Computed Tomographic Pulmonary Angiography (CTPA) has become the procedure of choice in evaluating patients with suspected pulmonary embolism (PE). The Wells score as a clinical decision rule (CDR), combined with a highly sensitive D-dimer has been useful in a mostly outpatient population. Experience in hospitalized patients is limited. The purpose of this study was to describe the experience with this approach in mostly hospitalized patients with suspected PE.
Guidelines using a CDR and highly sensitive D-dimer were incorporated into a computerized order entry menu. All requests for CTPA required results of a CDR and D-dimer if indicated before proceeding to CTPA. After guideline implementation, all CTPAs per formed from December 2006 to November 2008 at our institution were reviewed.
A total of 261 CTPAs were performed with 43 (16.5%) positive for PE, compared to a previous 5% yield. The CDR score and D-dimer were 5.5 ± 2.4 and 4965 ± 2892 ng/ml respectively for those with PE compared to 4.5 ± 2.1 and 2398 ± 2100 ng/ml (both p < 0.01) without PE. The negative predictive value of a CDR score < 4 and D-dimer <500 and <1000 ng/ml was 1.0. A logistic regression model that combined a CDR score > 4 and D-dimer > 3000 ng/ml was the best at identifying patients with PE (72%). A CDR > 4 and increasing D-dimer levels were associated with increasing percent of PE as outlined in the table.
Guidelines combining a CDR with D-dimer can increase the yield of CTPA in hospitalized patients. This increased yield will reduce the number of negative CTPAs performed. A CDR score < 4 and D-dimer <1000 ng/ml may be useful in identifying patients who require no further evaluation for PE. A CDR > 4 and elevated D-dimer > 3000 ng/ml may identify patients more likely to have a PE.
A CDR and highly sensitive D-dimer may be useful in identifying hospitalized patients with high and low likelihood for PE.
Guy Soo Hoo, No Financial Disclosure Information; No Product/Research Disclosure Information