Annals of Internal MedicineT

In the ClinicT

Deep Venous
Thrombosis
Prevention

V
enous thromboembolism (VTE) is the third
most common cardiovascular disorder,
affecting up to 5% of the population. VTE
commonly manifests as lower-extremity deep ve- Diagnosis
nous thrombosis (DVT) or pulmonary embolism.
Half of these events are associated with a transient
risk factor and may be preventable with prophy-
laxis. Direct oral anticoagulants are effective and Treatment
safe and carry a lower risk for bleeding than vita-
min K antagonists. Many patients with VTE will
have a chronic disease requiring long-term antico- Practice Improvement
agulation. Postthrombotic syndrome affects 25% to
40% of patients with DVT and significantly impacts
function and quality of life.

CME/MOC activity available at Annals.org.

Physician Writer doi:10.7326/AITC202209200

Lisa Duffett, MD
The Ottawa Hospital, Ottawa, This article was published at Annals.org on 13 September 2022.
Ontario, Canada CME Objective: To review current evidence for prevention, diagnosis, treatment,
and practice improvement of deep venous thrombosis.
Funding Source: American College of Physicians.
Acknowledgment: The author thanks John Spandorfer, MD, and Taki Galanis,
MD, authors of the previous version of this In the Clinic.
Disclosures: All relevant financial relationships have been mitigated. Disclosures
can also be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.
do?msNum=M22-1364.

With the assistance of additional physician writers, the editors of Annals of

Internal Medicine develop In the Clinic using MKSAP and other resources of
the American College of Physicians. The patient information page was written by
Julia Thayer from the Center for Quality at the American College of Physicians.
In the Clinic does not necessarily represent official ACP clinical policy. For ACP
clinical guidelines, please go to https://www.acponline.org/clinical_information/
guidelines/.

© 2022 American College of Physicians

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

 Venous thromboembolism (VTE) is a systematic review of 5 randomized trials
1. Heit JA. The epidemiology
of venous thromboembo- relatively common and potentially life- and 24 507 patients with acute VTE, the
lism in the community.
Arterioscler Thromb Vasc threatening condition. The annual inci- pooled rate of recurrent fatal VTE dur-
Biol. 2008;28:370-2. dence is estimated at 1 per 1000 per- ing the initial 3-month treatment period
[PMID: 18296591]
2. Stevens SM, Woller SC, sons (1). About one third of patients was 0.14% (95% CI, 0.05% to 0.27%) for
Kreuziger LB, et al.
Antithrombotic therapy for with VTE present with pulmonary em- VKAs and 0.16% (CI, 0.05% to 0.27%)
VTE disease: second update bolism (PE), and two thirds present for DOACs. Rates of fatal major bleed-
of the CHEST guideline and
expert panel report. Chest. with deep venous thrombosis (DVT). ing events with these treatments were
2021;160:e545-e608.
[PMID: 34352278] Fifty to sixty percent of VTE events are 0.33% (CI, 0.21% to 0.48%) and 0.14%
3. Ortel TL, Neumann I, provoked by recent surgery or hospi- (CI, 0.07% to 0.24%), respectively (4).
Ageno W, et al. American
Society of Hematology talization, 20% are associated with
2020 guidelines for man- Select patients should be considered
agement of venous throm- active cancer, and the remaining 20%
for extended-duration anticoagulation
boembolism: treatment of to 30% are considered unprovoked (1).
deep vein thrombosis and for secondary prevention of recurrent
pulmonary embolism. Diagnosis of DVT combines an assess-
Blood Adv. 2020;4:4693- VTE. The most common complications
4738. [PMID: 33007077] ment of pretest probability, D-dimer
of VTE are postthrombotic syndrome
4. Wu C, Alotaibi GS, Alsaleh testing, and compression ultrasonogra-
K, et al. Case fatality of and venous ulcer, with overall inci-
bleeding and recurrent ve- phy (CUS). The mainstay of DVT treat-
nous thromboembolism dence of 76.1 and 18.0 per 100 000
during, initial therapy with ment is anticoagulation, with direct oral
person-years, respectively (1). This
direct oral anticoagulants: a anticoagulants (DOACs) being the pre-
systematic review. Thromb review focuses on the diagnosis and
Res. 2014;134:627-32. ferred choice in most patients (2, 3).
[PMID: 25047174] management of patients presenting
5. Konstantinides SV, Meyer
G, Becattini C, et al; ESC
Anticoagulation with low-molecular- with DVT; however, discussion of risk
Scientific Document Group. weight heparin (LMWH) followed by a factors, prevention, and some aspects
2019 ESC Guidelines for
the diagnosis and manage- vitamin K antagonist (VKA) or a DOAC of treatment applies to patients with
ment of acute pulmonary
embolism developed in
is an effective treatment for VTE. In a DVT and PE.
collaboration with the
European Respiratory
Society (ERS). Eur Heart J.
2020;41:543-603. [PMID:
31504429]
6. Timp JF, Braekkan SK,
Prevention
Versteeg HH, et al. What factors increase risk for VTE? include non-O blood type and heterozy-
Epidemiology of cancer-
associated venous throm- Surgery and hospitalization for acute ill- gous gene mutations for factor V Leiden
bosis. Blood.
2013;122:1712-23. [PMID: ness account for half of VTE events. and prothrombin (Appendix Table 1,
23908465]
Although epidemiologic studies have available at Annals.org) (8, 9). Despite
7. Key NS, Khorana AA,
Kuderer NM, et al. Venous identified many associated risk factors the availability of genetic thrombophilia
thromboembolism prophy-
laxis and treatment in for VTE, it is essential to consider the testing, identifying a family history of
patients with cancer: ASCO
strength of this association when decid- VTE is more valuable, as only 30% of
clinical practice guideline
update. J Clin Oncol. ing on prophylactic treatment (Table 1) persons with a first-degree relative with
2020;38:496-520. [PMID:
31381464] (5). Frequently, VTE results from multi- VTE have a positive result on genetic
8. Middeldorp S, van
ple risk factors that, when combined, thrombophilia screening (10). For this
Hylckama Vlieg A. Does
thrombophilia testing help exceed the “thrombosis threshold.” reason, genetic thrombophilia testing of
in the clinical management
of patients. Br J Haematol. Cancer is one of the most substantial patients and family members is not
2008;143:321-35. [PMID:
18710381] risk factors, with a 7-fold increased risk recommended.
9. Connors JM.
Thrombophilia testing and
and a 5% to 20% absolute risk for VTE
venous thrombosis. N Engl within the first year after a cancer diag- Antiphospholipid syndrome (APS) is an
J Med. 2017;377:1177-87.
nosis (6). Furthermore, 4% to 10% of autoimmune disorder and a type of
[PMID: 28930509]
10. Bezemer ID, van der Meer
patients presenting with a first unpro- acquired thrombophilia. Patients with
FJ, Eikenboom JC, et al.
The value of family history voked VTE will ultimately have cancer APS are at significantly increased risk for
as a risk indicator for ve-
diagnosed within 1 year, making VTE arterial and venous thromboembolic
nous thrombosis. Arch
Intern Med.
an important first sign of cancer (7). complications and pregnancy-related
2009;169:610-5. [PMID:
19307525]
complications. Antiphospholipid anti-
11. Tektonidou MG, Andreoli A family history of VTE, especially a first- bodies include anticardiolipin antibod-
L, Limper M, et al. EULAR
recommendations for the degree relative with an unprovoked ies, b2-glycoprotein antibodies, and the
management of antiphos-
pholipid syndrome in VTE before age 50 years, is a significant lupus anticoagulant. APS requires clini-
adults. Ann Rheum Dis.
2019;78:1296-1304.
risk factor for VTE. The most common cal and laboratory criteria as well as
[PMID: 31092409] genetic risk factors (thrombophilia) demonstration of persistent antibodies

© 2022 American College of Physicians ITC2 In the Clinic Annals of Internal Medicine

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

 Table 1. Transient Risk Factors for Venous Thrombosis

Strong risk factor (odds ratio >10)

Hip or leg fracture
Hip or leg joint replacement
Major general surgery
Major trauma
Spinal cord injury 12. Kearon C, Akl EA,
Moderate risk factor (odds ratio of 2 to 9) Comerota AJ, et al.
Antithrombotic therapy
Arthroscopic knee surgery for VTE disease:
Cancer Antithrombotic Therapy
and Prevention of
Central venous lines Thrombosis, 9th Ed:
American College of
Congestive heart or respiratory failure Chest Physicians evi-
Hormone replacement therapy dence-based clinical prac-
tice guidelines. Chest.
Oral contraceptive therapy 2012;141:e419S-e496S.
Paralytic stroke [PMID: 22315268]
13. Quinlan DJ, Eikelboom
Postpartum JW, Dahl OE, et al.
Association between
Previous venous thromboembolism asymptomatic deep vein
Thrombophilia thrombosis detected by
venography and sympto-
Weak risk factor (odds ratio <2) matic venous throm-
Bed rest >3 d boembolism in patients
undergoing elective hip
Immobility due to sitting (e.g., prolonged road or air travel) or knee surgery. J
Thromb Haemost.
Increasing age 2007;5:1438-43. [PMID:
Laparoscopic surgery (e.g., cholecystectomy) 17425687]
14. Sch€unemann HJ,
Obesity Cushman M, Burnett AE,
Pregnancy (antepartum) et al. American Society of
Hematology 2018 guide-
Varicose veins lines for management of
venous thromboembo-
lism: prophylaxis for hos-
pitalized and
nonhospitalized medical
patients. Blood Adv.
over a span of 12 weeks (because of the What factors should be considered 2018;2:3198-3225.
[PMID: 30482763]
possibility of clinically insignificant tran- when deciding on prophylaxis? 15. Barbar S, Noventa F,
sient antibodies). Unlike hereditary Rossetto V, et al. A risk
VTE prophylaxis may be given phar- assessment model for the
thrombophilia, testing for APS is recom- identification of hospital-
macologically (for example, anticoa- ized medical patients at
mended because most patients should risk for venous throm-
gulation) or via mechanical means
remain on extended-duration anticoa- boembolism: the Padua
(for example, graduated compression Prediction Score. J Thromb
gulation for secondary prophylaxis if the Haemost. 2010;8:2450-7.
stockings and intermittent pneumatic [PMID: 20738765]
result is positive (11). 16. Spyropoulos AC,
compression devices). Decisions on Anderson FA Jr ,
Should clinicians screen specific which patients should receive pro- FitzGerald G, et al;
IMPROVE Investigators.
patients for DVT? phylaxis should include an assess- Predictive and associative
models to identify hospi-
Primary prophylaxis with anticoagula- ment of the risk for VTE, the benefit of talized medical patients at
tion is preferred over withholding pharmacologic prophylaxis, and the risk for VTE. Chest.
2011;140:706-14.
prophylaxis and screening high-risk risk for bleeding. The goal of prophy- [PMID: 21436241]
17. Greene MT, Spyropoulos
patients for asymptomatic DVT (12). laxis should be to reduce fatal PE and AC, Chopra V, et al.
Although asymptomatic DVT detected symptomatic VTE without increasing Validation of risk assess-
ment models of venous
by screening venography or CUS is a major bleeding. thromboembolism in hos-
pitalized medical patients.
common efficacy outcome in thrombo- Am J Med.
Which patients should receive
prophylaxis trials, there is no evidence 2016;129:1001.e9-1001.

