ADIOLOGY -----·-­ 1,.11 NCOLOGY December 2002 Vol. 36 No. 4 Ljubljana ISSN 1318-2099 SIEMENS SiemensMedical.com/oncology o N <( u ,; § o u C, U >­ . <( Q(/) "'::::, ->sentinel bleeding« because it often precedes, by 6 hours to 10 days, massive hemorrhage from the erosion of a large arteri­al branch. Rumstat et a/3 suggest immediate surgical intervention with a revision of the pancreatic-digestive anastomosis when >>sen­tinel bleeding<< appears at the drainage or the gastro-intestinal tract. When the dehiscence of the anastomosis occurs, the massive hem­orrhage causes mortality in 15% to 58% of cas­es.3·17 The frequent presence of anastomotic dehiscence requires an embolization tech­nique similar to that used in acute pancreati­tis. The most commonly involved vessels are the gastroduodenal artery stump or the com­mon hepatic artery.3·17·20 The embolization with coils leads to definite stoppage of the bleeding only when the gastroduodenal ar­tery is occluded in a tract not involving septic maceration or, alternatively, the common he­patic artery is directly occluded. Embolization is sometimes used as a temporary procedure to stop or slow down bleeding so that the pa­tient can be operated on electively rather than in emergency. Considering the possible complication of the endovascular embolization, the vessels that can generally be embolized safely in this region include the left gastric, gastroduode­nal, gastroepiploic, and pancreaticoduodenal arteries.21 The occlusion of the common he­patic artery, with normal patency of the por­tal vein has no clinical consequences; em­bolization is, however, unadvisable in the presence of thrombosis or compression of the portal vein.22 Moreover, the presence of a bil­iary-enteric anastomosis is considered a risk factor for developing a hepatic abscess fol­lowing a hepatic artery embolization.23 The treatment of bleeding from the common he­patic artery or from the short stump of the gastroduodenal artery following a pancreato­duodenectomy is often problematic for the surgeon, and the radiologist should decide to go ahead with classic embolization. But com­pression, even to the point of thrombosis, of the portal vein is common due to the pres­ence of adjacent hematic collection. In these cases, the alternative treatment to emboliza­tion is positioning covered stents that main­tain the patency of the hepatic artery in emer­gency (Figure 1). Although very significant, there are still only a few reports in literature concerned with the use of this technique.24.25 Occlusion of the stent over time due to hy- Radio/ Oncol 2002; 36(4): 281-90. perplasia of intima is predictable. Never­theless, the treatment stops the bleeding im­mediately and maintains the hematic contri­bution to the liver. The slow occlusion of the stent can then be compensated by the re­cruitment of collateral intra-hepatic arterial circulation and the return to normal portal flow or the growth of a collateral portal net­work. Conclusions Considering the high mortality rate of splanchnic artery bleeding in pancreato-bil­iary diseases and the poor results of surgical intervention, the endovascular approach, with embolization or repair of the bleeding vessel, has to be considered as the treatment of first choice. In our series, the endovascular treatment of splanchnic artery bleeding in pancreato-biliary disease resulted as clinically successful in up to 75% of cases at the three­month follow-up. References 1. Shibata T, Sagoh T, Ametani F, Maetani Y, ltoh K, Konishi J. Transcatheter microcoil embolotherapy for ruptured pseudoaneurysm following pancreat­ic and biliary surgery. Cardiovasc lntervent Radio/ 2002; 25: 180-5. 2. Rumstadt B, Schwab M, Korth P, Samman M, Trede M. Hemorrhage after pancreatoduodenecto­my. Ann Surg 1998; 227: 236-41. 3. Messina LM, Shanley CJ. Visceral artery aneurysms. Surg Clin North Am 1997; 77: 425-42. 4. Burke JW, Erickson Sj, Kellum CD, Tegtmeyer CJ, Williamson BRJ, Hansen MF. Pseudoaneurysms complicating pancreatitis: detection by CT. Radiology 1986; 161: 447-50. 5. Frey CF, Stanley JC, Eckhauser F. Hemorrhage. In: Bradley EL, editor. Complications of pancreatitis. Philadelphia: WB Saunders Co; 1992. p. 96-123. 6. Kiviluoto T, Kivisaari L, Kivilaakso E, Lempinen M. Pseudocysts in chronic pancreatitis. Surgical results in 102 consecutive patients. Arch Surg 1989; 124: 240-3. 7. Sankaran S, Wait AJ. The natural and unnatural history of pancreatic pseudocyst. Br] Surg 1975; 62: 37-44. 8. Bresler L, Boissel P, Grosdidier J. Major haemor­rhage from pseudocysts and pseudoaneurysms caused by chronic pancreatitis: surgical therapy. World] Surg 1991; 15: 649-52; 652-3. 9. Ammori BJ, Alexander DJ, Madan M. Haemorrhagic complications of pancreatitis: pres­entation, diagnosis and management. Ann R Col/ Surg Eng/ 1998; 80: 316-25. 10. Lendrum R. Chronic pancreatitis. In: Misiewicz JJ, Pounder RE, V enables CW, editors. Diseases of the gut and pancreas. London: Blackwell Scientific Publications; 1994. p. 441-54. 11. Schoder M, Cejna M, Langle F, Hittmaier K, Lammer J. Glue embolization of a ruptured celiac trunk pseudoaneurysm via the gastroduodenal ar­tery. Eur Radio! 2000; 10: 1335-7. 12. Stabile BE, Wilson SE, Dibas HT. Reduced mortal­ity from bleeding pseudocysts and pseudoa­neurysms caused by pancreatitis. Arch Surg 1983; 118: 45-51. 13. El Hamel A, Pare R, Adda G, Bouteloup PY, Huguet C, Malafosse M. Bleeding pseudocysts and pseudoaneurysms in chronic pancreatitis. Br] Surg 1991; 78: 1059-63. 14. Stanley JC, Frey CF, Miller TA, Lindenauer SM, Child CG. Major arterial hemorrhage. Arch Surg 1976; 111: 435-8. 15. Golzarian J, Nicaise N, Deviere J, Ghysels M, Wery D, Dussaussois L, et al. Transcatheter em­bolization of pseudoaneurysms complicating pan­creatitis. Cardiovasc lntervent Radio/ 1997; 20: 435­40. 16. Yamakado K, Nakatsuka A, Tanaka N, Takano K, Matsumura K, Takeda K. Transcatheter arterial em­bolization of ruptured pseudoaneurysms with coils and n-butyl cyanoacrylate. ]VIR 2000; 11: 66-72. 17. Sato N, Yamaguchi K, Shimizu S, Morisaki T, Yokohata K, Chijiiwa K, et al. Coil embolization of bleeding visceral pseudoaneurysms following pan­createctomy: the importance of early angiography. Arch Surg 1998; 133: 1099-102. 18. Aranha GV, Prinz RA, Greenlee HB, Freeark RJ. Gastric outlet and duodenal obstruction from in­flammatory pancreatic disease. Arch Surg 1984; 119: 833-5. Radio/ Oncol 2002; 36(4): 281-90. 19. Brodsky JT, Turnbull AD. Arterial hemorrhage af­ter pancreatoduodenectomy. The »Sentinel bleed«. Arch Surg 1991; 126: 1037-40. 20. Balladur P, Christophe M, Tiret E, Pare R. Bleeding of the pancreatic stump following pancreatoduo­denectomy for cancer. Hepatogastroenterology 1996; 43: 268-70. 21. Rosen RJ, Sanchez G. Angiographic diagnosis and management of gastrointestinal hemorrhage Current Concepts. Radio/ Clin North Am 1994; 32: 951-67. 22. Cardella JF, Vujic I, Tadavarthy SM, Beltran M, Castafieda-Zufiiga WR. Gastrointestinal bleeding. Part 1. Vasoactive drugs and embolotherapy in the management of gastrointestinal bleeding. In: Castafieda-Zuiiniga WR, editors. Interventional ra­diology. 3'd ed. Baltimore: Williams & Wilkins; 1997. p. 207-52. 23. Okajima K, Kohno S, Tamaki M, Hosono M, Kawamoto M, Nishiyama Y, et al. Bilio-enteric anastomosis as a risk factor for postembolic he­patic abscess. Cardiovasc lntervent Radiol 1989; 12: 128-30. 24. BLrger T, Halloul Z, Meyer F, Grote R, Lippert H. Emergency stent-graft repair of a ruptured hepatic artery secondary to local postoperative peritonitis. ] Endovasc Ther 2000; 7: 324-7. 25. Paci E, Antico E, Candelari R, Alborino S, Marmorale C, Landi E. Pseudoaneurysm of the common hepatic artery: Treatment with a stent­graft. Cardiovasc lntervent Radio/ 2000; 23: 472-84. Radio/ Oncol 2002; 36(4): 281-90. Radiol Oncol 2002; 36(4): 291-5. case report Cerebral hyperperfusion syndrome after carotid angioplasty Zoran Miloševic1, Bojana Žvan2, Marjan Zaletel2, Miloš Šurlan1 1Institute of Radiology, 2University Neurology Clinic, University Medical Center, Zaloška cesta 7, Ljubljana, Slovenia Background. Cerebral hyperperfusion syndrome after carotid endarterectomy is an uncommon but well-de­fined entity. There are only few reports of »hyperperfusion injury« following carotid angioplasty. Case report. We report an unstable arterial hypertension and high-grade carotid stenosis in a 58-year-old, right-handed woman. After a stroke in the territory of middle cerebral artery carotid angioplasty was per­formed in the patient. Among risk factors, the long lasting arterial hypertension was the most pronounced. Immediately after the procedure, the patient was stable without any additional neurologic deficit. The sec­ond day, the patient had an epileptic seizure and CT revealed a small haemorrhage in the left frontal lobe. Conclusions. The combination of a high-grade carotid stenosis and unstable arterial pressure is probably an important prognostic factor in the pathogenesis of hyperperfusion syndrome. Key words: carotid artery diseases; angioplasty - adverse effects; reperfusion injury, hyperperfusion syn­drome; hypertension Introduction Cerebral hyperperfusion after carotid en-darterectomy is an uncommon, but well-de­fined entity.1 Despite the increasing use of carotid anigoplasty, there are only few re­ports of »hyperperfusion injury« following carotid angioplasty in the literature.2-5 The syndrome occurs in a number of clinical set­tings and is characterised by the diagnostic triad of unilateral headache, seizures, and in-tracranial haemorrhage. In our case, the pa- Received 25 October 2002 Accepted 4 November 2002 Correspondence to: Zoran Miloševic, MD, University Medical Center, Institute of Radiology, Zaloška 7, SI­1525 Ljubljana, Slovenia; Phone: +386 1 432 2346; E­mail: zoran.milosevic@guest.arnes.si tient developed typical signs of hyperperfu­sion syndrome detected from typical comput­ed tomography (CT) findings. Case report In September 2001, a 58-year-old, right-hand­ed woman was referred to our Department af­ter an ischemic stroke. In 1995, she had an an­terior circulation cerebrovascular accident but she has made a good recovery. Her medical history included long-lasting arterial hyper­tension for more than 20 years and hyper-lipoproteinemia. There was no history of ciga­rette smoking or excessive alcohol intake. In spite of regular ACE-inhibitor treatment, her blood pressure fluctuated. She had been treat­ed with Aspirin 100 mg daily and statin 20 mg. When assessed in our hospital before carotid angioplasty she had a residual expres­sive dysphasia, as well as a mild weakness in the face and limb on the right. Her blood pressure was 180/110 mm Hg. Computed to­mography scan (CT) of the brain revealed widened liquor spaces, pre-existed ischemic lesions up to 1 mm in size, located deep in the left cerebral hemisphere, in the left frontal lobe and subcortically in the left parietal lobe. There was no evidence of haemorrhage. All haematological and biochemical tests were normal with a normal platelet count and co­agulation screen. Duplex ultrasonography made in 2001 revealed a 90-per-cent stenosis of the left internal carotid artery (ICA) pro­duced by echolucent plaque, type I (according to the accepted international classification). The plaque was unstable and had an irregular surface (Figure 1). The patient underwent the left carotid an­gioplasty through the femoral approach un­der local anaesthesia. Intra-arterial digital subtraction angiography confirmed a 95-pe-cent stenosis of the left ICA (Figure 2a). The patient was given 5000IU of heparin IV. The stenosis was crossed with a flexible coronary guidewire (V-18 Control Wire; Boston Scientific Corp). Glycopyrrolate and 0.5 mg atropine IV were administered during the procedure. The stenosis was predilated with a low-profile coronary balloon (4 x 20 mm Bypass Speedy Monorail Catheter, Boston Scientific Corp) and stented with a 7 x 30 Carotid Wallstent Monorail (Boston Scientific Corp). The stent was dilated with a 5.5 x 20 mm Bypass Speedy Monorail Catheter (Boston Scientific Corp) that embedded it firmly into the vessel wall. The blood pressure varied between 160/90 mm Hg and 175/105 mm Hg during the pro­cedure, but there were no residual adverse neurological sequelae. Postprocedural an-giogram showed no significant stenosis or dissection (Figure 2b). In the following 24 hours, the patient was treated with Aspirin Radiol Oncol 2002; 36(4): 291-5. and clopidogrel, her blood pressure varied be­tween 140 and 160/95 mm Hg and she was clinically stable. On the following day she was dismissed. She did not continue antihy­pertensive therapy when she was at home be­cause she was convinced that she did not need this therapy after carotid stenting. After 2 days she was urgently re-admitted because of a grand mal type epileptic seizure. After the seizure she had a transient left right-sided hemiplegia. Blood pressure at the time of admission was 180/100 mm Hg. An urgent brain CT revealed a small haemor­rhage in the left frontal lobe (Figure 3). Colour Doppler ultrasound of the ICA revealed a vis­ibly patent vessel (Figure 4). The peak systolic velocity rose to 2.3 m/s, with the end diastolic velocity of 1.2 m/s. The patient was managed conservatively. Hypertension was easily con­trolled with 10 mg enalapril twice daily. The antiepileptic therapy was introduced. She re­covered completely after two weeks. Discussion Cerebral hyperperfusion syndrome (CHS) may manifest as ipsilateral headaches, seizures, or intracerebral haemorrhages. Risk factors such as high-grade stenosis, contralat­eral carotid occlusion, poor collateral flow, chronic ipsilateral hypoperfusion, preopera­tive and postoperative hypertension, and pe­rioperative use of anticoagulant or an-tiplatelet agents have been reported.1 In our case, we do not have pathologic evidence to support hyperperfusion injury as a cause of the haemorrhage after CAS, but clinical fea­ture and postprocedural systemic hyperten­sion, together with the lobar appearance of the haemorrhage, indicate to the mechanism of hyperperfusion injury. Angioplasty per­formed in the patients with high degree carotid stenosis proved that the stenosis was associated with poststenotic drop in perfu­sion pressure. Therefore, it is likely that the patient suffered a chronic ischemia, which can cause a loss of autoregulation. This could be an important pathophysiologic mecha­nism of hyperperfusion injury.6 CHS has been defined as cerebral blood flow (CBF) in excess of that required for meta­bolic needs or a postoperative increase greater than 100% of the preoperative cere­bral blood flow.7 Therefore, when cerebral autoregulation is impaired, the elevated blood pressure could increase CBF. In our case, the patient had unstable arterial pres­sure with tendency toward high pressure. During hospitalisation, the arterial pressure did not exceed 145/90 mm Hg and the patient did not use antihypertensive medication. According to our experiences, the heart rate Radiol Oncol 2002; 36(4): 291-5. and arterial pressure in most of the patients decrease after the carotid angioplasty. The reason could be iatrogenic stimulation of carotid baroreceptors, which seems to im­prove in the next few days. Thus, the arterial pressure could rise to hypertensive levels in the days after carotid stenting, thereby in­creasing regional CBF in the presence of im­paired autoregulation and causing CHS. CHS has also been widely reported in the surgical literature as an infrequent complica­tion of carotid endarterectomy (CEA) with an incidence of approximately 0.6%.6 In our opinion and previous experiences, it may also occur after percutaneous transluminal carotid angioplasty and stenting with causal mecha­nism and clinical features similar to those of CEA.5,8 One cannot completely exclude the possibility of embolism and silent cerebral in­farction with subsequent haemorrhagic trans­formation in response to hyperperfusion, but the CT scan appearances do not indicate to such a mechanism. Conclusions In conclusion, CHS may occur after carotid stenting. The combination of a high-degree carotid stenosis and unstable arterial pres­sure is probably an important cause in the pathogenesis of hyperperfusion syndrome. The arterial blood pressure monitoring seems to be important after carotid angioplasty of high-degree stenosis. Figure 3a, 3b, 3c. CT of the brain demonstrates small haemorrhage in the left frontal region. Radiol Oncol 2002; 36(4): 291-5. Figure 4. Colour Doppler ultrasound of the internal carotid artery shows a visibly patent vessel. References 1. Keunen R, Nijmeijer HW, Tavy D, Stam K, Edelenbosch R, Muskens E, et al. An observation­al study of pre-operative transcranial Doppler ex­aminations to predict cerebral hyperperfusion fol­lowing carotid endarterectomies. Neurol Res 2001; 23: 593-8. 2. Milosevic ZV, Zvan B, Zaletel M, Surlan M. Cerebral hyperperfusion syndrome after carotid stenting. [Online]. Mar 22, 2002. URL: http:// www.eurorad.org/case.cfm?UID=1563 Luxembourg, Euromedia. 3. Chuang YM, Wu HM. Early recognition of cerebral hyperperfusion syndrome after carotid stenting - a case report. Kaohsiung J Med Sci 2001; 17: 489-94. 4. Morrish W, Grahovac S, Douen A, Cheung G, Hu W, Farb R, et al. Intracranial hemorrhage after stenting and angioplasty of extracranial carotid stenosis. AJNR Am J Neuroradiol 2000; 21: 1911-6. 5. McCabe DJ, Brown MM, Clifton A. Fatal cerebral reperfusion hemorrhage after carotid stenting. Stroke 1999; 30: 2483-6. 6. Hosoda K, Kawaguchi T, Shibata Y, Kamei M, Kidoguchi K, Koyama J, et al. Cerebral vasoreac­tivity and internal carotid artery flow help to iden­tify patients at risk for hyperperfusion after carotid endarterectomy. Stroke 2001; 32: 1567-73. 7. Nakase H, Sakaki T, Kempski O. A scanning tech­nique to measure regional cerebral blood flow and oxyhemoglobin level. Neurosurgery 2001; 48: 1335­42. 8. Mansoor GA, White WB, Grunnet M, Ruby ST. Intracerebral hemorrhage after carotid endarterec­tomy associated with ipsilateral fibrinoid necrosis: a consequence of the hyperperfusion syndrome? J Vasc Surg 1996; 23: 147-51. Radiol Oncol 2002; 36(4): 291-5. Radiol Oncol 2002; 36(4): 297-303. Cortico-basal ganglionic degeneration: radiological and functional features Maja Ukmar1, Rita Moretti2, Paola Torre3, Rodolfo M. Antonello3, Renata Longo4, Antonio Bava2, Roberto Pozzi Mucelli1 1Department of Radiology, 2Department of Physiology and Pathology, 3Department of Internal Medicine and Clinical Neurology, 4Department of Physics, University of Trieste Background. Cortico-basal ganglionic degeneration is a rare degenerative pathology that involves parietal areas and gradually determines frontal involvement. The aim of our work was to describe the main radio­logical findings in this pathology and to evaluate the cortical activation in these patients by f-MRI during simple and complex movements. Patients and methods. We have evaluated eight patients with morphological and functional magnetic res­onance by using a 1.5 T imager. Results. Morphological evaluation: We found an asymmetric perirolandic cortical atrophy in seven pa­tients, a mild hyperintensity in the perirolandic cortex in five patients, a mild atrophy of the basal ganglia in seven patients and, in one, a hypointensity in the lenticular nuclei. In one patient the morphological as­pect was normal. Functional evaluation: The most important aspect was the hypoactivation of the parietal areas during the movement with the affected hand in all the patients. Conclusions. We consider f-MRI a helpful tool for the diagnosis and follow-up of this pathology. Key words: basal ganglia diseases, cortico-basal degeneration; magnetic resonance imaging; fluorine ra­dioisotopes, f-MRI; movement, apraxia, parietal lobe Introduction Cortico-basal ganglionic degeneration (CBGD) has become a more widely recog­nized entity: it has been considered as a de­generative movement disorder since its first Received 25 October 2002 Accepted 4 November 2002 Correspondence to: Maja Ukmar, MD, Department of Radiology, University of Trieste, Trieste, Italy description by Rebeiz thirty years ago.1 Since then, one hundred cases have been reported,2 but it remains a rare disease of unknown in­cidence and prevalence.3,4 CBGD usually presents after the fifth decade of life, with a varied combination of symptoms including stiffness, clumsiness, jerking, segmental dystonia, bradykinesia and usually ideomotor apraxia which can progress to the complete development of an »alien hand syndrome«. Other symptoms, such as action-induced and stimulus sensitive focal reflex myoclonus may precede or ac­company the development of dystonic pos­tures.5 Invariably, it leads to a progressive disability (both motor and cognitive) which gradually leads to immobilization and institu­tionalisation. Logistic regression analysis identified two models that contributed to distinguish these disorders and predicted the diagnosis of CBGD. The first one included asymmetric Parkinsonism as symptom onset and instabil­ity and falls at first clinic visit. The other one included cognitive disturbances, asymmetric parkinsonism within the first year of symp­tom onset and speech disturbances at the first clinic visit.6 A recent evaluation7 sug­gests that cognitive signs of disruption, which comprise a marked impairment of daily living functions, may be the commonest presenta­tion of CBGD, rather than the better recog­nised perceptual-motor syndrome, described previously. Radiological evaluation seems to be impor­tant as a diagnostic and clinical follow-up in­strument. The most constant expression of the pathological features are: an important at­rophy of the perirolandic gyri, particularly in the postrolandic cortex, associated with a mild atrophy of the basal ganglia,8-10 and an almost evident, though not constant hy­pointensity of the lenticular nuclei.8-10 The possibility of a direct and functional evaluation of cortical activation during hand motion, registered in CBGD patients, seems to us very interesting. This was the aim of our study: we discussed the results with an overview of literature. Patients and methods During the period between 1st January 1997 and 1st January 2002, eight patients (three males and five females) were included into our observation. Their mean age was 62.1 years old (SD + 6.9), all of them were right-handed (average score at Briggs and Nebes Test: + 22.56).11 The past history of all the patients was completely mute for cerebrovascular disease, hypertension and metabolic disorders. No signs of addiction could be found. Their most common complain was the relatively recent development of an asymmetric akinetic syn­drome (55% of cases), affecting in all the eight patients their left superior limb, ideomotor apraxia (43% of cases), bradykinesia (36% of cases), alien-limb syndrome (16% of cases), slurred speech (5% of cases) and gait difficul­ty (5% of cases). Only two of the subjects (35% of cases) showed action tremor and one of them (13% of cases) supranuclear gaze palsy affecting vertical and horizontal gazes (evi­denced at oculomotor evaluation). All the pa­tients underwent a complete oculomotor evaluation: a normal saccadic velocity (con­sidering anti-saccades, reflexive saccades and voluntary saccades), with increased latency of saccades (especially of voluntary saccades) and preserved pursuit and optokinetic nys­tagmus were globally evidenced. The mean duration of symptoms dated from 7.62 + 5.32 months prior to their admis­sion. From the cognitive perspective, intelli­gence performances were within normal ranges (111 + 2.34) as stated by the average score obtained in Raven Standard Progressive Matrices;12 the patients recog­nised right/left personal and extrapersonal hemispace, and no signs of tactile agnosia and of bucco-facial apraxia were found. Wechsler Adult Intelligence Scale (WAIS) av­erage results demonstrated a mild general tendency to global deterioration (21.1% SD + 2.34%).13 All the patients could reproduce Koh’s Block quite well, by copy and by mem­ory. Wechsler Memory Scale (WMS)14 put in evidence an MQ average score of 56.35 (SD + 5. 