Investigation of suspected pulmonary embolism in pregnancy

1. PE in Pregnancy
Dr Chantelle Badawy
ED Registrar – CME - SCGH

2. PE in Pregnancy

3. Epidemiology
The risk of PE in the non-pregnant female risk is 1 in 100,000, this
increases to 1 in 10,000 in pregnant age-matched females.
PE occurs in 1 of 1600 pregnancies
PE during pregnancy may be fatal in almost 15% of patients, and
in 66% of these, death will occur within 30 minutes of the embolic
event.(RCOG)
PE is a leading cause of Maternal mortality in the developed
world
In Australia, the leading causes of direct maternal deaths
 (when combined = ¾ all maternal deaths)
 amniotic fluid embolism (9),
 thromboembolism (8),
 obstetric haemorrhage (7)
 eclampsia (6),
Sepsis and cardiac conditions were the leading causes of
maternal death for Indigenous Australians.

4. Why is this so important?
VTE is a significant cause of morbidity and
mortality during pregnancy. The clinical
evaluation alone cannot confirm or refute a
diagnosis of VTE in the non-pregnant population,
and diagnosing VTE during pregnancy is more
challenging.
Ultimately - nobody “WANTS”
to perform a peri-mortem cesarean

5. Why is it difficult to evaluate?
Difficulty and confusion arises in the work up of PE in the
pregnant patient due to 3 things:
1.The normal physiological changes in pregnancy;
dyspnoea, tachycardia and leg swelling are also
common PE symptoms
2.The pre-test probability score, (Wells & PERC), cannot be
used in a pregnant patient as they were excluded from
the analysis group for criteria validation.
3.The d-dimer will rise in the second trimester and remain
elevated for 4-6 weeks post-partum.
The diagnosis of VTE during pregnancy often is impaired
by a reluctance to expose a pregnant patient and her
fetus to radiation.

6. Multiple Risk factors at play…
1)Pregnancy itself is a risk factor
2)Prev VTE in pt on hormonal contraception
3)Prev VTE during pregnancy (risk 12%)
4)Rheumatic heart disease and pts with heart prosthesis
5)Thrombophilia's
6)Age > 35yrs
7)Obesity BMI > 30
8)Multi-parirty >4
9)Recurrent miscarriage
10)Bed rest & physical immobility
11)Hyperemesis gravidarum
12)Shock & dehydration
13)Infection
Why is it difficult to evaluate?

7. The pregnant patient fulfills all of Virchow’s
triad
Emboli cause respiratory compromise due
to lack of perfusion of ventilated lung and
heamodynamic compromise due to
increased pulmonary arterial resistance
Pathophysiology

8. Diagnosis of PE & DVT
Diagnosis of pulmonary embolism
Clinical presentation
Pre-test probability
Routine lab evaluation
ABG
ECG
CXR
D Dimer
Ultrasound
CTPA
VQ Scan

9. Diagnosis – Clinical Sympotms
The classic symptoms of VTE are less specific during
pregnancy than during the non-pregnant state.
Sweating (18%) Dyspnea (62%)
Pleuritic chest pain (55%) Cough (24%)
Tachycardia & palpitations Tachypnoea
Leg swelling & pain Haemoptysis
There are no clinical S&S that are specific for PE, esp in
pregnancy. Consider that dyspnoea occurs in up to 70% of
all normal pregnancies
Clinical suspicion must remain high, and the appropriate
diagnostic workup should ensue.

10. Diagnosis – Pre-test Probability
There is no evidence for the use of (WELLS) pretest probability
assessment in the management of acute VTE in pregnancy.
Pregnant patients were excluded form the validation trials
and therefore no supporting data
Modified Wells which takes into account leg circumference
and trimester has a high sensitivity & specificity though there
is not enough data to put this to use at this point.
Unfortunately, the PERC rule has not been validated in
pregnancy either.
Kline is in the process of validating a modified PERC for
pregnant patients, to be used with HR >105 & applied to
low risk patients to risk – however this lacks evidence at this
stage

11. Diagnosis - ABG
Limited diagnostic value (for pregnant & non-pregnant)
Resp alkalosis is common in both pregnancy and PE.
In one study, ABG analysis showed that only 10% had
arterial PO2 levels less than 60 mmHg and 2.9% had oxygen
saturation levels less than 90%.
The presence of hypoxaemia and normal CXR should raise
the suspicion for PE in pregnancy

