Curtis Caldwell

The impact of 18 FDG-PET on target and critical organs in CT-based treatment planning of patients with poorly defined non-small-cell lung carcinoma: a prospective study ☆

PURPOSE To prospectively study the impact of coregistering (18)F-fluoro-deoxy-2-glucose hybrid positron emission tomographic (FDG-PET) images with CT images on the planning target volume (PTV), target coverage, and critical organ dose in radiation therapy planning of non-small-cell lung carcinoma. METHODS AND MATERIALS Thirty patients with poorly defined tumors on CT, referred for radical radiation therapy, underwent both FDG-PET and CT simulation procedures on the same day, in radiation treatment position. Image sets were coregistered using external fiducial markers. Three radiation oncologists independently defined the gross tumor volumes, using first CT data alone and then coregistered CT and FDG-PET data. Standard margins were applied to each gross tumor volume to generate a PTV, and standardized treatment plans were designed and calculated for each PTV. Dose-volume histograms were used to evaluate the relative effect of FDG information on target coverage and on normal tissue dose. RESULTS In 7 of 30 (23%) cases, FDG-PET information changed management strategy from radical to palliative. In 5 of the remaining 23 (22%) cases, new FDG-avid nodes were found within 5 cm of the primary tumor and were included in the PTV. The PTV defined using coregistered CT and FDG-PET would have been poorly covered by the CT-based treatment plan in 17--29% of cases, depending on the physician, implying a geographic miss had only CT information been available. The effect of FDG-PET on target definition varied with the physician, leading to a reduction in PTV in 24-70% of cases and an increase in 30-76% of cases. The relative change in PTV ranged from 0.40 to 1.86. On average, FDG-PET information led to a reduction in spinal cord dose but not in total lung dose, although large differences in dose to the lung were seen for a few individuals. CONCLUSION The coregistration of planning CT and FDG-PET images made significant alterations to patient management and to the PTV. Ultimately, changes to the PTV resulted in changes to the radiation treatment plans for the majority of cases. Where possible, we would recommend that FDG-PET data be integrated into treatment planning of non-small-cell lung carcinoma, particularly for three-dimensional conformal techniques.

Beyond the hippocampus: MRI volumetry confirms widespread limbic atrophy in AD

Objective: To examine volumetric changes in limbic structures in patients with probable AD using planimetric measures on MRI. Method: Limbic structures (i.e., hippocampus, amygdala, anterior thalamus, hypothalamus, mamillary bodies, basal forebrain, septal area, fornix, and cingulate, orbitofrontal, and parahippocampal cortices) were traced on 3D T1-weighted MR images of 40 patients with mild to moderate AD and 40 age-, sex-, and education-matched normal control subjects. Limbic volumes were compared between groups and the predictive ability was assessed. Results: Overall, limbic structures showed significant atrophy in AD patients compared with normal control subjects. Differences (p < 0.05) were found in all limbic regions except the anterior cingulate cortex. The greatest percentage volumetric losses occurred in the septal area (34%), hippocampus (28%), amygdala (21%), parahippocampal cortex (21%), and posterior cingulate cortex (20%). Combining volumetric measures of amygdala and septal area distinguished patients with AD from normal control subjects with 93% accuracy. Conclusions: These results verify that system-wide limbic degeneration occurs in patients with AD. In addition, atrophy in selected limbic structures was used to distinguish patients with AD from normal elderly individuals with over 90% accuracy in this select clinical sample. The measures require further exploration in samples more representative of those seen by primary care physicians before their utility can be accurately assessed.

Clinical investigation: lungObserver variation in contouring gross tumor volume in patients with poorly defined non-small-cell lung tumors on CT: the impact of 18FDG-hybrid PET fusion☆

PURPOSE To quantify interobserver variation in gross tumor volume (GTV) localization using CT images for patients with non-small-cell lung carcinoma and poorly defined tumors on CT and to determine whether variability would be reduced if coregistered 2-[18F]fluoro-2-deoxy-d-glucose (FDG)-hybrid positron emission tomography (PET) with CT images were used. METHODS AND MATERIALS Prospectively, 30 patients with non-small-cell lung carcinoma had CT and FDG-hybrid PET examinations in radiation treatment position on the same day. Images were coregistered using eight fiducial markers. Guidelines were established for contouring GTVs. Three radiation oncologists performed localization independently. The coefficient of variation was used to assess interobserver variability. RESULTS The size of the GTV defined showed great variation among observers. The mean ratios of largest to smallest GTV were 2.31 and 1.56 for CT only and for CT/FDG coregistered data, respectively. The addition of PET reduced this ratio in 23 of 30 cases and increased it in 7. The mean coefficient of variation for GTV based on the combined modalities was significantly smaller (p < 0.01) than that for CT data only. CONCLUSIONS High observer variability in CT-based definition of the GTV can occur. A more consistent definition of the GTV can often be obtained if coregistered FDG-hybrid PET images are used.

