Ventilation/Perfusion scanning for acute pulmonary embolism: effect of direct communication on patient treatment...

Although the synthesis of ventilation-perfusion (V/Q) scan result and clinical pretest probability is vital to the diagnostic accuracy of V/Q scanning, there is no updated pretest/posttest tabular decision tool since that presented in the Prospective Investigation of Pulmonary Embolism Diagnosis I data. Because this verbal communication between the nuclear medicine physician and referring clinician is indispensable in this process, we sought to study the relationship between documentation of such communication (DCOMM) of the result of a V/Q scan and patient treatment outcome in patients with suspected acute pulmonary embolus (PE).

Electronic medical records were searched for patients who underwent V/Q scan for the workup of acute pulmonary embolism from 2000 through 2009 at Harborview Medical Center in Seattle, Washington.

Sin hundred eighty-two patients were included in the analysis. Ventilation-perfusion scan outcomes are reported categorically as normal (N), very low (V), low (L), intermediate (I), or high (H) probability. The distribution of V/Q results was 58 (N), 58 (V), 454 (L), 83 (I), and 29 (H). The treatment rates by group were 1.7% (N), 1.7% (V), 1.1% (L), 25.3% (I), and 96.6% (H). The rates of electronic documentation of communication between the nuclear medicine physician and referring physician were 16% (N), 22% (V), 19% (L), 30% (I), and 69% (H). In a logistic regression analysis, odds ratios for treatment relative to the N group were 1596 (P < 0.001) (H), 19 (P < 0.01) (I), and 0.63 (not statistically significant) (L). DCOMM predicted a higher rate of treatment, independent of the outcome of the V/Q scan. In a multivariate logistic regression, the odds ratio for treatment after DCOMM is 2.7 (P < 0.05) and increases to 4.8 (P < 0.01) when on-call studies are excluded.

Review of 682 patients who underwent V/Q scan for suspected PE during 2000-2009 at a single institution demonstrates that DCOMM is a strong predictor of subsequent treatment for PE, across all probability outcomes. The results of this study demonstrate the potential effect of verbal communication from nuclear medicine physicians in this decision process, and a suggested precedent for critical results is presented.