause o maternal morbiity an mortality

(Abe, 2019). Trombotic pulmonary embolism accounte or

almost 10 percent o pregnancy-relate eaths in the Unite

States between 2011 an 2015 (Creanga, 2017; Petersen, 2019).

Te absolute incience o VE uring pregnancy is low—1

or 2 cases per 1000 pregnancies. However, the risk is approximately ve times higher than that among women who are not

pregnant (Greer, 2015). Approximately equal numbers o cases

are ientie antepartum an in the puerperium. But eep-vein

thrombosis alone is more requent antepartum, an pulmonary embolism is more common postpartum. During the rst

6 weeks o the puerperium, the estimate incience o a thromboembolic complication is 22 events per 100,000 eliveries.

Although still elevate, the risk alls to approximately 3 cases

per 100,000 eliveries uring the secon 6-week postpartum

perio (Kamei, 2014). As many as 2 percent o these women

have postthrombotic synrome (Govinappagari, 2020).

PATHOPHYSIOLOGY

Ruol Virchow (1856) postulate that stasis, local trauma to

the vessel wall, an hypercoagulability preispose to venous

thrombosis. During pregnancy, the risk or each o these rises.

Compression o the pelvic veins an inerior vena cava by the

enlarging uterus reners the lower extremity venous system particularly vulnerable to stasis. Marik an Plante (2008) cite a

50-percent reuction in venous ow velocity in the legs that

lasts rom the early thir trimester until 6 weeks postpartum.

Tis stasis is the most constant preisposing risk actor or

venous thrombosis. Venous stasis, elivery, preeclampsia, an

sepsis contribute to enothelial cell injury. Last, as liste in the

Appenix (p. 1228), the synthesis o most clotting actors is

markely enhance uring pregnancy an avors coagulation.

Risk actors or eveloping VE uring pregnancy are

shown in Table 55-1. Te most important o these is a personal history o thrombosis. Specically, 15 to 25 percent o

all VE cases uring pregnancy are recurrent events (American

College o Obstetricians an Gynecologists, 2020b). Another

important iniviual risk actor is a genetically etermine

thrombophilia. An estimate 20 to 50 percent o women who

evelop a venous thrombosis uring pregnancy or postpartum

have an ientiable unerlying procoagulant genetic isorer

(American College o Obstetricians an Gynecologists, 2020a).

Calculate risks or thromboembolism are approximately

ouble in women with multietal gestation, anemia, hyperemesis, hemorrhage, cesarean elivery, obesity, preeclampsia,

an postpartum inection (Walman, 2013). Scheres an

TABLE 55-1. Some Risk Factors Associated with an

Increased Risk for VTE

Obstetrical General

Cesarean delivery

Cesarean hysterectomy

Diabetes mellitus

Hemorrhage and anemia

Hyperemesis

Immobility—prolonged

bed rest

Multifetal gestation

Multiparity

Preeclampsia

Puerperal infection

Stillbirth

Age 35 years or older

Anatomical anomalya

Antiphospholipid antibodies

Connective tissue disease

Dehydration

Hormonal contraceptives

Hospitalization

Immobility

Infection and inflammatory

disease

Inflammatory bowel disease

Malignancy

Nephrotic syndrome

Obesity

Orthopedic procedures

Paraplegia

Prior VTE

Sickle-cell disease

Smoking

Surgery

Thrombophilia

aIncludes May-Thurner syndrome (iliac vein compression

syndrome).

VTE = venous thromboembolism.

absence o well-recognize risk actors. O greatest signicance

is a amily history o suen eath ue to pulmonary embolism

or a history o multiple amily members requiring long-term

anticoagulation therapy because o recurrent thrombosis (Connors, 2017).

Antithrombin Deficiency

Synthesize in the liver, antithrombin is one o the most important inhibitors o thrombin an inactivates thrombin an actor

Xa (Rhéaume, 2016). Notably, the rate o antithrombin interaction with its target is accelerate over 1000–o1 by heparin.

Antithrombin eciency may result rom hunres o ierent

mutations that are almost always autosomal ominant. ype I

eciency results rom reuce synthesis o biologically normal

antithrombin, an type II eciency is characterize by normal

levels o antithrombin with reuce unctional activity.

Homozygous antithrombin eciency is lethal. Heterozygous

eciency aects approximately 1 in 2500 iniviuals. It is the

most thrombogenic o the heritable coagulopathies. Te risk or

thrombosis correlates with the egree o eciency. Antithrombin eciency is associate with a 25- to 50-ol higher relative

risk o VE in the general population an a sixol increase

risk o thromboembolic complications uring pregnancy (Ilonczai, 2015). As shown in able 55-2, those aecte have an

inorinately high risk o VE uring pregnancy.

Sabaell an associates (2010) stuie the outcomes o 18

pregnancies complicate by antithrombin eciency. welve o

these women were treate with therapeutic oses o low-molecular-weight heparin (LMWH). Tree o the untreate patients

suere a thromboembolic episoe compare with none in the

treate group. Untreate women also ha a 50-percent risk o

stillbirth an etal-growth restriction. By comparison, none o

the treate women ha a stillbirth, but a ourth evelope etalgrowth restriction. Abbattista an coworkers (2020) reporte a

7-percent risk o VE or women treate with LMWH uring

pregnancy an a 12-percent risk or those not treate. Similar

results were reporte by Ilonczai an colleagues (2015). Last,

Garcia-Botella an associates (2016) escribe a mesenteric vein

thrombosis in a pregnant woman with antithrombin eciency.

Given such risk, aecte women are manage uring pregnancy with anticoagulation regarless o whether they have

ha a prior thrombosis (Abbattista, 2020; Bates, 2018). When

anticoagulation is necessarily withhel, such as uring surgery

or elivery, or i VE evelops in an anticoagulate woman,

treatment with recombinant human antithrombin may be protective (Paias, 2016).

Protein C Deficiency

When thrombin is boun to thrombomoulin on enothelial

cells o small vessels, its procoagulant activities are neutralize.

Tis bining also activates protein C, a natural anticoagulant.

It, in the presence o protein S, controls thrombin generation,

in part, by inactivating actors Va an VIIIa (see Fig. 55-1).

Protein C activity increases moestly but signicantly throughout the rst hal o pregnancy, an some have speculate that

this augmentation may play a role in maintaining early pregnancy through both anticoagulant an inammatory regulatory

pathways (Sai, 2010).

coworkers (2020) ha similar observations or women with preeclampsia. Risks were signicantly higher among women who

ha a stillbirth or who unerwent peripartum hysterectomy.

Last, as iscusse in Chapter 57 (p. 1021), inammatory bowel

isease increases the risk two- to threeol (Kim, 2019).

THROMBOPHILIAS

Several important regulatory proteins act as inhibitors in the

coagulation cascae (Fig. 55-1). Normal values or many o

these proteins uring pregnancy are oun in the Appenix

(p. 1228). Inherite or acquire eciencies o these inhibitory proteins are collectively reerre to as thrombophilias. Tese

can lea to hypercoagulability an recurrent VE (Connors,

2017). Although these isorers are collectively present in

approximately 15 percent o white European populations, they

are responsible or approximately 50 percent o all thromboembolic events uring pregnancy (Pierangeli, 2011). Some aspects

o the more common inherite thrombophilias relate to VE

are summarize in Table 55-2.

■ Inherited Thrombophilias

Patients with inherite thrombophilic isorers oten have a

amily history o thrombosis. Tese mutations are also oun in

up to hal o all patients who present with VE beore the age

o 45 years, particularly in those whose event occurre in theThromboembolic Disorders 977

CHAPTER 55

TABLE 55-2. Inherited Thrombophilias and Their Association with Venous

Thromboembolism (VTE) in Pregnancy

VTE Risk per Pregnancy (%)

No History Prior VTE Family History

Factor V Leiden heterozygote 0.5–3.1 10 0.06–1.2

Factor V Leiden homozygote 2.2–1.4 17 1–16

Prothrombin gene heterozygote 0.4–2.6 >10 0–0.73

Prothrombin gene homozygote 2–4 >17 1.6

Factor V Leiden/prothrombin double

heterozygote

8.2 >20 0–2.4

Antithrombin deficiency 0.2–1.16 40 0–8

Protein C deficiency 0.1–5.4 4–17 0–5

Protein S deficiency 0.1–6.6 0–22 0–1.5

Hyperhomocysteinemia 0 0 <1

Data from Abbattista, 2020; American College of Obstetricians and

Gynecologists, 2020a; Bates, 2018; Croles, 2017.

