Quantitation of SPECT performance: Report of Task Group 4, Nuclear Medicine Committee

A comprehensive performance testing program is an essential ingredient of high-quality single-photon emission computed tomography (SPECT). Many of the procedures previously published are complicated, time consuming, or require a special testing environment. This Task Group developed a protocol for evaluating SPECT imaging systems that was simple, practical, required minimal test equipment, and could be performed in a few hours using processing software available on all nuclear medicine computers. It was designed to test rotational stability of uniformity and sensitivity, tomographic spatial resolution, uniformity and contrast, and the accuracy of attenuation correction. It can be performed in less than three hours and requires only a Co-57 flood source, a line source, and a tomographic cylindrical phantom. The protocol was used 51 times on 42 different cameras (seven vendors) by four different individuals. The results were used to establish acceptable ranges for the measured parameters. The variation between vendors was relatively small and appeared to reflect slight differences in basic camera performance, collimation, and reconstruction software. Individuals can use the tabulated values to evaluate the performance of individual systems.

Captopril-enhanced 99Tcm-MAG3 renal scintigraphy in subjects with suspected renovascular hypertension

This pilot study was undertaken to generate preliminary data on the accuracy of captopril-enhanced renal scintigraphy with a relatively new radiopharmaceutical, 99Tcm-mercaptoacetyltriglycine (99Tcm-MAG3) for detecting significant renal artery stenosis. Truth data was based either on arteriographic or outcome criteria (blood pressure response to therapy). Twenty-seven subjects with suspected renovascular hypertension were studied with baseline and captopril-enhanced 99Tcm-MAG3 renal scintigraphy and renal arteriography. Scan interpretations were expressed as a probability of a significant renal artery stenosis. Scan interpretations were compared with renal arteriographic results, renal vein renin levels, blood pressure values after renal artery repair, and blood pressure control after 4-26 months of clinical follow-up. Using > or = 50% luminal obstruction on arteriography as the reference standard for renal artery stenosis and a high probability scan representing a positive test, the test sensitivity and specificity were 33 and 97%, respectively (using high or indeterminate probability to represent a positive scan, the test sensitivity and specificity were 67 and 83%, respectively). The negative predictive value of a low probability scan for renal artery stenosis was 80%. However, including a measure of renovascular hypertension (blood pressure response to renal artery repair) as the reference standard, the accuracy of the scan improves, with the negative predictive value of a low probability scan for renovascular hypertension increasing to 97%. Scintigraphic results were also positively correlated with renal vein renin values in a statistically significant fashion (two-tailed Fisher exact test statistic = 6.43, P = 0.0219). Captopril-enhanced 99Tcm-MAG3 renal scintigraphy is a moderately accurate technique for detecting renal artery stenosis. More importantly, our preliminary findings suggest that the scintigraphic technique using 99Tcm-MAG3 appears to predict the blood pressure response to renal artery repair in subjects with suspected renovascular hypertension, thereby separating subjects with haemodynamically insignificant renal artery stenosis from those with renovascular hypertension.

Positron emission tomographic measurement of bone marrow blood flow to the pelvis and lumbar vertebrae in young normal adults

Ten young normal adults had pelvic and lumbar vertebral body bone marrow blood flow examined using [15O]water and positron emission tomography (PET) in a study designed to assess the feasibility and reproducibility of the PET technique for measuring marrow blood flow to various marrow regions. The procedure was well tolerated. Repeated blood flow measurements obtained from two consecutive [15O]water exams on each individual subject were highly reproducible. In addition, there was minimal variation in marrow blood flow from individual to individual and no gender differences were noted. In contrast, mean +/- SD bone marrow blood flows (expressed as milliliters per minute per 100 g) at selected anatomical sites were significantly different and were as follows: lower lumbar vertebral bodies, 17.6 +/- 3.1; most posterior and superior pelvis (conventional site of percutaneous bone marrow biopsy), 14.3 +/- 3.1; and total superior pelvis, 11.1 +/- 2.0. We conclude that PET is a relatively noninvasive, simple, and reproducible technique for measuring bone marrow blood flow. Marrow blood flow is consistent between normal young subjects, but varies significantly between different anatomic regions of the marrow.

