Deep Vein Thrombosis: Clinical Prediction Rules and their Application

Stephanie A. Weyrauch, SPTDPT/MSCI Candidate

Washington University in St. Louis9/3/2014

Deep Vein Thrombosis: Clinical Prediction Rules and their

Application

Overview

• Introduction• Diagnosis of Deep Vein Thrombosis (DVT)• Clinical Prediction Rules• Literature Review• Clinical Application• Conclusions

INTRODUCTION

Introduction

• Deep Vein Thrombosis– Annual incidence: 2/1000 (van der Velde et al, 2011)

– Lead to pulmonary embolism, post-phlebitic syndrome, chronic thromboembolic pulmonary hypertension (Tan et al, 2009; Wells et al, 2006; Oudega et al, 2005; Tamariz et al, 2004; Wells et al, 2003)

– Unnecessary anti-coagulation therapy (Tan et al, 2009; Wells et al, 2006; Oudega et al, 2005; Wells et al, 2003)

Natural History of DVT General History

Most start in calf Rarely cause leg symptoms

25% untreated symptomatic calf DVTs extend to proximal veins within 1 week of presentation

80% symptomatic DVTs involve proximal veins Risk highest within 2 weeks of surgery

Remains elevated ~2-3 monthsKearon C, 2003

Natural History of DVT General History

DVTs associated with surgery start intraoperatively

50% resolve spontaneously within 72 hours Risk for progression is greater in those with

continuing risk factors for DVT i.e. immobilization, large initial DVT

Kearon C, 2003

Natural History of DVT

Specific to Orthopedic Surgery 2x greater risk for DVT compared to general

surgery 75% of DVT occur in operated leg Timing of post-op DVT depends on surgery

i.e. risk higher s/p TKA initially but drops more rapidly compared to s/p THA

Kearon C, 2003

DIAGNOSIS OF DVT

Diagnosis of DVT• Medical history• Physical examination

– Homan's sign (Urbano, 2001)

• Risk assessment– Use of clinical prediction rules

• Potential referral for further testing– D-dimer testing– Ultrasonography

» False negative: about 2% (Kearon, Ginsberg & Hirsch, 1998)

– Venography

Risk Factors for DVTMajor Risk Factors* Minor Risk Factors*

Active Malignancy Obesity

Recent major surgery or trauma Smoking

Recent hospitalization Long distance flights

Prolonged immobilization

Pregnancy

Hormonal therapy

Positive family history

Known thrombophilic factor

Previous venous thrombosis*Table adapted from Tan et al, 2009

Signs/Sxs of DVT

• Signs– Tenderness and swelling in affected leg– Increased warmth– Changes in skin color

• Symptoms– Pain in affected leg

• Not sensitive or specific for diagnosis

Tan et al, 2009; Urbano et al, 2001; Tamariz et al, 2004

Differential Diagnosis for DVT• DVT may have similar signs/sxs as:

– Ruptured calf muscle/tendon– Ruptured popliteal synovial membrane or

Baker's cyst– Muscle cramp– Muscle hematoma– Cellulitis– Chronic venous insufficiency– Lymphodema Tan et al, 2009

CLINICAL PREDICTION RULES

Clinical Prediction Rules

• Definition (Toll et al, 2008; Childs & Cleland, 2006; Tamariz et al, 2004)

– Designed to improve clinical decision making by assisting clinicians in making particular diagnosis, establishing prognosis, or matching patients to optimal interventions

– Based on subset of predictor variables from history and physical examination

Wells Clinical Prediction RuleClinical Characteristic Score

Active cancer (treatment within the last 6 months) 1

Paralysis, paresis, recent plaster immobilization of LE 1

Recently bedridden for 3+ days or major surgery within previous 12 weeks requiring general/regional anesthesia 1

