Respiratory-gated 18F-FDG PET imaging in lung cancer: effects on sensitivity and specificity.

BACKGROUND: Respiratory motion is known to deteriorate positron emission tomography (PET) images and may lead to potential diagnostic errors when a standardized uptake value (SUV) cut-off threshold is used to discriminate between benign and malignant lesions. PURPOSE: To evaluate and compare ungated and respiratory-gated 18F-fluorodeoxyglucose PET/computed tomography (CT) methods for the characterization of pulmonary nodules. MATERIAL AND METHODS: The list-mode acquisition during respiratory-gated PET was combined with a short breath-hold CT scan to form the CT-based images. We studied 48 lesions in 43 patients. PET images were analyzed in terms of the maximum SUV (SUV(max)) and the lesion location. RESULTS: Using receiver-operating characteristic (ROC) curves, the optimal SUV cut-off thresholds for the ungated and CT-based methods were calculated to be 2.0 and 2.2, respectively. The corresponding sensitivity values were 83% and 92%, respectively, with a specificity of 67% for both methods. The two methods gave equivalent performance levels for the upper and middle lobes (sensitivity 93%, specificity 62%). They differed for the lower lobes, where the CT-based method outperformed the ungated method (sensitivity values of 90% and 70%, respectively, and a specificity of 73% with both methods) - especially for lesions smaller than 15 mm. CONCLUSION: The CT-based method increased sensitivity and did not diminish specificity, compared with the ungated method. It was more efficient than the ungated method for imaging the lower lobes and smallest lesions, which are most affected by respiratory motion.

Initial clinical results for breath-hold CT-based processing of respiratory-gated PET acquisitions.

PURPOSE: Respiratory motion causes uptake in positron emission tomography (PET) images of chest structures to spread out and misregister with the CT images. This misregistration can alter the attenuation correction and thus the quantisation of PET images. In this paper, we present the first clinical results for a respiratory-gated PET (RG-PET) processing method based on a single breath-hold CT (BH-CT) acquisition, which seeks to improve diagnostic accuracy via better PET-to-CT co-registration. We refer to this method as "CT-based" RG-PET processing. METHODS: Thirteen lesions were studied. Patients underwent a standard clinical PET protocol and then the CT-based protocol, which consists of a 10-min List Mode RG-PET acquisition, followed by a shallow end-expiration BH-CT. The respective performances of the CT-based and clinical PET methods were evaluated by comparing the distances between the lesions' centroids on PET and CT images. SUV(MAX) and volume variations were also investigated. RESULTS: The CT-based method showed significantly lower (p = 0.027) centroid distances (mean change relative to the clinical method = -49%; range = -100% to 0%). This led to higher SUV(MAX) (mean change = +33%; range = -4% to 69%). Lesion volumes were significantly lower (p = 0.022) in CT-based PET volumes (mean change = -39%: range = -74% to -1%) compared with clinical ones. CONCLUSIONS: A CT-based RG-PET processing method can be implemented in clinical practice with a small increase in radiation exposure. It improves PET-CT co-registration of lung lesions and should lead to more accurate attenuation correction and thus SUV measurement.

Contribution of single photon emission computed tomography to the differential diagnosis of dementia in a memory clinic.

To evaluate the contribution of single photon emission computed tomography (SPECT) to the differential diagnosis of dementia, we studied 48 consecutive patients (median age: 63) with a degenerative or vascular dementia, a 99mTc-HMPAO SPECT imaging, and a diagnostic confirmation (autopsy or genetic mutation). The SPECT scans were visually rated by two nuclear medicine physicians (first blinded to the clinical data, then with the data). Comparisons between clinical diagnoses and/or SPECT imaging and neuropathology were performed. At the time of SPECT was performed, the clinical diagnosis of Alzheimer's disease (AD) sensitivity was 83%, specificity was 76%, and diagnostic accuracy was 79%. The blinded SPECT sensitivity was 57%, specificity 92%, and diagnostic accuracy 75%. The SPECT associated with clinical data sensitivity was 65%, specificity 84%, and accuracy 75%. The clinical diagnosis of frontotemporal-lobar degeneration (FTLD), progressive supranuclear palsy (PSP), and corticobasal degeneration syndrome (CBDs) sensitivity was 83%, specificity 87%, and accuracy 85%. The blinded SPECT sensitivity was 50%, specificity 97%, and accuracy 79%. The SPECT associated with clinical data sensitivity was 61%, specificity was 93%, and accuracy 81%. Whenever the blinded SPECT interpretation agreed with the clinical diagnosis of AD and FTLD/PSP/CBDs, the condition was confirmed by neuropathological assessment in all cases. Compared with clinical diagnosis alone, SPECT imaging improved the specificity of the etiological diagnosis in degenerative dementia, although its sensitivity was not as good as that of clinical diagnosis. For AD and FTLD/PSP/CBDs, agreement between the clinical and SPECT-based diagnoses was always confirmed by neuropathological assessment.

Improved imaging of intrahepatic colorectal metastases with 18F-fluorodeoxyglucose respiratory-gated positron emission tomography.

BACKGROUND: In 18F-fluorodeoxyglucose PET/CT, respiratory motion induces bias in image interpretations (i.e. organ misregistration or lesion omission/underestimation). OBJECTIVE: The present study applied our custom gating method (referred to as 'CT-based gated PET' and that has already been validated by our research group) to PET imaging of the liver and compared its per lesion sensitivity with that of standard clinical (i.e. ungated) PET. METHODS: A total of 13 patients scheduled for liver surgery were referred to our department for PET/CT imaging. Each patient underwent both ungated and CT-based gated PET imaging protocols. Two independent, blinded observers interpreted the two sets of PET images and reached a consensus when necessary. Image interpretations were combined with histological analysis and/or intraoperative ultrasound examination to compute each method's per lesion sensitivity and true positive fraction. Analyses were also performed by considering lesions according to their size (longest axis over 10 mm or over 15 mm). RESULTS: Forty-eight lesions were confirmed by pathology reports. When considering all the uptakes, the ungated and the CT-based gated PET methods had sensitivities of 54.2 and 64.6% (P=0.025) and true positive fractions of 83.9 and 86.1%, respectively. When considering uptakes greater than 10 mm in size, ungated and CT-based gated PET had sensitivities of 74.3 and 88.6% (P=0.025), respectively. For lesions greater than 15 mm in size, the corresponding sensitivities were 85.2 and 100% (P=0.046). CONCLUSION: CT-based gated PET yielded a higher lesion-based sensitivity than routine, ungated PET did. Hence, this method improves the detection of intrahepatic colorectal metastases--especially for lesions that are close in size to the detection limit of the PET gantry.