Contribution of routine prone acquisition to quality of reporting for myocardial perfusion imaging

Objectives Attenuation artifacts reduce our ability to evaluate perfusion of effected myocardial segments. The aim of this study was to evaluate the impact of routine prone position image evaluation on stepwise interpretation of myocardial perfusion SPECT images.

Methods We have included 279 consecutive patients that were referred for evaluation of myocardial ischemia. Average age was 57.4±10.8, 185/279 (66%) of the patients were female. All patients had routine ECG-gated supine SPECT imaging along with half-time non-gated prone position SPECT imaging. The images were interpreted by three nuclear medicine physicians in a stepwise fashion, polar maps, supine tomographic images, MIP images, motion frozen gated images and then prone tomographic images. Images were interpreted in five levels of abnormality (zero to four scale).

Results All physicians reported lower number of abnormal studies after evaluating prone images (from MIP to prone, 116/279 to 70/279, 95/279 to 63/279, 47/279 to 24/279; p<0.0001, p=0.0001 and p<0.0001). From MIP to prone, equivocal evaluations (two in the scale) were reclassified in 53/58 (91%), 34/52 (65%) and 21/25 (84%) for three physicians respectively. From probable (one and three in the scale) and equivocal evaluations, 106/182 (58%), 54/137 (39%) and 29/112 (26%) patients were reclassified as definitely normal or definitely abnormal. Of these reclassifications, 73/106 (69%), 39/54 (72%) and 23/29 (79%) of patients were female. In 23/221, 9/227, 3/254 patients the decision was reversed (from abnormal states to normal states or vice versa) after evaluating polar images. There were slight changes in inter-observer agreement weighted kappa values between physicians, (0.386 to 0.456, 0.441 to 0.480 and 0.595 to 0.593).

Conclusions Routine prone imaging changes the final evaluation of myocardial perfusion in a substantial number of patients, along with significant decreases in the number of equivocal evaluations. Whether these changes are concordant with proven ischemia in myocardium needs to be studied