to support screening of patients for prophylaxis, and what agents should e18. [PMID: 27107925]
18. Zayed Y, Kheiri B,
asymptomatic DVT as a strategy to pre- be used? Barbarawi M, et al.
Extended duration of
Hospitalized medical patients
vent symptomatic VTE. Asymptomatic thromboprophylaxis for
medically ill patients: a
DVT detected by routine screening Universal prophylaxis strategies in hos- systematic review and
meta-analysis of rando-
protocols is 5 to 21 times more com- pitalized medical patients have minimal mised controlled trials.
mon than symptomatic DVT, and its effect on VTE incidence because of the Intern Med J.
2020;50:192-9. [PMID:
clinical significance is unclear (13). heterogeneity of risk in this population 31276276]

Annals of Internal Medicine In the Clinic ITC3 © 2022 American College of Physicians

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

 (14). However, predicting which medi- reduces symptomatic VTE by up to 70%
19. Collins R, Scrimgeour A,
Yusuf S, et al. Reduction cal patients would benefit from prophy- (19). In recent years, improved surgical
in fatal pulmonary embo-
lism and venous thrombo- laxis is challenging. Clinical prediction techniques, early ambulation, and wide-
sis by perioperative rules, such as the IMPROVE score and spread use of pharmacologic prophy-
administration of subcuta-
neous heparin. Overview the Padua Prediction Score (Table 2), laxis have substantially decreased post-
of results of randomized
can be used to identify medical operative VTE (20). The Caprini Risk
trials in general, orthope-
dic, and urologic surgery. patients at increased risk, but fewer Assessment Model is the most widely
N Engl J Med.
1988;318:1162-73. than 20% of hospitalized medical used tool to assess thromboembolic
[PMID: 3283548]
patients will be identified as being at risk (Table 2) (21). The American
20. Chan NC, Siegal D, Lauw
MN, et al. A systematic College of Chest Physicians (ACCP) sug-
increased risk, and their impact on clini-
review of contemporary gests using the Caprini score to assess a
trials of anticoagulants in cal outcomes requires additional exter-
orthopaedic thrombopro- patient's VTE risk and adopting a
phylaxis: suggestions for nal validation (14–17). Clinical practice graded approach: no prophylaxis or
a radical reappraisal. J
Thromb Thrombolysis.
guidelines recommend that pharmaco- early ambulation for very low risk, me-
2015;40:231-9. [PMID: logic prophylaxis be considered in chanical prophylaxis for low risk (Caprini
25403952]
21. Pannucci CJ, Laird S, hospitalized patients without increased score of 1 to 2), pharmacologic or me-
Dimick JB, et al. A vali-
dated risk model to pre-
risk for bleeding, and LMWH is recom- chanical prophylaxis for moderate risk
dict 90-day VTE events in mended over DOACs (14). In patients (Caprini score of 3 to 4), and combined
postsurgical patients.
Chest. 2014;145:567-73. who cannot receive pharmacologic pro- pharmacologic and mechanical prophy-
[PMID: 24091567]
22. Gould MK, Garcia DA, phylaxis because of high bleeding risk, laxis for high risk (Caprini score ≥5) (22).
Wren SM, et al. mechanical prophylaxis is suggested The American Society of Hematology
Prevention of VTE in non-
orthopedic surgical (14). Evidence for routine extended anti- (ASH) clinical practice guidelines offer a
patients: Antithrombotic
Therapy and Prevention of coagulation (30 to 45 days) with DOACs more general suggestion of pharmaco-
Thrombosis, 9th Ed: or LMWH to reduce VTE events after hos- logic prophylaxis in patients having
American College of Chest
Physicians evidence-based pitalization has been mixed and suggests major general surgery (such as nonor-
clinical practice guide-
increased risk for major bleeding (18). thopedic general, abdominal, and pel-
lines. Chest. 2012;141:
e227S-e277S. [PMID: Extended-duration prophylaxis (beyond vic surgery) (23). Unlike for hospitalized
22315263]
23. Anderson DR, Morgano hospital discharge) and prophylaxis in medical patients, extended anticoagula-
GP, Bennett C, et al.
the long-term care setting are not recom- tion for 3 to 4 weeks after discharge is
American Society of
Hematology 2019 guide- suggested for hospitalized surgical
mended (14).
lines for management of patients (22, 23). Further refinement of
venous thromboembo-
lism: prevention of ve- Surgical patients which subtype of surgical patients
nous thromboembolism
would benefit most from extended anti-
in surgical hospitalized The risk for VTE after surgery depends
patients. Blood Adv. coagulation is needed.
2019;3:3898-3944. on the type of surgery and patient fac-
[PMID: 31794602]
24. Anderson DR, Dunbar M, tors. Use of unfractionated heparin In orthopedic surgery, DOACs have
Murnaghan J, et al. (UFH) for postoperative VTE prophylaxis largely replaced LMWH for postoperative
Aspirin or rivaroxaban for
VTE prophylaxis after hip
or knee arthroplasty. N
Engl J Med. Table 2. Risk Scores for VTE Prophylaxis
2018;378:699-707.
[PMID: 29466159] Risk Scores, by Patient Population Risk Predicted Interpretation
25. Lyman GH, Carrier M, Ay
C, et al. American Society Hospitalized medical patients
of Hematology 2021
guidelines for manage-
Padua Prediction Score Risk for VTE at 3 mo Low risk (<4 points): 1.1%
ment of venous throm- High risk (≥4 points): 3.5%
boembolism: prevention IMPROVE score Risk for VTE at 3 mo Low risk (0 points): 0.4%
and treatment in patients
with cancer. Blood Adv.
High risk (≥4 points): 5.7%
2021;5:927-74. [PMID: Geneva score Risk for VTE at 3 mo Low risk (<3 points): 0.6%
33570602] High risk (≥3 points): 3.2%
26. Khorana AA, Kuderer NM,
Culakova E, et al. Surgical patients
Development and valida-
tion of a predictive model
Caprini Risk Assessment Model Risk for VTE at 3 mo Low risk (0 points): <0.7%
for chemotherapy-associ- High risk (≥9 points): 10.7%
ated thrombosis. Blood. Ambulatory patients with cancer
2008;111:4902-7. [PMID:
18216292] Khorana score Risk for VTE at 2.5 mo Low risk (0 points): 0.8%
27. Carrier M, Abou-Nassar K, High risk (≥3 points): 7.1%
Mallick R, et al; AVERT
Investigators. Apixaban to Postpartum patients
prevent venous throm- RCOG score Not studied Low risk (<2 risk factors)
boembolism in patients
with cancer. N Engl J
Intermediate/high risk (≥2 risk factors)
Med. 2019;380:711-9.
[PMID: 30511879] RCOG = Royal College of Obstetricians and Gynaecologists; VTE = venous thromboembolism.

© 2022 American College of Physicians ITC4 In the Clinic Annals of Internal Medicine

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

VTE prophylaxis, and aspirin is an emerg- anticoagulation outside of pregnancy,
28. Khorana AA, Soff GA,
ing alternative. A large randomized con- antepartum and postpartum anticoa- Kakkar AK, et al; CASSINI
Investigators. Rivaroxaban
trolled trial (RCT) compared extended gulation are recommended (30). For for thromboprophylaxis in
prophylaxis with a DOAC (rivaroxaban) other at-risk persons (Appendix Table high-risk ambulatory
patients with cancer. N
versus a DOAC for 5 days followed by as- 2, available at Annals.org), prophylaxis Engl J Med.
pirin after hip and knee arthroplasty and is targeted at the first 6 weeks post- 2019;380:720-8. [PMID:
30786186]
found no difference in symptomatic VTE partum, as VTE risk is 15- to 35-fold 29. Kotaska A. Postpartum ve-
nous thromboembolism
or bleeding (24). The ASH guidelines higher during this period versus the prophylaxis may cause
more harm than benefit:
suggest pharmacologic prophylaxis with nonpregnant state (31). LMWH is safe a critical analysis of inter-
either a DOAC or aspirin (23). during pregnancy and breastfeeding national guidelines
through an evidence-
and is the anticoagulant of choice based lens. BJOG.
Ambulatory patients with cancer 2018;125:1109-16.
in the antepartum and postpartum [PMID: 29512316]
The incidence of VTE in ambulatory 30. Bates SM, Rajasekhar A,
patients with cancer ranges from 5% periods. Risk factors for pregnancy- Middeldorp S, et al.

to 20%, with considerable variability associated VTE include varicose veins, American Society of
Hematology 2018 guide-
according to such factors as cancer emergency cesarean section, stillbirth, lines for management of
venous thromboembo-
type, prothrombotic anticancer treat- comorbidities (heart, kidney, inflamma- lism: venous throm-

ments, and patient-specific risk factors tory bowel disease), smoking, inherited boembolism in the
context of pregnancy.
(25). Ambulatory patients who have thrombophilia, preeclampsia, postpar- Blood Adv. 2018;2:3317-
59. [PMID: 30482767]
cancer-associated thrombosis (CAT) tum infection, and postpartum hemor- 31. Heit JA, Kobbervig CE,

have increased risk for bleeding, recur- rhage (32). How these risk factors James AH, et al. Trends in
the incidence of venous
rent VTE, and early death (25). The should be used to select patients who thromboembolism during
pregnancy or postpartum:
Khorana score (Table 2) can be used to would benefit from thromboprophy- a 30-year population-

estimate VTE risk and identify which laxis is unclear. Clinical trials have based study. Ann Intern
Med. 2005;143:697-706.
patients with cancer might benefit from focused on post–cesarean section pa- [PMID: 16287790]
32. Sultan AA, West J,
primary prophylaxis (26). DOACs are tients but are limited by low event rates Grainge MJ, et al.