Localized tenderness along distribution of deep venous system 1

Entire leg swollen 1

Calf swelling at least 3 cm larger than asymptomatic side 1

Pitting edema confined to symptomatic leg 1

Collateral superficial veins (nonvaricose) 1

Previously documented DVT 1

Alternative diagnosis at least as likely as DVT -2Wells et al, 1995

Kahn Clinical Prediction Rule

Clinical Characteristic Score

Male sex 1

Orthopedic surgery <6 months 1

Localized warmth 1

Superficial venous dilation 1

Cut off points:0 = low1-2 = moderate≥3 = high

Kahn et al, 1999

St. Andre Clinical Prediction Rule




Localized warmth 1


Superficial venous dilation 1

Previously documented DVT -1

Cut off points:≤0 = low1-2 = moderate≥3 = high

Constans et al, 2001

Ambulatory Constans Prediction RuleClinical Characteristic Score

Male sex 1


Recently bedridden for 3+ days or major surgery within previous 12 weeks requiring general/regional anesthesia

1

Unilateral lower limb pain 1


Alternative diagnosis at least as likely as DVT -2

Cut off points:≤0 = low1-2 = moderate≥3 = high Constans et al, 2003

LITERATURE REVIEW

Statistics Terminology

Likelihood Ratio – assesses usefulness of diagnostic test Positive (+LR) – corresponds to ruling in a

disease Negative(-LR) – corresponds to ruling out a

disease

Statistics Terminology

Likelihood RatioLR Interpretation

> 10 Large and often conclusive increase in the likelihood of disease

5 – 10 Moderate increase in the likelihood of disease

2 – 5 Small increase in the likelihood of disease

1 – 2 Minimal increase in the likelihood of disease

1 No change in the likelihood of disease

0.5 - 1.0 Minimal decrease in the likelihood of disease

0.2 - 0.5 Small decrease in the likelihood of disease

0.1 - 0.2 Moderate decrease in the likelihood of disease

< 0.1 Large and often conclusive decrease in the likelihood of disease

Statistics Terminology Area Under the Curve (AUC)

Measures accuracy of a test Correctly classify those with and without disease in

randomly selected pairs of patients

AUC Interpretation

0.90 – 1.0 Excellent

0.80 – 0.90 Good

0.70 – 0.80 Fair

0.60 – 0.70 Poor

0.50 – 0.60 No better than chance

Image: http://gim.unmc.edu/dxtests/roc3.htm

Comparison of Clinical Prediction Rules

Clinical Decision Rule AUC 95% Confidence Interval Accuracy

Wells 0.76 0.70-0.81 Fair

Kahn 0.57 0.50-0.63 No better than chance

St. Andre 0.67 0.61-0.73 Poor

Constans 0.79 0.74-0.84 Fair

Accuracy of 4 clinical prediction rules for classifying outpatients with and without DVT (adapted from Tan et al, 2009)

Conclusion: Wells clinical prediction rule is recommended Extensively validated Superior accuracy

Wells Clinical Prediction Rule

N% with

Temporary Risk

Factor

% with Cancer

Wells Pretest Probability Category

% Pts with DVT

+LR -LR AUC

4855 16 12

High 19 7.6 0.3

0.79Medium 10 1.1 1.0

Low 3 0.3 3.9

Average likelihood of DVT in outpatient setting using Wells Clinical Prediction Rule (adapted from Tamariz et al, 2004)

Conclusion: Wells clinical prediction rule is useful in identifying patients at low risk of being diagnosed with DVT

Wells Clinical Prediction Rule Wells et al, 2006

14 studies involving 8239 patients

Clinical Probability of

DVT

Prevalence of DVT

95% Confidence

Interval

High 53% 44% - 61%

Moderate 17% 13% - 23%

Low 5% 4% - 8%

Pooled 19% 16% - 23%

Prevalence of DVT within Wells clinical probability groups

Conclusion: Routine use of Wells clinical prediction rule may help us screen outpatients who are at risk for DVT

CLINICAL APPLICATION

Case Examples

CA 22 year old male PMH: obesity, L ACLR x 6 years ago Fell off mechanical bull, rupture ACLR,

medial/lateral menscus tears, patellar tendon tear

Case Examples CA

s/p L knee arthroscopy revision, ACLR bone-patellar-bone autograph, medial & lateral meniscal repair, patellar tendon repair, removal of medal hardware/screws/staples within bone (6/5/14)

Precautions• TDWB x 6 weeks (brace locked in extension)• Brace locked in extension x12 weeks• ROM 3x/week by PT only 0-90 deg x 12 weeks

Case Examples CA

– 6/6/14 (1 day s/p surgery)• 3-4/10 pain (Narco, Oxycontin, Keflex)

– 6/10 (5 days s/p surgery)• New c/o lateral calf pain (rest, TTP) 2/10• No warmth, swelling, redness, (-) Homans• Differential Diagnosis: DVT vs. muscle cramp

Use Wells Clinical Prediction Rule to determine likelihood of DVT. Would you refer him out?