41), underlying a mild deterioration of log­ical, procedural and verbal memory strate­gies. All the subjects could not pantomime to Radiol Oncol 2002; 36(4): 297-303. verbal command of the examiner: all together showed signs of ideomotor apraxia with the left, affected hand. On the contrary, they did not show signs of ideational apraxia. All the patients showed moderate insight into their general situation, but principally into their motion disruption. All the patients underwent brain-MRI, per­formed by a 1.5 T magnet. For the morpho­logical evaluation an axial SE PD/T2 (TR/TE=2709/20-80) and a turbo-FLAIR (TR/TE/TI=9832/150/2000) sequences were performed. In order to obtain a dynamic acquisition of cortical activation during complex and ac­quired motor process, we decided to study our patients with f-MRI. After training, sub­jects had to oppose the thumb to the other fingers in a sequential task, in a 2, 3, 4, 5 se­quence and in a complex, alternating se­quence, 1-2, 1-4, 1-3 and 1-5 sequence. The total acquisition time was equally di­vided into three-motor task periods, alternat­ed with a three-rest period. Seven images per period were collected; so, in each measure­ment, 42 images were acquired. The images were oriented transversally. The major pa­rameters of the 2D gradient-echo MR pulse sequence were the following: TR=60 ms, TE=40ms, Flip Angle=25, FOV=160x144 mm2, Slice Thickness=4 mm, Scan Matrix=128x 128. The T1 contrast enhancement option was activated.15 An MR angiography acquisition was per­formed per each T1-GRE f-MRI acquisition. The major parameters of the angiographic se­quence were the following: TR=shortest, Flip Angle=20, FOV=the same of the T1-gre, Slice Thickness=1 mm, Scan Matrix=256x256, Slices=12, Slice Thickness=1 mm, Phase Contrast Technique. The image analysis is performed by a pro­gram developed in IDL environment (Interactive Data Language, Research System Inc., USA). The basic analysis consists of the calculation of the correlation coefficient be­tween the time-intensity behavior of each pix­el and the square wave model function. In order to exclude transient hemodynam­ic responses, 5 images per block (from the 3rd to the 7th of each block) are included in the analysis. A raw activation map was obtained by applying a correlation analysis (p<0.001) and a cluster filtering (at least 5 pixels). The raw map was affected by flow artefacts; to eliminate these artefacts, activation map and MR angiography were compared and activa­tion clusters related to the vessels were re­jected. Whole-head high-resolution T1­weighted images (TR/TE=500/15) were then acquired to be used as an anatomical refer­ence for the transformations into the Talairach space.16 Results In seven patients, the morphological exami­nation showed an important, asymmetric perirolandic and postrolandic cortical atro­phy (Figure 1) associated to a mild atrophy of the basal ganglia. Subtle MRI T2 hyperin- Radiol Oncol 2002; 36(4): 297-303. tense lesions in the primary motor cortex, compatible with underlying gliosis, were found in five patients (Figures 2a, 2b), hy­pointensity in the lenticular nuclei in one case only. In one patient the morphological aspect was normal. During simple motor task carried out with the non-affected hand, we could observe a good activation of the contralateral rolandic cortex, associated with a discrete activation of the supplementary motor area (SMA) and of the parietal regions, as well as with a dis­crete activation of the right prefrontal region, but without significant difference from those in healthy population. On the contrary, during a simple task of opposing the fingers executed with the af­fected hand, an evident hypo-activation of the homolateral and contralateral periro­landic cortex, associated with an evident hy-po-activation of the contralateral SMA and with the affected parietal region, was ob­served. It was significantly different from the healthy control subjects. During the complex sequence execution by the non-affected hand, an obvious bilater­al activation of rolandic areas, of the parietal areas, of the SMA, and of the contralateral frontal region could be seen. These observa­tions are similar to those in healthy controls. When the complex sequence was executed by the affected hand, the observed activation was limited to the bilateral rolandic region, to SMA, and to a very modest activation of the parietal regions. Quantitatively, hypoactiva­tion was significantly different from that in healthy population. Qualitatively, the movement of the affect­ed hand was impaired and not fluent at all de­spite more training exercises. Moreover, it was obvious that the patients had to see the entire procedure when they were performing the exercise with the affected hand. Without receiving the visual input that controlled the motor act the movement was even more im­paired. Discussion The patients with progressive focal cortical syndromes are being recognized with increas­ing frequency. CBGD is a degenerative disor- Radiol Oncol 2002; 36(4): 297-303. der, involving primarily parietal areas, and gradually extending to frontal areas. The pa­tients with CBGD revealed a mild to moder­ate global deficits including a frontal dysex­ecutive syndrome, explicit learning deficits without retention difficulties and displayed prominent deficits on the tests of sustained attention/mental control and verbal fluen­cy.17 Different neuroimaging studies have been conducted on CBGD, using conventional computed tomography (CT) and magnetic resonance imaging (MRI),4,18 and demonstrat­ing an asymmetric pericentral cortical atro­phy in approximately 50% of cases. On the contrary, another recent MRI study, compar­ing clinically diagnosed cases of CBGD with progressive supranuclear palsy (PSP), found that 87.5% patients with CBGD (but none of the PSP group) had asymmetric fronto-pari­etal atrophy while midbrain atrophy was seen in 6.3% and 89.3% of the same cases respec­tively.18 Another radiological observation4 revealed that pathologically proven PSP, frontal lobe dementia and AD with clinical features of CBGD had similar MRI cortical changes. The only possible conclusion is that, due to the distribution of pathological changes, the static imaging simply correlates with the pre­dominant clinical presentation and not with the specific underlying pathological sub­strate. Functional imaging studies were done with positron emission tomography (PET), but re­sults did not seem to be conclusive.18FDG­PET studies demonstrated a greater hetero­geneity of the metabolic patterns including diffuse hypometabolism despite asymmetri­cal clinical features. F-Dopa PET demonstrat­ed a severe asymmetric reduction in striatal F-Dopa uptake which tended to be equal in putamen and caudate consistent with wide­spread substantia nigra neuron loss.19 From these studies, the general suggestion is that (18F)-fluorodeoxyglucose (FDG) and (18F)-flu­oro-dopa (F-Dopa)-PET must be utilized at the same time, in the same patient.19,20 Our study is the first one on f-MRI in CBGD: the most interesting part is the dy­namic acquisition which gives information on how the cortex works when the affected arm is moving. The results obtained from this selected group of eight patients confirmed the results, reported previously.21-25 Our f-MRI evaluation revealed, while the patients executed the simple task of opposing the fingers with the affected hand, a hypo-ac­tivation of the homolateral and contralateral perirolandic cortex, associated with an evi­dent hypo-activation of the contralateral SMA and of the affected parietal region. During the complex motor skills with the affected hand, a drastic reduction of activation of premotor and motor areas (perirolandic region) of the contralateral cortex, associated with a very modest activation of the supplementary mo­tor area of the same hemisphere were noted. It is interesting to see a very modest acti­vation of motor areas, associated with the re­duced activity of the parietal area: the latter is largely preventable, but the former is rather unexpected. The imaging studies on both monkeys and humans report of the activation in the region of the supplementary motor area, globus pal-lidus, and parietal cortex during the perform­ance of sequential movements.26 Neurons of the globus pallidus also discharge during spe­cific phases of a sequential performance.26 This does not involve uniquely motor speed and correct execution, but also attention, co­ordination and dynamic adjustment to the situation. The hypoactivity of the cortical ar­eas we have reported does match with previ­ous finding. Nevertheless, there is an unresolved ques­tion: our patients indisputably presented with a parietal (postrolandic) atrophy: consid­ering that the parietal damage is the cause, why should the pure motor areas should hy- Radiol Oncol 2002; 36(4): 297-303. poactivated? We maintain that the brisk stop due to the disconnection of the parietal cortex and the SMA because of degenerative alter­ation of parietal areas causes an interruption of the neural net to putamen and pallidus, ending in frontal and prefrontal areas. Therefore, f-MRI, with its constant demon­stration of hypoactivity of motor cortical ar­eas, SMA and parietal areas, while affected arm is moving, might be a valid supportive tool for the diagnosis of CBGD: a positive cor­relation with hypoactivation of motor areas could be found even in modest, and initial stages of disease. Having assessed that the accuracy of neu­rologists’ clinical diagnosis of CBGD is very low, even in specialized centers2 (at first visit mean sensitivity for CBGD 35%, specificity 99.6% and at the last visit mean sensitivity 48.3% while specificity remained stable), then MRI and f-MRI might be suggested as an ad­junctive tool of the diagnosis refinement. References 1. Rebeiz JJ, Kolodny EH, Richardson EP. Corticodentatonigral degeneration with neuronal achromasia. Arch Neurol 1968; 18: 20-33. 2. Litvan I, Agid Y, Goetz C, Jankovic J, Wenning GK, Brandel JP, et al. Accuracy of the clinical di­agnosis of corticobasal degeneration: a clinico-pathologic study. Neurology 1997; 48: 119-25. 3. Lippa CF, Smith TW, Fontneau N. Corticonigral degeneration with neuronal achromasia. J Neurol Sci 1990; 98: 301-10. 4. Lang A. Cortico-basal ganglionic degeneration. San Diego: American Academy of Neurology; 2000. 5. Thompson PD, Day BL, Rothwell JC, Brown P, Britton TC, Marsden CD. The myoclonus in corti­cobasal degeneration. Evidence for two forms of cortical reflex myoclonus. Brain 1994; 117: 1197­207. 6. Litvan I, Grimes DA, Lang AE, Jankovic J, McKee A, Verny M, et al. Clinical features differentiating patients with postmortem confirmed progressive supranuclear palsy and corticobasal degeneration. J Neurol 1999; 246(Suppl 2): 1-5. 7. Grimes DA, Lang AE, Bergeron C. Dementia is the most common presentation of cortical-basal gan­glionic degeneration. Neurology 1998; 50(Suppl 4): a96. 8. Hauser R., Murtaugh F, Akhter K, Gold M, Olahow C. Magnetic resonance imaging of corti­cobasal degeneration. J Neuroimaging 1996; 6(4): 222-6. 9. Otsuki M, Sama Y, Yoshimuro N, Tsuji S. Slowly progressive limb-kinetic apraxia. Eur Neurol 1997; 37(2): 100-3. 10. Tokumaru AM, O’uchi T, Kuru Y, Maki T, Murayama S, Horichi Y. Corticobasal degenera­tion: MR with histopathologic comparison. ASNR 1996; 17(10): 1849-52. 11. Briggs GC, Nebes RD. Patterns of hand preference in a student population. Cortex 1975; 11: 230-8. 12. Raven JC. Standard progressive matrices. London: Lewis; 1938. 13. Wechsler D. Wechsler adult intelligence scale manual. New York: Grune & Stratton; 1976. 14. Wechsler D. A standardized memory scale for clinical use. J Psychol 1945; 19: 87-97. 