The safety and pharmacokinetics in adult subjects of an intravenously administered 99mTc-labeled 17 amino acid peptide (CYT-379)

A phase I study was designed to evaluate the safety and pharmacokinetics of a novel platelet reactive peptide, peptide acetyl-SYGRGDVRGDFKCTCCA-amide (CYT-379), which binds to the fibrinogen receptor of activated platelets and also binds to 99mTc. Eleven subjects with suspected deep venous thrombosis had 0.1, 0.5 or 1.0 mg of the peptide infused intravenously. Pharmacokinetics were determined by assaying blood samples in 6 of the 11 subjects and by urine sampling in 5 of these 6 subjects. Plasma and whole blood time-activity curves demonstrated an initial fast component with half-time clearance of 0.2 +/- 0.01 and 0.2 +/- 0.02 h and a slow component with half-time clearance of 2.8 +/- 0.3 and 2.7 +/- 0.2 h (mean +/- SEM for plasma and whole blood, respectively). Urine clearance was 22.6 +/- 3.3 and 10.8 +/- 1.6 mL/min when normalized to body surface area. The cumulative excretion of 99mTc-CYT-379 in the urine was 16.6 +/- 3.6, 45.6 +/- 16.9 and 45.6 +/- 1.8% of the administered dose over 0-2, 0-12 and 0-24 h after radiopharmaceutical injection, respectively. Images obtained in 11 subjects immediately, at 1-2, and 4-6 h after injection were evaluated for abnormalities and were compared with duplex Doppler ultrasonography. 99mTc-CYT-379 images were positive in only 3 of 7 subjects who had a positive duplex Doppler examination in at least one lower extremity. One subject with negative duplex Doppler had also negative 99mTc-CYT-379 scintigraphy. One subject with negative scintigraphy and two other subjects with positive scintigraphy had no other imaging studies of the deep venous system performed. No adverse reactions were observed during or after the infusion of 99mTc-CYT-379. 99mTc-CYT-379 appears to be a safe radiopharmaceutical and demonstrates rapid clearance from plasma in human subjects.

An emission-based technique for obtaining attenuation correction data for myocardial SPECT studies

In this paper, a technique is described for obtaining the information needed to perform attenuation correction in the thorax entirely from an emission study. This technique is based on the observation that the variation in soft tissue and lung attenuation coefficients is small among individuals. Thus only the outer contours of the body and lungs need be determined for obtaining the attenuation map. The contours are determined by using 99Tcm-macroaggregated albumin (MAA) to locate the lungs and an external source wrapped about the chest to locate the body boundary. Simulation studies were performed to investigate how errors in the presumed tissue attenuation coefficients affect the accuracy of the correction. Body and lung attenuation coefficients were varied from 20% less to 20% more than the coefficients used in the corrections. Over this range, there was less than a 15% alteration in the relative distribution of counts in the left ventricle. To test clinical feasibility, seven patients referred for clinical myocardial perfusion studies were scanned before and after the placement of the body source and the administration of 99Tcm-MAA. Reconstructed images from these studies showed clear demarcation of all body and lung boundaries. The presence of 99Tcm-MAA in the lungs had no significant effect on the clinical interpretation of the single photon emission computed tomographic (SPECT) studies. It is concluded that this technique is feasible for clinical application and that it offers important advantages over other current methods.

Dual isotope brain SPECT imaging for monitoring cognitive activation: physical considerations

The physical considerations of using dual isotope brain SPECT imaging to monitor blood flow changes during cognitive activation studies were investigated. These factors included field uniformity, spatial resolution and crosstalk. Serial dual isotope single photon emission computed tomographic (SPECT) studies of a test tube phantom and an anthropomorphic brain phantom filled with 99Tcm and 123I were made over a 10 h period. The reconstructed counts in the 99Tcm and 123I windows were corrected for crosstalk and were plotted as a function of time. The plotted data from each window decreased over time with a half-life characteristic of each radionuclide. The relative difference between true 123I and 99Tcm region counts has to be of the order of 10% to be statistically significant at the P < 0.05 level.

Dual isotope brain SPECT imaging for monitoring cognitive activation: initial studies in humans

A dual isotope, single photon emission tomography (SPECT) technique using 99Tcm-hexamethylpropyleneamine oxime (HMPAO) and 123I-iodoamphetamine (IMP) was investigated to determine its suitability for assessing regional cerebral blood flow (rCBF) changes resulting from cognitive activation. The similarity of the 123I-IMP and 99Tcm-HMPAO distributions under the same physiological conditions was first investigated by administering the two agents to human subjects (n = 8) either simultaneously or at different times but during the performance of the same task. Normalized ratio images generated from the 99Tcm and 123I data showed that the two tracers distributed similarly in the left and right cerebral hemispheres when administered under similar physiological conditions. There was, however, a significant anterior/posterior gradient that appears to be the result of partial volume effects due to small differences in spatial resolution of the two agents. In two subjects, 99Tcm-HMPAO was administered during a resting period with eyes-closed and 123I-IMP was injected later during visual checkerboard stimulation. Ratio images showed a localized increase in the occipital lobes during the visual stimulation consistent with the expected increase in rCBF. The dual isotope strategy appears promising for study of changes in rCBF due to cognitive activation.

