Pulmonary SPECT imaging and the stripe sign

CONCLUSION Submandibular scintigraphy revealed excellent diagnostic ability to predict the prognosis in patients with acute peripheral facial nerve paralysis in its early symptomatic period. Conse quently, submandibular scintigraphy should be added as one of the routine tests to predict the prognosis of peripheral facial nerve paralysis and, especially, to determine of the indications for surgical therapy. 10. Fisch U, Esslen F.. Total intratemporal exposure of the facial nerve. Arch Otolaryngol 1972:95:335-341. 11. Magielski JE. Blatt IM. Submaxillary salivary flow: a test of chorda tympani nerve function as an aid in diagnosis and prognosis of facial nerve paralysis. Laryngoscope 1958:68:1770-1789. 12. Mizukoshi K. Watanabe Y, Aso S. Asai M. Prognostic value of blink test in patients with facial paralysis. Acta Otolaryngol (Stockh) 1988;446(suppl|:70-75. 13. May M, Klein SR, Taylor FH. Idiopathic (Bell's) facial palsy: natural history defies steroid or surgical treatment. Laryngoscope 1985:95:406-409. 14. Mishkin FS. Radionuclide salivary gland imaging. Semin NucÃ-Med 1981:11:258-265. 15. Rosen G, Vered IY, Fedchteyn SC. Submandibular salivary gland scan: a prognostic indicator of Bell's palsy. J Laryngol Olol 1980:94:1021-1024. REFERENCES 16. Yamashita T, Ino C, Tomoda K, Kumazawa T. Prognostic determination submandibular function in Bell's palsy. Arch Otolaryngol 1989:111:244-248. 1. Peiterson E. The natural history of Bell's palsy. Am J Olol 1982;4:107-111. 2. Devriese PP, Schumacher T, Scheide A, DeJongh RH, Houtkooper JM. Incidence, prognosis and recovery of Bell's palsy; a survey of about 1000 patients. Clin Otolaryngol 1990; 15:15-27. 3. Robillard RB, Hilsinger RL, Adour KK. Ramsay Hunt facial paralysis: clinical analysis of 185 patients. Otolaryngol Head Neck Surg 1986;95:292-297. 4. Heathfield KWG, Mee AS. Prognosis of the Ramsay Hunt syndrome. Br Med J 1978; 1:343-344. 5. Fisch U. Maximal nerve excitability testing versus electroneurography. Arch Otolar yngol 1980; 106:352-357. 6. May M, Hardin WB, Sullivan J, Wette R. Natural history of Bell's palsy: the salivary flow test and other prognostic indicators. Laryngoscope 1976:86:704-712. 7. Sinba PK, Keith RW, Pensak ML. Predictability of recovery from Bell's palsy using evoked electromyography. Am J Otol 1994;I5:769-771. 8. Fisch U. Surgery for Bell's palsy. Arch Otolaryngol 1981:107:1-11. 9. Tojtma H, Aoyagi M, Inamura H, Koike Y. Clinical advantages of electroneurography in patients with Bell's palsy within two weeks after onset. Acta Otolaryngol (Stockh) I994;511(suppl):147-149. and 17. May M. Facial paralysis, peripheral type: a proposed method of reporting. Laryngo scope 1970:80:331-390. 18. May M. Blumenthal F, Taylor F. Bell's palsy: surgery based upon prognostic indicators and results. Laryngoscope 1981:91:2092-2103. 19. Hughes GB. Practical management of Bell's palsy. Otolaryngol Head Neck Surg 1990:102:658-663. 20. Stankiewicz JA. Steroid and idiopathic facial nerve paralysis. Olol Head Neck Surg 1983:91:672-677. 21. Stafford FW, Welch AR. The use of acyclovir in Ramsay Hunt syndrome. J Laryngol Otol 1986:100:337-340. 22. Inamura H. Aoyagi M. Tojima H. Koike Y. Effects of acyclovir in Ramsay Hunt syndrome. Ada Otolaryngol (Siockh) 1988;446(suppl):l 11-113. 23. Adour KK, Ruboyianes JM. Von Doerstein PG, et al. Bell's palsy treatment with acyclovir and prednisone compared with prednisone alone: a double-blind, random ized, controlled trial. Ann Otol Rhinol Laryngol 1996:105:371-378. 24. Aoyagi M. Koike Y, Ichige A. Results of facial nerve decompression. Ada Otolar yngol (Siockh) 1988;466(suppl):IOI -105. Pulmonary SPECT Imaging and the Stripe Sign William M. Pace and Michael L. Goris Department of Nuclear Medicine, Stanford University, Stanford, California A patient with high clinical suspicion for pulmonary embolism underwent a diagnostic scintigraphic ventilation/perfusion scan. The planar images revealed an unmatched perfusion defect with a stripe sign in the right middle lobe. A stripe sign is the appearance of normally perfused tissue between the defect and the pleural surface suggesting a nonpleural-based abnormality. SPECT images ac quired in the same