15. Sobol WT, Gauntt DM. On the stationary states in gradient echo imaging. J Magn Reson Imaging 1996; 6: 384-98. 16. Talairach J, Tornoux P. Co-planar stereotaxic atlas of the human brain. Thieme: Stuttgart; 1988. 17. Kertesz A, Munoz DG. Clinical and pathological overlap between frontal dementia, progressive aphasia and corticobasal degeneration - the Pick complex. Neurology 1997; 48: 293-300. 18. Gimenez-Roldan S, Mateo D, Benito C, Grandas F, Perez-Gilabert Y. Progressive supranuclear palsy and corticobasal ganglionic degeneration: differ­entiation by clinical features and neuroimaging techniques. J Neural Transm 1994; 98(Suppl 42): 79-90. 19. Brooks DJ. Pet studies on the early and differential diagnosis of Parkinson’s disease. Neurology 1993; 43(Suppl 6): S6-16. 20. Eidelberg D. Differential diagnosis of parkinson-ism with (18-f)fluorodeoxyglucose FDG and PET. [abstract]. Mov Disord 1996; 11: 349. 21. Moretti R, Ukmar M, Torre P, Antonello RM, Longo R, Nasuelli D, et al. Cortical-basal ganglion­ic degeneration: a clinical, functional and cogni­tive evaluation (1-year follow-up). J Neurolog Sci 2000; 182: 29-35. Radiol Oncol 2002; 36(4): 297-303. 22. Moretti R, Torre P, Antonello RM, Ukmar M, Cazzato G, Bava A. Valutazione con risonanza magnetica funzionale dell’attivazione corticale du­rante compiti motori fini in soggetti con degener­azione cortico-basale. Nuova Rivista di Neurologia 2001; 11(3): 73-9. 23. Moretti R, Torre P, Antonello RM, Bava A. Deterioramento cognitivo nella sindrome cortico­basale. Dementia Update 2002; 11: 49-52. 24. Moretti R, Torre P, Antonello RM, Ukmar M, Longo R, Cazzato G, et al. Complex distal move­ment in cortico-basal ganglionic degeneration. A functional evaluation. Funct Neurol 2002; 17(2): 71­6. 25. Jeannerod M, Arbib MA, Rizzolatti G, Sakata H. Grasping objects: the cortical mechanisms of vi-suomotor transformation. A review. Trends Neurosci 1995; 18: 314-20. 26. Lu MT, Preston JB, Strick PL. Interconnections be­tween the prefrontal cortex and the premotor ar­eas in the frontal lobe J Comp Neurol 1994; 341: 375-92. Radiol Oncol 2002; 36(4): 297-303. Radiol Oncol 2002; 36(4): 305-12. Breast MRI of ductal carcinoma in situ: Is there MRI role? Giuliana E. Francescutti, Viviana Londero, Ilaria Berra, Chiara Del Frate, Chiara Zuiani, Massimo Bazzocchi Department of Medical and Morphological Research, Institute of Radiology, University of Udine Background. The purpose of this study is to report our personal experience of 22 cases of ductal carcino­ma in situ (DCIS) studied with magnetic resonance imaging (MRI). Patients and methods. From September 1995 to December 2001, 22 women diagnosed with DCIS lesions underwent contrast enhanced MRI within 7 days after mammographic examination. Dynamic MRI was performed with a 1 T system, using a three dimensional fast low angle shot (FLASH) pulse sequence before and after contrast media administration. We evaluated the morphologic features of the enhancement, the enhancement rate and the signal time intensity curve. Pathology was obtained in all cases. Results. The results of histopatological examination included: 15 DCIS and 7 DCIS with associated mi-croinvasive component or microfoci of invasive ductal carcinoma (IDC). On MRI, 21 of 22 (95%) DCIS lesions showed contrast enhancement. Fourteen out of 15 pure DCIS lesions demonstrated respectively a low (3), undeterminate (5), and strong (6) enhancement. Morphologically, the enhancing lesion was focal in 7, segmental in 4, and with linear branching in 3 cases. Wash out was found in 4 cases, plateau curve in 8 and Type I curve in 2 cases. Multifocality was present in 5 cases. All DCIS with associated microinvasion demonstrated contrast enhancement: 1/7 cases showed a low en­hancement, 2/7 showed an indeterminate enhancement and 4/7 showed a strong enhancement. Morphologically, the enhancing lesion was focal in 3/9, segmental in 5 and with linear branching in 1 case. The wash out was demonstrated in 3/7 cases, plateau curve in 3 and Type 1 curve in 1case. Multifocality was present in 3 cases. Conclusions. In conclusion, the sensitivity of MRI for DCIS detection is lower than that achieved for in­vasive breast cancer; however, contrast-enhanced MRI can depict foci of DCIS that are mammographically occult. The MRI technique is of complementary value for a better description of tumor size and detection of additional malignant lesions. Key words: breast neoplasms; carcinoma, ductal, noninfiltrative; carcinoma in situ, minimally invasive breast cancer; magnetic resonance imaging Correspondence to: Professor Massimo Bazzocchi, Istituto di Radiologia, APUGD via Colugna ni 50, Received: 7 June 2002 33100 Udine, Italy; Phone: +39 43 25 59 266; Fax: +39 Accepted: 17 June 2002 43 25 59 867; E-mail: massimo.bazzocchi@med.uniud.it Introduction Ductal carcinoma in situ (DCIS) is histologi­cally not considered as a single entity, but as a heterogeneous group of lesions that differ in their histopathologic features, growth pat­tern, clinical presentation and biological be­havior. Before the advent of widespread mammographic screening, DCIS was rarely detected and accounted for only 0.8%-5.0% of all breast cancers.1 With the introduction of mammographic screening, DCIS account­ed for 15-20% of all detected breast cancers, and for 25%-56% of all clinically occult can­cers.1,2 Seventy percent of DCIS presents as a cluster of microcalcifications; therefore, mammography is the primary and most sen­sitive technique to identify DCIS. Never­theless, in many cases, it is not accurate ei­ther in assessing the real cancer’s extent (un­derestimation of 46% of cases) or detecting multifocal lesions.3 The potential of magnetic resonance imag­ing (MRI) in the detection of DCIS is well doc­umented in many recent trials,4-6 where en­couraging data about the role of MRI have been shown. Unfortunately, these data are not always concordant with others in differ­ent studies, reporting of varying in sensitivi­ties. The explanation is likely related to the extreme variability in histologic features of a tumor, tumor size, tumor grade, different MRI parameters used, different technical fac­tors involved in performing breast MR imag­ing and image interpretation. The purpose of this study is to report our personal experience on 22 cases of DCIS studied with MRI. Patients and methods We retrospectively reviewed the MRI and mammographic (Mx) examinations per­formed from September 1995 to December 2001on 22 women (aged between 75 and 43 years, mean 53 years) affected by DCIS. Bilateral mammograms (mediolateral oblique and craniocaudal views) were ob­tained on a standard mammographic unit (Mammo DIAGNOST UC, Philips medical Systems Inc., Best Netherlands). In most cas­es, additional mammograms (e.g. spot views) were obtained in both projections. The following mammographic features were specifically assessed in each examination: -Focal nodular mass; -Microcalcifications: distribution, charac­ teristics, size, association with mass (sus­picious of malignancy if: clustered, pleo­morphic, mixed density or associated with mass or area of architectural distortion); -Associated features: architectural distor­tion, parenchymal distortion. Breast MRI was performed within 7 days from the Mx, with a 1 T system (Magneton Impact Siemens, Erlangen Germany), with a dedicated bilateral breast surface coil. A three-dimensional fast low-angle shot (FLASH) pulse sequence was used: 14 ms repetition time (TR), 7 ms echo time (TE), 25° flip angle, 2.5 mm effective slice thickness, 192x256 matrix, and 84 sec acquisition time. Images were acquired in coronal plane, with rectangular FOV (4/8). The entire breast was imaged before and five times after intra­venous injection of 0.1-mmol of Gd-DTPA/Kg body weight (Magnevist; Schering Berlin, Germany). The post-processing procedures included: -Digital image subtraction: subtraction of pre-contrast from the second acquisition of the post-contrast images. The subtrac­tion enables to obtain the signal suppres­sion of fat tissue and to identify the en­hancing areas (malignant lesions with neo­angiogenetic activity); -Maximum intensity projection (MIP) per­mits to obtain a 3D-image rotating on axial and on sagittal plane, based on the sub­tracted images; Radiol Oncol 2002; 36(4): 305-12. -Multiplanar reconstruction (MRP) permits to obtain axial images, based on the coro­nal acquisition. Semi-quantitative analysis of the signal in­ tensity to time relation was performed with the region of interest technique. The region of interest (ROI) (2-5 pixel) was placed within the tumoral area, where the highest signal in­tensity enhancement was seen. The percentage of signal intensity increase was defined as: SI Increase lesion = (SI post-SI pre)/SI pre x 100 SI = signal intensity; »pre« and »post« mean before and after contrast administration. With respect to enhancement kinetics, the enhancement rate that is referred to as a rela­tive signal intensity increase that occurs in a certain period of time (usually identified in the first contrast minute) was calculated. The optimal threshold value above which the enhancement level should be considered suggestive of malignancy is still debated. According to recent studies, a relative signal increase below 70% is usually considered as an index of no or minimal enhancement; a relative signal increase ranging between 70% and 140% is considered as intermediate, and a relative signal increase over 140% is consid­ered as strong.4 The three types of time-intensity curve, which have been previously described,7,8 were used: -Type 1 (continuous signal intensity in­ crease): a persistent increase in SI was present beyond 2 minutes after the con­trast media injection; -Type II (plateau): the maximum signal in­tensity was achieved in the first 2 minutes and then remained fairly constant; -Type III (wash out): the maximum signal intensity was achieved in the first 2 min­utes and went decreasing over time The morphology of the enhancement7 was classified as: -Focal enhancement corresponding to a well-defined mass; -Ductal enhancement corresponding to a linear/linear branching configuration, which can be related to a single enhancing duct (»galactogram«); -Segmental enhancement usually present in the lesions confined to the territory of a duct or of a ductal system, of a triangular shape with the tip pointing toward the nip­ple; -Regional or diffuse enhancement corre­sponding to the areas of confluent en­hancement that do not respect the borders of ductal system. In 6 patients, the histological diagnosis was obtained by surgical biopsy, following the placement of a mammorep under Mx or US guidance, while in the other 16 patients, histological diagnosis was provided by core needle biopsy. The patients underwent defin­itive surgical treatment within 14 days after MRI. Finally, in all patients, the definitive his­tological diagnosis was made from the patho­logic specimen (mastectomy or breast-conser­vative surgery). Results Pathology demonstrated the presence of DCIS in 15 cases (68%), and of DCIS with as­sociated microinvasive component or micro-foci of invasive ductal carcinoma (IDC) in 7 cases (32%). In 12 patients an isolate focus was diagnosed. Twenty-one out of 22 lesions (95%) showed enhancement after contrast media administration at MRI. DCIS The diagnosis of pure DCIS was made by histopathology in 15 lesions; 2 lesions were Radiol Oncol 2002; 36(4): 305-12. classified as comedo type and 13 as non-comedo type DCIS. Mammography demonstrated clusters of exclusively pleomorphic round and branch­ing microcalcifications in 12 cases (3 associat­ed with mass) (86%), and achitectural distor­tion or opacity in 2 (14%). One patient had negative mammograms, but ultrasonography (US) revealed abnormal findings. On MRI, 14/15 (93%) lesions revealed the uptake of contrast media, while 1 (7%) lesion was not identified at MR imaging. Nevertheless, Mx demonstrated