Clinical question: Does a negative computed tomography pulmonary angiography rule out venous thromboembolism (VTE)?
Background: Computed tomography pulmonary angiography (CTPA) is the most common diagnostic modality used to diagnose pulmonary embolism (PE) and has a high negative predictive value in patients with a low 3-month risk of VTE. In patients with higher pretest probability of PE, it is unknown whether CTPA is sufficient to rule out VTE.
Study design: Meta-analysis.
Setting: Published prospective outcome studies of patients with suspected PE using CTPA as a diagnostic strategy.
Synopsis: The authors reviewed 3,143 publications from MEDLINE, EMBASE, and the Cochrane Library and identified 22 prospective outcome studies to include in their meta-analysis. A VTE was diagnosed in 3,923 out of 11,872 participants (33%) using CTPA. Of the 7,863 patients with a negative CTPA, 148 patients had an acute VTE confirmed by venous ultrasound, ventilation/perfusion scan, or angiography, and 74 patients experienced VTE during a 3-month follow-up period, yielding an overall proportion of 2.4% of patients (95% confidence interval, 1.3%-3.8%).
Subgroup analysis showed that cumulative occurrence of VTE was related to pretest prevalence. In the subgroup of patients with a VTE prevalence greater than 40%, VTE was observed in 8.1% of patients with a negative CTPA (95% CI, 3.4%-14.5%).
Bottom line: CTPA may be insufficient to rule out VTE in patients with a high pretest probability of PE.
Citation: Belzile D et al. Outcomes following a negative computed tomography pulmonary angiography according to pulmonary embolism prevalence: a meta-analysisof the management outcome studies.
Dr. Jenkins is assistant professor of medicine and an academic hospitalist, University of Utah, Salt Lake City.