the preferred anticoagulant for pre- and poor recruitment. Recommen- Development and valida-
tion of risk prediction
venting CAT because of their oral route dations for postpartum prophylaxis model for venous throm-
boembolism in postpar-
and established safety profile (25). Two from several professional societies are tum women:
summarized in Appendix Table 2 (30, multinational cohort
RCTs have compared DOACs versus no study. BMJ. 2016;355:
prophylaxis in ambulatory patients with 33–37). i6253. [PMID:
27919934]
cancer with high-risk Khorana scores Should patients with thrombophilia 33. ACOG practice bulletin
no. 196 summary:
(≥2) (27, 28). The CASSINI trial showed a routinely receive pharmacologic thromboembolism in
pregnancy. Obstet
trend toward reduced VTE events but prophylaxis? Gynecol. 2018;132:243-
did not reach statistical significance (28). Patients may be identified as carriers
8. [PMID: 29939933]
34. American College of
In contrast, the AVERT trial showed a sta- of inherited thrombophilia because of Obstetricians and
Gynecologists' Committee
tistical reduction in VTE events (27). Both a prior VTE event or family screening. on Practice Bulletins—
trials showed a slight increase in major Prophylaxis for VTE in patients with
Obstetrics. ACOG practice
bulletin no. 197:
bleeding of about 1%. Major practice thrombophilia without a history of
inherited thrombophilias
in pregnancy. Obstet
guideline recommendations are consist- VTE is not recommended. Because Gynecol. 2018;132:e18-
ent, however, and suggest the use of e34. [PMID: 29939939]
the incidence of VTE in the general 35. Chan WS, Rey E, Kent NE,
pharmacologic prophylaxis in high-risk population is low, even if some
et al; VTE in Pregnancy
Guideline Working
patients (Khorana score ≥2) while con- thrombophilia is associated with in- Group. Venous throm-
sidering bleeding risk, treatment cost, boembolism and antith-
creased risk for VTE, the absolute risk rombotic therapy in
and patient preference (7, 25). pregnancy. J Obstet
remains small (Appendix Table 1). Gynaecol Can.
2014;36:527-53. [PMID:
Pregnant patients Similarly, the presence of thrombo- 24927193]
Although peripartum VTE occurs in 1 philia should not influence the dura- 36. Bates SM, Greer IA,
Middeldorp S, et al. VTE,
in 1000 deliveries and is the causal fac- tion of therapy after an initial VTE thrombophilia, antithrom-
botic therapy, and preg-
tor in 15% (1 to 5 maternal deaths per event. However, during transient nancy: Antithrombotic
100 000 live births), consensus on who high-risk periods, such as hospitaliza- Therapy and Prevention
of Thrombosis, 9th Ed:
should receive prophylaxis is elusive tion or postpartum, the presence of American College of
Chest Physicians evi-
(29). For those with a history of unpro- thrombophilia may warrant consider- dence-based clinical prac-
voked VTE or estrogen-associated ation, especially if other high-risk con- tice guidelines. Chest.
2012;141:e691S-e736S.
VTE or those receiving long-term ditions are present. [PMID: 22315276]

Annals of Internal Medicine In the Clinic ITC5 © 2022 American College of Physicians

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

 How should physicians counsel aspirin, or a single dose of LMWH 2 to 4
37. Nelson-Piercy C,
MacCallum P, Mackillop patients about prevention of VTE hours before a flight (39). The trial
L; Royal College of
Obstetricians and during travel? showed a reduction in DVT in LMWH-
Gynaecologists. Reducing
the Risk of Venous Air travel is a weak risk factor for VTE, treated patients compared with pla-
Thromboembolism during with a 2- to 4-fold increased risk for cebo or aspirin; however, 60% of events
Pregnancy and the
Puerperium. Green-top symptomatic DVT after a flight lasting 4 were asymptomatic DVT and thus of
Guideline No. 37a. April
2015. hours or longer and an absolute risk uncertain clinical importance.
38. Chandra D, Parisini E, increase of 1 in 4600 flights for sympto-
Mozaffarian D. Meta-anal- Although data are limited, expert guide-
ysis: travel and risk for ve- matic events (14, 38).
nous thromboembolism. lines state that patients at high risk may
Ann Intern Med.
2009;151:180-90. [PMID:
In the LONFLIT3 RCT, patients at high benefit from graduated compression
19581633] risk for DVT (previous DVT, coagulation stockings (below the knee, pressure of
39. Cesarone MR, Belcaro G,
Nicolaides AN, et al. disorder, obesity, limited mobility, can- 15 to 30 mm Hg) or pharmacologic pro-
Venous thrombosis from phylaxis, with consideration of patient
air travel: the LONFLIT3
cer, or varicose veins) were randomly
study—prevention with as- assigned to no prophylaxis, 400 mg of values and preferences (14, 40).
pirin vs low-molecular-
weight heparin (LMWH)
in high-risk subjects: a
randomized trial.
Angiology. 2002;53:1-6. Prevention... DVT has many risk factors, and risk scores are helpful for clinicians to
[PMID: 11863301] select which patients might benefit from prophylaxis. Determining the clinical benefit of
40. Kahn SR, Lim W, Dunn prophylaxis should include assessment of risk for fatal PE and symptomatic VTE versus
AS, et al. Prevention of
VTE in nonsurgical risk for major bleeding. DOACs are the agents of choice for prophylaxis in orthopedic
patients: Antithrombotic patients and patients with cancer, whereas LMWH is the treatment of choice in pregnant
Therapy and Prevention of
Thrombosis, 9th Ed: patients and hospitalized medical patients.
American College of Chest
Physicians evidence-based
clinical practice guide-
lines. Chest. 2012;141: CLINICAL BOTTOM LINE
e195S-e226S. [PMID:
22315261]
41. Goodacre S, Sutton AJ,
Sampson FC. Meta-analy-
sis: The value of clinical
assessment in the diagno-
sis of deep venous throm-
bosis. Ann Intern Med. Diagnosis
2005;143:129-39. [PMID:
16027455] When should clinicians suspect DVT? What is the role of D-dimer testing?
42. Siccama RN, Janssen KJ,
Verheijden NA, et al.
Clinical features, such as pain, warmth, D-dimer, a degradation product of
Systematic review: diag- and erythema, have limited value in the cross-linked fibrin, is typically elevated
nostic accuracy of clinical
decision rules for venous diagnosis of DVT. The differential diag- in patients with acute VTE and some
thromboembolism in el-
derly. Ageing Res Rev. nosis for suspected DVT is extensive nonthrombotic disorders. D-dimer lev-
2011;10:304-13. [PMID: (Table 3). Therefore, when DVT is clini- els have good sensitivity but poor spec-
21130902]
43. Oudega R, Moons KG, cally suspected, a pretest probability of ificity for diagnosis of VTE; therefore,
Hoes AW. Ruling out
deep venous thrombosis DVT should be determined using vali- their utility is in a negative result. In
in primary care. A simple dated clinical prediction rules (Table patients with a low or “unlikely” pretest
diagnostic algorithm
including D-dimer testing. 4). Although the Wells score is the probability of DVT (<10%), a negative
Thromb Haemost.
2005;94:200-5. [PMID: most extensively evaluated tool (41), it D-dimer result can safely rule out DVT
16113804] has not been validated in pregnant or and obviate the need for additional
44. Geersing GJ, Zuithoff NP,
Kearon C, et al. Exclusion hospitalized patients and may have investigation. In fact, when a clinical
of deep vein thrombosis
using the Wells rule in lower specificity in elderly patients (42). prediction rule and D-dimer testing are
clinically important sub- An alternative prediction rule, the pri- used together, DVT can be excluded in
groups: individual patient
data meta-analysis. BMJ. mary care rule, was derived and vali- 29% of patients with suspected DVT
2014;348:g1340. [PMID:
24615063] dated in ambulatory patients presen- without the need for additional diag-
45. Freund Y, Chauvin A, ting with symptoms of DVT and may be nostic imaging (44). The sensitivity of
Jimenez S, et al. Effect of
a diagnostic strategy more appropriate in this context (43). D-dimer testing for PE can be improved
using an elevated and
age-adjusted D-dimer No matter which tool is used, a clinical when clinical probability and age are
threshold on thromboem-
bolic events in emergency
prediction rule should not be the sole included (45). There is an ongoing clini-
department patients with criterion for diagnosis; rather, it should cal trial of age-adjusted D-dimer in DVT
suspected pulmonary em-
bolism: a randomized be used in conjunction with D-dimer diagnosis (NCT02384135). D-dimer
clinical trial. JAMA.
2021;326:2141-9. [PMID:
testing to determine whether diagnos- should not be used as a screening test
34874418] tic imaging is necessary. when DVT is not clinically suspected.

