The Dept of Anaesthesia & Intensive Care, CUHK thanks

for an unrestricted education grant
Intensive Care Nephrology - beyond BASIC courses: London (June), Singapore (July)
Click here for details
Asian Intensive Care Conference, Hong Kong, November 21-22nd 2013

PE & DVT

Updated May 2009 by Charles Gomersall

Thromboembolic disease

70% of patients with confirmed PE have DVT
40% of patients with DVT have silent PE
DVT limited to calf veins (distal DVT) seldom results in clinically obvious PE
similarity in clinical outcome of patients with DVT or with PE during long-term follow up
PE and DVT should probably be regarded as a single clinicopathological entity

Risk

~6.5% of patients have DVT on admission to ICU
further 20-30% develop DVT during ICU stay
risk highest amongst patients who have suffered major trauma (40-70%) and spinal cord injury (60-80%)

Mechanism of risk

Changes in blood flow

venous stasis
- immobilization
- raised CVP
- valvular damage due to previous thromboembolic disease

Changes in properties of blood

increased coagulation and/or platelet activity eg lupus anticoagulant
decrease in physiological anticoagulants and/or fibrinolytic activity common in critically ill patients
- antithrombin III, protein S and protein C deficiencies
- acquired activated protein C resistance
- high levels plasminogen activator inhibitor I

Changes in vessel wall

endothelial damage triggers coagulation
- trauma
- central venous catheters

Risk factors

Strong predisposition

lower limb fracture or joint replacement
major general surgery
major trauma
spinal cord injury

Moderate predisposition

arthroscopic knee surgery
CVC
chemotherapy
chronic heart or respiratory failure
hormone replacement therapy
malignancy
oral contraceptive
paralytic stroke
pregnancy: post-partum
previous venous thromboembolism
thrombophilia
- antithrombin III deficiency
- protein C & S deficiencies
- inherited abnormalities of fibrinogen and plasminogen

Weak predisposition

bed rest >3 days
immobility due to sitting (eg prolonged car or air travel)
increased age
laparoscopic surgery
obesity
varicose veins

Pathophysiology

massive PE increases RV afterload, enlarges RV and deviates septum to left thus decreasing LV volume and compliance
cardiac output and coronary blood flow (especially to right heart) are diminished resulting in decreased ventricular contractility and a cycle of cardiac decompensation. Death is usually due to right heart failure not refractory hypoxia.
neurogenic and humoral influences may result in catastrophic pulmonary hypertension with occlusion of as little as 30% of the pulmonary vascular bed
critically ill patients often have reduced cardiorespiratory reserve and small pulmonary emboli which would be well tolerated by less sick patients may be poorly tolerated by ICU patients
hypoxia is due to a combination of:
- decreased cardiac output resulting in a fall in oxygen delivery and hence mixed venous oxygen saturation
- zones of reduced flow in obstructed areas and zones of luxury perfusion of non-obstructed areas resulting in shunting
- right to left shunt through patent foramen ovale resulting from raised right heart pressures

Clinical features

Small "heraldic" emboli commonly precede a major embolus and 90% of fatal emboli are recurrent emboli

Symptoms

dyspnoea or tachypnoea in 91% of patients; dyspnoea, tachypnoea or pleuritic pain present in 97% of patients with acute PE. Pleuritic pain is more common in patients with pulmonary infarction which is more common in patients with peripheral PE
non-productive cough
haemoptysis only occurs if infarction has occurred
syncope (suggests massive PE though may occur following minor PE)

Physical signs

RS

tachypnoea common
cyanosis usually only occurs after massive PE
pleural effusion
pleural rub (in patients with pulmonary infarction)
wheeze and crackles

CVS

HR: usually increased, bradycardia ominous
+/- signs of pulmonary hypertension (eg loud P2, gallop rhythm and parasternal heave, especially following massive PE
+/- raised JVP with prominent "a" waves
+/- shock after massive PE. Small sharp peripheral pulse may be palpated

Lower limbs

clinical evidence of DVT found in only 30%. DVT found in "normal" leg in 36%

Others

sweating and fever infrequent

Clinical prediction rules

	Revised Geneva score		Wells score
Predisposing factors	Age>65 Previous DVT or PE Surgery or fracture within 1 month Active malignancy	+1 +3 +2 +2	Previous DVT or PE Recent surgery or immobilization Cancer	+1.5 +1.5 +1
Symptoms	Unilateral lower limb pain Haemoptysis	+3 +2	Haemoptysis	+1
Clinical signs & judgement	Heart rate 75-94/min Heart rate ≥95/min Pain over lower limb deep vein on palpation and unilateral oedema	+3 +5 +4	Heart rate >100/min Clinical signs of DVT Alternative diagnosis less likely than PE	+1.5 +3 +3
Clinical probability	Low Intermediate High	Total 0-3 4-10 ≥11	Low Intermediate High	Total 0-1 2-6 ≥7

Differential diagnosis

MI
LVF
aspiration pneumonitis
pleural effusion
pneumonia
fat embolism
pneumothorax
aortic dissection

Investigations

Most screening tests have not been adequately validated for use in ICU patients and comments on sensitivity and specificity are based on studies in other groups of patients

To diagnose DVT

Compression ultrasonography
- Sensitivity >90%, specificity ~95% for proximal DVT
- Examination of common femoral and popliteal vessels only appears to be adequate
Impedence plethysmography: low sensitivity and specificity
MRI: accuracy of 96% in detecting pelvic thromboses