Protein S

deficiency

Inactivates

factor Va

Inactivates

factor VIIIa

PROTEIN S PROTEIN S

Protein C

deficiency

Factor V Leiden mutation

Factor V resistant to

degradation by protein C

Hyperhomocysteinemia

(inhibits activation of protein C)

G20210A mutation

Increased prothrombin

levels

Antithrombin deficiency

Controls thrombin

generation

Decreased thrombin

neutralization

Thrombin binds to

thrombomodulin on

endothelial cells

Activated protein C Protein C

Prothrombin Thrombin Coagulation

FIGURE 55-1 Inherited thrombophilias and their effect(s) on the coagulation cascade.978

Section 12

Medical and Surgical Complications

More than 160 ierent autosomal ominant mutations

or the protein C gene have been escribe (Louis-Jacques,

2016). Te prevalence o heterozygous protein C eciency is

2 to 3 per 1000, but many o these iniviuals o not have a

thrombosis history because the phenotypic expression is highly

variable. Tese prevalence estimates correspon with unctional

activity threshol values o 50 to 60 percent, which are use

by most laboratories. As shown in able 55-2, these are associate with a six- to twelveol higher risk or VE (Lockwoo,

2012). Te rare cases o homozygous protein C eciency may

cause atal neonatal purpura ulminans. Last, sepsis can result

in purpura ulminans in the ault with heterozygous protein C

eciency (Benapui, 2021).

Protein S Deficiency

Tis circulating anticoagulant is activate by protein C, which

enhances the capacity o protein S to inactivate actors Va an

VIIIa (see Fig. 55-1). Protein S eciency may be cause by

more than 130 ierent mutations, with an aggregate heterozygous prevalence o approximately 0.3 to 1.3 per 1000 iniviuals (Louis-Jacques, 2016). Protein S eciency may be measure

by antigenically etermine ree, unctional, an total S levels.

All three ecline substantively uring normal gestation (Appen-

ix, p. 1228). Tus, the iagnosis in pregnant women—as well

as in those taking certain oral contraceptives—is ifcult. I

screening uring pregnancy is necessary, threshol values or

ree protein S antigen levels in the secon an thir trimesters

have been ientie at <30 percent an <24 percent, respectively. Te risk o VE in pregnancy is increase severalol

(see able 55-2). Among those with a positive amily history,

the VE risk in pregnancy is 6 to 7 percent (American College

o Obstetricians an Gynecologists, 2020a).

Conar an coworkers (1990) escribe thrombosis in ve

o 29 pregnant women with protein S eciency. Tey, as well

as Burneo an colleagues (2002), reporte maternal cerebral

vein thrombosis. Neonatal homozygous protein S eciency

is usually associate with purpura ulminans, the atal clinical

phenotype.

Factor V Leiden Mutation

Tis is the most prevalent o the known thrombophilia syn-

romes an is characterize by resistance o plasma to the

anticoagulant eects o activate protein C. Tere are several

mutations that create this resistance, but the most common is

the actor V Leien mutation. Tis missense mutation results

rom a substitution o glutamine or arginine at position 506

in the actor V polypeptie. As a result o this mutation, activate actor V is neutralize approximately tenol more slowly

by activate protein C (see Fig. 55-1). Tis leas to enhance

thrombin generation.

Heterozygous inheritance o actor V Leien is the most

common heritable thrombophilia. It is oun in 3 to 15 percent

o select European populations, 1.2 percent o Arican Americans, an 2.2 o Hispanic Americans, but is virtually absent in

Arican blacks an Asians (American College o Obstetricians

an Gynecologists, 2020a). Women who are heterozygous

or actor V Leien account or approximately 40 percent o

VE cases uring pregnancy. However, the actual risk among

pregnant women who are heterozygous an who o not have a

personal history or a rst-egree relative with a thrombotic episoe beore age 50 years is 5 to 12 events per 1000 births (see

able 55-2). In contrast, this risk is at least 10 percent among

pregnant women with a personal or amily history. Pregnant

women who are homozygous without a personal or amily history have a 1- to 4-percent risk or VE, whereas those with

such a history have an approximately 17-percent risk (American College o Obstetricians an Gynecologists, 2020a).

Diagnosis uring pregnancy is perorme by DNA analysis or the mutant actor V gene. Bioassay is not use because

o the normal resistance that evelops ater early pregnancy

rom alterations in other coagulation protein concentrations.

O note, activate protein C resistance can also be cause by

antiphospholipi synrome, which is escribe on page 979

an also etaile in Chapter 62 (p. 1114).

o assess the prognostic signicance o maternal actor V

Leien mutation uring pregnancy, Kjellberg an associates

(2010) compare the outcomes o 491 carriers with those o

1055 controls. All three o the thromboembolic events occurre

among the carriers. But, preterm birth rates, birthweights, or

hypertensive complication rates i not ier between the two

groups. In a prospective observational stuy o almost 5000

women conucte by the Maternal-Fetal Meicine Units Network, the heterozygous mutant gene incience was 2.7 percent

(Dizon-ownson, 2005). O three pulmonary emboli an one

eep-vein thrombosis cases—a rate o 0.8 per 1000 pregnancies—none was among these carriers. Moreover, in the heterozygous women, the risks o preeclampsia, placental abruption,

etal-growth restriction, or pregnancy loss were not elevate.

Te investigators conclue that universal prenatal screening

or the Leien mutation an prophylaxis or carriers without a

prior VE is not inicate.

Prothrombin G20210A Mutation

Tis missense mutation in the prothrombin gene leas to excessive accumulation o prothrombin, which may be converte

to thrombin. Prothrombin levels are increase approximately

30 percent in heterozygotes an 70 percent in homozygotes

(MacCallum, 2014). As with actor V Leien, a personal or

amily history o VE in a rst-egree relative beore age 50

years raises the risk o VE uring pregnancy (see able 55-2).

For a heterozygous carrier with such a history, the risk excees

10 percent. Without such a history, heterozygous carriers o

the mutation have less than a 1-percent risk o VE uring

pregnancy (American College o Obstetricians an Gynecologists, 2020a).

Silver an coworkers (2010) teste nearly 4200 women

or the prothrombin G20210A mutation. A total o 157—or

3.8 percent—o the women carrie the mutation, an only

one o these was homozygous. Carriers ha similar rates o

pregnancy loss, preeclampsia, etal-growth restriction, an placental abruption compare with noncarriers. Tree thromboembolic events occurre in women who teste negative or the

mutation.

Homozygous patients, or those who are oubly heterozygous or a G20210A mutation with a actor V Leien mutation, have a greater thromboembolism risk than heterozygousThromboembolic Disorders 979

CHAPTER 55

carriers (Connors, 2017). Others have provie etaile inormation on pregnancy outcomes in women with such rare compoun thrombophilias (Carroll, 2018; Lim, 2016).

Hyperhomocysteinemia

Te most common cause o elevate homocysteine is the

C667 thermolabile mutation o the 5, 10– methylene–tetrahyroolate reuctase (MHFR) enzyme. Inheritance is autosomal recessive. Elevate homocysteine levels may also result

rom eciency o one o several enzymes involve in methionine metabolism an rom correctible nutritional eciencies

o olic aci, vitamin B6, or vitamin B12. During normal pregnancy, mean homocysteine plasma concentrations ecline an

make a iagnosis ifcult (Lockwoo, 2002).

A metaanalysis by en Heijer an associates (2005) reporte

an association between MHFR polymorphisms an a slightly

greater risk or thrombosis in international stuies, but no such

association was oun in North American stuies (en Heijer,

2005). Te authors speculate that olic aci supplementation coul explain the ierence. Recall that olic aci serves

as a coactor in the remethylation reaction o homocysteine to

methionine. Similarly, the American College o Chest Physicians posite that the lack o an association with thromboembolism coul reect the physiological reuctions in homocysteine

levels associate with pregnancy as well as the eects o wiesprea prenatal olic aci supplementation (Bates, 2012). Te

American College o Obstetricians an Gynecologists (2020a)

has conclue that there is insufcient evience to support

assessment o MHFR polymorphisms or measurement o

asting homocysteine levels in the evaluation or VE.

Other Thrombophilia Mutations

Protein Z is a vitamin K–epenent protein that serves as a

coactor in actor Xa inactivation. Stuies have oun that low

protein Z levels are associate with an elevate thromboembolism risk in nonpregnant patients an may be implicate in

the pathogenesis o poor pregnancy outcomes (Almawi, 2013).

Similarly, plasminogen activator inhibitor type 1 (PAI-I) is an

important regulator o brinolysis. Certain polymorphisms in

the gene promoter have been associate with slightly higher

VE risks. Although these thrombophilias may exacerbate risk

among patients when coinherite with other thrombophilias, the American College o Obstetricians an Gynecologists

(2020a) has conclue that evience to recommen screening

is insufcient.

■ Acquired Thrombophilias

Some examples o acquire hypercoagulable states inclue

antiphospholipi synrome (APS), heparin-inuce thrombocytopenia, an cancer.