1







Alternative diagnosis at least as likely as DVT -2Cut off points:≤0 = low1-2 = moderate≥3 = high

Wells et al, 1995

CA: Wells Score of ?


Active cancer (treatment within the last 6 months) -




Entire leg swollen -

Calf swelling at least 3 cm larger than asymptomatic side -

Pitting edema confined to symptomatic leg -

Collateral superficial veins (nonvaricose) -

Previously documented DVT -


A. C.: Wells Score of 1 + 1 +












A. C.: Wells Score of 1 + 1 + 1












A. C.: Wells Score of 1 + 1 + 1 – 2 = ?












A. C.: Wells Score of 1 = moderate probability DVT












Case Example

We referred CA to MD to rule out DVTGiven Narco, (-) Doppler US

Case Examples CA

– 6/16 (1.5 weeks s/p surgery)• Less calf pain but still TTP lateral gastroc• Insurance restrictions: PT 1x/week; ROM by

family daily

– 6/20 (2 weeks s/p surgery)• No calf pain

– 6/26 (3 weeks s/p surgery)• Only performed ROM once

Case Examples CA

– 7/10 (5 weeks s/p surgery)• L ankle pain (standing, walking, 2/10) for one

week, resolved prior to PT session• Mild edema posterior knee

– 7/18 (6 weeks s/p surgery)• New c/o inner thigh pain (walking, laying on R

side, 3/10), TTP adductors• Took Narco, no pain in knee/patellar tendon• Mild edema throughout LE• Able to reduce LE pain (1/10) with gait

training, adding padding to brace

Case Examples CA

– 7/24 (7 weeks s/p surgery)• No pain in LE after replacing padding in brace• Mild edema and redness throughout LE, no

warmth (reduced with elevation, STM)

Case Examples CA

– 7/28 (8 weeks s/p surgery)• 7/10 pain, pressure inner thigh, groin (rest,

movement, with/without brace), took Narco• Moderate-severe edema, redness throughout

LE, pitting edema near lateral patella• Hardness in calf, inner thigh, groin• Stopped activity, rested in bed during

weekend

Use Wells Clinical Prediction Rule to determine likelihood of DVT. Would you refer him out?





1








Cut off points:≤0 = low; 1-2 = moderate; ≥3 = high Wells et al, 1995

CA: Wells Score of 5 = high probability of DVT



Paralysis, paresis, recent plaster immobilization of LEimmobilization of LE 1

Recently bedridden for 3+ days or major surgery within previous 12 major surgery within previous 12 weeks requiring general/regional anesthesiaweeks requiring general/regional anesthesia

1

Localized tenderness along distribution of deep venous systemLocalized tenderness along distribution of deep venous system 1

Entire leg swollenEntire leg swollen 1


Pitting edemaPitting edema confined to symptomatic leg 1




Case Example

We referred CA to MD to rule out DVT

Case Examples CA

– Arrival to ED• Tachycardic, unable to

WB• Doppler LLE US: (+) DVT

L common femoral vein extending to L popliteal vein, profundus and calf veins

– Vascular surgeon US found iliofemoral DVT as well

Image: Wells & Scarvelis, 2006

Case Examples CA

– 7/31/14 (8.5 weeks s/p surgery)• Thrombolysis vascular surgery via venogram

(catherization of IVC)

– 8/25 (12 weeks s/p surgery)• US – non-occlusive obstruction of femoral and

popliteal vein, gastro vein; occlusive obstruction of soleal vein at mid calf. Resolved obstruction of common, external iliac, common, distal femoral veins

Case Examples

R. J. (Wells et al, 2006)

60 year old male Seeing you for knee pain Cut plantar aspect of L foot on glass 10 days

ago Resting in bed since

History of diabetes No history of previous DVT

Case Examples

R. J. (Wells et al, 2006)

Presentation L leg pain, mild calf swelling, redness, heat Febrile Pitting edema of L calf Erythema is hot, tender, well demarcated

Differential diagnosis: cellulitus vs. DVT Use Wells Clinical Prediction Rule to determine

likelihood of DVT. Would you refer him out?