Brain-behavior correlations in hemispatial neglect using CT and SPECT The Sunnybrook Stroke Study

Objective: Structural and functional lesion localization in patients with hemispatial neglect. Design: Location and severity of brain damage on CT and SPECT correlated with neglect performance as assessed with a battery of drawings, line bisection, and line and shape cancellation subtests. Patients: Participants included 120 consecutive stroke patients with a single right-hemisphere-damaged lesion on CT who were admitted to the Acute Stroke Care Unit at Sunnybrook Health Science Centre. Of these, 88 also had a SPECT. Results: On CT, 82 patients with neglect (compared with 38 without neglect) had more extensive damage in the parietal and sensorimotor cortex and white matter fiber bundles, including the posterior-superior longitudinal and inferior-frontal fasciculi (p < 0.05). Parietal and anterior cingulate damage best predicted neglect score using the CT data (p< 0.05), and regional blood flow in the parietal cortex best predicted neglect score using the SPECT data (p < 0.05) after controlling for the influence of age and lesion size on multiple linear regression. Conclusions: Damage in the parietal and anterior cingulate cortex and posterior white matter fiber bundles correlated with hemispatial neglect. Combining structural- and functional-imaging techniques with neurobehavioral analysis can elucidate brain-behavior relationships.

Coregistered FDG PET/CT-Based Textural Characterization of Head and Neck Cancer for Radiation Treatment Planning

Coregistered fluoro-deoxy-glucose (FDG) positron emission tomography/computed tomography (PET/CT) has shown potential to improve the accuracy of radiation targeting of head and neck cancer (HNC) when compared to the use of CT simulation alone. The objective of this study was to identify textural features useful in distinguishing tumor from normal tissue in head and neck via quantitative texture analysis of coregistered 18 F-FDG PET and CT images. Abnormal and typical normal tissues were manually segmented from PET/CT images of 20 patients with HNC and 20 patients with lung cancer. Texture features including some derived from spatial grey-level dependence matrices (SGLDM) and neighborhood gray-tone-difference matrices (NGTDM) were selected for characterization of these segmented regions of interest (ROIs). Both K nearest neighbors (KNNs) and decision tree (DT)-based KNN classifiers were employed to discriminate images of abnormal and normal tissues. The area under the curve (AZ) of receiver operating characteristics (ROC) was used to evaluate the discrimination performance of features in comparison to an expert observer. The leave-one-out and bootstrap techniques were used to validate the results. The AZ of DT-based KNN classifier was 0.95. Sensitivity and specificity for normal and abnormal tissue classification were 89% and 99%, respectively. In summary, NGTDM features such as PET coarseness, PET contrast, and CT coarseness extracted from FDG PET/CT images provided good discrimination performance. The clinical use of such features may lead to improvement in the accuracy of radiation targeting of HNC.

Predictors of Hemorrhagic Transformation Occurring Spontaneously and on Anticoagulants in Patients With Acute Ischemic Stroke

