Abstract
1425
Objectives: Pulmonary thromboembolism (PTE) is a serious medical condition with high mortality if diagnosis is delayed. Ventilation-perfusion (V/Q) scintigraphy is an established diagnostic test for suspected PE but frequently yields non-diagnostic results, has limited availability and low specificity. Hybrid single photon emission computed tomography/ computed tomography (SPECT/CT) imaging can provide simultaneous assessment of perfusion and ventilation (using low dose CT and omitting ventilation scans). In this study, we assessed the incremental value of perfusion (Q) SPECT/CT over conventional planar imaging interpreted with chest radiograph. Other objectives were to compare the diagnostic accuracy of modified PIOPED II and PISAPED criteria for interpretation of SPECT/CT (Q) lung scintigraphy.
Methods: Clinical records of 180 consecutive patients clinically suspected with acute PTE and referred for V/Q scintigraphy were retrospectively analyzed. The inclusion criteria were: a) performance of perfusion planar and SPECT/CT scans, b) availability of chest radiograph and clinical data over a 6 month follow up period. Number of eligible patients were 143 (80 males and 63 females). Wedge shaped areas of reduced perfusion without corresponding morphologic changes were considered true perfusion defects. To evaluate the diagnostic efficacy of perfusion planar and SPECT/CT imaging, comparison was made with a composite reference standard (consisting of Wells' criteria, D-dimer assay, lower extremity Doppler ultrasonography, CT pulmonary angiography and clinical follow up). Results: PTE was confirmed in 57 (40%) patients and ruled out in 86 (~60%) patients using the composite reference standard. The sensitivity and specificity values of each method were as follows: planar (Q) 96% and 67%, SPECT/CT(Q) interpreted using modified PIOPED II criteria 98% and 96% excluding non diagnostic results which occurred in 9% (n=13). Twelve out of these 13 non diagnostic results were confirmed to have PTE based on the composite reference standard. SPECT/CT (Q) interpreted using PISAPED criteria had a sensitivity and specificity of 100% and 83% respectively with 0% non diagnostic results. When only segmental (involving >75%) wedge shaped perfusion defects were considered as true defects, the sensitivity and specificity of PISAPED criteria became 98% and 95% with 0% non diagnostic results. Conclusion: Hybrid SPECT/CT (Q) imaging has a high diagnostic accuracy for detecting PTE than planar (Q) scans interpreted with chest radiographs. In addition, low dose CT used in SPECT/CT was able to identify other non-embolic pathologies (eg: pneumonia, emphysematous bullae, tumor or lymphadenopathy) responsible for perfusion defects which could have been missed on chest radiography thereby increasing the specificity. Modified PIOPED II had a higher number of non-diagnostic results compared to PISAPED which increases the latter's clinical utility.