study period, however, failed to demonstrate normally perfused tissue between the defect and the pleural sur face. Previous studies have compared planar ventilation/perfusion studies with stripe sign perfusion defects to pulmonary angiography. The results suggest that stripe sign perfusion defects are generally not due to emboli. However, planar imaging is projectional and may miss pleural contact in some perfusion lesions depending on the projection. In the absence of SPECT data, the significance of the stripe sign may need to be reassessed. Key Words: pulmonary embolism; pulmonary angiography; venti lation/perfusion scan J NucÃ-Med 1998; 39:721-723 A. stripe sign in a ventilation/perfusion (V/Q) scan is the appearance of normally perfused tissue between a lesion and the pleural surface. The presence of a stripe sign may imply that there is a region of normally perfused tissue distal to the defect. In contradistinction, embolie lesions are believed to be pleural based, with the defect contacting the pleural surface. Nonpleu ral-based lesions with a stripe sign are thought to be less likely Received Mar. 12, 1997; revision accepted Jul. 17, 1997. For correspondence or reprints contact: Michael L Gorls, MD, PhD, 300 Pasteur Dr., Stanford, CA 94305-5465. due to pulmonary embolism (PE). We report a case of a young woman with a clinical history suggestive of PE. A scintigraphic V/Q study was ordered to evaluate segmental ventilation and perfusion. Planar images revealed a large perfusion lesion with a stripe sign in the right middle lobe on the right posterior oblique (RPO) projection. This defect was not matched with a ventilation abnormality. Using slmKr as the ventilation agent, our laboratory performed dual-isotope SPECT acquisitions in all patients who could safely undergo the procedure. This technique eliminates the positioning discrepancies created when the perfusion and ventilation images are acquired separately. Images were visualized in the transaxial, sagital and coronal planes. The perfusion defect and its ventilation mismatch were well visualized, but a rim of normally perfused tissue could not be identified between the lesion and the pleural surface. This suggests that the limited number of views acquired in planar imaging may fail to locate the point where a perfusion defect contacts the pleural surface. Also, normally perfused tissue adjacent to the lesion, or in the contralateral lung, could mimic a stripe between the defect and the pleural surface. This could be incorrectly interpreted as a less-suspicious, nonpleural-based lesion, and the patient may not get proper treatment. The definition of a stripe sign may need to be reassessed to include the caveat that the stripe must be seen in all of the available perspectives to be labeled a nonpleural-based abnormality. Under these conditions, SPECT imaging should be considered because it is less likely to misrepresent pleural contact in perfusion defects than traditional planar imaging. CASEREVIEWFORPLANARANDSPECT PULMONARY IMAGING• Pace and Goris 721 V/Q scan with perfusion defect and stripe sign VENT'LATIONSPECT P ERFUSION SPEC T RFO RPO Perfusion RPO Ventilation CORONAL FIGURE 1. Planar perfusion and ventilation images in the RPO projection. A middle lobe perfusion defect is seen with a stripe sign. A subtle abnormality is seen on the ventilation image. CASE REPORT Patients A 42-yr-old woman with a 7-day history of left arm swelling and pleuritic chest pain was presented to the emergency room. The patient had a history of endometriosis for which she had taken oral contraceptives intermittently for 20 yr. She was given a combina tion pill I mo before her evaluation. Physical exam revealed a swollen and tender left upper arm with a circumference I in. greater than the right upper arm. Pulse and blood pressure were normal in both arms and oxygen saturation was 96% on room air. The patient denied any cough or hemoptysis. A ventilation/perfusion (V/Q) scan was ordered for suspicion of pulmonary embolism. The perfusion study was performed using WmTc-MAA and was acquired simultaneously with a 8lrnKr gas ventilation study. The major abnormality seen in the planar images was a right middle lobe perfusion defect