Sensitivity, resolution and image quality with a multi-head SPECT camera

This investigation sought to determine which collimation factors were most important in providing superior image quality with a three-headed SPECT device. The relationship between sensitivity, resolution and SPECT image quality was studied. Two different sets of parallel-hole collimators were used. The ultrahigh-resolution collimators have higher spatial resolution (8.9 versus 11.0 mm), but only 55% of the sensitivity of the high-resolution collimators. A phantom with hot rods was imaged with both collimator sets. Observers compared images with the ultrahigh-resolution collimators to images of varying counts with the high-resolution collimators and determined which high-resolution images matched the ultrahigh-resolution images in image quality. Eleven patient studies were acquired with both collimator sets for equal time, and observers chose which image set they preferred. Transverse images of brain and liver studies were simulated with varying resolution and counts and subjectively compared. The phantom study indicated that the improvement in resolution led to image quality comparable to increasing the number of counts by a factor of 2.5 to 3.4. The clinical studies showed that the ultrahigh-resolution collimators were preferred in a large majority of the cases. These trends were also seen in the simulation study. These results confirm that higher resolution collimators should be used with multihead SPECT devices. The improvement in resolution more than compensates for the loss in sensitivity, leading to an overall improvement in image quality.

Spatial resolution and count density requirements in brain SPECT imaging

A set of simulations has been performed to investigate the spatial resolution and count density requirements for brain SPECT imaging. Projections were drawn from a matrix representation of the Hoffman brain phantom. These projections were convolved with realistic point spread functions and Poisson noise was added to simulate a wide range of imaging situations normalized to a fixed imaging time. The projections were optimally smoothed with a Wiener filter and were reconstructed with a ramp filter. The quality of the reconstructed images was determined objectively from the normalized mean square between the simulated data and the true distribution. This ranking was validated against the preferences of a group of trained observers. The results from this study indicate that the optimal choice of spatial resolution (collimation) depends on the available count density. As the count density (normalized to 10 mm resolution) increases by a factor of 2.7, results from the simulations indicate that the optimal spatial resolution improves by 1 mm. For brain studies in which the administered activity is limited (such as 123I IMP), the optimal spatial resolution is approximately 8 to 9 mm. With 99Tcm labelled brain agents the amount of administered radioactivity can be increased six-fold and the optimal spatial resolution is predicted to fall to about 6 to 7 mm. If sensitivity is further increased by the use of a dedicated SPECT unit with multiple detectors, the optimal spatial resolution will be on the order of 4 to 5 mm.

An improved method for the quantification of left-to-right cardiac shunts

We have investigated a technique for quantifying QP/QS in left-to-right cardiac shunts. In this method, the gamma variate, which is fitted to the first-pass portion of the lung curve, is used to generate a curve, which simulates the response of a normal lung curve with systemic recirculation. The difference between this curve and the observed lung curve is then used to calculate QP/QS. This method was evaluated on a set of simulated lung time-activity curves with precisely known QP/QS values on a group of 11 patients with no clinical suspicion of cardiac shunt and on a group of 30 patients referred for cardiac shunt studies. The QP/QS in each of these studies was determined by three individuals using both the Maltz-Treves method and the new method. This method yielded QP/QS values that were more accurate on the simulated lung data and had less interobserver variation on all the studies than those obtained from the Maltz-Treves method.

Gestational changes in electrolyte transport, electrical activity, and permeability of the porcine placenta

1. The electrolyte transport capacities of the porcine placenta and fetal membranes (amnion, chorion, and allantois) during gestation (47-112 days) were assessed in vitro and in the absence of electrochemical, osmotic, or hydrostatic driving forces. Net transmural transport of 22Na and 36Cl across sections of porcine allantochorion-endometrium (ALCE), amniochorion-endometrium (AMCE), and allantoamnion (AA) mounted in Ussing chambers could not be detected at any stage of gestation. 2. These tissues were characterised throughout gestation by low electrical potential difference (PD; less than 5 mV, fetal side positive with respect to maternal side) and short circuit current (SCC; less than 16 microA.cm-2) which were amiloride and iodoacetate sensitive. 3. A periparturitent reversal in the polarity of PD to fetal side negative and a change in the direction of the SCC was observed in tissues obtained from sows (n = 3) at 111-112 day gestation. These changes were accompanied by a 73% increase in electrical resistance (R) and a 2-10-fold decrease in the permeability of tissues to Na and Cl. 4. Morphological studies revealed a decrease in the thickness of allantoamnion and a decrease in the height of epithelia from ALCE, AMCE, and AA during gestation, indicating a decrease in activity. These morphological changes only partly explained the pre-term electrophysiological changes in these tissues.