© 2022 American College of Physicians ITC6 In the Clinic Annals of Internal Medicine

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

 46. Kraaijpoel N, Carrier M,
Table 3. Differential Diagnosis of DVT Le Gal G, et al. Diagnostic
accuracy of three ultraso-
Disease Characteristics Notes nography strategies for
deep vein thrombosis of
Venous insufficiency (venous Usually due to venous hyperten- Obtain ultrasonography of venous the lower extremity: a sys-
reflux) sion from such causes as venous reflux tematic review and meta-
analysis. PLoS One.
reflux or obesity 2020;15:e0228788.
Superficial thrombophlebitis Firm, tender, varicose vein Superficial thrombosis is rarely [PMID: 32045437]
associated with DVT 47. Ageno W, Squizzato A,
Wells PS, et al. The diag-
Muscle strain, tear, or trauma Pain occurring with a range of Perform appropriate orthopedic nosis of symptomatic
motion more characteristic of evaluation recurrent pulmonary em-
orthopedic problem due to bolism and deep vein
thrombosis: guidance
trauma; usually a history of leg
from the SSC of the ISTH.
injury J Thromb Haemost.
Swelling in a paralyzed leg History of paraplegia Patients with paralyzed limb may 2013;11:1597-602.
[PMID: 23682905]
develop edema without DVT
48. van Es N, Le Gal G, Otten
Baker cyst Pain localized to popliteal region Seen on ultrasonography HM, et al. Screening for
of leg occult cancer in patients
with unprovoked venous
Cellulitis Skin erythema and warmth Consider antibiotic treatment thromboembolism: a sys-
tematic review and meta-
Lymphedema Toe edema is more characteristic Lymphedema can occur in 1 or
analysis of individual
of lymphedema than of venous both legs patient data. Ann Intern
edema Med. 2017;167:410-7.
[PMID: 28828492]
49. Aujesky D, Obrosky DS,
DVT = deep venous thrombosis. Stone RA, et al. Derivation
and validation of a prog-
nostic model for pulmo-
What is the role of CUS? vein DVT, which is of uncertain clinical nary embolism. Am J
Respir Crit Care Med.
Owing to safety, ease of application, significance. 2005;172:1041-6.
[PMID: 16020800]
and absence of contrast and ionizing How should pregnant patients be 50. Kahn SR, Shapiro S, Wells
PS, et al; SOX trial investi-
radiation, CUS is the preferred diag- evaluated for suspected DVT? gators. Compression
nostic imaging modality for DVT. CUS Clinical prediction rules have not been
stockings to prevent post-
thrombotic syndrome: a
can be performed according to 3 pro- formally validated in pregnant patients. randomised placebo-con-
trolled trial. Lancet.
tocols: 1) continuous visualization with Furthermore, D-dimer assays are asso- 2014;383:880-8. [PMID:
compression of proximal leg deep ciated with poor specificity in preg-
24315521]
51. Franco L, Giustozzi M,
veins only, stopping at the calf trifurca- nancy, as levels naturally increase Agnelli G, et al.
Anticoagulation in
tion; 2) 2-point visualization with com- during the third trimester. CUS is thus patients with isolated dis-
pression limited to the common recommended as the initial test when
tal deep vein thrombosis:
a meta-analysis. J Thromb
femoral vein and popliteal trifurcation; DVT is suspected, and follow-up ultra- Haemost. 2017;15:1142-
54. [PMID: 28316124]
or 3) continuous visualization with com- sonography is suggested in pregnant 52. Ageno W, Bertuª L,
pression from common femoral veins patients with an initial normal ultra-
Bucherini E, et al; RIDTS
study group. Rivaroxaban
to calf veins. For protocols that limit sound. Patients with symptoms sugges-
for the treatment of
symptomatic isolated distal
imaging to the proximal leg vein, serial tive of iliac venous thrombosis (whole- deep vein thrombosis –
RIDTS study [abstract]. Res
CUS is needed in patients with a high leg edema or discomfort in the flank, Pract Thromb Haemost.
or “likely” clinical probability and a pos- back, or buttock) should have the pel-
2021;5 (Suppl 2).
53. Mulder FI, Bosch FTM,
itive D-dimer result (Appendix Figure, vic vessels evaluated with ultrasonogra- Young AM, et al. Direct
oral anticoagulants for
available at Annals.org). phy or magnetic resonance imaging cancer-associated venous
thromboembolism: a sys-
A meta-analysis comparing serial lim- (30, 36). tematic review and meta-
analysis. Blood.
ited proximal CUS and whole-leg CUS 2020;136:1433-41.
When should clinicians consider [PMID: 32396939]
showed similar failure rates (VTE diag- 54. Carrier M, Blais N,
recurrent DVT?
nosed within 45 days of follow-up) of Crowther M, et al.
Treatment algorithm in
1.4% and 1.0%, respectively. The pro- After acute treatment of DVT, approxi- cancer-associated throm-

portion of isolated distal DVT identified mately half of patients have residual bosis: Canadian expert
consensus. Curr Oncol.
by whole-leg CUS ranged from 23% to vein occlusion on follow-up imaging, 2018;25:329-37. [PMID:
30464682]
62% (46). and 25% to 40% have persistent pain, 55. Bates SM, Middeldorp S,
skin discoloration, and swelling, known Rodger M, et al. Guidance
for the treatment and pre-
Although a potential advantage of as postthrombotic syndrome (PTS) vention of obstetric-associ-
ated venous
whole-leg CUS may be that it precludes (47). The presence of PTS makes distin- thromboembolism. J
the need for repeated testing, it may guishing the symptoms and the imag- Thromb Thrombolysis.
2016;41:92-128. [PMID:
identify more patients with isolated calf ing findings of DVT challenging when 26780741]

Annals of Internal Medicine In the Clinic ITC7 © 2022 American College of Physicians

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

 Table 4. Clinical Prediction Rules for DVT
56. Pengo V, Denas G,
Zoppellaro G, et al. Clinical Prediction Rule Score
Rivaroxaban vs warfarin in
high-risk patients with Wells score*
antiphospholipid syn-
drome. Blood. Active cancer (treatment ongoing, within 6 mo, or palliative) 1
2018;132:1365-71.
[PMID: 30002145] Paralysis, paresis, or recent plaster immobilization of the lower extremities 1
57. Ordi-Ros J, Sáez-Comet L, Recently bedridden for >3 d or major surgery within 12 wk requiring general or regional 1
Perez-Conesa M, et al. anesthesia
Rivaroxaban versus vita-
min K antagonist in anti- Localized tenderness along the distribution of the deep venous system 1
phospholipid syndrome: a
randomized noninferiority Entire leg swollen 1
trial. Ann Intern Med. Calf swelling 3 cm larger than the asymptomatic side (measured 10 cm below the tibial 1
2019;171:685-94. [PMID: tuberosity)
31610549]
58. Cohen H, Cuadrado MJ, Pitting edema confined to the symptomatic leg 1
Erkan D, et al. 16th
International Congress on Collateral superficial veins (nonvaricose) 1
Antiphospholipid Previously documented DVT 1
Antibodies Task Force
report on antiphospholi- Alternative diagnosis at least as likely as DVT –2
pid syndrome treatment Primary care rule†
trends. Lupus.
2020;29:1571-93. [PMID: Male sex 1
33100166]
59. Douxfils J, Adcock DM, Use of hormonal contraceptives 1
Bates SM, et al. 2021 Active cancer in past 6 mo 1
update of the
International Council for Surgery in previous month 1
Standardization in Absence of leg trauma 1
Haematology recommen-
dations for laboratory Distention of collateral leg veins 1
measurement of direct
oral anticoagulants. Difference in calf circumference ≥3 cm 2
Thromb Haemost. Abnormal D-dimer assay result 6
2021;121:1008-20.
[PMID: 33742436]
60. Poller L, Keown M, DVT = deep venous thrombosis.
Ibrahim S, et al. An inter- * Using the original score, <0 points indicates low probability, 0–2 points indicates intermediate proba-
national multicenter bility, and >2 points indicates high probability. Using the dichotomized score, ≤1 point indicates that
randomized study of com- DVT is unlikely and ≥2 points indicates that DVT is likely. From Wells PS, Anderson DR, Rodger M, et al.
puter-assisted oral antico- Evaluation of D-dimer in the diagnosis of suspected deep-vein thrombosis. N Engl J Med.
agulant dosage vs.
medical staff dosage. J 2003;349:1227-35.
Thromb Haemost. † Very low risk (score ≤3) or increased risk (score ≥4). From Toll DB, Oudega R, Vergouwe Y, et al. A
2008;6:935-43. [PMID: new diagnostic rule for deep vein thrombosis: safety and efficiency in clinically relevant subgroups.
18489430] Fam Pract. 2008;25:3-8.
61. Stergiopoulos K, Brown
DL. Genotype-guided vs
clinical dosing of warfarin
and its analogues: meta- considering DVT recurrence. The most month prevalence of cancer after VTE
analysis of randomized
clinical trials. JAMA Intern reliable way to diagnose recurrent DVT diagnosis of 5.2% (CI, 4.1% to 6.5%).
Med. 2014;174:1330-8.
[PMID: 24935087] is to compare repeated imaging with a Overall, this was higher in patients with
62. Rodger MA, Miranda S, extensive screening than in those with
Delluc A, et al.
posttreatment image. For this reason,
Management of sus- CUS is suggested in all patients at the limited screening (odds ratio, 2.0 [CI,
pected and confirmed
recurrent venous throm- time of withdrawal of anticoagulation 1.2 to 3.4]) (48). Patients aged 50 years
bosis while on anticoagu-
lant therapy. What next.
(baseline). The finding of a new non- or older were 7 times more likely to be
Thromb Res. compressible venous segment or an diagnosed with cancer (odds ratio, 7.1
2019;180:105-9. [PMID:
31279158] increase of 4 mm in clot diameter from [CI, 3.1 to 16]).
63. Khan F, Rahman A, Carrier
M, et al; MARVELOUS baseline imaging suggests recurrent Although there is insufficient evidence
Collaborators. Long term DVT (47). to inform a specific routine screening
risk of symptomatic recur-
rent venous thromboemb-
olism after discontinuation What other underlying conditions strategy, clinicians should keep in mind
of anticoagulant treatment
and clinical manifestations should that 1 in 20 patients with unprovoked
for first unprovoked ve-
nous thromboembolism clinicians look for? VTE may have cancer diagnosed within
event: systematic review
and meta-analysis. BMJ. Within 12 months of an unprovoked the following 12 months and the high-
2019;366:l4363. [PMID:
VTE diagnosis, 4% to 10% of patients est risk is in patients aged 50 years or
31340984]
64. Douketis J, Tosetto A,
are diagnosed with cancer. older. Existing age- and sex-appropri-
Marcucci M, et al. Risk of
recurrence after venous ate cancer screening guidelines should
thromboembolism in A systematic review and patient-level be followed, and clinicians should have
men and women: patient
level meta-analysis. BMJ. meta-analysis of 10 studies evaluating a low index to investigate cancer in
2011;342:d813. [PMID:
21349898] cancer screening strategies found a 12- older patients.

