• 
• 

Three Part Question

In [a clinically well pregnant patient with a suspected Pulmonary Embolism] is [a negative D-dimer sensitive enough] to [exclude Pulmonary Embolism]

Clinical Scenario

A patient attends the emergency department (ED) with atraumatic pleuritic chest pain. She is 12 weeks pregnant with no other medical history. A junior doctor has dutifully followed the ED guideline, noted that the patient is at ‘low clinical risk’ of pulmonary embolism (PE) and requested a D-dimer level, which has returned within normal limits. The junior doctor is now keen to discharge the patient, who has remained well in the ED, but wants to ‘run it by you’ first. You are surprised by the normal D-dimer level in pregnancy but wonder whether the sensitivity and negative predictive values are as high in pregnant patients as they are in low risk non-pregnant patients.

Search Strategy

Medline 1950–23rd November 2010) and Embase (1980–2nd November 2010) via NHS Evidence (http://www.library.nhs.uk/). The Cochrane Library, 23rd November 2010.
MEDLINE and EMBASE: (exp pulmonary embolism/OR exp venous thromboembolism/OR venous thromboembolism.mp. OR pulmonary embolism.mp.) AND (exp pregnant women/OR exp pregnancy/OR pregnant.mp. OR pregnancy.mp.) AND (exp fibrin fibrinogen degradation products/OR exp cross-linking reagents/OR D-dimer.mp. OR cross linking reagents.mp.). Limit to humans and English language. Cochrane: D-dimer AND pregnan*.

Search Outcome

One hundred and forty-three relevant papers were identified using MEDLINE and EMBASE and eight in the Cochrane Library. Five papers were directly relevant to the three-part question (table). No additional papers addressing the three-part question were found in the Cochrane database.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Damodaram et al, 2009, United Kingdom	37 patients suspected of PE who underwent ventilation/perfusion diagnostic scanning and also had D-dimer testing.	Retrospective cohort study	Sensitivity of D-dimer as a stand alone diagnostic test for PTE	73%	Retrospective analysis based on incidental measurement of D-dimer without guidance. Ventilation/perfusion scan result appears to have been used in isolation as reference standard for disease. This is against current guidelines.
			Specificity of D-dimer as a stand alone diagnostic test for PTE	15%
			Negative likelihood ratio using D-dimer as a stand alone diagnostic test for PTE	1.8
To et al, 2008, United Kingdom	Solo case report of pregnant patient with large right sided PE and serial negative antenatal D-dimers	Case report	Serial D-dimer estimation	Initially positive at 9 weeks with proven iliofemoral DVT, followed by sequentially negative measurements throughout pregnancy despite developing and confirmed PE	Isolated case report
Nijkeuter et al, 2006, Netherlands	Four studies eligible for inclusion, assessing diagnostic methods in pregnant patients with clinical suspicion of DVT/PE.	Systematic review of diagnostic technique	Studies utilising D-dimer estimation as a diagnostic test	None	Low number and poor quality of studies. No prospective studies on investigative strategies identified for PTE in pregnancy. Strict predefined inclusion criteria caused rejection of >2/3 potentially eligible papers.
			Venous thromboembolic event rates at 20 months in untreated patients following normal/non-diagnostic ventilation/perfusion scans in pregnant patients with clinical suspicion of PE	0% (95% CI 0.0-1.0%)
Kline et al, 2005, USA	50 healthy women who indicated their desire to become pregnant at an obstetric office. Serial D-dimer testing at pre-conception and in each trimester (23 women completed the protocol)	Prospective observational cohort study	Proportion of women with D-dimer below the reference range (500 ng/ml)	79% of women had normal D-dimer pre-conception compared with 50% in the 1st trimester, 22% in the 2nd trimester and 0% in the 3rd trimester	Small numbers with high dropout rate. Recruited at an obstetrics office pre-conception so the population may have been high risk at baseline
Kovac et al, 2010, Serbia	* 89 healthy pregnant patients referred for haemostasis testing and tested negative for thrombophilia (reference group). Serial D-dimer testing in each trimester. 12 pregnant women admitted to hospital with suspected venous thromboembolism, 10 of whom had confirmed DVT (thrombosis group)	Prospective observational cohort study	Proportion of women in the reference group with D-dimer <230 ng/ml (diagnostic cut-off)	1st trimester:84%. 2nd trimester:33%. 3rd trimester:1%	Small numbers with thrombosis (n=10), none of whom had PE Lack of statistical detail regarding derivation of suggested trimester specific reference ranges
			Suggested trimester specific diagnostic cut-offs	1st trimester: 286 ng/ml, 2nd trimester: 457 ng/ml; 3rd trimester: 644 ng/ml
			Sensitivity for VTE in the thrombosis group	100% (no women with DVT had D-dimer <230 ng/ml)

