ESC Guidelines on management of acute pulmonary embolism - is an update needed? Arnaud PERRIER, MD Division of General Internal Medicine Geneva University Hospitals Geneva, Switzerland Conflicts of interest: consulting fees for bioMérieux
ESC Guidelines on management of acute pulmonaryembolism - is an update needed?
Arnaud PERRIER, MDDivision of General Internal Medicine
Geneva University HospitalsGeneva, Switzerland
Conflicts of interest: consulting fees for bioMérieux
2008
Guidelines update?main topics to be addressed
• Diagnosis
• Prognosis (risk stratification)
• Treatment
What is new?
Guidelines update?
Diagnosis
• Scores for clinical assessment
• D-dimer
• MRI
Assess clinical probability of PEImplicit or prediction rule
NegativeNo treatment
Low/intermediate clinical probabilityor "PE unlikely"
D-dimer
No PENo treatment or
investigate further
MultidetectorCT
PETreatment
High clinical probabilityor "PE likely"
No PENo treatment
PETreatment
PositiveMultidetector CT
SUSPECTED NON HIGH-RISK PEno shock or hypotension
Clinical prediction rules for PEThromb Haemost 2000;83:416-20. Ann Intern Med 2006;144:65-71
Revised Geneva score
Age > 65 years +1
Previous DVT or PE +3
Surgery or fracture (< 1 month) +2
Cancer +2
Unilateral lower limb pain +3
Hemoptysis +2
Heart rate
75 to 94 beats per minute +3
≥ 95 beats per minute +5
Clinical signs of DVT* +4
Low 0 to 3; intermediate 4 to 10; high >11
Wells score
Previous DVT or PE + 1.5
Immobilization or surgery (< 4
weeks)
+ 1.5
Cancer + 1
Alternative diagnosis less probable + 3
Hemoptysis + 1
Heart rate > 100/min + 1.5
Clinical signs of DVT* + 3
*limb edema and pain on palpation of deep veins
• Low 0 to 1; intermediate 2 to 6; high > 7
• PE unlikely: 0 to 2; PE likely: 2.5 or more
Accuracy of clinical prediction rules for PECeriani E et al., J Thromb Haemost 2010;8:957-970.
Score Prevalence of PE, % [95% CI]
3-level rule Low probability Intermediateprobability
High probability
Wells 6 [4-8] 23 [18-28] 49 [43-56]
Geneva revised 9 [8-11] 26 [24-28] 76 [69-82]
2-level rule PE unlikely - PE likely
Wells 8 [6-11] - 34 [29-40]
All accurate, little used: too complicated?
Simplified clinical prediction rules for PEThromb Haemost 2009; 101: 197–200; Arch Intern Med 2008;168:2131-6.
Revised Geneva score
Age > 65 years +1
Previous DVT or PE +1
Surgery or fracture (< 1 month) +1
Cancer +1
Unilateral lower limb pain +1
Hemoptysis +1
Heart rate +1
75 to 94 beats per minute +1
≥ 95 beats per minute +1
Clinical signs of DVT* +1
• Low 0 to 1; intermediate 2 to 4; high 5 or more
• PE unlikely: 0 to 2; PE likely: 3 or more
Wells score
Previous DVT or PE + 1
Immobilization or surgery (< 4
weeks)
+ 1
Cancer + 1
Alternative diagnosis less probable + 1
Hemoptysis + 1
Heart rate > 100/min + 1
Clinical signs of DVT* + 1
*limb edema and pain on palpation of deep veins
• PE unlikely: 0 to 1; PE likely: 2 or more
Simplified revised Geneva score for PEArch Intern Med 2008;168:2131-6.
• 1049 patients from 2 large prospective PE diagnostic trials that
• Still to be prospectivelyvalidated!
Which prediction rule should we select?Ceriani E et al., J Thromb Haemost 2010;8:957-970.
• Similar accuracy
• Prevalence of PE in suspected patients
• > 20%: revised Geneva score
• < 20%: any score
• Setting
• Outpatients, emergency ward: all scores
• Inpatients: Wells score
• D-dimer
• Highly sensitive: 3-level scores
• Less sensitive: 2-level scores
Ruling out PE with point-of-care (POC) D-dimer assaysBMJ 2009;339:b2990
Qualitative (interobserver variability ++)Sn = 85%
6796 patientsPrevalence VTE 18%
5730 patientsPrevalence VTE 8%
Quantitative, only tested in DVTSn= 96%
925 patients Prevalence DVT 34%
Cardiac D-dimer
SimpliRED
Simplify
Specificity01.0
Sen
siti
vity
Combining D-dimer and clinical probability
Clinical prob. low40%
Intermediate55%
High5%
PE unlikely67%
PE likely33%
Less sensitive D-dimer Highly sensitive D-dimer
3-levelscheme
2-levelscheme
Qualitative POC assays
MRI for suspected PE: the PIOPED III studyAnn Intern Med. 2010;152:434-443.
• Multicenter US study including 371 consecutive patients
• MR angiography technically inadequate in 25% (11 to 52%)
• Performance of technically adequate tests:
– Sensitivity 78%, specificity 99%
Guidelines update?
Diagnosis
• Scores for clinical assessment– Extensively validated and simplified
• D-dimer– Point-of-care assays useful but not highly sensitive rule out
PE in low clinical probability or PE unlikely patients
• MRI– Not yet helpful
Guidelines update?