© 2022 American College of Physicians ITC8 In the Clinic Annals of Internal Medicine

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

 Diagnosis... Clinical features have little predictive value in diagnosing DVT. Clinicians
should use a clinical prediction rule that incorporates D-dimer to stratify risk for throm-
bosis. Various clinical prediction rules exist, of which the Wells score is the most studied.
Whole-leg ultrasonography may limit the need for repeated testing but identifies more
patients with isolated distal DVT. The use of an extensive cancer screening and thrombo-
philia testing strategy is controversial and is not routinely recommended. Clinicians
should consider consulting a specialist if DVT recurrence or breakthrough is suspected.

CLINICAL BOTTOM LINE 65. Rodger MA, Le Gal G,

Anderson DR, et al;
REVERSE II Study
Investigators. Validating
the HERDOO2 rule to
guide treatment duration

Treatment for women with unpro-

voked venous thrombosis:
multinational prospective
As shown in the Figure, DVT treat- How should clinicians decide whether cohort management study.
BMJ. 2017;356:j1065.
ment is divided into 3 phases. For to treat patients on an outpatient or [PMID: 28314711]
66. Broderick C, Watson L,
most patients, a DOAC is the pre- inpatient basis? Armon MP. Thrombolytic
strategies versus standard
ferred treatment (see the discussion Most patients with DVT can be safely anticoagulation for acute
of choice of anticoagulant later in this treated with either a DOAC or LMWH
deep vein thrombosis of
the lower limb. Cochrane
section). Some DOACs (rivaroxaban as outpatients. Patients with associated Database Syst Rev.
2021;1:CD002783.
and apixaban) can be used immedi- PE at the time of diagnosis of DVT [PMID: 33464575]
67. Rabinovich A, Kahn SR.
ately during the initial phase of treat- should be assessed for risk for short- How I treat the post-
thrombotic syndrome.
ment, whereas others (dabigatran term PE-related mortality, such as with Blood. 2018;131:2215-
and edoxaban) require 5 to 10 days the Pulmonary Embolism Severity 22. [PMID: 29545327]
68. Kearon C, Akl EA, Ornelas
of LMWH before starting. Table 5 Index (49). Those with hemodynamic J, et al. Antithrombotic
therapy for VTE disease:
summarizes the various anticoagu- instability, defined as systolic blood CHEST guideline and
expert panel report.
lants used for DVT treatment, includ- pressure below 90 mm Hg for 15 Chest. 2016;149:315-52.
ing mechanisms of action, pharma- minutes or more, should be consid- [PMID: 26867832]
69. Witt DM, Nieuwlaat R,
cokinetics, prescribing information, ered for thrombolysis. Those with high- Clark NP, et al. American
Society of Hematology
and availability of reversal agents. risk features, based on either a risk 2018 guidelines for man-
agement of venous
thromboembolism: opti-
mal management of anti-
coagulation therapy.
Blood Adv. 2018;2:3257-
Figure. Phases of anticoagulation. 91. [PMID: 30482765]
70. Lim W, Le Gal G, Bates SM,
et al. American Society of
Hematology 2018 guide-
lines for management of
venous thromboembolism:
Initial Long-term Extended diagnosis of venous throm-
boembolism. Blood Adv.
2018;2:3226-56. [PMID:
30482764]
71. Qaseem A, Chou R,
Apixaban, 10 mg BID for 7 d Apixaban, 5 mg BID Apixaban, 5 mg BID or Humphrey LL, et al;
Rivaroxaban, 15 mg BID for Dabigatran, 150 mg BID 2.5 mg BID‡ Clinical Guidelines
21 d Edoxaban, 60 mg OD† Aspirin, 81–100 mg OD, if Committee of the
Rivaroxaban, 20 mg OD anticoagulation not possible American College of
LMWH/fondaparinux* for
Physicians. Venous throm-
≥5 d and INR ≥2 for 2 d VKA (warfarin), INR 2–3 Dabigatran, 150 mg BID boembolism prophylaxis in
Edoxaban, 60 mg OD† hospitalized patients: a clin-
Rivaroxaban, 20 mg OD or ical practice guideline from
10 mg OD‡ the American College of
VKA (warfarin), INR 2–3 Physicians. Ann Intern
Med. 2011;155:625-32.
[PMID: 22041951]
72. American College of
0 to 7 d 8 d to 3 mo 3 mo to indefinitely Obstetricians and
Gynecologists' Committee
on Practice Bulletins—
BID = twice daily; INR = international normalized ratio; LMWH = low-molecular-weight heparin; OD = Obstetrics. ACOG practice
once daily; VKA = vitamin K antagonist. bulletin no. 196:
thromboembolism in
* LMWH is needed for 5 to 10 days before starting dabigatran or edoxaban. pregnancy. Obstet
† 30 mg/d if creatinine clearance is 30 to 50 mL/min or weight is <60 kg. Gynecol. 2018;132:e1-
‡ Dose reduction may be considered after 6 months of therapy. e17. [PMID: 29939938]

Annals of Internal Medicine In the Clinic ITC9 © 2022 American College of Physicians

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

 Table 5. Characteristics of Common Anticoagulants
Drug Target Peak Effect Half-Life Renal Protein Use in Renal Laboratory Reversal/
Elimination Clearance, % Binding, % Failure Measurement Bypassing Agent

Warfarin Vitamin K– 5–7 d 20–60 h* None 99 Safe INR Vitamin K and/or

(vitamin K dependent PCC, or frozen
antagonist) factors II, VII, IX, X; plasma
protein C and S
Unfractionated Antithrombin/ Immediate 30–90 min Minimal Very high Safe Anti-Xa† or aPTT Protamine
heparin indirect factor (intravenous)
IIa, factor Xa
(other factors to
lesser extent)
Low-molecular- Factor Xa 3–5 h 2–7 h 40 None Avoid if GFR Not routinely Partial reversal
weight heparin <30 mL/min required; anti-Xa† with protamine
when needed
Apixaban Factor Xa 3h 8–14 h 25 85 Avoid if GFR Not routinely Andexanet alfa;
<25 mL/min required; anti-Xa† PCC may be
when needed effective if not
available
Dabigatran Factor IIa 1.5 h 14–17 h >80 35 Avoid if GFR dTT, anti-IIa Idarucizumab
<30 mL/min
Rivaroxaban‡ Factor Xa 2–3 h 7–11 h 33 90 Avoid if GFR Not routinely Andexanet alfa;
<30 mL/min required; anti-Xa† PCC may be
when needed effective if not
available
Edoxaban Factor Xa 4h 8–11 h 33 55 Avoid if GFR Not routinely Andexanet alfa;
<30 mL/min required; anti-Xa† PCC may be
when needed effective if not
available

aPTT = activated partial thromboplastin time; dTT = diluted thrombin time; GFR = glomerular filtration rate; INR = international nor-
malized ratio; PCC = prothrombin complex concentrate.
* Duration of clinical effect is 2 to 5 days.
† Requires drug-specific calibration.
‡ 15- and 20-mg tablets should be taken with food for maximum absorption and efficacy.

prediction rule or concerning clinical Should clinicians treat isolated

73. Kahn SR, Comerota AJ,
Cushman M, et al;
American Heart
Association Council on
Peripheral Vascular
Disease, Council on
Clinical Cardiology, and
Council on Cardiovascular
and Stroke Nursing. The
postthrombotic syndrome:
evidence-based preven-
tion, diagnosis, and treat-
ment strategies: a
scientific statement from
the American Heart
Association. Circulation.
2014;130:1636-61.
[PMID: 25246013]
presentations, such as severe dyspnea
or syncope, should be admitted in a
closely monitored setting. Patients with
uncomplicated isolated DVT who do
not need admission for parenteral anti-
coagulation (for example, UFH in renal
failure) or pain management can be
treated as outpatients (3). Patients
with clinical signs suggestive of risk
for limb ischemia (phlegmasia ceru-
lea dolens), severe pain, or a high risk
for bleeding should be admitted for
initial management.
What local measures should clinicians
recommend?
Data on the efficacy of compression
therapy in reducing risk for PTS are
conflicting. For example, elastic com-
pression stockings did not prevent PTS
in an RCT (50). However, clinicians
should consider early ambulation, leg
elevation, and compression therapy for
symptom relief.
distal DVT?
Isolated distal (calf) DVT carries a signif-
icantly lower risk for PE and recurrence.
When a whole-leg CUS is performed to
investigate DVT, half of all DVTs will be
isolated to the calf veins (peroneal,
posterior tibial, anterior tibial) (12). A
diagnostic strategy of imaging only
the proximal leg veins is therefore
one way to reduce overdiagnosis and
treatment of isolated distal DVT. If an
isolated distal DVT is diagnosed,
treatment options include either no
anticoagulation and close serial CUS
or therapeutic anticoagulation.
A meta-analysis of randomized and non-
randomized studies suggested that anti-
coagulation (both prophylactic and
therapeutic dose) for distal DVT may
reduce recurrent VTE without increasing
bleeding risk (51). A recent cohort study
showed that stopping anticoagulation

© 2022 American College of Physicians ITC10 In the Clinic Annals of Internal Medicine