with a stripe sign (Fig. I). Only a subtle defect was seen on the ventilation images in the same region and the impression was a mismatched perfusion lesion in the right middle lobe with a stripe sign. The SPECT images failed to show normally perfused tissue between the right middle lobe defect and the pleural space (Fig. 2) but showed other unmatched perfusion lesions the right lung. PIOPED criteria applied to these findings would be "high probability for pulmonary embolism" (/). An ultrasound evaluation of the left upper extremity showed a possible thrombus in the midsubclavian region. This finding was confirmed by venography, which demonstrated an extensive clot from the subclavian vein to the forearm. The patient underwent urokinase thrombolysis with a heparin drip, and resolution of the clot was confirmed by daily venograms. She experienced a rethrombosis on day 4 at the hospital when the heparin was discontinued for several hours. A bilateral thoracic outlet syndrome was diagnosed during her admission. Complete clot lysis was achieved by hospital day 9 and lower extremity ultrasound was negative for deep vein thrombosis. A repeat V/Q study showed resolution of the right lung perfusion lesions (Fig. 3). Unfortunately, no pulmonary angiography was performed to confirm the V/Q findings. DISCUSSION The stripe sign was first described by Sostman and Gottschalk (2) and validated in a retrospective study. The authors proposed the hypothesis that nonpleural-based perfu sion defects are not PE. Specifically, "a perfusion defect is considered nonpleural based if it shows a stripe of perfused lung between the defect and the adjacent pleural surface." The hypothesis was validated by reviewing cases where the stripe sign was present, and where the presence or absence of pulmonary emboli in the affected segment was demonstrated by pulmonary angiography. In the first study by Sostman and Gottschalk (2), there were nine 722 SAGITAL TRANSAMAL FIGURE 2. The perfusion defect is in contact with the pleural surface in the coronal, sagital and transaxial planes. The ventilation study shows a minimal defect only. The RPO volume rendered view mimics the findings in the planar images. segments with the stripe sign and a matched ventilation abnormal ity. None were associated with artériographieevidence of PE in the same segment. There were six segments with a stripe sign perfusion defect and no ventilation abnormality. Of these, only one was shown to be associated with a PE in the same segment (2). The authors later revisited the issue using the data acquired in the PIOPED study (3). In that review, however, the ventilation status of the perfusion lesions is not reported. Nevertheless, from a total of 85 segments with the stripe sign, only six were associated with a PE in the same segment or region. The data suggest that perfusion abnormalities with a stripe sign are generally not embolie in nature. There are two caveats: (a) the PIOPED study and the Sostman and Gottschalk (7,2) studies have no general data concerning the segment-bysegment association of mismatched perfusion defects and pul monary emboli; and (b) that the stripe sign was deemed present if seen in any projection of the planar images. The presence of the stripe sign, however, in any single projection does not necessarily indicate that a layer of perfused lung is present between the defect and the pleural surface. Planar imaging typically includes only four to six views in most laboratories including anterior/posterior, left posterior oblique/ RPO and, occasionally, anterior oblique images. This strategy provides only three to four images of each lung, which may not give enough information to accurately evaluate perfusion defects for pleural contact. If the lesion contacts the pleural surface in any single projection, it is a pleural-based lesion. The limited number of views acquired in planar imaging may fail to locate the point where a perfusion defect contacts the pleural surface. A stripe sign THE JOURNALOF NUCLEARMEDICINE• Vol. 39 • No. 4 • April 1998 Repeat Y/Q scan after throabolyais RPO Perfusion with resolution of perfusion defect RPO Ventilation could also be mimicked by normally perftised tissue in segments adjacent