To diagnose PE

blood tests: raised WCC, raised D-dimer (non specific but sensitive; sensitivity varies from 80-99% depending on assay used)
CXR
- changes neither sensitive nor specific.
- localized infiltrates, consolidation, atelectasis in 2/3
- pleural effusion in 1/2
- classical "plump" pulmonary arteries with peripheral pruning relatively rare and non-specific. Tends to be associated with massive PE
ECG
- normal in 30%.
- changes of R heart strain most common
- S1Q3T3 pattern infrequent
- LAD > RAD
- P pulmonale, RBBB, atrial arrhythmias occasionally present. Persistent if PE massive
ABG
- hypoxaemia frequent but PaO2 >10.6 in 1/4 and A-a gradient may be normal
- metabolic acidosis in shocked patients

Multi-detector CT angiography (MDCT)
- imaging method of choice
  
  Clinical probability Negative predictive value Positive predictive value
  
  Low 96% 58%
  
  Intermediate 89% 92%
  
  High 60% 96%
- negative MDCT adequate criterion for excluding PE in patients with low or intermediate clinical probability of PE
- MDCT showing segmental or more proximal PE adequate proof of PE in patient with intermediate or high probability of PE
echo:
- not recommended as part of elective diagnostic strategy in haemodynamically stable patients
- may be useful in haemodynamically unstable patient
  - absence of echo signs of RV overload or dysfunction practically excludes PE as cause ro haemodynamic instability
  - unequivocal signs of RV pressure overload and dysfunction in patient with suspected PE are highly suggestive of PE
  - In patients who are too unstable to undergo CT some advocate treatment of PE for those with:
    - high clinical probability
    - shock index (HR/SBP) ≥1
    - RV dysfunction on echo
- transoesophageal echo may demonstrate central PE or right heart thrombus
ventilation perfusion scintigraphy
- use currently restricted to those without a high risk of death and who have a contraindication to CT arteriography or if multi-detector CT is unavailable

Management of pulmonary embolus

Appropriate management requires assessment of risk of death

high risk:
- shock or
- SBP <90 mmHg or pressure drop ≥40 mmHg for 15 mins, not due to new onset arrhythmia, hypovolaemia or sepsis
intermediate risk
- not shocked AND
- right ventricular dysfunction and/or myocardial injury

Supportive

Cautious fluid loading – marked rise in RVEDP will result in a marked decrease in right sided coronary perfusion pressure and a shift of the interventricular septum to the left, reducing LV preload
Norepinephrine ± dobutamine may be best combination of vasoactive drugs.
- Isoproterenol is arrhythmogenic and a systemic vasodilator and is detrimental.
- Nitroprusside decreases RV coronary perfusion
hypoxaemia can usually be reversed with supplemental oxygen. If mechanical ventilation is required care should be taken to avoid exacerbating RV failure

Definitive therapy

Thrombolytic therapy

recommended for patients with high risk of death. Note that absolute contra-indications to thrombolysis in acute myocardial infarction (eg GI bleeding within 1 month) may be considered only a relative contra-indication in a patient with an immediately life threatening PE
may also be used for patients with significant swelling associated with proximal DVT in absence of contraindications (rare in ICU patients).
contraindicated in pregnancy
regimes:
- streptokinase: 1.5 million IU over 2h
- urokinase: 3 million IU over 2h
- rtPA: 100 mg over 2 h
check APTT after infusion of thrombolytic is complete. Start heparin without loading dose when APTT <2 times upper limit of normal. If APTT is above this level repeat measurement every 4 h until safe to start heparin.

Catheter extraction of clot

requires technical expertise
limited case series suggest that this is a useful technique in those in whom thrombolysis is contraindicated or fails to improve haemodynamics

Surgical embolectomy

Rarely indicated
Mortality approaches 40%
Consider if thrombolysis contraindicated or unsuccessful and there is persistent hypotension, oliguria, hypoxia and metabolic acidosis with radiological confirmation of >50% occlusion

Anticoagulation with heparin and then warfarin.

Heparin should be started as soon as the diagnosis is suspected in the absence of specific contraindications
Heparin should be continued for 5 days after the start of warfarin even if INR is in therapeutic range. This is because warfarin not only inhibits vitamin K dependent clotting factors, it also reduces protein C levels. Thus a hypercoagulable state may exist during early stages of warfarin therapy, before the longer half-life factors (II, IX, X) are inhibited.
Low molecular weight heparin is as least as effective as unfractionated heparin in preventing recurrent venous thromboembolism and at least as safe with respect to major bleeding. Agent of choice for intermediate or low risk PE.

Unfractionated heparin indicated for patients with renal failure. There is a wide discrepancy in recommended unfractionated heparin dosage regimes. The American College of Chest Physicians recommend an initial bolus of 80 U/kg followed by 18 U/kg/h. Subsequent changes in infusion rate are based on APTT (see table 1).

APTT (sec)	APTT (x control)	Adjustment
<35	<1.2	80 U/kg bolus & infusion by 4 U/kg/h
35-45	1.2-1.5	40 U/kg bolus & infusion by 2 U/kg/h
46-70	1.5-2.3	No change
71-90	2.3-3	¯ infusion by 2 U/kg
>90	>3	Stop infusion for 1 h then ¯ infusion by 3 U/kg/h

Table 1. Heparin dose adjustment

Secondary prevention

IVC filter

Indications:

absolute contra-indication to anti-coagulation and high risk of recurrent venous thromboembolism
consider in pregnant women who develop extensive venous thrombosis in weeks before delivery

No data to support routine use of filters in patients with free-floating proximal DVT (risk of PE with adequate anticoagulation ~3%)