A method for obtaining an approximate Wiener filter

The Wiener restoration filter yields the minimum mean-square error between the restored image and the true object function. However, it has found limited use because, in its usual formulation, it requires information about the object power spectrum which is generally unknown. In this paper, it is shown that the Wiener filter can be derived from the noise-free image power spectrum, and a method is presented for estimating this from the observed data. From this estimate an approximate Wiener filter was calculated. The method was tested on three sets of simulated data which included a constant background, rectangular defects, and Gaussian defects at varying contrast and noise levels. The performance of the approximate Wiener filter was compared both to the true Wiener filter and to the standard 1-2-1 three-point smooth. The results confirmed that the approximate Wiener filter adapted to the information content of the observed data and closely matched the performance of the true Wiener filter. The approximate Wiener filter outperformed the three-point smooth in all cases, especially at low contrast and high noise levels. The approximate Wiener filter can be calculated without operator intervention and requires little additional computation time over conventional Wiener filter techniques.

A 3-D anthropomorphic brain phantom for ECT applications

A three-dimensional (3-D) anthropomorphic brain phantom was developed for ECT (emission computed tomography) applications. This phantom is made up of a stack of interlaced plates as an insert in a water tank containing radioactive solution. The thickness of the plates in the stack is chosen to create the desired spatial frequency content and object contrast between simulated grey and white matter regions for tracers commonly used to study regional blood flow or metabolism. The transverse, sagittal, and coronal SPECT images acquired with a rotating camera system show a realistic simulation of the grey matter distribution in a human brain.

Radioiodinated fatty acid esters in the management of hepatocellular carcinoma: preliminary findings

Radioiodinated fatty acid esters, such as lipiodol or ethiodol, are localized in the hypervascular hepatocellular carcinoma (HCC) for a long time following intra-arterial hepatic injection, enabling delivery of high internal radiation to the tumor. The desired radiation can easily be delivered to small HCC, less than 5 cm in diameter, in single or multiple procedures with an 8-week interval. For larger tumors, [131I]lipiodol or [131I]ethiodol in conjunction with chemotherapy emulsion, Ivalon embolization or all three combinations should be considered for maximal clinical results. A strong beta emitter with shorter physical half-life, i.e. 90Y will be more effective in the management of HCC if one can label lipiodol with 90Y.

Dosimetry of iodine-131 ethiodol in the treatment of hepatoma

The in vivo distribution and kinetics of [131I]Ethiodol injected through the hepatic artery have been measured on a group of four patients with hepatocellular carcinoma. The [131I]Ethiodol was distributed predominantly in the liver (70-90%) and lungs (10-20%) and was selectively concentrated and retained in the patients with massive and multinodular hepatomas with approximately 10% of the administered activity localizing in tumor. The radioactivity in the blood 2 hr postinjection was less than 0.1% and was never higher than 0.9% of the administered activity. The radioactivity cleared from normal liver tissue with an effective half-life of approximately 4 days while the clearance time from the tumor was 20-25% longer. Activity in the lungs initially increased and then cleared with a 5-day effective half-life. Based on these measurements, the estimated dose per mCi of [131I]Ethiodol administered is 31 rad to the liver, 22 rad to the lungs, 1.9 rad to the total body and 239 rad to a 4-cm diameter tumor. These results suggest that [131I]Ethiodol has the potential to deliver curative radiation doses to hepatomas with acceptable radiation burdens to normal tissues.