BACKGROUND AND PURPOSE Hemorrhagic transformation (HT) is a common evolution of large-volume ischemic lesions, particularly of cardioembolic origin. We used transcranial Doppler ultrasound (TCD), single-photon emission computed tomography (SPECT) with 99mTc-hexamethylpropyleneamine oxime (HMPAO), and the Toronto Embolic Scale (TES) to decide (1) whether TCD, HMPAO-SPECT, and TES can improve on clinical and CT tests to predict spontaneous HT and (2) whether SPECT can help to predict the outcome of symptomatic HT. METHODS Prognostic criteria included Canadian Neurological Scale (CNS) scores < or = 50 on admission, early ischemic changes on CT, M1 middle cerebral artery occlusion on TCD, the focal absence of brain perfusion on SPECT, and a high risk of cardiogenic embolism on TES. RESULTS In part 1, 85 consecutive patients admitted within the first 6 hours were studied. No patient received thrombolysis. HT was found in 11 patients (13%) at 3 to 5 days. Admission CNS and CT were not predictive of HT: odds ratios (95% confidence intervals) were 0.49 (0.18 to 1.23) (P = .1) and 0.88 (0.23 to 3.45) (P = .8), respectively, TCD, SPECT, and TES were significant predictors of HT (P < .05), as follows: TCD, 8.67 (1.42 to 70.59); SPECT, 17.40 (2.69 to 170.89); and TES, 18.13 (2.6 to 406.86). In part 2, 490 consecutive patients were studied and 21 (4%) had symptomatic HT, of which 12 had focal hypoperfusion on SPECT at 4 days after stroke onset and 9 had focal hyperperfusion. Patients with hypoperfusion had larger CT lesions (115 +/- 97 versus 42 +/- 29 cm3; P = .04) and poorer outcome at 2 weeks (CNS, 38 +/- 45 versus 96 +/- 10; P = .001), including death (6/12 versus 0/9; P = .04); compared with those with hyperperfusion on SPECT. CONCLUSIONS High risk of cardioembolism, M1 middle cerebral artery occlusion, and absence of collateral flow evaluated by TES, TCD, and SPECT help to identify patients at risk for spontaneous HT. Although TES was the most powerful predictor of HT, SPECT is the best single adjunct to the triage of clinical and CT tests. Patients with brain hyperperfusion on HMPAO-SPECT after symptomatic HT have better chances for recovery.

Simple Visual Analysis of Brain Perfusion on HMPAO SPECT Predicts Early Outcome in Acute Stroke

BACKGROUND AND PURPOSE Single-photon emission computed tomography (SPECT) is used in patients with acute stroke but as yet is of controversial value. We investigated an association of brain perfusion changes in stroke patients with stroke severity, volume of brain damage, and recovery. METHODS Consecutive patients with hemispheric stroke were studied prospectively with serial neurological examinations using the Canadian Neurological Scale (CNS), CT. and 99mTc-hexamethylpropyleneamine oxime (HMPAO) SPECT. Visual SPECT patterns of brain perfusion (normal, high, mixed, low, and absent) were correlated with the severity of stroke, lesion volume, and short-term outcome. RESULTS SPECT studies were performed in a total of 458 consecutive acute stroke patients within 2 weeks after the onset (mean time, 5 days; range, 1 to 12 days). SPECT perfusion patterns correlated with stroke severity (CNS score) during the first 2 weeks (P < .001). Focal absence of brain perfusion on SPECT was associated with the largest volume of brain damage: 104 +/- 84 mL (P < .0001). SPECT perfusion patterns predicted the shortterm outcome: 97% of patients with normal and increased HMPAO uptake made good recovery, 52% of those with decreased perfusion had moderate stroke, and 62% of patients with absent patterns fared badly. In a multiple logistic regression model, admission CNS scores had the strongest predictive value (P = .0001). SPECT had its own prognostic value independent of clinical judgment (P = .03). SPECT statistically improved predictive power of the CNS score (+1% receiver operating characteristic curve area, [X2]2 = 20, P < .001) because of distinction between focal decrease or absence of brain perfusion in patients studied within the first 72 hours of stroke. CONCLUSIONS Visual brain perfusion patterns correlate with the extent, severity, and short-term outcome of hemispheric stroke. HMPAO SPECT may improve the prognostic value of clinical examination if performed during the first 72 hours of stroke.

A geometrically accurate vascular phantom for comparative studies of x-ray, ultrasound, and magnetic resonance vascular imaging: construction and geometrical verification.

A technique for producing accurate models of vascular segments for use in experiments that assess vessel geometry and flow has been developed and evaluated. The models are compatible with x-ray, ultrasound, and magnetic resonance (MR) imaging systems. In this paper, a model of the human carotid artery bifurcation, is evaluated that has been built using this technique. The phantom consists of a thin-walled polyester-resin replica of the bifurcation through which a blood-mimicking fluid may be circulated. The phantom is surrounded by an agar tissue-mimicking material and a series of fiducial markers. The blood- and tissue-mimicking materials have x-ray, ultrasound, and MR properties similar to blood and tissue; fiducial markers provide a means of aligning images acquired by different modalities. The root-mean-square difference between the inner wall geometry of the constructed model and the desired dimensions was 0.33 mm. Static images were successfully acquired using x-ray, ultrasound, and MR imaging systems, and are free of significant artifacts. Flow images acquired with ultrasound and MR agree qualitatively with each other, and with previously published flow patterns. Volume-flow measurements obtained with ultrasound and MR were within 4.4% of the actual values.

