May 28 – 30, 2015, Montréal, Québec

All Pulmonary Emboli Should Be Treated

Carole Dennie MD FRCPCProfessor of Radiology and Medicine

University of OttawaSection Head, Cardiac and Thoracic Radiology

The Ottawa Hospital

The Case AGAINST

No

Yes

May 28 – 30, 2015, Montréal, Québec

I do not have an affiliation, financial or otherwise, with a pharmaceutical company, medical device or communications organization.

I have no conflicts of interest to disclose ( i.e. no industry funding received or other commercial relationships).

I have no financial relationship or advisory role with pharmaceutical or device-making companies, or CME provider.

I will not discuss or describe in my presentation at the meeting the investigational or unlabeled ("off-label") use of a medical device, product, or pharmaceutical that is classified by Health Canada as investigational for the intended use.

The Lung is a Natural Filter• Has dual blood supply• Prevents small emboli formed

in systemic venous circulation from travelling to systemic arterial circulation

• Capillary endothelium has complete complement of fibrinolytic enzymes to break down clot without clinical effect

Arterioscler Thromb Vasc Biol 2008, 28:387-391

The Lung as a Natural FilterThe Evidence

Incidental pulmonary emboli• >50% autopsies1

• 4% contrast-enhanced CT2 • 17% inpatients > 80 yo3

• 24% asymptomatic trauma patients4

1Goodman LR. Radiol 2005;234(3):654-58 2Storto ML et al AJR 2005;184(1):264-67

3Ritchie G et al. Thorax 2007;62:536-404Schultz DJ et al. J Trauma 2004;56:727-31;discussion 731-33

Wiener RS et al. Ann Intern Med 2011;171:831-37

Time Trends in PE Diagnosis

Additional cases of PE may be associated with a lower severity of illness

72% increase in incidence of PE as primary diagnosis

No change in mortality

36% decrease in case fatality

When a Test is Too Good

Wiener RS et al. Ann Intern Med 2011;171:831-37

Evidence for Overdiagnosis• Anderson DR et al. JAMA 2007

– Randomized controlled trial comparing utility of CTPA and V/Q

– Greater # of PE diagnosed with CTPA than V/Q– Rate of VTE at 3 months identical in untreated patients – Additional PE cases on CTPA clinically unimportant

Anderson DR et al. JAMA 2007;298:2743-2753

Evidence for OverdiagnosisMeta-analysis

22 clinical trials

• Subsegmental PE (SSPE) diagnosis 4.7% for single vs 9.4% for MDCT

• False negative (FN) rate - 0.9% vs 1.1% respectively

• MDCT increases diagnosis of SSPE without change in FN rate SSPE may not be clinically relevant

Carrier M et al. J Thromb Haemosst 2010;8:1716-1722

Overdiagnosis• Definition

Diagnosis of clinically unimportant disease

• Does not mean diagnosis is WRONG

• CT frequently depicts “pathology” of uncertain significanceLung nodules, atherosclerosis, pancreatic, liver,

adrenal and thyroid

• Paradox of improved CT spatial and temporal resolution

ALL Dots are not Necessarily Clots!• PIOPED II1

• Positive predictive value for SSPE only 25% • Low interobserver agreement for SSPE diagnosis on

CTPA • Ghanima W, Nielssen BE, Holmen LG et al. Acta

Radiol 2007 k=0.38; 95% CI, 0.0-0.89• Pena E, Kimpton M, Dennie C et al. J Thromb Hemost

2012 k=0.51; 95% CI, 0.39-0.64

1Stein PD et al. NEJM 2006;354:2317-327

Overdiagnosis Leads to Overtreatment

• Accepted practice to treat ALL patients with PE with anticoagulation

• Majority diagnosed with SSPE on CT anticoagulated• Majority with probable SSPE on V/Q (low or

intermediate prob) not anticoagulated

Donato AA et al. Thromb Res 2010;126:e266-70Eyer BA et al. AJR 2005;184:623-38

Overtreatment May Lead to Harm

71% increase post-CTPA!