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

(therapeutic LMWH or rivaroxa-
ban) in patients with resolution
of symptoms and no extension
on repeated CUS at 2 weeks was
associated with a low rate of
recurrent VTE at 3 months (1.3%
[CI, 0.05% to 4.85%]). Further
data on DOAC therapy and du-
ration for treatment of distal DVT
are expected soon (52).
ACCP clinical practice guidelines
suggest anticoagulation only in
patients with severe symptoms
or risk factors for progression (D-
dimer, extensive disease, proxim-
ity to proximal veins, unprovoked
event, active cancer, history of
VTE, or hospitalization) (2, 12).
When therapeutic anticoagula-
tion is chosen, ACCP suggests 3
months' duration; however, pub-
lished data on the duration for
distal DVT vary from 6 weeks to 3
months (51). In low-risk patients,
withholding anticoagulation and
performing serial CUS weekly for
2 weeks is a reasonable clinical
approach.
When should clinicians start
anticoagulants?
There are no randomized studies
investigating the timing of anti-
coagulant therapy initiation in
patients with suspected DVT. In
patients with a high pretest prob-
ability and low risk for bleeding,
clinicians should initiate a short-
acting anticoagulant while await-
ing the results of the diagnostic
work-up. In patients diagnosed
with acute proximal DVT, clini-
cians should initiate anticoagula-
tion with a parenteral antico-
agulant, apixaban, or rivaroxaban
immediately unless they are con-
traindicated (Figure). If a VKA is
chosen for the long-term phase of
therapy, it should be started on
the same day as the parenteral
anticoagulant.
Annals of Internal Medicine
Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.
Which anticoagulants should
clinicians use?
Anticoagulation for confirmed
DVT is divided into 3 phases: the
initial 7 days, 8 days to 3 months,
and extended therapy beyond 3
months (Figure). Cancer- and
pregnancy-associated DVT are
considered separately. Table 5
summarizes the standard antico-
agulation options for treatment.
Historically, VKAs (such as warfa-
rin) were the only oral treatment;
warfarin remains the preferred
agent in patients with advanced
renal failure, those with APS, or
those for whom medication cost
is important. For VKAs, concur-
rent parenteral anticoagulation
is required for at least 5 days,
and 2 consecutive international
normalized ratios (INRs) must be
between 2 and 3 before paren-
teral therapy is withdrawn. UFH
is used for this initial period in
patients with advanced renal
impairment, and LMWH is used
in most others. Except for the
special populations discussed in
the following sections, DOACs
are the preferred treatment for
acute DVT (2, 3). DOACs are
given at fixed doses, do not
require laboratory monitoring,
and are noninferior to VKAs in
phase 3 RCTs for the acute treat-
ment of DVT to prevent sympto-
matic recurrent VTE.
Several patient populations require
attention for treatment decisions:
patients with VTE associated with
active cancer, pregnant patients,
and patients with APS.
Patients with CAT
Treatment of CAT is distinct from
other causes of VTE because of
the high risk for recurrent VTE,
drug interactions, and increased
bleeding complications. CAT
treatment differs from noncancer
VTE in several ways. For example,
clinical trials show that LMWH is
In the Clinic
more effective than VKAs at
reducing recurrent VTE in CAT.
Similarly, RCTs in patients with
CAT have shown that DOACs
have similar or increased efficacy
compared with LMWH but may
be associated with increased
bleeding (7, 25, 53). CAT DOAC
trials have studied the oral direct
Xa inhibitors (rivaroxaban, apixa-
ban, edoxaban) only; there have
been no trials of oral direct
thrombin inhibitors (such as dabi-
gatran) for treatment of CAT. An
increase in the overall trend of
bleeding during CAT treatment is
observed for patients with gastro-
intestinal or genitourinary cancer.
DOACs are safe and affordable
and have an oral administration
route, making them the preferred
choice for many patients with
CAT (7, 25, 54). There is insuffi-
cient evidence to guide the choice
between DOACs for CAT, and
individualized decisions should be
made with patients. Based on ana-
lysis of current trials, caution with
DOACs is suggested in patients
with an unresected gastrointestinal
lesion, thrombocytopenia (platelet
count <50
109/L), recent life-
threatening bleeding, high-risk in-
tracranial lesions, hepatic impair-
ment (Child–Pugh B or C), and
concomitant use of antiplatelets. In
such patients, LMWH is the pre-
ferred therapy (54).
Pregnant patients
UFH and LMWH do not cross the
placenta. They are safe during
pregnancy, with LMWH being the
mainstay of treatment because it
can be given as a self-adminis-
tered subcutaneous injection.
DOACs, VKAs, and fondaparinux
cross the placenta and should be
avoided in pregnancy. Duration
of therapy should be 6 months or
6 weeks postpartum, whichever is
longer (30, 36). Patients on antico-
agulation during pregnancy may
benefit from scheduled induction
ITC11 © 2022 American College of Physicians
of labor, as this would enable anti-
coagulation to be held before
delivery, allowing for neuroaxial
anesthesia while reducing peri-
partum hemorrhage. If the preg-
nancy-associated VTE event occu-
rred 4 weeks or more before
delivery, the last dose of LMWH
can be administered 24 hours
before labor induction, and it can
be restarted 6 hours after delivery
in the absence of any ongoing
postpartum hemorrhage. If the
VTE event is within 4 weeks of
delivery, UFH at a therapeutic
dose should be started at induc-
tion of labor and held once active
labor begins. An inferior vena
cava (IVC) filter should be consid-
ered if a VTE event occurs within
2 weeks of delivery (55). IVC filters
should be removed as soon as
anticoagulation can be resumed
postpartum.
Patients with APS
Patients with VTE and APS are at
high risk for recurrent VTE, and
APS is an indication to remain on
extended-duration or lifelong
anticoagulation. TRAPS, an RCT
of warfarin compared with rivarox-
aban in triple-positive patients
with APS, was terminated early
when an interim analysis showed
significantly more thrombotic ev-
ents in the rivaroxaban-treated
patients (19%) compared with
those receiving VKAs (3%) (56).
A second RCT, in which 60%
of patients were triple-positive,
found a higher annualized recur-
rent thrombosis rate in the rivar-
oxaban-treated patients (3.9%)
compared with those receiving
VKAs (2.1%), with stroke more
common in rivaroxaban-treated
patients (57). Given these data,
VKAs are the preferred anticoagu-
lant for patients with APS (2, 58).
How should clinicians monitor
anticoagulation?
Clinicians should use a cali-
brated anti-Xa assay or activated
partial thromboplastin time to
© 2022 American College of Physicians
Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.
adjust the UFH dose. LMWH and
DOACs do not require routine
laboratory monitoring; however,
in exceptional circumstances,
such as major bleeding or the
need for urgent, life-threatening
surgery, laboratory testing to
estimate drug levels may be
necessary (Table 5). It is impor-
tant to note that the correlation
between these laboratory assays
and anticoagulation effects or
bleeding risk has not been firmly
established. At best, these tests
can be used for determination
of the presence or absence of
anticoagulant effects rather than
precise quantification (59). The
INR, a calculated adjustment of
the prothrombin time, can help
monitor VKA therapy. For treat-
ment of VTE, the target thera-
peutic range of the INR should
be between 2 and 3. The mecha-
nism of action of VKAs involves
reducing the synthesis of new
vitamin K–dependent clotting
factors only; therefore, VKA dose
changes take approximately 3
days to affect INR measurement.
A systematic and coordinated
INR monitoring process is recom-
mended, and software may also
assist in dose adjustments (60). A
patient's time in the therapeutic
range (TTR) should be reviewed
periodically, with the goal of an
average TTR greater than 60% to
80%. Compared with standard
dosing protocols, a genotype-
guided dosing strategy does not
increase TTR or reduce bleeding
or thromboembolic complica-
tions with VKAs (61).
How should clinicians manage
recurrent DVT during
anticoagulant therapy?
Recurrent VTE during anticoagu-
lant therapy is uncommon, occur-
ring in about 2% of patients
during the acute treatment phase
ITC12 In the Clinic
and even less often in the exten-
ded treatment phase (62). When
confronted with this, clinicians
should examine drug adherence
and risk factors for therapeutic
failure, such as underlying can-
cer, APS, and heparin-induced
thrombocytopenia (if there was
recent heparin exposure). There
are no high-quality studies to
support one management strat-
egy for therapeutic failures;
therefore, referral to a consultant
is recommended. Patients using
VKAs are generally switched to
LMWH for 4 weeks before
resuming their prior therapy or
switching to a DOAC. Failures
with LMWH can be managed by
increasing the LMWH weight-
based dose to 120% to 125% for
4 weeks and then resuming prior
treatment. There is no evidence
to guide treatment decisions af-
ter DOAC failure. Switching to
VKAs, switching to LMWH for 4
weeks, or reinitiating a higher “ini-
tial” treatment phase of DOAC
dosing are reasonable (Table 5).
When should clinicians stop
anticoagulation?
Acute DVT should be treated
for a minimum of 3 months,
regardless of the cause (3, 12).
Extension of therapy beyond this
period should be based on the
risk for recurrence. The Inter-
national Society on Thrombosis
and Haemostasis Scientific and
Standardization Subcommittee
suggests that patients with an
estimated recurrence risk less
than 5% at 1 year or 15% at
5 years should stop anticoagulation
after 3 months. All others should
be considered for extended-dura-
tion (indefinite) therapy. VTE pro-
voked by major transient risk
factors, such as major surgery, car-
ries a less than 1% risk for recur-
rence at 1 year, and antico-
agulation can be stopped after 3
Annals of Internal Medicine
months as long as the transient
risk factor is resolved. However,
patients with minor transient risk
factors, such as hormonal treat-
ments, have a 15% risk at 5 years
and may still benefit from extended
therapy.
A systematic review and meta-
analysis of 18 studies and 7515
patients with unprovoked VTE
showed that after anticoagula-
tion was stopped, VTE recurred
in 10% of patients after 1 year
and 36% after 5 years (63).
Guidelines recommend that pa-
tients with unprovoked VTE who
are not at high bleeding risk
continue anticoagulation indefi-
nitely. The risk for recurrent VTE
is 2.2-fold higher for men than
for women (64), and the “Men
continue, and HERDOO-2” clinical
prediction rule is one strategy
to safely identify a subset of
women with a low risk for recur-
rent VTE who can stop anticoa-
gulation (65).
Table 6. The Villalta Scale for Diagnosis and Grading of PTS
Variable
Symptoms†
Pain
Cramps
Heaviness
Paresthesia
Pruritus
Clinical signs‡
Pretibial edema
Skin induration
Hyperpigmentation
Redness
Venous ectasia
Pain on calf compression
Venous ulcer
Score
No PTS
Mild PTS
Moderate PTS
Severe PTS
PTS = postthrombotic syndrome.
* For symptoms and clinical signs: 0 = absent, 1 = mild, 2 = moderate, and 3 = severe.
† Self-rated by the patient.
‡ Rated by the clinician.
Annals of Internal Medicine
Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.
What options are available for
patients who cannot tolerate
anticoagulants?
An IVC filter may be used in
patients with an acute proximal
DVT or PE who cannot receive
full-dose anticoagulation because
of uncontrollable active bleeding,
a high risk for life-threatening
bleeding, or urgent surgery
requiring interruption of anti-
coagulation. Anticoagulation
should be started as soon as it is
safe to do so. Once full-dose
anticoagulation is resumed with-
out recurrence of major bleed-
ing, the IVC filter should be
removed to reduce the risk for
filter-related complications, which
increases over time.
When should clinicians use
thrombolysis?
Thrombolytic therapy for iso-
lated DVT is generally reserved
for patients with threatened limb
ischemia (phlegmasia cerulea
dolens) (12). Although some
In the Clinic
studies have suggested a possi-
ble benefit of reduced incidence
of PTS in patients treated with
thrombolytic therapy, the ATTRACT
trial found that the incidence of
PTS at 24 months was the same
between patients treated with
local thrombolytic therapy com-
bined with clot aspiration versus
anticoagulation alone. ATTRACT
also confirmed no benefit in
mortality or VTE recurrence out-
comes between the strategies.
An updated Cochrane review of
19 RCTs showed that thromboly-
sis improved imaging assess-
ment of complete clot lysis at
early and intermediate follow-up
but not late follow-up (reported
in only 2 studies). There was a
significant increase in bleeding
(6.7% vs. 2.2%; risk ratio, 2.45
[CI, 1.58 to 3.78]) and a modest
reduction in PTS (50% vs. 53%;
risk ratio, 0.78 [CI, 0.66 to 0.93])
among those who received
thrombolysis (66).
Score*
0, 1, 2, 3
0, 1, 2, 3
0, 1, 2, 3
0, 1, 2, 3
0, 1, 2, 3
0, 1, 2, 3
0, 1, 2, 3
0, 1, 2, 3
0, 1, 2, 3
0, 1, 2, 3
0, 1, 2, 3
Absent or present
0–4
5–9
10–14
15 or presence of venous ulcer
ITC13 © 2022 American College of Physicians
How should clinicians treat
PTS?
Approximately 25% to 40% of
patients with symptomatic DVT
develop PTS. Symptoms of PTS
include pain in the affected limb,
heaviness, swelling, stasis der-
matitis, and ulceration and can
affect quality of life and lead to
disability. The Villalta score
should be used to diagnose and
grade the severity of PTS (Table
6) (67). Treatment includes leg
exercises, avoiding dependent
positions for lengthy periods,
and using elastic compression
stockings. Elastic compression
stockings for symptom control
are typically prescribed at 20 to
40 mm Hg, and below-the-knee
stockings are better tolerated
than those that cover the leg
higher. Contraindications to com-
pression stockings include severe
peripheral artery disease and
allergies to the material.
When should clinicians
consider consulting a specialist
for DVT?
Clinicians should consider consult-
ing a specialist with expertise in
thrombosis when considering a
diagnosis of recurrent DVT or
breakthrough DVT during anticoa-
gulation. A referral should also be
considered if imaging is nondiag-
nostic but suspicion for recurrence
remains high, complications nece-
ssitating alternatives to anticoagu-
lation emerge, or VTE in the set-
ting of pregnancy is encountered.
Life-threatening bleeding and the
need for urgent reversal of antico-
agulation also benefit from consul-
tation. Management of severe
PTS associated with venous ulcers
should involve a multidisciplinary
team that includes internists, der-
matologists, vascular surgeons, and
wound care nurses.