Antiphospholipid Syndrome

Tis prothrombotic isorer can aect both the venous an

arterial circulations. Te eeper veins o the lower limbs an

the cerebral arterial circulation are the most requent sites o

venous an arterial thrombosis, respectively (Garcia, 2018;

Konkle, 2018). Besies thrombosis, the other major clinical

maniestations o APS are averse obstetrical outcomes (able

11-4, p. 205). Criteria or APS inclue (1) at least one otherwise unexplaine etal eath at or beyon 10 weeks; (2) at

least one preterm birth beore 34 weeks’ gestation because o

eclampsia, severe preeclampsia, or placental insufciency; or (3)

at least three unexplaine consecutive spontaneous abortions

beore 10 weeks. Te synrome is iscusse in etail in Chapter

62 (p. 1114).

esting or antiphospholipi antiboies shoul be per-

orme in women with an unexplaine arterial or venous

thromboembolism or a history o one etal loss or three or

more recurrent embryonic or etal losses (American College

o Obstetricians an Gynecologists, 2020a). At this time, preterm severe preeclampsia an early-onset placental insufciency

o not support such testing. Tat sai, there is evience that

antiphospholipi positivity coners a worse prognosis or pregnancies complicate by HELLP synrome (Pécourt, 2021).

Tese patients shoul be teste or the presence o three actors:

(1) lupus anticoagulant, (2) anticariolipin immunoglobulin G

an M (IgG an IgM) antiboies, an (3) anti-β2 glycoprotein-I

IgG an IgM antiboies. I any o these laboratory test results

are positive, a conrmatory test is perorme 12 weeks later

(Connors, 2017).

Base on their stuy o 750 singleton pregnancies complicate by APS, Saccone an colleagues (2017) oun that

anticariolipin antiboy is the most common sole antiphospholipi antiboy present. Women with APS have a 5- to

12-percent risk o thrombosis uring pregnancy or the puerperium. Inee, up to 25 percent o these thrombotic events

occur uring pregnancy or in the puerperium.

■ Thrombophilias and Pregnancy

Complications

Contrary to past teaching, no enite causal link between

inherite thrombophilias an averse pregnancy outcomes

exists (Ormesher, 2017). Te conusion was cause by a number o observational stuies that were later ollowe by large

prospective cohort stuies. Currently, the American College o

Obstetricians an Gynecologists (2020a) oes not recommen

inherite thrombophilia screening or women with a history o

etal loss or averse pregnancy outcomes such as etal-growth

restriction, placental abruption, an preeclampsia.

Tere are no convincing ata that link inherite thrombophilias an etal loss. It ollows that there is no benet to aspirin or heparin therapy as might accrue or recurrent pregnancy

loss ue to antiphospholipi antiboies (Chap. 62, p. 1116).

Stillbirths also are not more common with heterozygous mutations, but there is a weak association with homozygous actor

V Leien mutation (Silver, 2016). Last, there is insufcient evi-

ence to screen or inherite thrombophilias in women with a

etal eath (Arachchillage, 2019).

Fetal-growth restriction also is not more common with inherite thrombophilias (American College o Obstetricians an

Gynecologists, 2020a; Inante-Rivar, 2002). Likewise, there

have been a number o prospective cohort stuies that showe

no association o inherite thrombophilias with placental

abruption (Dizon-ownson, 2005; Silver, 2010). Conversely,980

Section 12

Medical and Surgical Complications

Alrevic an coworkers (2002) reporte an association with

these an both heterozygosity an homozygosity or actor V Leien mutations as well as heterozygous prothrombin

G20210A mutation.

Stuies showing an association o inherite thrombophilias

an preeclampsia also are conicting (American College o Obstetricians an Gynecologists, 2020a). One large systematic review

an metaanalysis o nearly 22,000 women i not show an association between actor V Leien or prothrombin gene mutations

with preeclampsia (Roger, 2010). Similarly, a more recent prospective stuy aile to show an association between preeclampsia

an 12 hereitary thrombophilias (Fernanez Arias, 2019).

Tus, because o uncertainties in the magnitue o risk an

in the benets o prophylaxis given to prevent pregnancy complications in women with heritable thrombophilias, it remains

unproven that universal screening is inicate (Louis-Jacques,

2016). In contrast, the association between APS an averse

pregnancy outcomes—incluing etal loss, recurrent pregnancy

loss, an preeclampsia—is much stronger (Liu, 2019).

■ Thrombophilia Screening

Given the relatively high incience o thrombophilia in the

population an the low incience o VE, universal screening

uring pregnancy is not cost-eective. Tus, a selective screening strategy is require. Te American College o Obstetricians

an Gynecologists (2020a) recommens that thrombophilia

screening be consiere in the ollowing clinical circumstances:

(1) a personal history o VE with or without a recurrent risk

actor such as ractures, surgery, or prolonge immobilization;

an (2) a rst-egree relative (parent or sibling) with a history

o a high-risk inherite thrombophilia. As iscusse previously,

testing or inherite thrombophilias in women who have experience recurrent etal loss or averse pregnancy outcomes is

not recommene.

Methos o screening or the more common inherite

thrombophilias are shown in Table 55-3. Whenever possible, laboratory testing is perorme at least 6 weeks ater the

thrombotic event, while the patient is not pregnant, an when

she is not receiving anticoagulation or hormonal therapy. Last,

screening or hyperhomocysteinemia is not recommene

(American College o Obstetricians an Gynecologists, 2020a).

DEEP-VEIN THROMBOSIS

■ Clinical Presentation

During pregnancy, most venous thromboses are conne to

the eep veins o the lower extremity. Most o these are locate

in the ilioemoral an iliac veins without involvement o the

cal veins (Chan, 2010). In contrast, in the general population, more than 80 percent o eep-vein thromboses involve

cal veins (Huisman, 2015). Te vast majority are let sie

(Blanco-Molina, 2007). Greer (2003) hypothesizes that this

results rom compression o the let iliac vein by the right iliac

an ovarian arteries, both o which cross the vein only on the

let sie. Yet, as escribe in Chapter 56 (p. 994), the ureter is

compresse more on the right sie.

Signs an symptoms vary epening on the egree o occlusion an the intensity o the inammatory response. Classically,

thrombosis involving the lower extremity is abrupt in onset, an

there is pain an eema o the leg an thigh. Te thrombus

typically involves much o the eep-venous system to the ilio-

emoral region. Occasionally, reex arterial spasm causes a pale,

cool extremity with iminishe pulsations. Cal measurements

may ientiy a iscorance in size. Alternatively, there may be

appreciable clot, yet little pain, heat, or swelling. Importantly,

cal pain, either spontaneous or in response to squeezing or to

Achilles tenon stretching—Homans sign—may be cause by a

straine muscle or contusion. Importantly, 30 to 60 percent o

women with a confrmed lower-extremity acute deep-vein thrombosis (DVT) have an asymptomatic pulmonary embolism (p. 986).

■ Diagnosis

Clinical iagnosis o DV is ifcult without testing an is

conrme in only 10 percent o cases (Hull, 1990). Another

TABLE 55-3. How to Test for Thrombophilias

Thrombophilia Testing Method

Is Testing

Reliable

During

Pregnancy?

Is Testing

Reliable

During Acute

Thrombosis?

Is Testing

Reliable with

Anticoagulation?

Factor V Leiden mutation Activated protein C resistance

assay (second generation)

If abnormal: DNA analysis

Yes

Yes

Yes

Yes

No

Yes

Prothrombin gene mutation G20210A DNA analysis Yes Yes Yes

Protein C deficiency Protein C activity (<65%) Yes No No

Protein S deficiency Functional assay (<55%) Noa No No

Antithrombin deficiency Antithrombin activity (<60%) Yes No No

aIf screening in pregnancy is necessary, cutoff values for free protein S antigen levels in the second and third trimesters have

been identified at less than 30% and less than 24%, respectively.

Reproduced with permission from American College of Obstetricians and Gynecologists’ Committee on Practice Bulletins–

Obstetrics. ACOG Practice Bulletin No. 197: Inherited Thrombophilias in Pregnancy, Obstet Gynecol 2018 Jul;132(1):e18-e34.Thromboembolic Disorders 981

CHAPTER 55

challenge is that many o the

common iagnostic tests that

have been investigate extensively in nonpregnant patients

have not been valiate appropriately in pregnancy (Huisman,

2015). Shown in Figure 55-2

is one iagnostic algorithm that

can be use or evaluation o

pre gnant women. With a ew

moications, we ollow a

similar evaluation at Parklan

Hospital.

Compression

Ultrasonography

With suspecte DV, the initial

iagnostic test recommene is

compression ultrasonography o

the proximal leg veins (American College o Obstetricians

an Gynecologists, 2020b). Te

iagnosis is base on the noncompressibility an typical echoarchitecture o a thrombose

vein. For nonpregnant patients

with suspecte thrombosis, the

saety o withholing anticoagulation or 1 week has been establishe in the setting o a normal

compression ultrasoun examination (Birwell, 1998). Serial

compression examinations are

then perorme because isolate

unetecte cal thromboses that

ultimately exten into the proximal veins will o so within 1 to

2 weeks o presentation in approximately a ourth o patients.

