William H. Howard

Scintigraphic appearance of the piriformis muscle syndrome.

This is the first report in the nuclear medicine literature of the scintigraphic appearance of the piriformis muscle syndrome. This syndrome previously has been thought to be a purely clinical diagnosis and imaging modalities have been ignored. However, its confusing clinical presentation can lead to unnecessary surgical exploration. This case is presented to illustrate the characteristic scintigraphic pattern and suggest the role of nuclear medicine scanning in establishing the diagnosis.

Biliary scintigraphy. The "hot rim" sign.

A new cholescintigraphic finding, the “hot rim” sign, is reported in a case of acute cholecystitis. Local inflammation in the gallbladder fossa may be the cause of this phenomenon.

Optimal technetium-99m RBC labeling for gastrointestinal hemorrhage study.

Four methods of Tc-99m red blood cell labeling were tested for labeling efficiency and kinetics in normal adults. From a perspective of the gastrointestinal hemorrhage study, the in vivo method was least appropriate because of extravascular loss of pertechnetate. The modified in vitro method was disadvantageous because of long labeling intervals in a syringe fixed to the patients forearm (45 minutes). The pure in vitro methods produced the highest labeling efficiency (95% +) and are preferred by the authors.

Fourier phase analysis in non-cyclic dynamic studies.

Single harmonic Fourier amplitude and phase (FA/P) analyses of 129 radionuclide angiographic (RNA) studies were found to reflect accurately the sequence and quantity of perfusion determined in the independently interpreted RNA studies. In addition, the anatomic detailing in peripheral studies (hands, feet, etc.) was considered superior to that obtained in RNA studies.

Dual isotope scanning with gallium-67 citrate and technetium-99m radiopharmaceuticals

The interpretation of Ga-67 scans is limited by the complex anatomic and physiologic distribution of Ga-67 and by the low resolution possible with this isotope. The use of a second isotope Tc-99m, tagged to an appropriate pharmaceutical allows the identification of anatomic landmarks. Computer acquisition of both the Ga-67 and Tc-99m images allows for subtraction of the Tc-99m images from the Ga-67 images and leads to precise localization of abnormal Ga-67 accumulations. Only a small percentage of the counts in the Tc-99m window are due to Ga-67 scatter and do not degrade the Tc-99m images. Cases illustrating the value of the dual-isotope technique are presented.

False aneurysm of ascending aorta.

An unsuspected false aneurysm of the ascending aorta was discovered during a first pass, left-to-right shunt study.

Chest radionuclide angiography in the evaluation of pulmonary masses

A retrospective analysis of 104 chest radionuclude angiography (CRNA) studies establishes the utility of this procedure in the evaluation of pulmonary malignancies and benign masses. Remarkable abnormalities in the perfusion of the lungs were identified in 36 (58%) of 62 cases that were not predictably normal or predictably abnormal form the clinical setting. While 60 of these 62 cases involved malignancy being evaluated for metastases, only 21 (34%) had metastases Identified by radionuclide bone or liver-spleen scintigraphy. In addition to diagnosis of superior vena cava obstruction, the CRNA may have a more frequent application as an adjunct to routine scintigraphic studies directed to the identification of metastatic disease.

LEFT VENTRICULAR VOLUMES FROM SINGLE PHOTON EMISSION COMPUTED TOMOGRAPHY (SPECT)

A prospective study was performed in 25 patients, comparing SPECT left ventricular (LV) end-diastolic (ED) volume (V) to that of contrast ventriculography (ANGIO). SPECT examination utilized 25 mCi Tc-99m RBC labeled in vivo. Sixty images were acquired with a Picker Dyna Camera 5 over 180° from RAO to LPO utilizing a 50 msec ED window. The 30-second images were preprocessed using a 5x5 spatial filter with a cutoff of 0.10 and then reconstructed into transaxial slices by filtered back-projection utilizing a low resolution proprietary filter. The apex and base of the LV were determined by review of additional long-axis and short-axis slice reconstructions. LVEDV was subsequently calculated from count-corrected regions of interest (ROIs) generated about the transaxial LV slices. All patients underwent contrast ventriculography (ANGIO) within 72 hours of the SPECT study. The correlation coefficient between SPECT and ANGIO LVEDV over a range of 87 to 506 cc was 0.97 with a standard error of the estimate (SEE) of 23 cc. We conclude that SPECT allows for noninvasive and accurate determination of LVEDV without the need for blood sample counting or assumptions about LV geometry and chest wall attenuation.