microcalcifi-cations typical for malignancy in this false negative at MRI. In five out of the 14 enhanc­ing lesions patients were affected by multiple foci of DCIS. Among the 14 enhancing lesions, 3 (21%) showed a low, 5 (36%) an indeterminate, and 6 (43%) a strong enhancement (Table 1a). Morphologically, the enhancing lesion was focal in 7 (50%), segmental in 4 (27%) and with linear branching in 3 (21%) cases (Table 1b). Wash out was found in 4 (29%) cases, a plateau curve in 8 (57%), and type I curve in 2 (14%) cases (Table 1c). In detail, the 2 comedo type DCISs showed a morphological pattern (ductal or segmental) and a enhancement behavior, strong typical for malignancy. DCIS + DCI In 7/22 lesions (32%), histology detected DCIS with associated minimally invasive car­cinoma. Mx was abnormal in all these patients (100%) who were suspicious for microcalcifi-cations present in all cases (in one case, mi-crocalcifications were associated with opaci­ty). All lesions demonstrated contrast en­hancement at MRI: 1/7 (14%) showed a low enhancement, 2/7 (29%) an indeterminate en­hancement, and 4/7 (57%) a strong enhance­ment (Table 1a). Morphologically, the enhancing lesion was focal in 3/7 (43%) cases, segmental (Figure 1) in 3 (43%), and with linear branching in 1 (14%) case (Table 1b) (Figure2). Wash out was demonstrated in 3/7 (43%) cases, plateau curve in 3 (43%), and type I curve in 1 (14%) case (Table 1c). Three out of 7 cases (43%) presented mul­tiple foci of DCIS with associated microinva­sion (Figure 3). Table 1a. Enhancement rates in14 DCIS and 7 DCIS with associated minimum invasion. Percentage of signal intensity increase <70% 70%-140% >140% DCIS 3 (21%) 5 (36%) 6 (43%) DCIS+DCI 1(14%) 2 (29%) 4 (57%) Total 19% 33% 48% Table 1b. Enhancement configConfiguration uration in 14 DCIS and 7 DCIS with associated minimum invasion. Focal mass like Segmental Linear-branching Table 1c. Signal intensity curve Signal intensity curve 7 (50%) 4 (27%) 3 (21%) types in 14 DCIS and 7 D 3 (43%) 3 (43%) 1 (14%) CIS with associated minimum invasion. 48% 33% 19% Type i 2 (14%) 1 (14%) 14% Type ii 8 (57%) 3 (43%) 52% Type iii 4 (29%) 3 (43%) 33% Radiol Oncol 2002; 36(4): 305-12. Figure 1. Segmental enhancement. A 54 year-old asymptomatic woman. Mammogram of the left breast, in medio-lateral oblique view, did not show abnormalities (a). MRI revealed an area of segmental enhancement, well visible both in the coronal and in the MPR images (b,c). Histology proved DCIS with a focal area of minimally invasive component. Discussion Ductal carcinoma in situ is a heterogeneous group of histopathologic lesions, traditionally classified in two main subgroups, non-come-do (cribriform, micropapillary, clinging and solid), and comedo type, on the basis of the architectural growth pattern and cell type, and the presence or absence of comedo type necrosis within the ducts. Recently, the new pathologic classifica­tions proposed by Holland et al.9 distinguish among well-, intermediately-, and poorly dif­ferentiated DCIS subtypes on the basis of cy­tonuclear differentiation and architectural growth pattern. Silverstain et al10 have intro­duced the so-called Van Nuys classification in which the presence or absence of high nu­clear grade and the presence or absence of comedo type necrosis is considered. A new pathologic classification, consider­ing the biological behavior of DCIS and play­ing an important role in recognizing more ag­gressive lesions for an optimal management of DCIS, should be recommended. The value of mammography is well known seeing that 70% of DCIS become evident as a cluster of microcalcifications (usually linear, with linear branching or granular; Le Gal type IV-V), and that, in well-differentiated lesions, mammography is also very useful. Several investigations demonstrated that contrast enhanced MRI has a very high sensi­tivity in invasive breast cancer detection, with the values reaching 100%, whereas the sensi­tivities for the identification of DCIS on MR images have been reported to be variable, ranging from 33 to 100%.4 There are many po­tential explanations for the wide variability in reported sensitivities; according to one, it may be due to the differences in the size of DCIS lesions studied. It is obvious that MRI is Radiol Oncol 2002; 36(4): 305-12. Figure 2. Linear branching enhancement. Images of a 58-year-old woman with a retracted mammary nipple on the left side. Mammography, in medio-lateral oblique and cranio-caudal views, demonstrated an area of subtle archi­tectural distortion, better seen in the second projection (a, b). MPR image on axial plane revealed a typical linear branching enhancement pattern (c). Histology proved a pure DCIS. not able to visualize tumors smaller than the slice thickness due to partial volume effect. Nevertheless, the size of lesions does not seem to be the only explanation for the vari­able detection of DCIS with MR imaging. In fact, false-negative results with tumor sizes ranging from 2 mm to 9 cm have been report­ed. Another factor that could affect the sensi­tivity of MRI in DCIS detection is the histo­logical type of the tumor. However, the histo­logical subtype by itself is not sufficient to ex­plain the presence or absence of contrast en­hancement on MR images because false neg­ative cases of both comedo type and non-comedo type DCIS have been reported. The degree of tumor angiogenesis is another his­tological variable that could influence the sensitivity of MRI in the depiction of DCIS. It is well known that malignant lesions release angiogenetic factors (e.g. vascular endothelial growth factor, VEGF) that induce sprouting and growth of pre-existing capillaries, and the ex novo formation of new vessels (angio-genetic activity). In dynamic breast MR imag­ing, invasive breast cancer is detectable due to its strong enhancement, whereas a certain degree of angiogenetic activity seems to be a prerequisite for tissue invasion, and it is not needed as long as the tumor stands in the preinvasive state (in situ). While invasive growth is almost invariably associated with contrast enhancement, this is not necessarily true for the in situ cancers. In fact, a weak tu­mor angiogenesis, found in the stroma and around the ducts involved in DCIS, can ex­plain the lack of a significant enhancement behavior. In addition to the size of the lesion, histo­logical subtype and neoangiogenesis, differ­ences in MRI technique (2D section selected sequence, 3D volume sequence, 3D volume Radiol Oncol 2002; 36(4): 305-12. Figure 3. Multifocal, multicentric enhancement. Images of a 47-year-old woman. Mammogram of the left breast, in the medio-lateral oblique view, did not show any abnomality (a). MIP and several coronal MR images revealed multifocal and multicentric enhancement (b,c,d,e). The SI/T curve, relative to the lesion, presents a plateau-type curve (f). Histology proved DCIS with associated microinvasion. fat suppressed technique, dynamic tech­nique,...), and the differences in image inter­pretation (morphologic criteria to establish the degree of suspicion, threshold value of enhancement above which a lesion should be considered suggestive for malignancy) may also explain the variability in reported sensi­tivities. Mammography, whose morphologic crite­ria of suspicion in a detected lesion are well known, does not use any criteria similar to those used in MR imaging. The enhancement configuration of DCIS is variable: DCIS le­sions can present a focal, mass-like enhance­ment with ill-defined borders or can exhibit a linear or linear branching enhancement (duct-like configuration) or even a segmental enhancement with a configuration correspon­ding to a ductal system.2,7 In a recent study reported by Viehweg et al.4, MRI detected 96% of DCIS lesions that exhibited at least some week enhancement. Only 4% of DCIS lesions did not show en­hancement at all. The morphology of en­hancement was focal (73%), diffuse (10%) or ductal (17%). In 65% of cases, the speed of en­hancement was considered as delayed. In high-grade DCIS lesions (according to Van Nuys classification), comparing comedo sub­type to non-comedo subtype, the behavior of enhancement was more often ill defined (83% vs. 43%), ductal (29% vs. 12%), and rapid (50% vs. 29%). Significant differences in the en­hancement behavior were not found between high-grade and low-grade DCIS, nor were there any between comedo and non-comedo subtype lesions. Although 96% of DCIS showed the contrast enhancement on MRI, only 50% of DCIS lesions showed a »typical« enhancement behavior suspicious for malig- Radiol Oncol 2002; 36(4): 305-12. nancy characterized by a strong and early en­hancement, focal ill-defined enhancement or an enhancement with ductal configuration. Moreover, in the same work, MRI allowed the detection of 25 additional foci of DCIS. In our study, the majority (95%) of DCIS le­sions demonstrated at least some enhance­ment. The only false negative lesion at MRI was a pure non-comedo DCIS. On the other hand, MRI identified additional foci of DCIS that were mammographically occult in 8 cas­es. However, using the usual diagnostic algo­rithm, the enhancement rate was considered typical of malignancy, e.g. strong or at least indeterminate in 79% of pure DCIS and in 86% of microinvasive DCIS. The configuration of DCIS was variable: linear branching enhancement, which is con­sidered to be an important feature of malig­nancy, was present in only 21% of pure DCIS and in 14% of microinvasive DCIS. Early wash out, known as a typical feature of inva­sive malignancy was present in 29 % of pure DCIS and 43% of microinvasive DCIS. The non-enhancing DCIS was mammographically identified by the presence of microcalcifica­tions. MR imaging, however, may contribute to the diagnosis of DCIS by detecting the le­sions not visible on mammography. In conclusion, the sensitivity of MRI for DCIS detection is lower than that achieved for invasive breast cancer; however, contrast enhanced MRI can depict mammographically occult foci of DCIS. Mammography remains the main diagnostic technique for breast ex­amination. The MRI technique is of comple­mentary value for better description of tumor size, in the detection of additional malignant lesions, and in the study of the dense breasts, poorly visible by mammography. References 1. Stomper PC, Margolin FR. Ductal carcinoma in situ: the mammographer’s perspective. AJR 1994; 162: 585-91. 2. Orel SG, Mendonca ME, Reynolds C, Schnall MD, Solin LJ, Sullivan DC. MR imaging of ductal carci­noma in situ. Radiology 1997; 202: 413-20. 3. Holland R, Hendriks JH, Vebeek AL, Mravunac M, Schuurmans Stekhoven JH. Extent, distribution and mammographic/ histological distribution of breast ductal carcinoma in situ. Lancet 1990; 335: 519-522. 4. Viehweg P, Lampe D, Buchmann J, Heywang-Kobrunner SH. In situ and minimally invasive breast cancer: morphologic and kinetic features on contrast-enhanced MR imaging. MAGMA 2000; 11: 129-37. 5. Soderstrom CE, Harms SE, Copit DS, Evans WP, Savino DA, Krakos PA, et al. Three-dimensional RODEO breast MR imaging of lesions containing ductal carcinoma in situ. Radiology 1996; 201: 427­32. 6. Gilles R, Meunier M, Lucidarme O, Zafrani B, Guinebretiere JM, Tardivon AA, et al. Clustered breast microcalcifications: evaluation by dynamic contrast enhanced subtraction MRI. J Comput Assist Tomogr 1996; 20 (1): 9-14. 7. Kuhl CK. MRI of breast tumors. Eur Radiol 2000; 10: 46-58. 8. Kuhl CK, Mielcareck P, Klaschik S, Leutner C, Wardelmann E, Gieseke J, et al. Dynamic breast MR imaging: are signal intensity time course data useful for differential diagnosis of enhancing le­sions? Radiology 1999; 211: 101-10. 9. Holland R, Peterse JL, Millis RR, Eusebi V, Faverly D, van de Vijver MJ, et al. Ductal carcinoma in situ: a proposal a new classification. Semin Diag Pathol 1994; 11 (3): 167-80. 