In pregnant women, however, normal nings with ultrasonography o not always exclue a pulmonary embolism. Tis is

because the thrombosis may have alreay embolize or because

it arose rom iliac or other eep-pelvic veins, which are less accessible to ultrasoun evaluation.

wo stuies are helpul or evaluating whether serial

examinations are necessary or pregnant women suspecte

o having a DV. Chan an associates (2013) stuie 221

pregnant an postpartum women presenting with a suspecte

DV. Te 205 women with a negative initial stuy result

unerwent serial testing, which was negative in all cases. O

these, one woman ha a pulmonary embolism 7 weeks later.

Le Gal an colleagues (2012) stuie 210 pregnant an postpartum women with a suspecte DV. O these, 177 women

without a DV were not anticoagulate an i not unergo

serial testing. wo ha an objectively conrme thrombosis

iagnose within 3 months. In sum, these ata suggest that a

negative single complete compression ultrasonography stuy

may saely exclue the iagnosis o DV in most pregnant

women.

FIGURE 55-2 Algorithm to evalate suspected venous thromboembolism in pregnancy depending upon number of YEARS criteria met. Evidence of DVT should prompt compression ultrasound.

aThe three YEARS criteria are clinical signs of DVT, hemoptysis, and PE as the most likely diagnosis.

bThe American Thoracic Society and Society of Thoracic Radiology recommend CXR; cThe American

Society of Hematology recommends pulmonary V/Q scanning. CT = computed tomography;

CTA = computed tomography angiography; CXR = chest x-ray; DVT = deep-vein thrombosis;

MR = magnetic resonance; PE = pulmonary embolism; V/Q scan = ventilation perfusion scan.

(From Bates, 2018; Guyatt, 2012; Leung, 2011; van der Hulle, 2019; van der Pol, 2019.)

Magnetic Resonance Imaging

Tis imaging technique allows excellent elineation o anatomical etail above the inguinal ligament. Tus, in many

cases, magnetic resonance (MR) imaging is immensely useul

or iagnosis o ilioemoral an pelvic vein thrombosis. Te

venous system can also be reconstructe using MR venography (Chap. 49, p. 875). MR imaging is 100-percent sensitive

an 90-percent specic or etection o venographically proven

DV in nonpregnant patients (Erman, 1990). Importantly,

almost hal o those without DV were oun to have nonthrombotic conitions that inclue cellulitis, myositis, eema,

hematomas, an supercial phlebitis.

d-dimer Screening Tests

Tese specic brin egraation proucts are generate when

brinolysin (plasminogen) egraes brin, as occurs in thromboembolism. Teir measurement is usually incorporate into

iagnostic algorithms or VE in nonpregnant patients (Righini,

2018). Tis metho was shown to be accurate in the YEARS stuy

(van er Hulle, 2017). Screening with the d-imer test in pregnancy, however, is problematic or several reasons. Depening982

Section 12

Medical and Surgical Complications

on assay sensitivity, d-imer serum levels rise with gestational

age along with substantively elevate plasma brinogen concentrations (Appenix, p. 1228) (Johnson, 2019). Levels are also

aecte by multietal gestation an cesarean elivery (Hu, 2020;

Miyamoto, 2020). Last, d-imer concentrations can also be elevate in pregnancy complications such as placental abruption,

preeclampsia, an sepsis (Cunningham, 2015).

Tree large prospective stuies have evaluate d-imer testing in pregnant women to iagnose VE. Righini an colleagues (2018) oun that d-imer measurement, bilateral leg

ultrasoun, an compute tomography-angiogram saely rule

out VE. Subsequently, van er Pol an coworkers (2019)

use a pregnancy-aapte YEARS algorithm to iagnose suspecte pulmonary embolism. Te algorithm use d-imer

measurements an clinical suspicion to conrm VE. With

none o the clinical criteria present, an with the d-imer level

<1000 ng/mL, or with one or more clinical criteria present an

with the d-imer level <500 ng/mL, pulmonary embolism was

saely exclue (see Fig. 55-2). Last, the DiPEP stuy showe

that neither clinical risk actors nor d-imer levels were accurate in ientiying pulmonary embolism in pregnant women

(Gooacre, 2019). Tey conclue that pregnant women with

suspecte pulmonary embolism shoul all unergo imaging

stuies.

TABLE 55-4. Anticoagulation Regimen Definitions

Anticoagulation Regimen Definition

Prophylactic LMWHa Enoxaparin, 40 mg SC once daily

Dalteparin, 5000 units SC once daily

Nadroparin, 2850 units SC once daily

Tinzaparin, 4500 units SC once daily

Therapeutic LMWHb Enoxaparin, 1 mg/kg every 12 hours

Dalteparin, 200 units/kg once daily

Tinzaparin, 175 units/kg once daily

Dalteparin, 100 units/kg every 12 hours

Target an anti-Xa level in the therapeutic range of 0.6–1.0 units/mL 4 hours after last injection

for twice daily regimen; slightly higher doses may be needed for a once-daily regimen.

Prophylactic UFH UFH, 5000–7000 units SC every 12 hours in first trimester

UFH 7500–10,000 units SC every 12 hours in the second trimester

UFH, 10,000 units SC every 12 hours in the third trimester, unless the aPTT is elevated

Adjusted dose

(therapeutic) UFHb

UFH, 10,000 units or more SC every 12 hours in doses adjusted to target aPTT in the

therapeutic range (1.5–2.5 × control) 6 hours after injection

Postpartum

anticoagulation

Prophylactic, intermediate, or adjusted-dose LMWH for 6–8 weeks as indicated. Oral

anticoagulants may be considered postpartum based upon planned duration of therapy,

lactation, and patient preference.

Surveillance Clinical vigilance and appropriate objective investigation of women with symptoms suspicious

of deep vein thrombosis or pulmonary embolism. VTE risk assessment should be performed

prepregnancy or early in pregnancy and repeated if complications develop, particularly

those necessitating hospitalization or prolonged immobility.

aAlthough at extremes of body weight, modification of dose may be required.

bAlso referred to as weight adjusted, full treatment dose.

aPTT = activated partial thromboplastin time; INR = international normalized ratio; LMWH = low-molecular-weight heparin;

SC = subcutaneously; UFH = unfractionated heparin.

Reproduced with permission from American College of Obstetricians and Gynecologists’ Committee on Practice Bulletins–

Obstetrics. ACOG Practice Bulletin No. 197: Inherited Thrombophilias in Pregnancy, Obstet Gynecol 2018 Jul;132(1):e18-e34.

Te nings o the YEARS stuy an subsequent others

will no oubt guie care. However, currently the American

College o Obstetricians an Gynecologists (2020b) recommens

against the use o d-imer assays to iagnose VE in pregnant

women. We agree, an in our view, these positive results nee to

be uplicate beore aaptation o their wiesprea use in pregnant women. Still, a negative d-imer assay shoul be reassuring,

unless VE is the most likely iagnosis (Viau-Lapointe, 2020).

■ Management of Venous Thromboembolism

Optimal management o VE uring pregnancy has not unergone major clinical stuy to provie evience-base practices.

Tere is, however, consensus or treatment with anticoagulation

an limite activity. Trombophilia testing oes not change

treatment (Connors, 2017). Anticoagulation is initiate with

either unractionate heparin (UFH) or LMWH (Table 55-4).

Although either type is acceptable, most recommen one o the

LMWHs (American College o Obstetricians an Gynecologists, 2020b; Kearon, 2016). Te American Society o Hematology recommens preerential use o LMWH uring pregnancy

because o better bioavailability, longer plasma hal-lie, more

preictable ose response, reuce risks o osteoporosis an

thrombocytopenia, an less requent osing (Bates, 2018).Thromboembolic Disorders 983

CHAPTER 55

Te uration o ull anticoagulation varies, an no stu-

ies have ene the optimal uration or pregnancy-relate

thromboembolism. In nonpregnant patients, evience supports a minimum treatment uration o 3 months (Kearon,

2012). For pregnant patients, the American College o Obstetricians an Gynecologists (2020b) recommens therapeutic

anticoagulation or 3 to 6 months ollowe by intermeiate

or prophylactic therapy or the uration o pregnancy an the

postpartum perio. I the woman is not breasteeing, irect

acting oral agents can be consiere (p. 985).

Recall that pulmonary embolism evelops in as many as 60

percent o patients with untreate venous thrombosis, an anticoagulation ecreases this risk to less than 5 percent. In nonpregnant patients, the mortality rate with a pulmonary embolism

approximates 1 percent (Douketis, 1998; Pollack, 2011).

