Iterative reconstruction incorporating background correction improves quantification of [18F]-NaF PET/CT images of patients with abdominal aortic aneurysm.

BACKGROUND: A confounding issue in [18F]-NaF PET/CT imaging of abdominal aortic aneurysms (AAA) is the spill in contamination from the bone into the aneurysm. This study investigates and corrects for this spill in contamination using the background correction (BC) technique without the need to manually exclude the part of the AAA region close to the bone. METHODS: Seventy-two (72) datasets of patients with AAA were reconstructed with the standard ordered subset expectation maximization (OSEM) algorithm incorporating point spread function (PSF) modelling. The spill in effect in the aneurysm was investigated using two target regions of interest (ROIs): one covering the entire aneurysm (AAA), and the other covering the aneurysm but excluding the part close to the bone (AAAexc). ROI analysis was performed by comparing the maximum SUV in the target ROI (SUVmax(T)), the corrected cSUVmax (SUVmax(T) - SUVmean(B)) and the target-to-blood ratio (TBR = SUVmax(T)/SUVmean(B)) with respect to the mean SUV in the right atrium region. RESULTS: There is a statistically significant higher [18F]-NaF uptake in the aneurysm than normal aorta and this is not correlated with the aneurysm size. There is also a significant difference in aneurysm uptake for OSEM and OSEM + PSF (but not OSEM + PSF + BC) when quantifying with AAA and AAAexc due to the spill in from the bone. This spill in effect depends on proximity of the aneurysms to the bone as close aneurysms suffer more from spill in than farther ones. CONCLUSION: The background correction (OSEM + PSF + BC) technique provided more robust AAA quantitative assessments regardless of the AAA ROI delineation method, and thus it can be considered as an effective spill in correction method for [18F]-NaF AAA studies.

MR characterization of mild hyperthermia-induced gadodiamide release from thermosensitive liposomes in solid tumors.

OBJECTIVES: Thermal dose in tumor tissue is a key factor for regional hyperthermia (HT) combined with chemotherapy and for drug delivery using thermosensitive liposomes (TSL). It influences therapy outcome, affects the accumulation of liposomes, and triggers the content release from TSL in the target tissue. For the development and clinical application of TSL, noninvasive visualization is of critical importance. For this purpose, TSL loaded with MRI contrast agent (CA) have been developed. With increase in temperature, the CA is released from TSL at the phase transition temperature Tm resulting in a relaxation time change, which allows MRI monitoring. The purpose of this study was to examine the feasibility of an in vivo application and MR characterization of Gd-DTPA-BMA-loaded phosphatidylglyceroglycerol-TSL (Gd-TSL) at mild HT conditions in tumor tissue using a clinically relevant setting. MATERIAL AND METHODS: Gd-TSL were characterized in vitro with varying thermal doses between 37 degrees C and 45 degrees C and distinct solvents by MR at 0.5 T and 1.5 T. In vivo studies were performed in C57BL/6 mice bearing BFS-1 fibrosarcomas at 1.5 T. One tumor-bearing leg was immersed in a temperature-controlled water bath (T). Gd-TSL (Tm = 43.5 +/- 0.2 degrees C) were injected either intratumorally or intravenously at T = 37.3 +/- 0.1 degrees C or T = 42.5 +/- 0.3 degrees C. As a control, nonliposomal Gd-DTPA-BMA was injected intravenously at T = 43.1 +/- 0.3 degrees C. A second tumor on the contralateral limb, which remained unheated, served as a control. CA release was monitored by T1-weighted spin-echo. RESULTS: The in vitro characterization demonstrated at heated and unheated samples a strong increase in T1-relaxivity of Gd-TSL solutions from 0.4 mM-1 s-1 (37.5 degrees C) to 4.2 mM-1 s-1 (43.3 degrees C) at 0.5 T. Thermal dose and solvent affected the rate of relaxation time change significantly. A fast and complete release was observed in samples with serum, whereas Gd-TSL in glucose was only partially released within 1 hour. A dedicated experimental setup was developed for standardized in vivo investigation. Tumor signal intensity changes were detectable in all animals. After intratumoral injection of Gd-TSL, the signal increased heterogeneously (max., +52% +/- 25%) within 3 minutes after temperature increase and decreased strongly thereafter, whereas after i.v. injection, the signal increased homogeneously (+19% +/- 3%) within 2 minutes persisting thereafter. The unheated control tumors on the contralateral legs showed a 10% +/- 3% signal increase within 2 minutes. Injection at 37 degrees C showed a continuous signal increase in "heated" and unheated tumors of up to 8% to 10%. Nonliposomal CA injection demonstrated that tumors were well perfused during HT. CONCLUSION: HT-induced CA release from Gd-TSL was monitored and characterized by MRI after i.v. injection in tumor-bearing mice. Higher temperatures resulted in higher signal changes. Immediately after i.v. injection, heated tumor tissue was distinguishable from unheated tumor tissue. The Gd-TSL appears to be suitable for MR monitoring of HT tumor treatment in a clinical MRI setting independent of field strength.