12. Diagnosis - ECG
The ECG changes associated with acute PE may be seen in any
condition that causes acute pulmonary hypertension, including
hypoxia causing pulmonary hypoxic vasoconstriction.
One study found that the ECG was abnormal in 41% of women
with acute PE; the most common abnormalities were
 T wave inversion (21%),
S1Q3T3 pattern (15%)
Right bundle branch block (18% during pregnancy and 4.2% in
the puerperium).
Given the increasing incidence of IHD in pregnancy, the ECG
may also be helpful in identifying alternative diagnoses

13. 1)
Sinus tachy
RBBB
T-wave inversions in the right precordial leads (V1-3) as well as lead III

14. 2)
RBBB
Extreme right axis deviation (+180 degrees)
S1 Q3 T3
T-wave inversions in V1-4 and lead III
Clockwise rotation with persistent S wave in V6

15. Diagnosis - CXR
Exclude or diagnose other pathology
Eg – pneumonia, pneumothorax or lobar collapse.
CXR is normal in >50%pt with proven PE.
Abnormal features caused by PE include atelectasis, effusion,
focal opacities, regional oligaemia or pulmonary oedema.
Normal CXR prior to V/Q scanning improves the likelihood of a
definitive V/Q result.
 If the CXR is abnormal with a clinical suspicion of PE, CTPA should
be performed.

17. Diagnosis – D Dimer
Currently, the general consensus is still that D-dimer is not an
adequate enough test to rule-out a pulmonary embolism in a
pregnant patient.
D Dimer
60% of healthy patients in normal pregnancy will have a raised D-
dimer
A ‘positive’ D-dimer test in pregnancy is not necessarily consistent
with VTE. The role of D-dimer testing in the investigation of acute
VTE in pregnancy remains controversial.
Kline proposes that an altered “normal value” in pregnancy is
useful
First trimester - 50% above upper limit
Second trimester – Double the upper limit
Third Trimester - 2.5 times the upper limit

18. Diagnosis – Compression Ultrasound
DVT are the source of more than 95% of PE and the prevalence of
PE correlates with predisposition to leg thrombosis - hence if you
find a leg DVT on US you can stop investigating (also given the DVT
& PE treatment are the same)
This strategy has been evaluated in a prospective cohort study of
221 pregnant women who presented with suspected DVT.’
Sensitivity of serial doppler - 94.1% (95% CI 69.2–99.7%)
Diagnosis of deep vein thrombosis
Ultrasound – Sensitivity 97% & Specificity 94%
No radiation exposure

19. Diagnosis - VQ Scan
The amount of radiation exposure varies
based on the isotope used, but VQ
scanning can be performed safely
during pregnancy.
It is possible to exclude the initial
Ventilatory (V) phase of the scan (if
normal CXR) and therefore reduce the
radiation exposure.
Many authorities continue to
recommend V/Q scanning as first-
line due to in its high negative
predictive value & its substantially
lower radiation dose to pregnant
breast tissue

20. Diagnosis - CTPA
CTPA has potential advantages over V/Q imaging
including:
CTPA is more readily available,
Delivers a low radiation dose to the fetus
Can identify other pathology
There is a theoretical risk of hypothyroidism for neonates who
have been exposed in utero contrast**
Suggest IDC & IV Fluid to “flush” the contrast rapidly and avoid
excessive exposure time to feotus - contrast crosses placenta &
collects in amniotic fluid.
 The radiation exposure is comparable
to that of VQ scan and is within the
amount considered to be safe in
pregnancy.

21. MRI
The use of MRI to diagnose
acute PE has been explored
but is not recommended for
routine diagnosis of PE.
Gadolinium should not be
used in pregnancy where
possible

22. Test Whole body
rad (mSv)
Risk to mother
(mGy)
Risk to Feotus
(mGy)
CXR 0.07 0.01
CTPA 1.6-8.3 Lung
39.5
Breast
10-60
 each
trimester
3.3 – 130
 risk breast
Ca >10mGy
by 13.6%*
above BG risk
VQ Scan
Q alone
1-2.5
0.8
Lung
5.7–13.5
Breast
0.98-1.07
Constant
each
trimester
0.3 – 7.4
Predictive
models show
 lifetime risk
Breast Ca
CTPA than
VQ
Bg rad / year 2.5 Dose 100mGy
= <1% risk
childhood Ca
Max allow rad
exp /yr
50
(Avge = 20)
Conversion
mSv-
effective
dose (tissue
wt factor/ or
absorbency)
mGy- qant
absorbed
dose
*13.6% of 0.1%
risk in 25yo =
0.0136% extra
risk (very little)

