PT - JOURNAL ARTICLE AU - Emmanuel Itti AU - Séverine Nguyen AU - Fabrice Robin AU - Serge Desarnaud AU - Jean Rosso AU - Alain Harf AU - Michel Meignan TI - Distribution of Ventilation/Perfusion Ratios in Pulmonary Embolism: An Adjunct to the Interpretation of Ventilation/Perfusion Lung Scans DP - 2002 Dec 01 TA - Journal of Nuclear Medicine PG - 1596--1602 VI - 43 IP - 12 4099 - http://jnm.snmjournals.org/content/43/12/1596.short 4100 - http://jnm.snmjournals.org/content/43/12/1596.full SO - J Nucl Med2002 Dec 01; 43 AB - Diagnosis of pulmonary embolism (PE) by visual interpretation of ventilation/perfusion (V/Q) scans is limited by the high percentages of patients classified in the intermediate- and low-probability categories. This study proposes a quantitative analysis of the distribution of V/Q ratios to better identify patients with PE. Methods: We studied 99 consecutive patients who underwent dual-isotope 81mKr/99mTc-macroaggregate V/Q scanning and arterial blood gas analysis within 48 h. The 8-view V/Q scans were visually analyzed by 2 observers according to the revised criteria of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) (normal scan or low, intermediate, or high probability of PE). Quantitative analysis of the posterior-view distribution histogram of V/Q ratios was performed using dedicated software. Briefly, regions of interest were drawn around the lungs on the matched V/Q images, smooth filtering was applied, normalized regional V/Q ratios were calculated within each pixel, and a distribution histogram was built. Results: Patients with normal scans (n = 16) had a predominance of V/Q ratios (63.3% ± 13.0%) between 0.8 and 1.2. They had only 9.8% ± 5.8% of ratios > 1.2, and the remaining 26.9% ± 7.5% of ratios were <0.8. By contrast, patients with PE (n = 34) were characterized by a significant increase (15.5 ± 10.0%, P = 0.04) in high V/Q ratios (>1.2) and a significant increase (34.5% ± 8.2%, P = 0.003) in low V/Q ratios (<0.8). Interestingly, a similar pattern was found in patients with a high PIOPED probability of PE, 21.3% ± 11.0% and 37.5% ± 9.2%, respectively. Within the nondiagnostic group (intermediate- + low-probability scans, n = 58), 17 patients were finally diagnosed with PE. Analysis of the distribution histogram in this group allowed the identification of 5 patients with PE (specificity, 78%). Conclusion: A quantitative approach to lung scan interpretation, based on the distribution histogram of V/Q ratios, may be helpful for categorizing patients with suspected PE.