With DV, over several ays, leg pain issipates. Ater

symptoms have abate, grae ambulation is begun. Elastic

stockings are tte, an anticoagulation is continue. Recovery

to this stage usually takes 7 to 10 ays. Grauate compression

stockings can be continue or 2 years ater the iagnosis to

reuce the incience o postthrombotic synrome (Branjes,

1997). Symptoms inclue chronic leg paresthesias or pain,

intractable eema, skin changes, an leg ulcers. Accoring to

the American Society o Hematology, catheter-irecte thrombolytic therapy oes not mitigate against this complication

(Bates, 2018).

Unfractionated Heparin

Tis agent shoul be consiere or the initial treatment o

thromboembolism an in situations in which elivery, surgery,

or thrombolysis may be necessary (American College o Obstetricians an Gynecologists, 2020b). UFH can be aministere

by one o two alternatives: (1) initial intravenous therapy ollowe by ajuste-ose subcutaneous UFH given every 12

hours; or (2) twice-aily, ajuste-ose subcutaneous UFH

with oses ajuste to prolong the activate partial thromboplastin time (aP) into the therapeutic range 6 hours post

injection (Table 55-5).

For intravenous therapy, several protocols are acceptable.

In general, i UFH is use, it is initiate with a bolus intravenous ose o 70 to 100 U/kg, which is 5000 to 10,000 U.

Tis is ollowe by continuous intravenous inusions beginning at 1000 U/hr or 15 to 20 U/kg/hr. Tis inusion rate is

titrate to achieve an aP 1.5 to 2.5 times control values

(Linnemann, 2016). Intravenous anticoagulation is maintaine or about 5 to 7 ays, ater which treatment is converte to subcutaneous heparin to maintain the aP to at

least 1.5 to 2.5 times control throughout the osing interval.

For women with lupus anticoagulant, aP oes not accurately assess heparin anticoagulation, an thus anti–actor Xa

levels are preerre.

Low-molecular-weight Heparin

Tis is a amily o erivatives o unractionate heparin, an

their molecular weights average 4000 to 5000 altons compare with 12,000 to 16,000 altons or conventional heparin.

None o these heparins cross the placenta, an all exert their

anticoagulant activity by activating antithrombin. UFH has

equivalent activity against actor Xa an thrombin compare

with LMWHs, which have greater activity against actor Xa

than against thrombin. Tey also have a more preictable

anticoagulant response an ewer bleeing complications than

UFH because o their better bioavailability, longer hal-lie,

ose-inepenent clearance, an ecrease intererence with

platelets (American College o Obstetricians an Gynecologists,

2020b; Bates, 2018). Teir longer hal-lie prohibits neuraxial

analgesia in laboring women. LMWH compouns are cleare

by the kineys an must be use cautiously when there is renal

ysunction.

Several stuies have shown that VE is treate eectively

with LMWH (Quinlan, 2004; apson, 2008). Using serial

venograms, Brein an associates (2001) observe that these

compouns were more eective than UFH in reucing thrombus size without increasing mortality rates or major bleeing

complications. Several ierent treatment regimens using

LMWH or treatment o acute VE are liste in able 55-4

(American College o Obstetricians an Gynecologists, 2020b).

Whether LMWH requires ajustments uring the course

o pregnancy is controversial (Berresheim, 2014). Te American College o Obstetricians an Gynecologists (2020b) recommens monitoring with an anti-Xa level 4 hours ater an

injection with ose ajustment to maintain a therapeutic

level. Large stuies using clinical en points that emonstrate

an optimal therapeutic range or show that ose ajustments

increase therapy, saety, or efcacy are lacking. Accoringly, the

American Society o Hematology an others have conclue

that routine monitoring with anti-Xa levels is ifcult to justiy

(Bates, 2018; McDonnell, 2017).

In general, heparin pharmacokinetics inclues more rapi

clearance o heparins as pregnancy progresses. Te peaks o

anti-Xa activity o enoxaparin (Lovenox) across pregnancy are

shown in Figure 55-3. Similarly, tinzaparin (Innohep) with a

osage o 75 to 175 U/kg/ was necessary to achieve peak anti–

actor Xa levels o 0.1 to 1.0 U/mL (Smith, 2004). In stuies

FIGURE 55-3 Anti–factor Xa activity with repeated daily administration of 40 mg enoxaparin subcutaneously in nonpregnant

women and those from each trimester (Data from Lebaudy, 2008).984

Section 12

Medical and Surgical Complications

TABLE 55-5. Some Recommendations for Thromboprophylaxis During Pregnancy

Clinical Scenario

Pregnancy Postpartum

ACOGa ASHb ACOGa ASHb

Prior single VTE

Risk factor no longer

present

Surveillance only Surveillance

only

Surveillance or prophylactic

heparin with additional

risk factorsf

Prophylactic or

intermediatedose heparin

Pregnancy- or estrogenrelated or no known

association

Prophylactic or

intermediate-dose

heparin

Prophylactic or

intermediatedose heparin

Prophylactic, intermediate,

or therapeutic-dose

heparin

Prophylactic or

intermediatedose heparin

Associated with a highrisk thrombophiliad or

affected first-degree

relative

Prophylactic, intermediate,

or therapeutic-dose

heparin

NSS Prophylactic, intermediate,

or therapeutic-dose

heparin

Prophylactic or

intermediatedose heparin

Associated with

a low-risk

thrombophiliae

Prophylactic or

intermediate-dose

heparin

NSS Prophylactic or

intermediate-dose

heparin

Prophylactic or

intermediatedose heparin

Two or more prior VTEs with or without thrombophilia

Not receiving

long-term

anticoagulation

Intermediate or therapeuticdose heparin

NSS Intermediate or therapeuticdose heparin

Prophylactic or

intermediatedose heparin

Receiving long-term

anticoagulation

Therapeutic-dose heparin Therapeuticdose heparin

Resume long-term

anticoagulation

Resume long-term

anticoagulation

No prior VTE

No thrombophilia Surveillance NSS Surveillance or prophylactic

heparin with additional

risk factorsf

NSS

High-risk

thrombophilia

Surveillance only or

prophylactic-dose heparin

Surveillance Prophylactic or intermediatedose heparin

Prophylactic

heparin

Positive family history

VTE and low-risk

thrombophiliac

Surveillance or prophylacticdose heparin

Surveillance Prophylactic or

intermediate-dose

heparin

Surveillance

Low-risk

thrombophilia

Surveillance Surveillance Surveillance; prophylacticdose heparin with

additional risk factorsf

Surveillance

Compound

heterozygous

thrombophilia

Prophylactic or

intermediate-dose

heparin

Prophylactic

heparin

Prophylactic or

intermediate-dose

heparin

Prophylactic

heparin

Antiphospholipid antibodies

History of VTE Prophylactic heparin (?plus

low-dose aspirin)

NSS Prophylactic heparin;

referral to specialistg

NSS

No prior VTE Surveillance or prophylactic

heparin plus low-dose

aspirin with recurrent

pregnancy loss or stillbirthf

NSS Prophylactic heparin plus

low-dose aspirin × 6 weeks

with recurrent pregnancy

loss or stillbirthg

NSS

aAmerican College of Obstetricians and Gynecologists 2020a,b.

bAmerican Society of Hematology (Bates, 2018).

cPostpartum treatment levels should be ≥ antepartum treatment.

dAntithrombin deficiency; doubly heterozygous or homozygous for prothrombin 20210A and factor V Leiden.

eHeterozygous factor V Leiden or prothrombin 20210A; protein S or C deficiency.

fFirst-degree relative with VTE at <50 years; other major thrombotic risk factors, e.g., obesity, prolonged immobility.

gWomen with antiphospholipid syndrome should not use estrogen-containing contraceptives.

hTreatment is recommended if the diagnosis of antiphospholipid syndrome is based on three or more prior pregnancy losses.

NSS = not specifically stated; VTE = venous thromboembolism.

Prophylactic, intermediate-, and adjusted-dose regimens are listed in Table 55-4.Thromboembolic Disorders 985

CHAPTER 55

o alteparin (Fragmin) pharmacokinetics, conventional starting oses o alteparin—100 U/kg every 12 hours—were

likely insufcient to maintain ull anticoagulation in pregnant

women (Barbour, 2004; Jacobsen, 2003). Tus, slightly higher

oses than that shown in able 55-4 may be require.

Any concerns regaring teratogenicity o enoxaparin have

been obviate (American College o Obstetricians an Gynecologists, 2020b; Bates, 2021). A comprehensive stuy o 1267

pregnant women treate with tinzaparin oun no cases o heparin-inuce thrombocytopenia, maternal eaths, or complications rom regional analgesia (Nelson-Piercy, 2011). A total

o 43 women (3.4 percent) require meical intervention or

bleeing. O 15 stillbirths, our were juge as possibly being

relate to tinzaparin use. LMWHs have been juge to be sae

uring breasteeing (Bates, 2018).