23. How to understand the Radiation exposure
to fetus (Kline)
Threshold for dangerous radiation 0.1Gy = $100
 (greater than this is believed to be excessive & possibly cause congenital
abnormalities)
Fetal absorbed CXR is 0.1 Cent
CTPA 25-50 Cents VS VQ – 50-75 Cents
BG radiation to fetus over 9 months $5
Radiation to increase lifetime Ca risk before age 20 by 1 in 10 000 is
$10
Argued that there is no reliable way to measure these doses and
effects in utero.
Exposure to Mum – age you about 3 years (ie equivalent to 3yrs
normal background radiation)

24. DOHWA
Imaging
Pathway

28. Suggested Pathway (ATS)

29. Treatment
UFH (Pregnancy Cat C) & LMWH (Pregnancy Cat C) can be
used in pregnancy
IV Heparin immediately before delivery is ideal given is short half
life. Heparin does not cross the placenta & does not carry risks for
teratogenesis and fetal hemorrhage, although bleeding at the
uteroplacental junction is possible.
- Warfarin (Pregnancy Cat D) crosses the placenta, is
teratogenic and are contraindicated during some stages of
pregnancy,
In the initial management of DVT, the leg should be elevated
and a graduated elastic compression stocking applied to
reduce oedema. Mobilization with graduated elastic
compression stockings should be encouraged.

30. Treatment – unstable
Massive PE may present with shock, refractory hypoxaemia
and/or right ventricular dysfunction on echocardiogram and is
a medical emergency.
Collapsed, shocked women who are pregnant or in the puerperium
should be assessed by a multidisciplinary resuscitation team of
experienced clinicians
Supportive care and intravenous anticoagulation should be instituted
without delay.
Subsequent treatment options are controversial and include IVC filter
placement, thrombolytics, and embolectomy.
Sudden death, cor pulmonale, or cardiovascular collapse occurs
when 60% or more of the pulmonary circulation is obstucted with
emboli

31. PIOPED II Study
 Study question – Clinically suspected acute PE, what is the
diagnostic accuracy of CTPA?
 PIOPED II (2006) is the largest study to date which compared CTPA
against a composite reference standard and demonstrated an 83%
sensitivity and 96% specificity in detecting an acute PE.
 PIOPED II suggested that the predictive value of CTPA is highly
concordant with the pretest clinical probability of PE using Well's
criteria.
 Design - Prospective, multicenter(8) , comparison study (7,284
screened, 3,262 eligible, 1,090 enrolled)over 2001-2003
 N=824 completed evaluation for diagnosis of PE by both:
 Composite reference standard, and
 CT angiography
***Interestingly, “Possible Pregnancy” was an exclusion
criteria for this study

32. PIOPED II Study
Made the following recommendations
1.D Dimer with clincal assessment should be
obtained
2.If D-Dimer is positive, venous Ultrasound is
recommended before using radiaiton imaging
3.69% of PIOPED II investigators recommend
Pulmonary VQ scan and 31% recommend CTPA.
 Stein PD, et al. "Multidetector Computed Tomography for Acute
Pulmonary Embolism". The New England Journal of Medicine.
2006. 354(22):2317-2327.

33. Kline : Systematic review and meta-analysis
of PE in Prengnacy (May 2014)
 Methods:
 Studies in all languages from multiple databases in Feb 2014, ED pts with
possible PE
 Outcome of VTE+ ( DVT or DVT&PE)
 Papers were assessed for selection and publication bias, and heterogeneity
 Findings:
 Seventeen full-length studies of 25,339 patients were analysed.
 The frequency of VTE+ rate among the 506 pregnant patients was 4.1% (2.6-
6.0) compared with 12.4% (9.0-16.3) among non-pregnant patients.
 The pooled RR of pregnancy for VTE+ diagnosis was 0.60 (0.41-0.87).
 Patients in the third trimester had an RR of 0.85 (0.40, 1.77) and patients of
childbearing age(≤45 years) had an RR of 0.56 (0.34 to 0.93).
 Interpretation:
 In the ED setting, physicians test for PE in pregnant patients at a low
threshold, resulting in a low rate of VTE diagnosis (4.1%, 95% CI: 2.6-6.0)
 Relative risk of VTE that is lower than non-pregnant women of childbearing
age who are tested for PE in the ED setting.

34. Summary
Higher risk of PE in pregnancy – must have clinical suspicion
Wells & Perc have not been proven in pregnancy
D Dimer is limited value unless alter cut off values
(controversial)
Always consider Lower limb ultrasound first – if POS then
treat. If NEG keep looking
If no limb symptoms or USS Neg perform CXR
VQ Scan (esp Perfusion phase only) presents less radiation
risk
CTPA is useful if CXR abnormal
When in doubt, treat until confirm/refute diagnosis
Engage multiple team members in decisions, keep the
patient well informed and discuss radiation risks openly

35. References