to the lesion or in the contralateral lung. This may be incorrectly interpreted as a less-suspicious, nonpleural-based le sion, and the patient may not get proper treatment. Under these circumstances, SPECT imaging is less likely to misrepresent pleural contact in perfusion lesions and should be considered when performing V/Q studies. FIGURE 3. A repeat V/Q study after thrombolysis showing resolution of the right middle lobe perfusion abnormality. REFERENCES 1. Sostman HD. Gottschalk A. The stripe sign: a new sign for diagnosis ot nonembolic defects on pulmonary scintigraphy. Ratltulog\' 1982;142:737-74l. 2. Sostman HD, Gottschalk A. Prospective validation of the stripe sign in ventilationperfusion scintigraphy. Radiology 1992:1X4:455-459. 3. The PIOPF.D investigators. Value of the ventilation/perfusion scan in acule pulmonary embolism: results of the prospective investigation of pulmonary embolism diagnosis (PIOPED). JAMA 1990:263:2753-2759. Early Detection of Bleomycin-Induced Lung Injury in Rat Using Indium- 111-Labeled Antibody Directed Against Intercellular Adhesion Molecule-1 Ronald E. Weiner, Daniel E. Sasso, Maria A. Gionfriddo, Sergei I. Syrbu, Henry M. Smilowitz, John Vento and Roger S. Thrall Departments of Diagnostic Imaging and Therapeutics, Medicine, Surgery and Pharmacology, University of Connecticut Health Center, Farmington, Connecticut; and Department of Radiology, VA Medical Center, Newington, Connecticut We have investigated whether an 1"In-labeled mouse monoclonal antibody to rat intercellular adhesion molecule-1 (111ln*alCAM-1) could detect lung injury early in rats treated with bleomycin. Methods: Rats received an intravenous injection of either 111ln*alCAM-1 or1111n-labeled normal mouse IgG (111ln*nmlgG) and were imaged and killed 24 hr later. Lung injury was induced by an intratracheal injection of bleomycin 4 or 24 hr before the rats were killed. After death, tissue was removed and activity was measured, lungs were cryostat-sectioned to detect the presence of ICAM-1 by immunofluorescence, and the up-regulation of LFA-1a was exam ined on blood polymorphonuclear leukocytes (PMNs) using fluores cence-activated cell-sorter (FACS) analysis. Results: In rats injected with111 ln*alCAM-1, the percent injected dose/organ in lungs both at 4 and 24 hr postbleomycin increased significantly compared to the values in either uninjured rats or rats that received 1111n'nmlgG.At 4 and 24 hr postinjury, the target-to-blood (T/B) ratio was 8/1 and 6/1, respectively. For111 ln*nmlgG, the T/B ratio at 4 hr was 0.5/1 and 0.4/1 at 24 hr. In 111ln*alCAM-1 rats injured at 4 or 24 hr, images could easily be distinguished from uninjured rats. All images of Received Jan. 9, 1997; revision accepted Jul. 16, 1997. For correspondence or reprints contact: Ronald Weiner, PhD, University of Connect icut Health Center, Nuclear Medicine MC-2804, 263 Farmington Ave., Farmington, CT 06030. 111ln*nmlgG rats showed only cardiac blood-pool and liver activity with little lung activity. Lung ICAM-1 immunofluorescence intensity increased in the bleomycin-treated samples compared to uninjured lungs. Expression of LFA-1a on PMNs increased 19% and 210% at 4 hr and 24 hr postinjury, respectively, compared to control values. Conclusion: Biodistribution and imaging data demonstrate that 111ln*alCAM-1 can detect early acute bleomycin-induced lung in jury. Immunofluorescence and FACS data suggest that 111ln*ICAM-1 uptake is a specific process. This antibody has poten tial as an early radionuclide detector of acute inflammations. Key Words: antiadhesion molecule antibody; acute respiratory distress syndrome; inflammation; ICAM-1 J NucÃ-Med 1998; 39:723-728 Indium-Ill or y9mTc-labeled autologous white blood cells (WBCs) and 67Ga-citrate are commonly used radiopharmaceuticals that are effective indicators of inflammatory processes in a variety of clinical settings (1,2). However, these agents are not without their limitations such as the time-consuming prepara tion and exposure to blood-borne pathogens with labeled WBCs, and low specificity and high bowel activity with 67Ga. To supplant or be an adjunct to these radiopharmaceuticals, INDIUM-11l*AlCAM-l IN BLEOMYCIN-TREATED RATS• Weiner et al. a 723