Wiener RS et al. Arch Intern Med 2011;171(9):831-37

Overdiagnosis Costs More

• Mean cost of admission for PE – increased from $25000 in 1998 to $44,000 in 2006

• Mean cost of subsequent warfarin anticoagulation, lab tests and clinic visits - $2694 per year

• Newer anticoagulants more expensive• Potential costs of major bleeding complications

Wiemer RS. BMJ 2013;347-354

Evidence for Safety of Withholding Anticoagulants for SSPE

• PIOPED I – 17% low prob V/Q had SSPE

• Prospective management cohort studies– Pts with low or intermed prob V/Q, low pre-test prob and

negative serial leg vein U/S safe to withhold anticoagulation (recurrent VTE = 0.5%, same as pts with neg CTPA = 1.7%)

• In practice patients with low or intermediate prob V/Q not anticoagulated Perrier A et al. Arch Intern Med 1996;156:531-36

Perrier A et al. Lancet 1999;353:190-95Salaun PY et al. Chest 2010;139:1294-98

Van Beek EJ et al. Clin Radiol 2001;56:838-42

Evidence for Safety of Withholding Anticoagulants for SSPE

Donato AA et al. 2010

• 93 patients with isolated subsegmental PE without DVT

• 3-month clinical outcomes (anticoagulation use, recurrence, death, hemorrhage)

• 24% observed – no recurrent PE, no deaths

• 76% treated – 5.3% major hemorrhage, no deaths

Donato AA et al.Thrombosis Research 2010;126 ;e266–e270

Isolated SSPETreatment vs. No Treatment

• Meta-analysis - over 60 patients• 0% rate of recurrent VTE at 3 months• 7% incidence of major bleeding on anticoagulants

Stein PD et al. Clinical and Applied Thrombosis/Hemostasis 2012;18(1):20-2

Current Guidelines

ACCP guidelines from 2012 do not differentiate between SSPE and more proximal PE

European Society of Cardiology guidelines from 2014 suggest individualized decision if SSPE and negative leg vein ultrasound

Kearon C et al. Chest 2012;2(Suppl):e419S-94SKomstantinides SV et al. Eur Heart J 2014;35:3033-73

SSPE: Who Not to Treat• Adequate cardiopulmonary reserve• No evidence of DVT• Major risk factor for PE no longer present, ie.

surgery, trauma, and no continuing risk factor • No history of central venous catheterization• No history of atrial fibrillation• Compliant and trustworthy patient who would return

for serial noninvasive leg tests

Stein PD et al. Clinical and Applied Thrombosis/Hemostasis 2012;18(1):20-2Hunsaker AR et al. Circ Cardiovasc Imaging 2010;3:491-506

Rebut

tal Rebuttal

• The lung is a natural filter• MDCTPA is too goodOVERDIAGNOSIS!• All dots are not clots• Overdiagnosis leads to overtreatment• Overdiagnosis may lead to more harm• Overdiagnosis costs more• Who not to treatSSPE

SSPE: Who Not to Treat• Adequate cardiopulmonary reserve• No evidence of DVT• Major risk factor for PE no longer present, ie.

surgery, trauma, and no continuing risk factor • No history of central venous catheterization• No history of atrial fibrillation• Compliant and trustworthy patient who would return

for serial noninvasive leg tests

Stein PD et al. Clinical and Applied Thrombosis/Hemostasis 2012;18(1):20-2Hunsaker AR et al. Circ Cardiovasc Imaging 2010;3:491-506

What is the Radiologist’s Role?• Recommend V/Q in patients with normal CXR• SSPE diagnosis on CTPA only when certain

– May recommend leg vein ultrasound in limited negative CTPA for added safety

“Placing the responsibility for overtreatment on the users rather than the providers of diagnostic

information amputates the clinical importance and value of radiologists”

Jha S. Radiol 2014;270(2):628-29

What is the Future of SSPE Diagnosis?

Outcome Trial • Prospective management cohort study

(NCT01455818)– Canada, France, Switzerland– Anticoagulation withheld

• Patients with SSPE diagnosis on CTPA• No DVT on serial leg vein ultrasound

Dr. Mayo

Questions?jmayo@vch.ca