Labor and Delivery

Women receiving either therapeutic or prophylactic anticoagulation shoul be converte rom LMWH to the shorter-hal-

lie UFH in the last month o pregnancy or sooner i elivery

appears imminent. Te purpose o conversion to UFH has less

to o with any risk o maternal bleeing at the time o elivery,

but rather with neuraxial blockae complicate by an epi-

ural or spinal hematoma (Chap. 25, p. 478). Te Society or

Obstetric Anesthesia an Perinatology recommens that UFH

low-ose thromboprophylaxis be withhel or 4 to 6 hours;

UFH intermeiate-ose thromboprophylaxis or 12 hours; an

UFH therapeutic ose or 24 hours beore neuraxial analgesia

(Leert, 2018). For LMWH, the Society recommens to withhol low oses or 12 hours; intermeiate oses 12 to 24 hours;

an therapeutic oses 24 hours.

I a woman begins labor while taking UFH, clearance can

be verie by an aP. Reversal o heparin with protamine

sulate—1 mg per 100 units heparin with a maximum ose o

50 mg—is rarely require an is not inicate with a prophylactic ose o heparin. For women in whom anticoagulation

therapy has temporarily been iscontinue, pneumatic compression evices are recommene.

Warfarin Compounds

Vitamin K antagonists are generally contrainicate because they

reaily cross the placenta an cause etal eath an malormations

rom hemorrhages. Tis is iscusse in etail in Chapter 8 (p.

156). Teir sole antepartum use is in women with a mechanical

heart valve (Daughety, 2020). Tese compouns o not accumulate in breast milk an are thus sae uring breasteeing.

Postpartum venous thrombosis is usually treate with intravenous heparin an oral wararin initiate simultaneously. Te

initial ose o wararin is 5 to 10 mg or the rst 2 ays. Subsequent oses are titrate to achieve an international normalize

ratio (INR) o 2 to 3. o avoi paraoxical thrombosis an

skin necrosis rom the early antiprotein C eect o wararin,

these women are maintaine on therapeutic oses o UFH or

LMWH or 5 ays an until the INR is in a therapeutic range

or 2 consecutive ays (American College o Obstetricians an

Gynecologists, 2020b).

reatment in the puerperium may require larger oses

o anticoagulant. Brooks an colleagues (2002) compare

anticoagulation in postpartum women with that o age-matche

nonpregnant controls. Te ormer require a signicantly larger

meian total ose o wararin—45 versus 24 mg—an a longer

time—7 versus 4 ays—to achieve the target INR.

Direct-acting Oral Anticoagulants

O these newer oral anticoagulants, abigatran (Praaxa)

inhibits thrombin an rivaroxaban (Xarelto) an apixaban

(Eliquis) inhibit actor Xa. Tese agents are becoming the

rugs o choice or nonpregnant subjects with thromboembolism (Kruger, 2019). Currently, very ew reports aress these

newer agents uring pregnancy an breasteeing, an thus the

human reprouctive risks are essentially unknown (Lameijer,

2018; Rosenbloom, 2019). Despite this, the GARFIELD-VE

stuy reporte that irect-acting oral anticoagulant use in pregnant women was gaining avor (Jerjes-Sanchez, 2021). Dabigatran crosses the placenta, however, it is unknown whether any

o these agents are excrete in breast milk (Bapat, 2014). At

this time, because o the potential or inant harm, a ecision

shoul be mae to either avoi breasteeing or use an alternative anticoagulant, such as wararin, in breasteeing women

(Bates, 2021; Kruger, 2019).

■ Complications of Anticoagulation

Tree signicant complications associate with anticoagulation

are hemorrhage, thrombocytopenia, an osteoporosis. Te latter two are unique to heparin, an their risk may be reuce

with LMWHs. Te most serious complication is hemorrhage,

which is more likely i there has been recent surgery or lacerations. roublesome bleeing is also more likely i the heparin

osage is excessive. Unortunately, management schemes using

laboratory testing to ientiy when a heparin osage is sufcient to inhibit urther thrombosis, yet not cause serious hemorrhage, have been iscouraging.

Heparin-induced Thrombocytopenia

Tere are two types—the most common is a nonimmune,

benign, reversible thrombocytopenia that evelops within

the rst ew ays o therapy an resolves in approximately 5

ays without therapy cessation. Te secon is the severe orm

o heparin-inuce thrombocytopenia, which results rom an

immune reaction involving IgG antiboies irecte against

complexes o platelet actor 4 an heparin. Te iagnosis o

heparin-inuce thrombocytopenia (HI) is base on a rop

in the platelet count o more than 50 percent or thrombosis

beginning 5 to 10 ays ater the start o heparin in association

with the appearance o platelet-activating HI antiboies. Te

all in platelet count in HI occurs rapily—over a perio o 1

to 3 ays—an is assesse relative to the highest platelet count

ater the start o heparin. Te typical nair is 40,000 to 80,000

platelets per microliter (Greinacher, 2015).

Although the incience o HI is approximately 3 to 5

percent in nonpregnant iniviuals, it is <0.1 percent in

obstetrical patients (American College o Obstetricians an

Gynecologists, 2020b). Fausett an coworkers (2001) reporte

no cases among 244 heparin-treate gravias compare with 10

among 244 nonpregnant patients. Accoringly, the American986

Section 12

Medical and Surgical Complications

College o Chest Physicians recommens against platelet count

monitoring when the risk o HI is consiere to be <1 percent (Linkins, 2012). In others, they suggest monitoring every

2 or 3 ays rom ay 4 until 14.

Women with a presumptive iagnosis o HI shoul have

laboratory testing or antiplatelet antiboies or by unctional

assays. However, these tests oten require several ays to return

results, so initial management shoul be irecte by clinical

nings. Heparin therapy is stoppe, an alternative anticoagulation initiate. LMWH may not be entirely sae because it

has some cross reactivity with UFH. Te American Society o

Hematology recommens use o onaparinux (Arixtra), which

is a pentasaccharie actor Xa inhibitor (Bates, 2018). Successul use in pregnancy has been reporte (De Carolis, 2015;

Elsaigh, 2015). Last, platelet transusions are avoie (Greinacher, 2015).

Heparin-induced Osteoporosis

Bone loss may evelop with long-term heparin aministration—usually 6 months or longer—an is more prevalent in

cigarette smokers. UFH can cause osteopenia, an this is less

likely with LMWHs. Women treate with any heparin shoul

be encourage to take an oral aily 1500-mg calcium supplement (Cunningham, 2005; Lockwoo, 2012).

■ Anticoagulation and Abortion

Te treatment o DV with heparin oes not preclue pregnancy termination by careul curettage. Ater the proucts are

remove without trauma to the reprouctive tract, ull-ose

heparin can be restarte in several hours.

■ Anticoagulation and Delivery

Te eects o heparin on bloo loss at elivery epen on several variables: (1) ose, route, an timing o aministration; (2)

number an epth o incisions an lacerations; (3) intensity o

postpartum myometrial contractions; an (4) presence o other

coagulation eects. Bloo loss shoul not be greatly increase

with vaginal elivery i there are no lacerations, an the uterus

promptly contracts. Unortunately, such ieal circumstances o

not always prevail. Tus, heparin therapy generally is stoppe

uring labor an elivery.

Te American College o Obstetricians an Gynecologists

(2020b) recommens restarting UFH or LMWH no sooner

than 4 to 6 hours ater vaginal elivery or 6 to 12 hours ater

cesarean elivery. From their review, Moriuchi an associates

(2019) oun that thromboprophylaxis starte within 24 hours

ater cesarean elivery was sae. At Parklan Hospital, we wait

at least 24 hours to restart therapeutic-ose heparin ater cesarean elivery or ater vaginal elivery with signicant lacerations.

SUPERFICIAL VENOUS THROMBOPHLEBITIS

Trombosis limite strictly to the supercial veins o the

saphenous system is typically associate with varicosities or as a

sequela o an inwelling intravenous catheter. It is treate with

analgesia, elastic support, heat, an rest. Supercial vein thrombosis raises the risk o DV our- to six-ol. In a metaanalysis

o nonpregnant patients with thrombophlebitis, 18 percent ha

a concomitant DV an 7 percent ha a pulmonary embolism (Di Minno, 2016). Although controversial, the American

Society o Hematology suggests LMWH therapy (Bates, 2018).

PULMONARY EMBOLISM

Although it causes approximately 10 percent o maternal

eaths, pulmonary embolism is relatively uncommon uring

pregnancy an the puerperium. Accoring to Marik an Plante

(2008), 70 percent o gravias presenting with a pulmonary

embolism have associate clinical evience o DV. An recall

that between 30 an 60 percent o women with a DV will

have a coexisting silent pulmonary embolism.

