|
If pretest probability of pulmonary embolism is high, probably not. The 2008 PIOPED II study demonstrated that computed tomographic pulmonary angiography (CTPA) had high specificity (96%) but only moderate sensitivity (83%) for identifying patients with pulmonary embolism (PE; NEJM JW Gen Med Jul 15 2006 and N Engl J Med 2006; 354:2317). With technology improvements, false-negative CTPA findings now occur in 1% or fewer patients with low-to-moderatepretest probability of PE (J Intern Med 2011; 269:433). However, whether CTPA can rule out PE effectively in patients with high pretest probability is unclear. In this meta-analysis of 22 studies (nearly 12,000 patients with suspected PE), CTPA was negative in two thirds of patients. Among those with negative CTPA findings, the frequency of objectively confirmed venous thromboembolism (VTE), found during concurrent admissions (mostly deep venous thromboses) or within 3 months of follow-up (mostly PEs), was 2.8%. In studies with high prevalences of PE (≥40%) — a reasonable surrogate for high pretest probability — VTE was confirmed in 8.1% of patients with negative CTPA findings. |
|
|
|
Leave a Reply Cancel reply
You must be logged in to post a comment.
COMMENT
As with other diagnostic testing, pretest probability can affect test-result interpretation dramatically. CTPA performs well for excluding PE in patients with low-to-moderate pretest probability. In patients with high pretest probability, CTPA alone is inadequate to rule out PE, as VTE will be missed in ≈1 in 12 of such patients. In patients with high pretest probability and negative CTPA, clinicians should employ additional diagnostic strategies (e.g., lower-extremity ultrasound) to search for surreptitious VTE and should educate patients on VTE signs and symptoms at discharge.