Limbic system perfusion in Alzheimer's disease measured by MRI-coregistered HMPAO SPET

Abstract The goal of this study was to perform a systematic, semi-quantitative analysis of limbic perfusion in patients with Alzheimers disease (AD) using coregistered single-photon emission tomography (SPET) images aligned to magnetic resonance (MR) images. Limbic perfusion in 40 patients with mild to moderate AD was compared with that of 17 age-, sex-, and education-matched normal controls (NC). HMPAO SPET scans and 3D T1-weighted MR images were acquired for each subject. Structures of the limbic system (i.e. hippocampus, amygdala, anterior thalamus, hypothalamus, mamillary bodies, basal forebrain, septal area and cingulate, orbitofrontal and parahippocampal cortices) were traced on the MR images and transferred to the coregistered SPET scans. Perfusion ratios for all limbic regions were calculated relative to cerebellar perfusion. General linear model multivariate analysis revealed that, overall, limbic structures showed significant hypoperfusion (F=7.802, P<0.00001, η2=0.695) in AD patients compared with NC. Greatest differences (d≥0.8) were found in the hippocampus, as well as all areas of the cingulate cortex. Significant relative hypoperfusion was also apparent in the parahippocampal cortex, amygdala/entorhinal cortex, septal area and anterior thalamus, all of which showed medium to large effect sizes (d=0.6–0.8). No significant relative perfusion differences were detected in the basal forebrain, hypothalamus, mamillary bodies or orbitofrontal cortex. Logistic regression indicated that posterior cingulate cortex perfusion was able to discriminate AD patients from NC with 93% accuracy (95% sensitivity, 88% specificity). The current results suggest that most, but not all, limbic structures show significant relative hypoperfusion in AD. These findings validate previous post-mortem studies and could be useful in improving diagnostic accuracy, monitoring disease progression and evaluating potential treatment strategies in AD.

Fractal dimension as a measure of altered trabecular bone in experimental inflammatory arthritis.

Our previous studies in experimental inflammatory arthritis (EIA) and in human rheumatoid arthritis demonstrated rapid remodeling with a 5‐fold increase in bone resorption and bone formation. Normal condylar trabecular bone is typically anisotropic, with its orientation along lines of stress; rapid remodeling in a pathological state could disturb the usual order of trabeculae. This study assessed change in the structure of trabecular bone of the distal femoral epiphysis after induction of EIA, using a measure of “fractal dimension,” which may be considered a quantitative description of the degree of irregularity of complex surfaces. Data was obtained from specimens in which EIA had been induced in the rabbit knee by 10 injections of carrageenan over 49 days. Photographic enlargements of embedded undecalcified cross‐sections of the distal femur were digitized, and software written on a Sun workstation was used to define repeatable regions of interest (ROIs) in the images. The ROIs were subjected to fractal analysis by a power law method. The fractal dimension of the trabecular bone pattern within the ROI was estimated by fitting an equation of the form A(ϵ) = λϵ(2–D) to the data. In this equation, A(ϵ) is the area of the “surface” formed by modeling the ROI data as a three‐dimensional structure with the grey‐level magnitude providing the third dimension, λ is a scaling constant, ϵ is the size of the measuring “tool” used to measure the area, and D is the fractal dimension. A Mann–Whitney U‐test applied to the average of the data from all ROIs showed that the two distributions of fractal dimension were significantly different (p < 0.005). There were only two overlaps between data points for arthritis (with these values higher) and normal groups (n = 11 for each group). Since Howships lacunae were too small to be resolved in the system utilized, we consider the difference in fractal dimension to be primarily related to trabecular surface orientation, rather than to the increased number of asperities (resorptive foci) occurring due to increased turnover in bone affected by inflammatory arthritis. The results suggest that fractal dimension may be a useful tool for assessing the degree of structural damage to trabeculae in conditions similar to EIA.