1








Wells et al, 1995

Case Examples R. J.: Wells Score of 0 = low probability of DVT



Paralysis, paresis, recent plaster immobilization of LE -


1

Localized tenderness along distribution of deep venous system -







CONCLUSIONS

Conclusions Highest risk of DVT: 2 weeks of surgery, stays elevated 2-3 mos 50% DVTs associated with surgery start intraoperatively but half

resolve spontaneously in ~72 hours Wells clinical prediction rule is recommended because of

superior validation, accuracy compared to others Routine use of Wells clinical prediction rule may help us screen

outpatients who are at risk for DVT Wells clinical prediction rule is useful in identifying patients at

low risk of being diagnosed with DVT Even with low probability of DVT, 5% risk of DVT

US false positive rate 2%

ReferencesChilds JD, Cleland JA. Development and application of clinical prediction rules to improve decision

making in physical therapist practice. Phys Ther. 2006;86:122-131.

Constans J, Boutinet C, Salmi LR, Salby JC, Nelzy ML, Baudouin P et al. Comparison of four clinical prediction scores for the diagnosis of lower limb deep venous thrombosis in outpatients. Am J Med. 2003;115:436-440.

Constans J, Nelzy ML, Salmi LR, Skopinski S, Salby JC, Le Metayer P et al. Clinical prediction of lower limb deep vein thrombosis in symptomatic hospitalized patients. Thromb Haemostasis. 2001;86:985-990.

Kahn SR, Abenhaim JL, Leclerc JR. Clnical prediction of deep vein thrombosis in patients with leg: symptoms. Thromb Haemostasis. 1999;81:353-357.

Kearon C, Ginsberg JS, Hirsch J, The role of venous ultrasonography in the diagnosis of suspected deep venous thrombosis and pulmonary embolism. Ann Intern Med. 1998;129:1044-1049.

Kearon C. Natural history of venous thromboembolism. Circulation. 2003;107:I22-I33.

Oudega R, Hoes AW, Moons KGM. The Wells Rule does not adequately rule out deep vein thrombosis in primary care patients. Ann Intern Med. 2005;143:100-107.

ReferencesTamariz LJ, Eng J, Segal JB, Krishnan JA, Bolger DT, Streiff MB, et al. Usefulness of clinical prediction

rules for the diagnosis of venous thromboembolism: a systematic review. Am J Med. 2004;117:676-684.

Tan M, van Rooden CJ, Westerbeek RE, Huisman MV. Diagnostic management of clinically suspected acute deep vein thrombosis. Brit J Hematol. 2009;46:347-360.

Toll DB, Janssen KJM, Yergouwe Y, Moons KGM. Validation, updating and impact of clinical prediction rules: a review. J Clin Epidemiol. 2008;61:1085-1094.

Urbano FL. Homans' sign in the diagnosis of deep venous thrombosis. Hosp Physician. 2001:22-24.

van der Velede EF, Toll DB, ten Cate-Hock, Oudega R, Stoffers HEJH, Bossuyt PM et al. Comparing the diagnostic performance of 2 clinical decision rules to rule out deep vein thrombosis in primary care patients. Ann Fam Med. 2011;9:31-36.

Wells PS, Anderson DR, Rodger M, Forgie M, Dreyer J, Kovacs G. Evaluation of D-dimer in diagnosis of suspected deep-vein thrombosis. New Eng J Med. 2003;349(13):1227-1235.

Wells PS, Hirsh J, Anderson DR, Lensing AWA, Foster G, Kearon C, et al. Accuracy of clinical assessment of deep-vein thrombosis. Lancet. 1998;345:1330-1332.

Wells PS, Owen C, Doucette S, Fergusson D, Tran H. Does this patient have deep vein thrombosis? JAMA. 2006;295(2):199-207.

Questions??

Deep Vein Thrombosis: Clinical Prediction Rules and their Application

Documents