Frequent Diagnostic Errors in Cardiac PET/CT Due to Misregistration of CT Attenuation and Emission PET Images: A Definitive Analysis of Causes, Consequences, and Corrections

• Published in: The Journal of nuclear medicine (1978) Vol. 48; no. 7; pp. 1112 - 1121
• Main Authors: Gould, K. Lance, Pan, Tinsu, Loghin, Catalin, Johnson, Nils P, Guha, Ashrith, Sdringola, Stefano
• Format: Journal Article
• Language: English
• Published: United States Soc Nuclear Med 01.07.2007; Society of Nuclear Medicine
• Subjects: Coronary Artery Disease - diagnostic imaging; Coronary Circulation; Diagnostic Errors; False Positive Reactions; Heart - diagnostic imaging; Heart failure; Human subjects; Humans; Image Interpretation, Computer-Assisted; Medical imaging; Medical treatment; Patients; Positron-Emission Tomography - methods; Risk factors; Scanners; Software; Tomography, X-Ray Computed - methods
• ISSN: 0161-5505; 1535-5667
• DOI: 10.2967/jnumed.107.039792

Cardiac PET combined with CT is rapidly expanding despite artifactual defects and false-positive results due to misregistration of PET and CT attenuation correction data-the frequency, cause, and correction of which remain undetermined. Two hundred fifty-nine consecutive patients underwent diagnostic rest-dipyridamole myocardial perfusion PET/CT using (82)Rb, a 16-slice PET/CT scanner, helical CT attenuation correction with breathing and also at end-expiratory breath-hold, and averaged cine CT data during breathing. Misregistration on superimposed PET/CT fusion images was objectively measured in millimeters and correlated with associated quantitative size and severity of PET defects. Misregistration artifacts were defined as PET defects with corresponding misregistration on helical CT-PET fusion images that resolved after correct coregistration using a repeat CT scan, cine CT averaged attenuation during normal breathing, or shifted cine CT data that coregistered with PET data. Misregistration of standard helical CT PET images caused artifactual PET defects in 103 of 259 (40%) patients that were moderate to severe in 59 (23%) (P = 0.0000) and quantitatively normalized on cine or shifted cine CT PET (P = 0.0000). Quantitative misregistration was a powerful predictor of artifact size and severity (P = 0.0000), particularly for transaxial misregistration >6 mm occurring in anterior or lateral areas in 76%, in inferior areas in 16%, and at the apex in 8% of 103 artifactual defects. Misregistration of helical CT attenuation and PET emission images causes artifactual defects with false-positive results in 40% of patients that normalize on cine CT PET using averaged CT attenuation data during normal breathing comparable to normal breathing during PET emission scanning and shifting cine CT images to coregister visually with PET.