10. Silverstein MJ, Poller DN, Waisman JR, Colburn WJ, Barth A, Gierson ED, et al. Prognostic classifi­cation of breast ductal carcinoma-in-situ. Lancet 1995; 345: 1154-7. Radiol Oncol 2002; 36(4): 305-12. Radiol Oncol 2002; 36(4): 313-8. Magnetic resonance microscopy of trabecular bone Maria Cova1, Renato Toffanin2,3, Agostino Accardo4, Igor Strolka5, Cristina Furlan1, Roberto Pozzi-Mucelli1 1 Department of Radiology, University of Trieste, Ospedale di Cattinara, 2 Department of Biochemistry, Biophysics and Macromolecular Chemistry, University of Trieste, 3PROTOS Research Institute, Trieste, 4DEEI, University of Trieste, Trieste, Italy 5 Institute of Measurement Science, Slovak Academy of Sciences, Bratislava Background. Bone diseases such as osteoporosis lead to changes in the trabecular bone mass and architec­ture. Improved methods for the quantitative assessment of trabecular bone are needed to better understand the role of trabecular architecture in bone strength. MR microscopy (MRM), with its ability to achieve res­olutions below 50 µm, has proved to be particularly useful for the ex vivo evaluation of the complex archi­tecture of trabecular bone. In this study, we describe the use of projection reconstruction (PR) with MRM for the quantitative evaluation of the three-dimensional structure of trabecular bone explants and for the pre­diction of their biomechanical properties. Material and methods. High-resolution 3D PR and trabecular bone explants were analysed to determine standard morphologic parameters such as trabecular bone volume fraction (BV/TV or Vv), trabecular thick­ness (Tb.Th) and trabecular separation (Tb.Sp). Segmentation of the high-resolution images into bone and bone marrow was obtained by using a Bayesian approach. The derived parameters were finally included in non-linear mathematical models for the prediction of Young’s modulus (YM). Results. The parameters derived from the PR spin-echo were found to be stronger predictor of YM (R2 = 0.86) than those derived from the conventional spin-echo images (R2 = 0.75) used for comparison. Conclusions. This ex vivo approach should be readily adaptable to the studies in human subjects. Key words: bones - ultrastructure; magnetic resonance imaging, magnetic resonance microscopy; projec­tion reconstruction, Young’s modulus Received 25 October 2002 Accepted 4 November 2002 Correspondence to: Maria Cova, Department of Radiology, University of Trieste, Ospedale di Cattinara, Strada di Fiume 447, I-34139 Trieste, Italy; Phone: +39 (0)40 399 4372; Fax: +39 (0)40 399 4500; Email: cova@gnbts.univ.trieste.it Introduction Bone diseases lead to changes in the trabecu­lar bone structure that are not only charac­terised by reduction of bone mass but also by modifications of the bone architecture often accompanied by atraumatic fractures.1 Osteoporosis is nowadays a problem of pri­mary importance, which mainly affects women and elderly people. In Europe, the number of bone fractures related to osteo­porosis amounts to more than one million per year and this number is expected to rise dra­matically over the coming years because the percentage of elderly people that is already high is still increasing. As a consequence, there is a great need to develop accurate methods for the evaluation of the status of bone tissue in order to achieve an early diag­nosis of the disease and to determine the lev­el of fracture risk as well as to provide thera­peutic intervention in high risk patients and to monitor the effects of therapy. Several in­vestigations have indicated that, besides bone mineral density (BMD), also trabecular archi­tecture can be an important factor in assess­ing bone strength.2,3 The accurate in vivo observation of the mi-crostructure of the bone is today not feasible, but the use of magnetic resonance imaging (MRI) techniques in the area of osteoporosis appears very promising.4-6 Among these tech­niques, the most powerful methodology for the ex vivo study of trabecular bone is cer­tainly magnetic resonance microscopy (MRM), which also able to provide detailed information on the trabecular bone structure. Recently, we have proposed the use of short-TE projection reconstruction (PR) MR mi­croscopy for the study of healthy and osteo­porotic bone explants.7 The aim of this study was to verify the potential of the PR method in the characterisation of trabecular bone ar­chitecture and in the prediction of its me­chanical properties. Material and methods Sixteen specimens consisting of cylindrical bone plugs (Ø=4mm) were obtained from load-bearing and no-load-bearing regions of porcine humeral heads. The explants were ex­amined at 7.05 T using a Bruker AM300 in­strument equipped with a vertical wide-bore magnet and a microimaging accessory. All the explants were studied in the air using a 5mm diameter radio frequency (RF) coil. Short-TE proton MR microimages were ac­quired using a 3D spin-echo (TE = 3.0 ms, TR = 1.0 s) sequence according to the projection reconstruction (PR) method with constant gradient step and partial echo-acquisition al­ready described.8 This method provided im­ages with a final voxel resolution of 41×41×82 µm3. Spin-echo microimages (TE = 6.2 ms, TR = 1.0 s) were obtained on the same explants by the standard Fourier transform imaging method. The Young’s modulus was measured on other sixteen bone specimens excised from adjacent areas to those used for the ex-plants subjected to MR microscopy. After interpolation to obtain an isotropic voxel resolution, the main standard structur­al parameters such as trabecular bone volume fraction (BV/TV or Vv), trabecular thickness (Tb.Th), and trabecular separation (Tb.Sp) were derived from contiguous cross sections of binary images using the t3m software.9 Segmentation of the high resolution images into bone and bone marrow was obtained by adopting a Bayesian approach based on the Markov random field model where the likeli­hood function was locally adaptive.10 The de­rived structural parameters were finally in­cluded in non linear mathematical models for the prediction of Young’s modulus (YM) of the trabecular bone explants. Results Figure 1 shows a 2D section of a 3D PR mi-croimage of an intact specimen of porcine trabecular bone. The morphologic parameters obtained from the PR images were compared with those extracted from the standard FT im­ages. In Table 1, the morphologic parameters extracted from the PR and standard FT im­ages of the load-bearing bone specimens are reported, whereas Table 2 presents the corre­sponding data derived for the bone speci- Radiol Oncol 2002; 36(4): 313-8. mens excised from no-load-bearing regions. The best prediction of the mechanical proper­ties of the examined trabecular bone explants were obtained by using the following non lin­ear model derived from the equation pro­posed by Hwang et al.:11 YM=Vv*(a+b/Tb.Sp+c/Tb.Sp2+d/Tb.Sp3)+Tb. Th*(e+f/Tb.Sp+g/Tb.Sp2+h/Tb.Sp3)+k [1] which included simultaneously Vv, Tb.Th and 1/Tb.Sp. Figures 2 and 3 show the rela­tionships between the Young modulus meas­ured experimentally and that predicted using the structural parameters derived from the PR and standard FT images, respectively. The parameters extracted from the PR images were found to be stronger predictors of YM (R2 = 0.86) than those derived from the stan­dard FT images (R2 = 0.75). Discussion In this study, short-TE PR method was adopt­ed to minimise the susceptibility effect at the bone-marrow interface, which may lead to overestimation of the trabecular thickness as Table 1. Standard morphologic parameters estimated from projection reconstruction (PR) and conventional FT spin-echo images of porcine trabecular bone explants from load-bearing regions. PR method FT method Sample Vv Tb.Th Tb.Sp Vv Tb.Th Tb.Sp % mm mm % mmmm 1 0.680 0.289 0.136 0.720 0.399 0.155 2 0.796 0.329 0.084 0.789 0.428 0.115 3 0.697 0.266 0.116 0.754 0.444 0.135 4 0.714 0.281 0.113 0.559 0.287 0.227 5 0.662 0.215 0.109 0.738 0.356 0.126 6 0.613 0.217 0.137 0.542 0.225 0.190 7 0.609 0.219 0.141 0.664 0.301 0.152 8 0.558 0.193 0.153 0.539 0.202 0.173 9 0.695 0.264 0.116 0.637 0.241 0.137 Mean 0.669 0.253 0.123 0.660 0.320 0.157 S.D. 0.070 0.044 0.021 0.097 0.090 0.035 Table 2. Standard morphologic parameters estimated from projection reconstruction (PR) and conventional FT spin-echo images of porcine trabecular bone explants from no-load-bearing regions. PR method FT method Sample Vv Tb.Th Tb.Sp Vv Tb.Th Tb.Sp % mm mm % mmmm 1 0.481 0.144 0.155 0.275 0.103 0.270 2 0.532 0.197 0.174 0.268 0.119 0.326 3 0.379 0.119 0.195 0.425 0.150 0.203 4 0.359 0.128 0.229 0.433 0.126 0.166 5 0.470 0.162 0.183 0.349 0.110 0.205 6 0.394 0.146 0.225 0.459 0.144 0.169 7 0.440 0.152 0.193 0.556 0.213 0.170 Mean 0.436 0.150 0.193 0.395 0.138 0.216 S.D. 0.062 0.025 0.027 0.104 0.037 0.061 Radiol Oncol 2002; 36(4): 313-8. reported by Majumdar and co-workers.12 The results reported in Table 1 and Table 2 do not seem to show significant differences in Tb.Th for the PR and FT method. This is probably due to the fact that the morphologic parame­ters were computed on isotropic image voxels with a 41 µm3 resolution. However, the PR-derived structural data appear to be more ac­curate as their SD resulted lower than those calculated for the FT-derived values. Even though a spin-echo scheme was adopted in this study, we foresee that the PR method can also be of great advantage in the case of gra-dient-echo sequences. This implies that the PR method can be readily implemented on modern clinical MRI scanners. Moreover, our best model for the prediction of Young’s Figure 2. Experimental Young’s modulus versus predicted Young’s modulus. Predicted YM values were calculated including the morphologic parameters Vv, Tb.Th and 1/Tb.Sp derived from PR images in the equation 1 (R2 = 0.86). Radiol Oncol 2002; 36(4): 313-8. Figure 3. Experimental Youngis modulus versus predicted Young’s modulus. Predicted YM values were calculat­ed including the morphologic parameters Vv, Tb.Th and 1/Tb.Sp derived from conventional FT images in the equa­ tion 1 (R2 = 0.75). modulus can contribute to a more accurate in vivo evaluation of the mechanical properties of the trabecular bone. Since the risk of bone fractures seems to be strongly related to bone architecture, the described PR-based approach may provide a relevant contribution to the clinical MRI in­vestigation of trabecular bone in ageing and osteoporosis. Acknowledgements Work supported by grants from the University of Trieste, from the Commissariato del Governo nella Regione Friuli-Venezia Giulia (Fondo Speciale per la Ricerca Scientifica e Tecnologica, P. 63-1997) and from the Slovak Academy of Sciences (VEGA 2/2040/22). References 1. Melton LJ 3rd, Thamer M, Ray NF, Chan JK, Chesnut CH 3rd, Einhorn TA, et al. Fractures at­tributable to osteoporosis: report from the National Osteoporosis Foundation. J Bone Miner Res 1997; 12: 16-23. 2. Mosekilde L. Vertebral structure and strength in vivo and in vitro. Calcif Tissue Int 1993; 53: 121-6. 3. Odgaard A, Kabel J, van Rietbergen B, Dalstra M, Huiskes R. Fabric and elastic principal directions are closely related. J Biomech 1997; 30: 487-95. 4. Gordon CL, Webber CE, Christoforou N, Nahmias Radiol Oncol 2002; 36(4): 313-8. C. In vivo assessment of trabecular bone structure at the distal radius from high-resolution magnetic resonance images. Med Phys 1997; 24: 585-93. 5. Wehrli F W, Hwang SN, Ma J, Song H K, Ford J C, Haddad J G. Cancellous bone volume and struc­ture in the forearm: noninvasive assessment with microimaging and image processing. Radiology 1998; 206: 347-57. 6. Link T M, Majumdar S, Augat P, Lin J C, Newitt D, Lu Y, et al. In vivo high resolution MRI of the cal-caneus: differences in trabecular structure in os­teoporosis patients. J Bone Miner Res 1998; 13: 1175-82. 7. Toffanin R, Szomolányi P, Jellúš V, Cova M, Pozzi-Mucelli RS, Vittur F. Magnetic resonance mi­croscopy of osteoporotic bone. In El-Genk MS, ed­itor. Space Technology and Applications International Forum - 2000. American Institute of Physics; 2000. p. 295-9. 