Pulmonary embolism: d-dimer may rule it out in low-risk cases

Clinical bottom line (level 1b)

A negative d-dimer test may help rule out pulmonary embolism in patients with a low clinical probability of a PE or with a normal or non-diagnostic lung scan.

Ginsberg et al: Annals of Internal Medicine 1998; 129: 1006-1011

Expires October 2003

The study

Setting: 4 university hospitals, Canada

1177 patients (aged 20 to 94; mean 54, 59% female) with a suspected pulmonary embolism

Excluded if

aged < 18

contraindication to contrast media

geographic inaccessibility

suspected upper extremity deep vein thrombosis

no symptoms within 48 hours of presentation

treated with anticoagulants for 72 hours or more

life-expectancy less than 3 months

Independent blinded reference standard, applied in all patients from a consecutive appropriate spectrum.
Reference standard:

ventilation-perfusion scan, followed by venography or bilateral compression ultrasound of the lower limbs if necessary, and at least 3 months follow-up

Diagnostic test: Simpli-Red d-dimer whole blood assay: abnormal if any erythrocyte agglutination detected

The evidence

pre-test probability of pulmonary embolism: 17%, (95% CI: 15% to 19%)
pre-test probability of PE with a low clinical probability: 3.4%, (95% CI: 2.1% to 4.8%)
pre-test probability of PE with a moderate clinical probability: 26%, (95% CI: 22% to 31%)
pre-test probability of PE with a high clinical probability: 78%, (95% CI: 70% to 87%)
pre-test probability of PE with a normal lung scan: 1.3%, (95% CI: 0.033% to 2.5%)
pre-test probability of PE with a non-diagnostic lung scan: 7.4%, (95% CI: 5.5% to 9.4%)
pre-test probability of PE with a high-probability lung scan: 88%, (95% CI: 83% to 93%)

diagnostic test	pulmonary embolism	no pulmonary embolism	LR+ (95% CI)	post-test probability	LR- (95% CI)	post-test probability
d-dimer positive	167	310	2.7 (2.4 to 3.0)	35%	0.22 (0.16 to 0.31)	4%
total	197	970

diagnostic test	PE	no PE	LR+ (95% CI)	post-test probability	LR- (95% CI)	post-test probability
low clinical probability: positive d-dimer	19	163	3.3 (2.6 to 4.2)	10%	0.27 (0.13 to 0.60)	1%
total	24	679

diagnostic test	PE	no PE	LR+ (95% CI)	post-test probability	LR- (95% CI)	post-test probability
moderate clinical probability: positive d-dimer	81	136	1.7 (1.4 to 1.9)	37%	0.38 (0.26 to 0.58)	12%
total	101	281

diagnostic test	PE	no PE	LR+ (95% CI)	post-test probability	LR- (95% CI)	post-test probability
high clinical probability: d-dimer positive	67	11	1.7 (1.1 to 2.5)	86%	0.15 (0.058 to 0.41)	36%
total	72	20

diagnostic test	PE		LR+ (95% CI)	post-test probability	LR- (95% CI)	post-test probability
normal lung scan: positive d-dimer	3	73	3.2 (1.8 to 5.9)	4%	0.33 (0.060 to 1.78)	0.41%
total	4	315

diagnostic test	PE	no PE	LR+ (95% CI)	post-test probability	LR- (95% CI)	post-test probability
non-diagnostic lung scan: positive d-dimer	40	227	2.2 (1.8 to 2.6)	15%	0.36 (0.22 to 0.59)	2.8%
total	52	646

diagnostic test	PE	no PE	LR+ (95% CI)	post-test probability	LR- (95% CI)	post-test probability
high-probability lung scan: positive d-dimer	124	10	1.7 (1.1 to 2.6)	93%	0.25 (0.13 to 0.49)	65%
total	141	19

between-assay agreement: K: 0.96
K _{interobserver} : 0.95

Comments

Note there is no change in the test characteristics for the different subgroups of patients.

Citation

Ginsberg JS, Wells PS, Kearon C, et al: sensitivity and specificity of a rapid whole-blood assay d-dimer in the diagnosis of pulmonary embolism. Annals of Internal Medicine 1998; 129: 1006-1011

Search Terms: ?
Contributor: Chris Ball and Musab Hayatli, October 1999
Reviewer:

_{Clinical Question.}

_Patient	_{suspected pulmonary embolism}
_{Intervention or Exposure}	_{whole blood d-dimer assay}
_Outcome	_{pulmonary embolism}