Comment(s)

In non-pregnant patients, there is a large evidence base to back the use of D-dimer testing to help ‘rule out’ venous thromboembolism, which has been widely adopted in clinical practice in the ED. However, the proven diagnostic accuracy of D-dimer in this population has perhaps led to overzealous use based on the inference that it will be equally effective in other situations. The key to safe use of D-dimer testing has always been to draw upon Bayesian principles by using it in tandem with a formalised clinical risk score, such as the Well's score. Thus D-dimer should only be used in those patients with a low pre-test probability of disease, in whom the negative predictive value is sufficiently high to confidently exclude PE.

In clinical practice, D-dimer is often used to exclude PE in pregnant patients. This approach is flawed on two accounts. Firstly, a risk score must be used in order to determine who is clinically at low/intermediate risk of disease prior to D-dimer evaluation. No validated risk scores exist at present for pregnant patients, who are subject to different physiology during each trimester. Secondly, pregnant women have a venous thromboembolism rate roughly four times that of non-pregnant controls(Heit).

One could therefore coherently argue that pregnant women will always be at least at moderate clinical risk by default. To apply an existing clinical risk score in this cohort, such as the Well's score, also ignores the fact that, during the derivation of this decision rule, pregnant women were specifically excluded. External validity is therefore compromised.

It is therefore only speculative opinion to suggest that the score will perform as well in pregnant patients. When a D-dimer is used as a tool to exclude PE in pregnancy without a formal clinical risk score, we have limited understanding of its sensitivity and specificity. The small amount of available evidence suggests that D-dimer may lack sufficient sensitivity to exclude the diagnosis in pregnant women. Certainly, reference ranging studies suggest that D-dimer has no value in late pregnancy, as levels are almost uniformly above standard diagnostic cut-offs.

At present, therefore, D-dimer is an inappropriate diagnostic test for suspected PE in pregnancy. Further research is clearly necessary, as the evidence-based use of biomarkers to exclude PE in pregnant women could obviate the need for alternative investigations involving radiation exposure, with their consequent risks to mother and baby alike.

Editor Comment

DVT, deep venous thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism.

Clinical Bottom Line

There is no current evidence to support the use of a single isolated negative D-dimer result as sufficient to rule out PE in the pregnant patient.

References

1. Damodaram M, Kaladindi M, Luckit J et al. D-dimers as a screening test for venous thromboembolism in pregnancy: Is it of any use? Journal of Obstetrics and Gynaecology 2009; 29: 101–103.
2. To MS, Hunt BJ and Nelson-Piercy C. A negative d-dimer does not exclude venous thromboembolism (VTE) in pregnancy. Journal of Obstetrics and Gynaecology 2008; 28: 222 – 240.
3. Nijkeuter M, Ginsberg JS, Huisman MV. Diagnosis of deep vein thrombosis and pulmonary embolism in pregnancy: a systematic review. Journal of Thrombosis and Haemostasis 2006; 4: 496–500.
4. Carrier M, Righini M, Djurabi RK, et al. VIDAS D-dimer in combination with clinical pre-test probability to rule out pulmonary embolism. A systematic review of management outcome studies. Thromb Haemost 2009;101:886–92.
5. Vinayakamoorthy V, Geary S, Ganatra R. A United Kingdom based survey of clinical practice in the diagnosis of suspected pulmonary embolism. Nucl Med Commun 2010;31:112–20.
6. Heit JA, Kobbervig CE, James AH, et al. Trends in the incidence of venous thromboembolism during pregnancy or postpartum: a 30-year population-based study. Ann Intern Med 2005;143:697–706.
7. Wells PS, Ginsberg JS, Anderson DR, et al. Use of a clinical model for safe management of patients with suspected pulmonary embolism. Ann Intern Med 1998;129:997–1005.
8. Kline JA, Williams GW, Hernandez-Nino J. D-dimer concentrations in normal pregnancy: new diagnostic thresholds are needed. Clin Chem 2005;51:825–9.
9. Kovac M, Mikovic Z, Rakicevic L, et al. The use of D-dimer with new cut-off can be useful in diagnosis of venous thromboembolism in pregnancy. Eur J Obstet Gynaecol Reprod Biol 2010;148:27–30.