8. Jellúš V, Latta P, Budinsky L, Toffanin R, Jarh O, Vittur F. Projection-reconstruction method with constant gradient step. In: Proceedings of the 5th Annual ISMRM Meeting. Vancouver; April 12-18; 1997. p. 1987. 9. Hipp J, Jansujwicz A, Simmons C, Snyder B. Trabecular bone morphology from micro-magnet­ic resonance imaging. J Bone Miner Res 1996; 11: 286-92. 10. Jansen M. Wavelet thresholding and noise reduc­tion. PhD dissertation. Katholieke Universiteit Leuven, 2000. 11. Hwang SN, Wehrli FW, Williams JL. Probability-based structural parameters from three-dimen­sional nuclear magnetic resonance images as pre­dictors of trabecular bone strength. Med Phys 1997; 24: 1255-61. 12. Majumdar S, Newitt D, Jergas M, Gies A, Chiu E, Osman, et al. Evaluation of technical factors af­fecting the quantification of trabecular bone struc­ture using magnetic resonance imaging. Bone 1995; 17: 417-30. Radiol Oncol 2002; 36(4): 313-8. Radio/ Onco/2002; 36(4): 319-25. Real time compound ultrasound of the shoulder Alessandro De Candia1, Stefano Doratiotto1, Francesco Pelizzo1, Elio Paschina2, Massimo Bazzocchi1 1Department of Radiology, 2Department of Orthopedic Surgery, University Hospital of Udine, Udine, Italy Background. The purpose of the study was to determinate the value of the ultrasound real time compound imaging in the evaluation of supraspinatus tendon in subacromial impingement disease. Patients and methods. Preoperative ultrasound was performed on 180 shoulders in 157 patients with clin­ical suspicion of rotator cuff disease; 71 patients were surgically treated with acrimioplasty and cuff repair. The sonograms were obtained under static and dynamic examination using a 5-12 MHz high frequency lin­ear array probe, and compound real time elaboration of multiple images from different viewing angles (Sono C'J©-ATL). The supraspinatus morphology was classified into the following groups: the absence of tears, the presence of partial thickness tears and the presence of full thickness tears. In the absence of tears, the supraspinatus tendinopathy was classified into three classes according to the Neer stages. The ultrasound findings were compared to the surgical inspection results. Results. Ultrasound showed 32 (96.9%) out of 33 full thickness tears, only one false negative in a patient with large body, and 9 (75%) out of 12 partial thickness tears. There were three false negative studies in supraspinatus tendons with superficial lesions. Ultrasound correctly showed 26 (100%) out of 26 rotator cuffs with the absence of tear. Conclusions. Real time compound ultrasound is a useful tool in the evaluation of the supraspinatus tendon diseases; this technique permits the reduction of the presence of conventional ultrasound acoustic artefacts and provides necessary information for preoperative planning. Key words: shoulder joint -ultrasonography; rotator cuff; shoulder impingement syndrome -ultrasonog­raphy; real time compound ultrasound; sono CT Introduction Rotator cuff tears are usually described as Received 7 June 2002 common pathologic conditions we can meet Accepted 17 June 2002 while performing the ultrasonographic imag­ ing of the shoulder.1 Technical development Correspondence to: Alessandro De Candia, MD, in the last few years allowed ultrasonography Department of Radiology, University Hospital of to be considered one of the best imaging Udine, APUGD Via Colugna 50, 33100 Udine, Italy; methods for studying those conditions. The Phone: +39 43 25 59 266; Fax: +39 43 25 59 867; E-mail: alessandro.decandia@med.uniud.it newest full-digital equipments in fact allow to obtain a high quality of the ultrasonographic images as never before. High-resolution digi­tal ultrasonography allows increasing the sensitivity, the specificity and the overall di­agnostic accuracy in the shoulder lesions characterisation if compared to the conven­ 6 tional ultrasonographic methods. 2­ The aim of this paper is to describe the fea­tures of real time compound ultrasound tech­nique and to introduce its potential benefits on the supraspinatus tendon study compar­ing sonographic images to surgical results. Patients and methods Between January 2000 and December 2000, we performed a prospective study on 180 shoulders in 157 patients (69 males and 88 fe­males aged between 17 and 82 years; mean age of males 49.7 years; mean age of females 58.2 years) with clinical suspicion of rotator cuff tear. Out of 157 patients 71, aged between 34 and 80 years (31 males, mean age 51.2 years; 40 females, mean age 55 years) underwent surgery on the day after sonography. Surgical intervention on the supraspinatus tendon with tear repair and acromioplasty was per­formed on all the 71 patients by the same op­erator who recorded the data using a stan­dard method. The study was performed in a clinical setting with the approval of the Ethics Committee of our Institute. Sonograms were obtained with an ATL 5000 scanner, using a 7 to 12 MHz linear ar­ray probe (Advanced Technology Labora­tories, Bothell, W A) applying the soon CT dig­ital algorithm. This algorithm allows the digital real time image capturing along nine different lines of view at the same time for every point on the patient's shoulder on which the operator places the transducer. The nine virtual im­ages that the transducer receives back at once from the patient are weighted in real time by Radio/ Oncol 2002; 36(4): 319-25. a digital beamformer unit and melted into one final image which is the result of the nine different frames compound processing.7•8 In all patients, both the static and the dy­namic evaluations were performed. The static examination was performed on the patient's arm in standard position of the shoulder joint to evaluate the supraspinatus tendon as it is described in the first part of the dynamic evaluation. Dynamic images were obtained by moving passively the pa­tients arm in intrarotated position first and then in extended position: during the first part of the dynamic evaluation, the patient's arm was placed in the abducted and intraro­tated position, with the forearm flexed to the arm, and the back face of the fingertips point­ing to the tip of the scapula. The second part of the dynamic evaluation was performed by moving the limb passively in adduction\ex­tension but keeping the intrarotation of the forearm during the movement.9•10 All the im­ages of the 71 patients were stored digitally on the US unit. Two radiologists performed separately the image evaluation. Both the observers evaluat­ed each image set for the presence or absence of tendon tear and for the characterisation of the supraspinatus tendon features in the ab­sence of sonographic proof of the tear. The patients in whom ultrasonography re­vealed the presence of tendon tear were di­vided into two groups: the first group com­prised the patients with complete rupture of the supraspinatus tendon, while the second group included the patients with partial rup­ture of the supraspinatus tendon. In the absence of sonographic appearance of any type of tendon tear, the lesion grade based on the sonographic appearance of the tendon was established. We described ten­dons by grouping them into three sonograph­ically determined classes. The first class in­cluded the tendons which appeared to be thickened with a hyperechogenic surface, but showing a normal fibrillar central region. The subacromial bursa was thickened but without intratendinous calcifications. The second class grouped the tendons that showed to be thickened and had an irregular and inhomo­geneous aspect of their central fibrillar re­gion. This class included the shoulders with the thickened subacromial bursa and intra­tendinous calcifications. The third class com­prised the patients in whom the supraspina­tus tendon appeared diffusely hypoechoic, in­homogeneous and thinner. In this class, in­tratendinous calcifications were also present. The ultrasonograpic findings were com­pared to surgery outcome, the latter being considered as the gold standard technique in characterising the presence or absence of tear. Results Considering the 71 patients who underwent surgery, the real time compound imaging cor- Figure 1. Transverse image of the supraspinatus ten­don. Real time compound imaging shows a full thick­ness tear of the supraspinatus tendon with the deltoid muscle laying on the humeral head. Note the severe humeral head arthrosic degeneration. rectly identified and characterized 32 (96.9%) out of 33 surgically confirmed full thickness tears of the supraspinatus tendon (Figures 1, 2). The study of the rotator cuff performed on one obese patient was not able to lead to the final right diagnosis (false negative). There was not any false positive study. The real Figure 2. Transverse images of the supraspinatus tendon. A) Conventional ultrasound shows a small full thickness tear on the front side edge of the tendon (arrow); B) Real Tine Compound lmaging shows a remarkable increase in lesion conspicuousness and the broken tendon fibres as well (arrow). The cartilage surface is better displayed (arrowhead). Radio/ On col 2002; 36(4): 319-25. Figure 3. Transverse image of the supraspinatus ten­don. Real Time Sono er imaging is better showing the tendon structure focusing on the hypehogenic area of tendon suffering on the front side of the tendon (ar­rows) mixed to the superficial area of abrasive injury (arrowheads). time compound imaging identified correctly 9 of 12 (75%) partial thickness tears diagnosed by surgery. There were 3 false negative stud­ies (superficial partial thickness tears of the supraspinatus tendon on surgery). There were no false positive studies. The absence of tear was detected in 26 patients by ultrasono­grapy {Figure 3). All these cases were con­firmed by surgery as tearless tendons. In 5 (29%) out of 26 patients, a first-class ultrasonographic aspect of the tearless ten­don were described (3 women, two of them aged 39 and the third 79 years old; 2 men aged 41 and 51 years). Surgery on these 5 pa­tients described one case as degeneration of the supraspinatus tendon with no tear, one case as abrasive lesion of the tendon due to impingement syndrome disease with no tear, and three cases without any lesion either in the tendon or in the bursa. In the first case, the surgeon described a hypertophic subacro­mial bursa, in the second one, the bursa was oedematous and swollen, and in the last three cases, the bursa was void of lesions. Fifteen (57%) out of 26 patients had a sec­ond class-like tendon type on ultrasonogra­phy (9 men aging between 34 and 61 years; 6 women aging between 54 and 74 years). The surgical examination showed 2 cases of supraspinatus tendon degeneration, 1 case of oedema of the whole rotator cuff, 2 cases of hyperaemic cuff, 2 cases of swollen and hy­peraemic cuff; 2 cases of hemorrhagic cuff. There were 6 cases without any lesion of the cuff. The subacromial bursa was described as swollen with oedema in 3 cases, hypertrofic in 4 cases, hypertrophic and hyperaemic in 4 cases, void of lesions in 3 cases. Both sonog­raphy and surgery showed one or more calci­fications in 9 cases, while 6 cases did not have any calcification. In 6 (24%) out of 26 patients, ultrasonogra­phy detected a third-class tendon (2 men aged 50 and 62 years and 4 women aged be­tween 44 and 55 years). On surgery, the rota­tor cuffs were described as void of lesions in 2 cases, in the third patient, hyperaemic spot was detected on the rotator cuff (impinge­ment site), the fourth had globally hyper­aemic cuff, the fifth inflammation of the cuff, and the sixth degenerated cuff. Both sonogra­phy and surgery showed one or more intra­tendinous calcifications in 3 cases, while the other 3 cases did not have any calcification. The comparison between surgery and sono CT findings is shown on Table 1. Table 1. Comparison between sono er findings and surgery (gold standard) in shoulders with no tears, partial thickness tear and full thickness tear SURGERY No tear Partial tear ComElete tear Total No tear 26 3 1 30 SONO CT Partial tear 9 9 Complete tear 32 32 Total 26 12 33 Radio/ Oncol 2002; 36(4): 319-25. Discussion MRl has always played an important role in the identification and characterization of musculoskeletal diseases. 