■ Clinical Presentation

In almost 2500 nonpregnant patients with a proven pulmonary

embolism, symptoms inclue yspnea in 82 percent, chest

pain in 49 percent, cough in 20 percent, syncope in 14 percent, an hemoptysis in 7 percent (Golhaber, 1999). Other

preominant clinical nings typically inclue tachypnea,

apprehension, an tachycaria. In some cases, an accentuate

pulmonic closure soun, rales, an/or riction rub is hear.

Right axis eviation an -wave inversion in the anterior

chest leas may be evient on the electrocariogram. In most

cases, chest raiography results are normal. In others, nonspecic nings may inclue atelectasis, an inltrate, cariomegaly,

or an eusion. Vascular markings in the lung region supplie

by the obstructe artery can be lost. Although most women

are hypoxemic, a normal arterial bloo gas analysis oes not

exclue pulmonary embolism. Approximately a thir o young

patients have PO2 values >80 mm Hg. Tus, the alveolar- arterial oxygen tension ierence is a more useul inicator o isease. More than 86 percent o patients with acute pulmonary

embolism will have an alveolar-arterial ierence >20 mm Hg

(Lockwoo, 2012). Even with massive pulmonary embolism,

signs, symptoms, an laboratory ata to support the iagnosis

may be eceptively nonspecic.

■ Massive and Submassive Pulmonary

Embolism

Dene as embolism causing hemoynamic instability, these

account or 5 to 10 percent o cases (Hanal-Örece, 2019).

Acute mechanical obstruction o the pulmonary vasculature

causes increase vascular resistance an pulmonary hypertension ollowe by acute right ventricular ilation. In otherwise

healthy patients, signicant pulmonary hypertension oes not

evelop until 60 to 75 percent o the pulmonary vascular tree

is occlue (Guyton, 1954). Moreover, circulatory collapse

requires 75- to 80-percent obstruction. Tis is epicte in

Figure 55-4 an emphasizes that most acutely symptomatic

emboli are large an likely a sale embolism (Singhal, 1973).

Tese are suspecte when the pulmonary artery pressure is substantively increase as estimate by echocariography.Thromboembolic Disorders 987

CHAPTER 55

Submassive embolism is iagnose when there is evience

o right ventricular ysunction (Hanal-Örece, 2019). Te

mortality rate approaches 25 percent, which compares with

a 1-percent rate without such ysunction (Kinane, 2008).

It is important in these cases to inuse crystallois careully

an to support bloo pressure with vasopressors. Aggressive

intravenous ui inusion has been associate with worsening right-sie ventricular ysunction (Konstanties, 2014).

As iscusse on page 988, oxygen treatment, tracheal intubation, an mechanical ventilation are complete preparatory to

thrombolysis, lter placement, or embolectomy.

■ Diagnosis

In most cases, recognition o a pulmonary embolism requires

a high inex o suspicion that prompts objective evaluation.

Exposure o the mother an etus to ionizing raiation is a concern when investigating a suspecte pulmonary embolism uring pregnancy. However, this concern is largely overrule by

the hazars o missing a potentially atal iagnosis. Moreover,

erroneously assigning a iagnosis o pulmonary embolism to a

pregnant woman is also raught with problems. It unnecessarily exposes the mother an etus to the risks o anticoagulation

treatment an will impact elivery plans, uture contraception, an thromboprophylaxis uring subsequent pregnancies.

Tereore, investigations shoul aim or iagnostic certainty

(Cohen, 2020).

Diagnosis o a pulmonary embolism ollows the same

algorithm as or DV shown in Figure 55-2. In aition to

compression ultrasoun o the extremities, i a pulmonary

embolism is suspecte, chest x-ray an electrocariogram may

be revealing. Echocariography is useul to etect other con-

itions that mimic pulmonary embolism—acute myocarial

inarction, pericarial tamponae, an aortic issection. Further imaging with compute tomography (C) scanning, MR

imaging, or ventilation-perusion lung scanning conrms the

iagnosis (American College o Obstetricians an Gynecologists, 2020b). As iscusse later (p. 988), the American Society o Hematology recommens lung scanning as the rst-line

iagnostic tool (Bates, 2018). At Parklan Hospital we preerentially use C-pulmonary angiography or suspecte pulmonary embolism.

Computed Tomographic Pulmonary Angiography

Multietector C-pulmonary angiography (CPA) is the most

commonly employe technique to iagnose pulmonary embolism in nonpregnant patients. Te technique is escribe urther in Chapter 49 (p. 874), an an imaging example is shown

in Figure 55-5. Te estimate etal raiation exposure averages

0.45 to 0.6 mGy. Te estimate maternal breast ose is 10 to

70 mGy (Waksmonski, 2014).

CPA has many avantages, but we n that the higher

resolution allows etection o previously inaccessible smaller

istal emboli that have uncertain clinical signicance. Similar

observations have been reporte by others (Anerson, 2007;

Hall, 2009). Bourjeily an colleagues (2012) perorme a

ollow-up stuy o 318 pregnant women who ha a negative

CPA perorme or a suspecte pulmonary embolism. All

Pulmonary trunk

Diameter = 3 cm; total area = 9 cm2

Segmental arteries (19)

Dia. 6 mm each;

total area = 36 cm2

Lobar arteries (5)

Dia. 8 mm each;

total area = 13 cm2

Subsegmental arteries (65)

Dia. 4 mm each;

total area = 817 cm2

Right and left pulmonary artery

Dia. 1.5 cm each; total area = 9 cm2

Left lobar arteries (2)

Right lobar arteries (3)

FIGURE 55-4 Schematic of pulmonary arterial circulation. Note that the cross-sectional area of the pulmonary trunk and the combined

pulmonary arteries is 9 cm. A large saddle embolism could occlude 50 to 90 percent of the pulmonary tree, causing hemodynamic instability. As the arteries give off distal branches, the total surface area rapidly increases, that is, 13 cm for the combined five lobar arteries,

36 cm for the combined 19 segmental arteries, and more than 800 cm for the total 65 subsegmental arterial branches. Thus, hemodynamic

instability is less likely with emboli past the lobar arteries.988

Section 12

Medical and Surgical Complications

were seen 3 months ollowing their initial presentation or at

6 weeks postpartum. None o these women were subsequently

iagnose with a thromboembolism.

Ventilation-Perfusion Scintigraphy Lung Scan

Tis technique involves a small ose o raiotracer such as

intravenously aministere technetium-99m macroaggregate

albumin. Tere is negligible etal an maternal breast raiation

exposure—0.1 to 0.4 mGy. Te scan may not provie a e-

nite iagnosis because many other conitions can cause perusion eects. Examples are pneumonia or local bronchospasm.

Chan an coworkers (2002) oun that a ourth o ventilationperusion scans in pregnant women were noniagnostic. In

these instances, CPA is preerre (romeur, 2017).

o compare the perormance o lung scintigraphy an

CPA, Revel an associates (2011) evaluate 137 pregnant

women with suspecte pulmonary embolism. Te two moalities perorme comparably an ha no signicant ierences

between the proportions o positive, negative, or ineterminate

results. Specically, the proportion o ineterminate results or

both approximate 20 percent, Similarly, one systematic review

conclue that both CPA an lung scintigraphy seem appropriate or exclusion o pulmonary embolism uring pregnancy

(van Mens, 2017).

Intravascular Pulmonary Angiography

Tis requires catheterization o the right sie o the heart an

is consiere the reerence test or pulmonary embolism. With

newer-generation multietector C scanners, however, the

role o invasive pulmonary angiography has been questione.

Tis is especially true given the higher raiation exposure or

the etus (Konstantinies, 2014). Other etractions are that it

can be time consuming, uncomortable, an associate with

ye-inuce allergy an renal ailure. It is reserve or conrmation when less invasive tests are equivocal.

■ Management

Immeiate treatment or pulmonary embolism is ull anticoagulation similar to that or DV as iscusse on page 982 an

shown in able 55-4. Tere are several complementary proce-

ures that may be inicate.

Vena Caval Filters

Te woman who has very recently suere a pulmonary embolism an who must unergo cesarean elivery presents a particularly serious problem. Reversal o anticoagulation may be

ollowe by another embolus, an surgery while ully anticoagulate requently results in lie-threatening hemorrhage or

troublesome hematomas. In these cases, placement o a vena

caval lter shoul be consiere beore surgery (Marik, 2008).

Moreover, in the very inrequent circumstances in which heparin therapy ails to prevent recurrent pulmonary embolism

rom the pelvis or legs or in which embolism evelops rom

these sites espite heparin treatment, a vena caval lter may also

be inicate. Such lters can also be use ollowing massive

emboli in patients who are not caniates or thrombolysis.

Te evice is inserte through either the jugular or emoral

vein an can be inserte uring labor (Jamjute, 2006). Routine

lter placement has no ae avantage to heparin given alone

(Decousus, 1998). Retrievable lters may be use as short-term

protection an then remove 1 to 2 weeks later (Liu, 2012).