Treatment... Most patients with DVT can be safely treated with DOACs as outpatients. In patients with renal
impairment or APS or those for whom medication cost is a concern, VKAs are more appropriate. Patients with a
transient reversible risk factor for VTE should be treated for 3 months. Extended treatment should be considered
when the patient has a low risk for bleeding and has had unprovoked VTE, has active cancer, or has had recurrent
VTE. IVC filters should be used only when full-dose anticoagulation is contraindicated. Thrombolytic therapy may
be considered for patients with a massive iliofemoral DVT and impending limb ischemia who are at low risk for
bleeding.

CLINICAL BOTTOM LINE

Practice Improvement
What measures do stakeholders
use to evaluate the quality of
care for patients with VTE?
The Centers for Medicare &
Medicaid Services has devel-
oped 11 categories of hospital-
acquired conditions, some of
which are preventable by apply-
ing evidence-based guidelines.
One of these is prevention of VTE
after total knee and hip replace-
ment. Hospital Compare (http://
medicare.gov/hospitalcompare)
compiles information about the
quality of VTE care at more than
4000 Medicare-certified hospi-
tals, including on the following
measures: 1) “patients who got
treatment to prevent blood
clots on the day of or day after
hospital admission or surgery,”
2) “patients who got treatment
to prevent blood clots on the
day of or the day after being
admitted to the intensive care
unit,” 3) “patients who developed
a blood clot while in the hospital
who did not get the treatment
that could have prevented it,”
and 4) whether “patients with
blood clots were discharged on
a blood thinner medicine and
received written instructions ab-
out that medicine.”
What do professional
organizations recommend
regarding the care of patients
with DVT?
ACCP offers a series of compre-
hensive guidelines, last published
in complete form in 2012, with
updated recommendations pub-
lished in shorter expert panel
reports since then (2, 12, 68).
ASH also offers guidance on
anticoagulation, diagnosis, can-
cer, pregnancy, and VTE preven-
tion in surgical and medical
patients (14, 23, 25, 30, 69, 70).
Specific guidelines for prophy-
laxis (71), cancer (7), pregnancy
(72), recurrent VTE (47), and PTS
(73) are also available. A sum-
mary of the recommendations
from major professional organi-
zations for the most common
clinical questions is provided in
Appendix Table 3 (available at
Annals.org).

© 2022 American College of Physicians ITC14 In the Clinic Annals of Internal Medicine

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

In the Clinic Patient Information

Tool Kit
https://medlineplus.gov/
deepveinthrombosis.html
https://medlineplus.gov/languages/
deepveinthrombosis.html
Information and handouts on deep venous
thrombosis in English and other languages
Deep Venous Thrombosis from the National Institutes of Health's
MedlinePlus.

www.nhlbi.nih.gov/health/venous-
thromboembolism
www.nhlbi.nih.gov/es/salud/
tromboembolia-venosa
Information on venous thromboembolism
in English and Spanish from the National
Heart, Lung, and Blood Institute.

www.heart.org/en/health-topics/venous-
thromboembolism
Information on venous thromboembolism
from the American Heart Association.

www.cdc.gov/ncbddd/dvt/index.html

In the Clinic
www.cdc.gov/NCBDDD/Spanish/dvt/
index.html
Information on venous thromboembolism
in English and Spanish from the Centers
for Disease Control and Prevention.
Information for Health Professionals
https://journal.chestnet.org/article/S0012-
3692(15)00335-9/fulltext
CHEST guideline and expert panel report on
antithrombotic therapy for venous
thromboembolism.

https://ashpublications.org/bloodadvances/
article/5/4/927/475194/American-Society-
of-Hematology-2021-guidelines-for
American Society of Hematology 2021
guidelines for prevention and treatment of
venous thromboembolism in patients with
cancer.

https://ashpublications.org/bloodadvances/
article/2/22/3317/16094/American-Society-
of-Hematology-2018-guidelines-for
American Society of Hematology 2018
guidelines for management of venous
thromboembolism in the context of
pregnancy.

Annals of Internal Medicine In the Clinic ITC15 © 2022 American College of Physicians

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

 In the Clinic
WHAT YOU SHOULD Annals of Internal Medicine
KNOW ABOUT VENOUS
THROMBOEMBOLISM
What Is Venous Thromboembolism?
Venous thromboembolism (VTE) is a common and
potentially life-threatening condition that causes
blood clots to form inside veins. VTE most com-
monly causes blood clots in the lower legs,
known as deep venous thrombosis (DVT), and it
can also cause blood clots in the lungs, known as
pulmonary embolism (PE). More than half of all
VTE disease results from transient risk factors
and may be preventable.
Risk factors for VTE include:
• Recent injury
• Recent surgery
• Recent bed rest, including prolonged How Is It Treated?
hospitalization Blood thinners are the most effective treatment for
• Active cancer VTE and may be given by injection and by
• Family history of VTE mouth. Most people can be successfully treated
• Autoimmune disorders as outpatients. Patients with VTE who are clini-
cally unstable or those who are at high risk for
How Can It Be Prevented? complications from DVT or PE may be treated in
• Blood thinners can prevent VTE in patients who the hospital. People with VTE may need to take
are at high risk. blood thinners for at least 3 months. Many
• Compression stockings may be applied to the patients with VTE may require lifelong treatment
legs to prevent VTE. with blood thinners. Moving the legs, elevating
• The risk for clots must be balanced with the risk the legs, and using compression stockings can
for bleeding. help keep the legs from swelling while the clot is
• Early mobilization also decreases VTE risk. being treated.

What Are the Warning Signs? Questions for My Doctor

• What are my risk factors for VTE?

In many cases of VTE, there may be no symptoms.
When symptoms do occur, they include:
• Chest pain
• Shortness of breath
• Swelling in the leg, including the calf, ankle, and
foot
• Pain in the leg
• Skin that feels warm to the touch
• Changes in skin color (redness)
Patient Information
• What can I do to prevent blood clots in the
future?
• What are the risks and benefits of blood
thinners?
• What symptoms require emergency care?
• How long will I need to stay on blood thinners?
• Are there any activities I should avoid?
• Can I take blood thinners if I am pregnant?

How Is It Diagnosed?
Your doctor will check your vital signs, including
your oxygen level, and examine your heart,
lungs, and legs for any swelling or tenderness.
Your doctor will also evaluate your risk factors
and may do blood tests to see if further testing is
needed. Additional tests may include an ultra-
sound of the legs or more advanced imaging of
the chest, such as computed tomography.