Iofetamine HCI I-123 brain scanning in stroke: a comparison with transmission CT

Although IMP scans fail to show fine anatomical details of the brain, because of poor resolution of a single head rotational system, adequate information is offered by the scans to localize most perfusion defects caused by stroke. The following conclusions can be drawn from our study: 1. The planar IMP brain scans processed through the computer are sensitive in the early diagnosis of acute stroke except for small and deeply localized lesions. 2. The SPECT IMP imaging is more sensitive than the planar or transmission CT scans in the early diagnosis of stroke. Semiquantitative evaluations are feasible with IMP SPECT. 3. Neither transmission CT nor IMP SPECT are sensitive in the detection of acute lacunar infarcts. 4. In acute infarction, the transmission CT is usually negative or minimally positive in the early stages, while impaired uptake of IMP occurs immediately after the onset of the stroke. In acute stroke, the extent of the perfusion defect on IMP is usually greater than the abnormality seen on the transmission CT. 5. On followup studies, IMP scans show improved perfusion reflecting physiologic changes, while transmission CT scans show further dense anatomical changes when compared to the initial studies. 6. Hyperemic changes are likely due to collateral circulation or luxury perfusion. This finding suggests that the IMP reflects local cerebral blood flow in strokes.

A precise method for correcting count-rate losses in scintillation cameras

Quantitative studies performed with scintillation detectors often require corrections for lost data because of the finite resolving time of the detector. Methods that monitor losses by means of a reference source or pulser have unacceptably large statistical fluctuations associated with their correction factors. Analytic methods that model the detector as a paralyzable system require an accurate estimate of the system resolving time. Because the apparent resolving time depends on many variables, including the window setting, source distribution, and the amount of scattering material, significant errors can be introduced by relying on a resolving time obtained from phantom measurements. These problems can be overcome by curve-fitting the data from a reference source to a paralyzable model in which the true total count rate in the selected window is estimated from the observed total rate. The resolving time becomes a free parameter in this method which is optimized to provide the best fit to the observed reference data. The fitted curve has the inherent accuracy of the reference source method with the precision associated with the observed total image count rate. Correction factors can be simply calculated from the ratio of the true reference source rate and the fitted curve. As a result, the statistical uncertainty of the data corrected by this method is not significantly increased.

Septal infarction demonstrated on technetium-99m PYP SPECT

The case of a 62-year-old man with an acute myocardial infarction detected by planar Tc-99m PYP imaging is presented. The use of SPECT imaging provided more information with regard to infarct localization by demonstrating uptake by the septum, a finding not apparent on the conventional planar images.

Technetium-99m pyrophosphate scintigraphy for the detection of acute myocardial infarction. How useful is it?

To evaluate the contribution of Tc-99m pyrophosphate scintigraphy (TPS) on the overall management of patients suspected of having acute myocardial infarction (AMI), hospital records of 58 consecutive patients who underwent TPS, were evaluated in depth. The results indicate that TPS was essential for the diagnosis of AMI in 16% of the patients. TPS was most rewarding in perioperative patients and in patients with borderline or uninterpretable electrocardiographic and enzyme changes. Also, in some cases, TPS was able to confirm or exclude the diagnosis of AMI prior to the confirmation by serial electrocardiograms (ECG) and serial enzyme changes. TPS was less rewarding in patients with clinically low index of suspicion for AMI. It may also be confusing in patients with high clinical likelihood of AMI and a history of prior myocardial infarction because of the possibility of persistently positive TPS in some of these patients. Considering the limitations of ECGs, the cardiac enzymes, and atypical clinical presentations in the patient population we evaluated, TPS appears to be fairly accurate when the scintigraphic findings are compared with the final diagnosis at the time of discharge from the hospital.

Enhancement of SPECT images by Fourier filtering the projection image set

Tomographic images from rotating gamma camera systems are often difficult to interpret because of poor contrast and high noise levels. A method is presented for improving the quality of these images by Fourier filtering the projection image set prior to reconstruction. A two-dimensional circularly symmetric Gaussian function is used as the spatial frequency filter. This filter can be optimized to enhance contrast and suppress noise in the projection image set in a straightforward and simple manner from the power spectra of representative projections. Preprocessing of the projections makes it possible to use a ramp reconstruction filter. The resulting tomographic sections show a dramatic improvement in image quality.

Neutrophil labeling: problems and pitfalls

The use of neutrophils labeled with gamma-emitting radionuclides has been shown to be acceptable for in vivo kinetic studies as well as for imaging inflammatory foci. Among the gamma-emitting radionuclides, indium-111 appears to be the agent of choice. Labeling neutrophils with 111In, however, is a relatively new technique. Although simple to perform, it involves several stages, none of which could be carried out without problems. These are discussed and the current research aimed at eliminating the problems is outlined. The knowledge of specific chemotactic receptors and surface antigens has stimulated investigations into selective neutrophil labeling that will continue to be challenging and exciting.