Natural History Conclusions Lancet, June 18, 1960, 13091312

1
Natural History ConclusionsLancet, June 18,
1960, 1309-1312
PE untreated has high mortality (5/19 in
randomized trial consistent with 23-87 in
retrospective analyses) Waiting is dangerous
since death happened in first two
weeks. Anticoagulation greatly decreased mortality
2
Timing of PE Death
Days after initial event in untreated group 4, 5,
12, 16, 6
3
Bleeding ComplicationsHow Common Are They Really?
Warfarin Bleeding Complications/Year Fatal
0.6 Major 3.0 Minor 9.6
Landefeld, Am J Med 199395315
4
Bleeding ComplicationsRule for Risk Prediction

• Outpatient Bleeding Risk Index
• gt 65 y.o.
• h/o stroke
• h/o GI bleed
• Comorbidity MI, anemia, kidney disease, DM
• 0 low risk, 1 or 2 moderate

Beyth,Am J Med 199810591
5
Bleeding Risk Index in PE

• Among 222 patients
• Risk Minor bleed Major bleed
• Low 128 7 0
• Moderate 92 5 5
• High 2 1 0
• Annualized major bleed in moderate 4.3

Wells, Arch Intern Med 2003163917
6
Treatment Threshold?
The bleeding risk to 90 treated who do not have
PE is balanced by the benefit of RX to 10
patients with PE (reduce death from 2.5 patients
to 0.2 patients)
90 without PE (1-2 major bleeds with Rx)
0.1
10 with PE (2.5 deaths without Rx)
7
Wells Criteria
Previous PE 1.5 HRgt100 1.5 Recent
surg/immob 1.5 Clin Signs DVT 3 Alt. Dx less
likely 3 Hemoptysis 1 Cancer 1 0-1 Low
likelihood, 2-6 Med, gt6 High
Wells Thromb Haemost 200083416
8
In Each Category with PE
High 78 Medium 28 Low 3
Wells includes multiple regression coefficients.
Allows programming you palm to spit out
probability.
9
Geneva (Wicki) Criteria
Arch Intern Med. 200116192
Previous PE/DVT 2 HRgt100 1 Recent Surgery 3 Age
60-79 1 gt80 2 PaCO2 lt36 2 36-39
1 PaO2 lt49 4, 49-60 3, 60-71 2, 71-81
1 Atelectasis 1 Elevated diaphragm 1 0-4 low,
5-8 Med, gt8 High
10
Combine Pretest and DD

• Wells Thromb Haemost 20008416
• Divided patients into two groups based on
  clinical criteria lt4 or gt 4 points and did
  D-dimer on all
• Bottom line The D-dimer successfully ruled out
  PE in the low likelihood group

11
Clinical Prediction Model( rounded)

• Score Pretest -DD DD
• lt4 5-8 2 12-18
• gt4 40 10-18 60
• (Caveat Not all D-D methods are equally
  effective)

12
From An Update on CUSKearon, Ginsberg, Hirsh,
Ann Intern Med 19981291044

• In patients with non-diagnostic lung scan, 5-10
  have CUS for DVT
• Among those with non-diagnostic lung scan and CUS
  -, 20 have PE BUT
• Not at high risk for recurrent PE unless
  redevelop proximal clot

13
Strategy for CUS Nondiagnostic (but not
negative) lung scan

• CUS - Treat
• CUS -, do serial CUS
• 2 will become
• Result lt2 recurrent PE in 6 months with this
  strategy

14
VQ ScanJAMA 19902632753
PIOPED study was an excellent study with enough
problems to make it a great journal club article.
Defined usefulness of VQ Divided into categories
of Negative, Intermediate, High
Probability Compared with angiogram or clinical
course
15
Likelihood Ratios for VQ Results

• Result with PE no PE LR
• High 102 14 7.3
• Intermed 105 217 0.48
• Low 39 273 0.14
• Nml/Near Nml 5 126 0.04
• Normal 0 21 0

16
Later Studies of Spiral CT

• Overall Sensitivity 88, Specificity 95
  (reviewed in Chest 20011191791)
• Less good for small subsegmental PEs (but some
  argue these are less clinically important)
• Likelihood ratios implied by above
• LR 18, LR- 0.1

17
What About Echocardiogram?
The issue Massive PE (gt50 reduction in blood
flow) may benefit from lytics. Can we define need
based on RV strain?
18
Massive PE
Further defined by severe hypoxia and vascular
collapse (shock and hypotension) In various
studies, massive PE represents from lt1 to 13 of
all PE patients.
19
The Randomized Controlled TrialJeres-Sanchez
Thromb Thrombolysis. 19952227
RCT Heparin or Heparin Streptokinase Pts.
With PE, systemic hypotension, PA hypertension,
RV dysfunction. All with streptokinase lived, all
without died. Four patients in each group.
20
What about lytics based on ECHO?Circulation
199796882
Pts with PE, RV dysfunction, no shock 169 got
lytics (L), 550 just heparin (H) Mortality 30
days L 4.7, H 11.1 Recurrent PE L 7.7, H
18.7 But choice of therapy was at discretion of
MD. Those getting lytics were younger with less
lung and heart disease.
21
Submassive PE and lyticsNEJM 20023471143
Pts with PE and one of ECHO-demonstrated RV
dysfunction Pulmonary hypertension (R heart
cath) ECG evidence of right heart strain Heparin
(n 118) vs. Heparin alteplase (128) 24.6
treatment escalation with heparin vs 10.2 in H
alteplase. No difference in mortality!
22
Recommendations

• Set an accurate pretest probability through use
  of prediction rule
• Use D-dimer to help R/O in patients with low
  probability
• Helical CT for all (or at least those with
  abnormal chest X-ray) moderate or high

23
Recommendations

• 4. Compression US helpful to find clot (5-10) or
  (serially) detect early reoccurence of clot (if
  CT negative)
• Follow negative Helical CT (and negative CUS)
  with angiogram for high or medium risk
• Echocardiogram the jury is still out!