For More Information

Centers for Disease Control and Prevention
www.cdc.gov/ncbddd/dvt/facts.html
Mayo Clinic
www.mayoclinic.org/diseases-conditions/deep-vein-thrombosis/
symptoms-causes/syc-20352557

© 2022 American College of Physicians ITC16 In the Clinic Annals of Internal Medicine

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

Appendix Table 1. Inherited and Acquired Thrombophilia
Thrombophilia Description Increased Risk Laboratory Prevalence
for First VTE Measurement

Factor V Leiden Point mutation (1691G!A Heterozygous: APCR, PCR Heterozygous:

Prothrombin gene mutation
Protein C deficiency
Protein S deficiency
Antithrombin deficiency
Antiphospholipid syndrome
substitution) resulting in 4- to 8-fold White: 5%
factor V resistance to cleavage Homozygous: Hispanic: 2.2%
by activated protein C 12-fold Black: 1.2%
Indigenous American: 1.2%
Homozygous: <1%
Point mutation (20210G!A Heterozygous: PCR Heterozygous:
substitution) resulting in 2- to 4-fold White: 2%
increased production of factor II Homozygous: Black: 0.5%
7-fold Indigenous American: 1.2%
Homozygous: <1%
Various mutations leading to loss 4- to 6-fold Activity assay <0.5%
of function of protein C, a
natural anticoagulant that
degrades activated factors V
and VIII
Associated with warfarin-induced
skin necrosis
Various mutations leading to loss 1- to 10-fold Activity assay <0.5%
of function of protein S, a co-fac-
tor for protein C
Clinical features similar to protein
C deficiency
Various mutations leading to loss 14-fold Activity assay 0.02%
of function of antithrombin, a
natural anticoagulant that inhib-
its thrombin and activated fac-
tors IX and X
Associated with heparin
resistance
Autoimmune disorder associated 1- to 10-fold Lupus anticoagulant† 0%–5%
with arterial and venous throm- (PTT-LA, dRVVT)
bosis as well as obstetric ELISA: Anticardiolipin antibodies
complications (IgG and/or IgM) in medium or
Criteria*: high titer (>40 GPL/MPL or
Clinical criteria: >99th percentile)
1. Vascular thrombosis ELISA: Anti–b2-glycoprotein
(≥1 arterial, venous, or 1 antibody (IgG and/or IgM)
small-vessel thrombosis in >99th percentile
any tissue or organ)
2. Pregnancy morbidity
(≥1 unexplained loss at or
beyond 10 wk, or prema-
ture birth before the 34th
week of gestation because
of eclampsia, severe pree-
clampsia, or placental insuf-
ficiency, or 3 consecutive
unexplained spontaneous
abortions before 10 wk)
Laboratory criteria: ≥1 laboratory
criterion, present on ≥2
occasions ≥12 wk apart

APCR = activated protein C resistance; dRVVT = dilute Russell’s viper venom time; ELISA = enzyme-linked immunosorbent assay;
GPL/MPL = IgG/IgM phospholipid units; PCR = polymerase chain reaction; PTT-LA = lupus anticoagulant–sensitive partial thrombo-
plastin time; VTE = venous thromboembolism.
* Revised Sapporo/Sydney classification.
† Lupus anticoagulant detected according to guidelines of the International Society on Thrombosis and Haemostasis.

Annals of Internal Medicine In the Clinic © 2022 American College of Physicians

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

Appendix Table 2. Summary of Guideline Recommendations for Postpartum Prophylaxis
Indication ACOG SOGC RCOG ACCP ASH

Thrombophilia
Heterozygous Clinical surveillance or Clinical surveillance or Consider prophylaxis for Clinical surveillance if no No prophylaxis regard-
FVL/PGM prophylaxis if risk factors 6-wk prophylaxis if any ≥10 d if 1 other risk factor; family history; 6-wk less of family history
additional risk factors consider extending to 6 prophylaxis if family
wk if family history of VTE history
Protein C or S Clinical surveillance or Clinical surveillance or 6-wk prophylaxis Clinical surveillance if no Prophylaxis if family
deficiency prophylaxis if risk factors 6-wk prophylaxis if any family history; 6-wk history of VTE
additional risk factors prophylaxis if family
history
Homozygous Prophylaxis 6-wk prophylaxis 6-wk prophylaxis Clinical surveillance if no Prophylaxis
or compound family history; 6-wk
heterozygous prophylaxis if family
FVL/PGM history or homozygous
Antithrombin Prophylaxis 6-wk prophylaxis 6-wk prophylaxis Clinical surveillance if no Prophylaxis if family
deficiency family history; 6-wk history of VTE
prophylaxis if family
history
APS 6-wk prophylaxis 6-wk prophylaxis 6-wk prophylaxis No recommendation No recommendation
Cesarean section Emergency and elective Emergency cesarean sec- Emergency cesarean Emergency cesarean No recommendation
cesarean section: pneu- tion: if ≥1 risk factor, section: 10-d prophylaxis section: prophylaxis if ≥1
matic compression devi- consider prophylaxis Elective cesarean section: risk factor
ces before delivery if not Elective cesarean section: consider 10-d thrombo- Elective cesarean section:
already receiving if ≥2 risk factors, consider prophylaxis if additional prophylaxis if ≥1 major
thromboprophylaxis prophylaxis risk factors or ≥2 minor risk factors
6-wk prophylaxis if risk
persistent; up to 2-wk
prophylaxis if risk
transient
Vaginal delivery Prophylaxis if multiple risk Prophylaxis if risk factors Prophylaxis for 10 d if ≥2 No recommendation No prophylaxis if
factors (6-wk prophylaxis if risk risk factors ≤1 risk factor
factor ongoing; 2-wk No recommendation
prophylaxis if transient) for >1 risk factor

ACCP = American College of Chest Physicians; ACOG = American College of Obstetricians and Gynecologists; APS = antiphospholi-
pid syndrome; ASH = American Society of Hematology; FVL = factor V Leiden; PGM = prothrombin gene mutation; RCOG = Royal
College of Obstetricians and Gynaecologists; SOGC = Society of Obstetricians and Gynaecologists of Canada; VTE = venous
thromboembolism.

© 2022 American College of Physicians In the Clinic Annals of Internal Medicine

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.

Appendix Table 3. Comparison of Major Society Guidelines for Diagnosis and Treatment of DVT
Clinical Question
How should PTP and D-dimer be used
for investigation of suspected DVT?
How should PTP and D-dimer be
adapted for use in pregnant patients
investigated for suspected DVT?
What anticoagulant should be used
for treatment of DVT (excluding
cancer and pregnancy)?
How should isolated distal DVT be
managed?
How should pregnancy-associated
DVT be managed?
How should cancer-associated VTE be
managed?
When should thrombolysis be used
for patients with DVT?
How long should patients be treated
after a first DVT event?
ACCP = American College of Chest Physicians; ASH = American Society of Hematology; CUS = compression ultrasonography;
DOAC = direct oral anticoagulant; DVT = deep venous thrombosis; GI = gastrointestinal; LMWH = low-molecular-weight heparin;
MRI = magnetic resonance imaging; PE = pulmonary embolism; PTP = pretest probability; VKA = vitamin K antagonist.
Annals of Internal Medicine
Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.
ACCP Recommendation
Recommends that in patients with a low
or intermediate PTP and a negative
D-dimer result, DVT can be excluded
without CUS (intermediate PTP must
use a highly sensitive D-dimer)
Recommends that patients with high
PTP undergo CUS as the initial test
No recommendation about age-
adjusted D-dimer
Recommends an initial evaluation with
proximal CUS followed by serial CUS
(days 3 and 7) if negative
Serial CUS can be omitted if D-dimer
result is negative
Suggests that if symptoms are sugges-
tive of isolated iliac vein thrombosis
and initial CUS is negative, the patient
should have imaging with Doppler of
iliac vein, venography, or MRI rather
than serial CUS
Recommends a DOAC over a VKA
Suggests anticoagulation if severe
symptoms or risk factors for extension;
otherwise, no treatment and serial
imaging for 2 wk
Recommends therapeutic, weight-
adjusted LMWH, with no recommen-
dation for dosing schedules or anti-Xa
monitoring
Recommends an anti-Xa inhibitor
DOAC (apixaban, edoxaban, rivaroxa-
ban) over LMWH, and recommends
that apixaban or LMWH may be pre-
ferred if luminal GI cancer is present
Suggests that DVT should not be
treated with interventions such as
thrombolytic, mechanical, or
pharmacomechanical
Recommends 3-mo treatment duration
after a first DVT that is provoked by a
major transient/reversible risk factor
Suggests extended treatment of indefi-
nite duration after a first DVT with no
identifiable risk factors if the patient is
not at high risk for bleeding
In the Clinic
ASH Recommendation
Suggests that in patients with a low PTP
and a negative D-dimer result, DVT
can be excluded without CUS
In patients with an intermediate or high
PTP, suggests CUS as the initial test
(for intermediate PTP, a potentially
acceptable alternative strategy is to
exclude DVT if highly sensitive D-
dimer result is negative and the preva-
lence of DVT is low)
Suggests that patients with high PTP
undergo CUS as the initial test
Suggestions for age-adjusted D-dimer
are made only for patients with sus-
pected PE, not DVT
Gives no recommendation for use of
PTP or D-dimer in pregnant patients
with suspected DVT
Suggests that all patients with sus-
pected DVT undergo a CUS with iliac
vein imaging and, if negative, also
undergo either a serial CUS or MRI
Suggests a DOAC over a VKA
Does not offer suggestions
Suggests therapeutic LMWH, either
once- or twice-daily dosing, and no
routine anti-Xa monitoring
Suggests a DOAC (apixaban, edoxa-
ban, rivaroxaban) over LMWH
Suggests that a DOAC should be used
with caution in patients with GI cancer
Suggests that most patients with proxi-
mal DVT should not be treated with
thrombolytic therapy
Suggests thrombolytic therapy be re-
served for patients with limb-threatening
DVT and select young patients at low
risk for bleeding with symptomatic DVT
involving the iliac and common femoral
veins; for these select patients, suggests
catheter-directed thrombolysis over
systemic therapy
Suggests 3 to 6 mo of treatment after a
first DVT that is provoked by a tran-
sient risk factor
Suggests extended treatment of indefi-
nite duration after a first DVT with a
chronic risk factor or no identifiable
risk factor if the patient is not at high
risk for bleeding
© 2022 American College of Physicians
Appendix Figure. Diagnostic management of outpatients with suspected DVT.

Suspected DVT

Calculate clinical probability score Whole-leg ultrasonography Limited ultrasonography

(from groin veins to calf trifurcation)

Likely clinical probability Normal Abnormal Normal Abnormal

Limited ultrasonography No anticoagulant Treat D-dimer testing Treat

therapy

Normal Abnormal

Negative Positive
D-dimer testing Treat

No anticoagulant Repeat ultrasonography

therapy in 7 d
Negative Positive

No anticoagulant Repeat ultrasonography

therapy in 7 d

Unlikely clinical probability

D-dimer testing

Negative Positive

No therapy Limited
ultrasonography

Normal Abnormal

No therapy Treat

DVT = deep venous thrombosis.

© 2022 American College of Physicians In the Clinic Annals of Internal Medicine

Downloaded from https://annals.org by Jose Alvarez Payares on 09/12/2022.