From their systematic review, Harris an colleagues (2016)

oun that complication rates in pregnant women with vena

caval lters are comparable to those in nonpregnant patients.

In another series o 24 lters place in obstetrical patients, 29

percent ha complications incluing removal ailure (Rottenstreich, 2019).

Thrombolysis

Compare with heparin, thrombolytic agents provie more

rapi lysis o pulmonary clots an improvement o pulmonary

hypertension in lie-threatening cases (Bates, 2021; Golhaber, 2018). In one stuy o nonpregnant patients receiving

heparin or an acute submassive pulmonary embolism, those

receiving the recombinant tissue plasminogen activator alteplase

ha a lower risk o eath or treatment escalation (Konstantinies, 2002). One metaanalysis o trials involving nonpregnant patients reporte that the risk o recurrence or eath was

signicantly lower in patients given thrombolytic agents an

heparin compare with those given heparin alone—10 versus

17 percent (Agnelli, 2002). Importantly, however, there were

ve—2 percent—atal bleeing episoes in the thrombolysis

group an none in the heparin-only group.

Tere have been several case reports an reviews o thrombolysis in pregnant women. Leonhart an coworkers (2006)

ientie 28 reports o tissue plasminogen activator use uring pregnancy. en cases were or thromboembolism. Complication rates were similar to those in nonpregnant patients,

an the authors conclue that such therapy shoul not be

withhel uring pregnancy i inicate. Akazawa an Nishia

FIGURE 55-5 Axial image of the chest from a four-channel multidetector spiral computed tomographic scan performed after

administration of intravenous contrast. There is enhancement of

the pulmonary artery with a large thrombus on the right (arrow)

consistent with pulmonary embolism. (Reproduced with permission from Dr. Michael Landay.)Thromboembolic Disorders 989

CHAPTER 55

(2017) reviewe 13 cases o systemic thrombolytic therapy

aministere uring the rst 48 hours ater elivery. Bloo

transusion was require in ve o the eight cesarean eliveries,

incluing three cases o hysterectomy an two cases o hematoma rainage. Sousa-Gomes an associates (2019) reviewe

outcomes in 141 pregnant women unergoing thrombolysis

or VE or stroke. Maternal eath complicate 3 percent o

cases, an major bleeing occurre in 9 percent.

Embolectomy

Given the efcacy o thrombolysis an lters, surgical embolectomy is uncommonly inicate. Publishe experience with

emergency embolectomy uring pregnancy is limite to case

reports (Colombier, 2015; Saee, 2014). From their review,

Ahearn an colleagues (2002) oun that although the operative risk to the mother is reasonable, the stillbirth rate is 20 to

40 percent.

THROMBOPROPHYLAXIS

A Cochrane review o guielines or thromboprophylaxis in

pregnancy conclue that there is a lack o overall agreement

about which women shoul be oere thromboprophylaxis

(Mileton, 2021). Te American Society o Hematology summarize that evience-base recommenations rely largely on

observational stuies an on ata extrapolate rom nonpregnant patients. In many cases, this le them to suggest—rather

than recommend—various schemes or thromboprophylaxis

(Bates, 2018). In an earlier stuy, Cleary-Golman an

coworkers (2007) surveye 151 ellows o the American College o Obstetricians an Gynecologists an reporte that intervention without a clear inication is common. able 55-5 lists

several consensus recommenations or thromboprophylaxis.

In some cases, more than one option is liste, thus illustrating

the conusion that currently reigns. It is important to emphasize that the ecision to treat with anticoagulation in pregnancy

is inuence by personal VE history, amily history o VE,

severity o thrombophilias, an aitional risk actors such as

obesity, cesarean elivery, or prolonge immobility (American

College o Obstetricians an Gynecologists, 2020a).

■ Prior Venous Thromboembolism

In general, either antepartum surveillance or heparin prophylaxis is recommene or women with prior VE but without a

recurrent risk actor. engborn an associates (1989), however,

suggeste that such management may not be eective. Tey

reporte outcomes in 87 pregnant Sweish women who ha

prior thromboembolic isease an were not teste or thrombophilias. Despite unractionate heparin prophylaxis, which

was usually 5000 U twice aily, 15 percent evelope antepartum VE recurrence. Tis compare with 12 percent not

given heparin.

Brill-Ewars an colleagues (2000) prospectively stuie

125 pregnant women with a single prior VE. Antepartum

heparin was not given, but anticoagulant therapy was given or

4 to 6 weeks postpartum. Six women ha a recurrent venous

thrombosis, but there were no recurrences in women without

a known thrombophilia or whose prior thrombosis was associate with a temporary risk actor. A stuy o 88 women

without antiphospholipi antiboies an who were not given

antithrombotic prophylaxis reporte a subsequent pregnancyor puerperium-relate VE in 22 percent (De Steano, 2006).

O 20 women whose original thrombosis was associate with a

transient risk actor—not incluing pregnancy or oral contraceptive use—there were no recurrences uring pregnancy, but

two uring the puerperium. Tese nings imply that women

with a prior thrombosis in association with a thrombophilia or

in the absence o a temporary risk actor generally shoul be

given both antepartum an postpartum prophylaxis (Connors,

2017). Tese ata also suggest that or women with a prior

VE, antithrombotic prophylaxis uring pregnancy coul be

tailore accoring to the circumstances o the original event.

It is important to emphasize that VTE may recur despite antithrombotic prophylaxis. A stuy o 270 women who ha at least

one previous VE oun that 10 percent suere a recurrent

VE (Galambosi, 2014). welve o these recurrences occurre

early in pregnancy beore the initiation o antithrombotic prophylaxis, an 16 occurre espite prophylactic use o LMWH.

Our practice at Parklan Hospital or many years or

women with a history o prior VE was to aminister subcutaneous UFH, 5000 to 7500 units two to three times aily. More

recently, we have use 40 mg o enoxaparin given subcutaneously aily with transition to UFH near elivery. With either

regimen, the recurrence o ocumente DV embolization has

been uncommon.

■ Cesarean Delivery

Te risk or DV an especially or atal thromboembolism

rises many ol ollowing cesarean elivery compare with

that ater vaginal elivery. When consiering that a thir o

women giving birth in the Unite States yearly unergo cesarean elivery, pulmonary embolism is unerstanably a major

cause o maternal mortality (Creanga, 2017; Petersen, 2019).

Tat sai, the lack o consensus escribe earlier by Mileton an coworkers (2021) creates consierable variation in the

current recommenations promulgate by the American College o Obstetricians an Gynecologists, the Royal College o

Obstetricians an Gynaecologists, an the American College o

Chest Physicians (Palmerola, 2016).

Aherence to iniviual society guielines ollowing cesarean elivery coners a greater risk in some instances. Accoring

to the Society or Maternal-Fetal Meicine (2020), i American College guielines were ollowe, only 1 percent o these

women woul be caniates or thromboprophylaxis. Almost

85 percent o the same cohort woul be given thromboprophylaxis ollowing Royal College guielines an about 35 percent i

American College o Chest Physician guielines were ollowe.

Some comparisons o these thromboprophylaxis guielines are

shown in Table 55-6.

Since 2011, the American College o Obstetricians an

Gynecologists (2020b) recommens placement o pneumatic

compression evices beore cesarean elivery or all women not

alreay receiving thromboprophylaxis. Tis recommenation was

base primarily on consensus an expert opinion. For patients990

Section 12

Medical and Surgical Complications

unergoing cesarean elivery with aitional risk actors or

thromboembolism, both pneumatic compression evices an

UFH or LMWH may be recommene. Te College stipulate

that cesarean elivery in an emergency setting shoul not be elaye because o the time necessary to begin thromboprophylaxis.

Implementation o this strategy by the Hospital Corporation

o America, the largest or-prot obstetrical health care elivery

system in the Unite States, was associate with a reuction in

eaths rom pulmonary embolism rom 7 o 458,097 cesarean

births to 1 o 465,880 cesarean births (Clark, 2011, 2014).

In 2016, the National Partnership or Maternal Saety publishe several consensus recommenations or the prevention

o maternal VE (D’Alton, 2016). Tese recommenations

inclue expane use o antenatal thromboprophylaxis or

women hospitalize 3 ays or longer, expane use o prophylaxis uring an ater vaginal elivery, an expane use

o pharmacological prophylaxis to most women ater cesarean

elivery. In response, Sibai an Rouse (2016) expresse concern that these new recommenations erive rom sparse ata

o questionable applicability to obstetrical patients. Tey calle

or better quality evience to measure the benets, harms, an

costs o increase pharmacological thromboprophylaxis. As

aptly expresse by Macones (2017), “an intervention, such

as increase postcesarean pharmacologic thromboprophylaxis,

where there are legitimate concerns about efcacy an saety,

requires a much higher egree o evience beore a national

guieline is implemente.” We agree with these sentiments.

Nhận xét