Prognosis
• Clinical scores
• Biomarkers
• CT angiography
Risk stratification for PE
ESC guidelines 2008
PESI rule (Pulmonary Embolism Severity Index)Am J Respir Crit Care Med 2005;172:1041-6
Items Points
Age, per year Age, in years
Male sex 10
History of cancer 30
History of heart failure 10
History of chronic lung disease 10
Pulse ≥110/minute 20
Systolic blood pressure < 100 mm Hg 30
Respiratory rate ≥30/minute* 20
Temperature < 36ºC 20
Altered mental status† 60
Arterial oxygen saturation < 90%* 20
≤65 class I; 66–85 class II; 86–105 class III; 106–125 class IV; and > 125 class V. Patients in risk classes I and II are defined as low-risk.
Validation of the PESI ruleThromb Haemost 2008; 100: 943–948
Multicenter prospective validation on 357 consecutive ED patients
Low-risk 1.1% (0.1–3.8)
Higher-risk 11.1% (6.8–16.8)
Pronostic value of biomarkers for PE
Reference n Bio- Assay Cut-off Test + NPV PPVmarkers value % % %
Troponin
Konstantinides et al 106 cTnI Centaur 0.07 ng/mL 41 98 14
Konstantinides et al 106 cTn TElecsys 0.04 ng/mL 37 97 12
Giannitsis et al 56 cTnT TropT 0.10 ng/mL 32 97 44
Janata et al 106 cTnT Elecsys 0.09 ng/mL 11 99 34
Pruszczyk et al 64 cTnT Elecsys 0.01 ng/mL 50 100 25
BNP or NT-proBNP
ten Wolde et al 110 BNP Shionoria 21.7 pmol/L 33 99 17
Kucher et al 73 NT-proBNP Elecsys 500 pg/mL 58 100 12
Kucher et al 73 BNP Triage 50 pg/mL 58 100 12
Pruszczyk et al 79 NT-proBNP Elecsys 153 to 334* pg/mL 66 100 23
PESI score vs. troponin in suspected PEJ Thromb Haemost 2009;8: 517–522
• 567 patients from a single center registry with confirmed acute PE
• PESI and Troponin I compared for prediction of 30-daymortality
• Overall 30-day mortality of 10% [95% CI, 7.6–12.5%].
PESI score vs. troponin in suspected PEJ Thromb Haemost 2009;8: 517–522
1,0%
9,2%
15,2%
12,0%
0,0%
2,0%
4,0%
6,0%
8,0%
10,0%
12,0%
14,0%
16,0%
PESI Troponin
30
-day
mo
rtal
ity
Low risk
High risk
Simplification of the PESI scoreArch Intern Med. 2010;170:1383-1389
Derivation: 995 patients prospectivelyincluded in single-center registry30-day mortality• Low-risk 1.0% (0.0-2.1)• High-risk 10.9% (8.5-13.2)
Validation: 7106 patients retrospectivelyanalysed in RIETE registry30-day mortality• Low-risk 1.1% (0.7-1.5)• High-risk 8.9% (8.1-9.8)
Low-risk, 0 points (30 to 36% of patients)High-risk, 1 or more points
Fatty-acid binding proteins in acute PE
• 126 non shocked patients with PE
• FABP cut-off defined by ROC analysis 6 ng/mL: – Above cut-off: 28% unfavorable outcomes (death, amines,
intubation or CPR)
– Below cut-off: 1%
• Tn and BNP not predictive
CT angiography for risk stratification in PEMoroni AL et al, Eur J Radiol 2010, in press
• 246 patients with PE
• Univariate analysis:
– RV/LV ratio > 1.0 not predictiveof increased mortality
– Must be combined to embolicburden > 40%
• Multivariate analysis
– CT not significant over clinicalpredictors
Guidelines update?
Prognosis
• Clinical scores– Extensively validated, useful for identifying low-rtisk
patients
• Biomarkers– No benefit over clinical assessment, for low-risk– Low positive ppredictive value for high-risk– Combination of biomarkers or with echocardiography?
• CT angiography– "one-stop" risk assessment?– Disappointing, no uniform criteria
Guidelines update?
Treatment
• New anticoagulants
• Duration of treatment
Treatment of PE: durationESC guidelines 2008
Better efficacy-safety profile of the new drugs?
New anticoagulants
FondaparinuxRivaroxaban/
dabigatran
Action via antithrombin yes no
Inactivation of fibrin-linked thrombin yes yes
Monitoring no no
Heparin-induced thrombocytopenia no no
Oral administration no yes
Animal origin no no
Good safety-efficacy ratio ? ?
Rivaroxaban: EINSTEIN-PE
Primary efficacy endpoint: symptomatic recurrent DVT or fatal or non-fatal PEMain safety endpoint: major and clinically relevant non-major bleeding
Dabigatran for acute DVT/PEThe RECOVER study
Schulman et al., N Engl J Med 2009;361:2342-52
• 2564 patients with acute DVT or PE (30%) included in a randomized comparison of
– Dabigatran (direct oral thrombin inhibitor) 2 x 150 mg/day vs.
– Initial parenteral anticoagulants plus warfarin (INR 2.0 to 3.0)
Recurrent VTE or death Bleeding
Dabigatran for acute DVT/PE: The RECOVER studySchulman et al., N Engl J Med 2009;361:2342-52
Clinically relevant non-major bleeding• Skin hematoma of at least> 25 cm2• Nose bleed > 5 min• Macroscopic hematuria, spontaneous or, if intervention, lasting > 24 hours• Rectal bleeding (more than spotting)• Gingival bleeding > 5 min• Bleeding leading to hospitalization and/or requiring surgical treatment• Bleeding leading to a transfusion (< 2 units of whole blood or red cells)• Any other bleeding event considered clinically relevant by the investigator
Guidelines update?
Treatment
• New anticoagulants– Dabigatran effective, safety identical to warfarin– Rivaroxaban?
• Duration of treatment– In unprovoked PE, prolonged treatment stil to be deceide
on an individual basis
Next guidelines for PE: when?
• Soon
Prof. Adam Torbicki, Warsaw, Poland