2•3 The technical drawback of the past conventional sono­graphic devices was limited possibility to reach the end of the diagnostic path using sonography by itself and forcing the radiolo­gist to get more hints from other imaging techniques because the specificity and sensi­tivity of ultrasound examination were too low. The ultrasonographic evaluation of the shoulder joint has been performed since late 80's to describe the rotator cuff tears. The sensibility of the sonographic images in those years was not comparable to the MRI ones.2•11-13 ln 1986, Middleton et al. per­formed a sonographic examination of the ro­tator cuff and described the drawbacks of sonography in evaluating the acute tears of the rotator cuff. The sensitivity and specifici­ty obtained by those Authors was lower than 90%. In late 80's, standard criteria to describe the tears of the rotator cuff were not encoded yet.12·14 In 1989, Miller et al. performed a study on 56 patients, but the sensitivity of 93%, the specificity of 53% and the overall ac­curacy of 77% of their study were still too low.t3 The technical development of the 90's with the introduction of the digital units allowed the image capture to be faster and rich in spa­tial details and time resolution with no loss of information and opened the possibility of ac­cessing the post-processing tasks.s,6,IO,I5,I6 Those new features allowed Moriggl and Steinlechner to encode the sonographic crite­ria to describe the features of normal rotator cuff tendonsY In our experience we noticed that, in par­ticular, the reduction of noise and the in­creased overall image contrast resolution, giv­en by the real time compound imaging tech­nique, allowed us to study the supraspinatus tendon with the highest resolution and to reach the best agreement between sonogra­phy and surgical findings. Our experience with Sono CT on 71 pa­tients demonstrates that the visualization of the rotator cuff tendon tears is possible with a level of accuracy that allowed us to estimate the features of the rotator cuff tendons and describe them in detail. The possibilities offered by the Sono CT al­gorithm permitted us to perform a precise measuring of the tendon thickness and of subacromial space, due to less image arte­facts, such as he speckle or the image blur­ring, which no more affected the image qual­ity. In our experience, it was also possible to describe in detail whether intratendinous cal­cifications and fluid collections in the sub­acromial bursa were present or not. In our experience, the full thickness acute tears were the lesions that we could better de­scribe because the borders of the tear into the broken tendon were always visible. It was al­so always possible to describe other signs of acute full thickness tear as peritendinous in­flammation or fluid collection in the subacro­mial bursa and obtain a good concordance with the surgical results. Partial thickness tears were visible, but it was not possible to describe any of them as well as the full thickness tears. Actually, par­tial lesions demonstrated to be better superfi­cial erosions than proper lesions of the ten­don and occurred strictly under the lower border of the acromion where the conflict was present. The drawbacks of the compound tech­nique are linked to those of sonography itself. In obese patients and in people with hyper­trophic musculature of the rotator cuff, the tendons are too deep to perform a correct di­agnostic study and the high frequency probes are obviously not able to visualize at best those deep structures. The shoulder is per­haps an anatomical district in which the op­erator can take the best advantage from the compound technique use. Radio/ 011col 2002; 36(4): 3 19-25. One of the major drawbacks of the com­pound technology is given by the reduction of the frame rate because of the delay in the multiple image processing. This reduction is sometimes a great limitation, in particular when deep structures are studied (i.e. study­ing the liver) because the image persistence is often the main artefact affecting the investi­gation. There are also other artefacts, like the image blurring that may slow down the ex­amination. This is not the case of the shoul­der investigation because the superficial posi­tion of the structures allows to proceed the study at a higher frame rate, maintaining the real time effect on.5•15•18 Conclusions The real time compound imaging technique is a full digital algorithm based on the digital beamformer unit. This unit permits the re­duction of conventional sonography acoustic artefacts because there is no analogic digital conversion during the image capturing process, but the system is all-digital. The pos­sibility to combine up to nine different digital images, captured in real time under nine dif­ferent view angles, into a single final image increases the overall image definition. In par­ticular, in the study of the rotator cuff, this new technique allows us to reach a better contrast resolution with smoother and more detailed images than conventional sonogra­phy ones. Those features, thanks to the new digital imaging capabilities, allow to obtain a good agreement between surgical findings and sonographic ones. References 1. King LJ, Healy JC, Baird P. lmaging of the rotator cuff and biceps tendon.] R Army Med Corps 1999. 145: 125-31. 2. Burk DL Jr, Karasick D, Kurtz AB, Mitchell DG, Radial Oncol 2002; 36(4): 319-25. Rifkin MD, Miller CL, et al. Rotator cuff tears: prospective comparison of MR imaging with arthrO!,'Taphy, sonography, and surgery. A]R Am] Roentgeno/1989; 153: 87-92. 3. D'Erme M, De Cupis V, De Maria M, Barbiera F, Maceroni P, Lasagni MP. [Echography, magnetic resonance and double-contrast arthrography of the rotator cuff. A prospective study in 30 pa­tients). [Article in Italian). Radio/ Med 1993; 86: 72­80. 4. Drakeford MK, Quinn MJ, Simpson SL, Pettine KA. A comparative study of ultrasonographic and arthrO!,'Taphy in evaluation of the rotator cuff. Clin Ortop 1990; 253: 118-22. 5. Fabis J, Synder M. [The sensitivity and specificity of sonographic examination in detection of rotator cuff tear). [Article in Polish). Chir Narzadow Ruchu Ortop Po/1999; 64: 19-23. 6. Farin PU, }aroma H. Acute traumatic tears of the rotator cuff: value of sonography. Radiology 1995; 197: 269-73. 7. Leotta DF, Martin RW. Three dimensional ultra­sound of the rotator cuff: spatial compounding and tendon thickness measurement. Ultrasound Med Bioi 2000; 26: 509-25. 8. Leotta DF. Three dimensional spatial compound­ing of ultrasound scans with weighting by inci­dence angle. U/trason Imaging 2000; 22: 1-19. 9. Martino F, Mocci A, Rizzo A, Dicandia V, Strada A, Macarini L, et al. [Echography of the supraspinatus tendon: forced passive adduction maneuver). [Article in Italian). Radio/ Med 1998; 95: 298-302. 10. Teefey SA, Middleton WD, Bauer GS, Hildebolt CF, Yamaguchi K. Sonographic differences in the appearance of acute and chronic full-thickness ro­tator cuff tears.] Ultrasound Med 2000; 19: 377-8. 11. Fabis J, Kowalska A. [The value of radiological ex­amination in the diagnosis of rotator cuff tear). [Article in Polish]. The value of radiological exam­ination in the diagnosis of rotator cuff tear. Chir Narzadow Ruchu Ortop Po/1999; 64: 447-52. 12. Middleton WD, Reinus WR, Melson GL, Totty WG, Murphy W A. Pitfalls of rotator cuff sonogra­phy. A]R Am] Roentgeno/1986; 146: 555-60. 13. Miller CL, Karasick D, Kurtz AB, Fenlin JM Jr. Limited sensitivity of ultrasound for the detection of rotator cuff tears. Skeletal Radio/ 1989; 18: 179­83. 14. Middleton WD, Reinus WR, Totty WG, Melson CL, Murph. Ultrasonographic evaluation of the ro­ tator cuff and biceps tendon. j Bone joint Surg Am 1986; 68: 440-5. 15. Teefey SA, Middleton WO, Yamaguchi K. Shoulder sonography. State of the art. Radio/ C/in Nort/1 Am1999; 37: 767-85. 16. Wallny T, Wagner UA, Prange S, Schmitt 0 , Reich H. Evaluation of chronic tears of the rotator cuff by ultrasound. Bone joint Surg Br 1999; 81: 675-8. 17. Moriggl B, Steinlechner M, Genser N. [Determining normal ultrasound values of the supra-and infraspinatus muscles]. [Article in German]. Ultraslzall Med 1993; 14: 52-7. 18. Thain L, Adler RS. Sonography of the rotator cuff and biceps tendon: technique, normal anatomy and pathology. j C/in Ultrasound 1999; 27: 446-58. Radio/ Oncol 2002; 36(4): 3 19-25. Radio/ Orzcol 2002; 36(4): 327-9. F-18-FDG PET in presurgical oro-maxillofacial carcinomas Reingard M. Aigner,1 Giinter Schultes,2 Gerald Wol£,1 Thomas Schwarz,l Mark Lorbach1 1 Department of Radiology, Division of Nuclear Medicine, 2Division of Maxillofacial Surgery, Karl-Franzens University Graz Background. We perfonned an analysis of the diagnostic impact of F-18-FDG-PET in presurgical oro-max­il/ofacial malignancies. Patients and methods. The diagnosis of the malignant primary was made clinically and was histological­ly verified before FOG-PET and the cervical CT examinations were perfonned in 25 patients of this study. For the FDG-PET investigation a full ring PET scanner was used (ECAT EXACT HR+, Siemens). Thoracic CT was performed only if pathological findings on FOG-PET scans required it. Results. The primary was clearly identified with FDG-PET in all patients. Active cervical lymph node sites were seen in 9/25 patients (ipsilateral: 8/25; ipsi-and contralateral: 1/25). Lung-metastases were found in 2/25 patients. Cervical CT: The primary was recognised in all patients. Artefacts caused by dental implants did not allow visualising the extension of the tumour in 9/25 patients. Ipsilateral lymph node sites were seen in 7/25 patients (size: 0.9-1.6 cm), and ipsi-and contralateral lymph node sites in 7/25 patients (size: 0.8-1.8 cm). The lung metastases primarily recognised with FOG were visualised with CT in both patients, too. Conclusion. FDG-PET is a sensitive diagnostic modality for the preoperative visualisation of active, i.e. sus­picious malignant lymph nodes. Distant metastases were demonstrated in 8% of the patients on whole body PET. The usefulness of FDG-PET and CT for establishing the diagnosis of the prim an; is limited. The pre­operative importance of er lies primarily in the accessibility of the algorithms for the intraoperative recon­struction of the facial structures. Key words: mouth neoplasms -diagnosis; maxillofacial neoplasms -diagnosis; tomography, emission ­computed, fluorine radioisotopes; F18-FDG, PET Introduction Received 25 October 2002 Diagnosis of maxillofacial malignancies re­Accepted 13 November 2002 mains a special problem for diverse radiolog­ical imaging modalities, even in the preopera­Correspondence to: Prof. Reingard M. Aigner, MD, tive stage. Artefacts caused by metallic Department of Radiology, Division of Nuclear crowns and dental implants reduce the diag­ Medicine, Karl-Franzens University Graz, nostic impact of magnetic resonance imaging Auenbruggerplatz 9, A-8036 Graz, Austria; E-rnail: reingard.aigner@uni-graz.at (MRI) and of computed tomography (CT). We analysed the diagnostic value of F-18-Fluoro­deoxy-glucose positron-emission-tomography (FOG-PET) as an alternative or additional method for preoperative evaluation of ora­maxillofacial malignant tumours. Patients and methods The series consisted of 25 patients, 17 males, and 8 females, aged between 38-69 years. The primaries were diagnosed clinically and his­tologically verified before FOG-PET and eT­examinations were performed. The serum glucose levels ranged between 70-110 mgldl. F-18-FOG was injected intravenously in a dose of 333 -370 MBq. The acquisition on a full ring PET scanner (ECAT EXACT HR+, Siemens, Medical Systems/CTI, Knoxville, USA) with an axial field of view of 15.5 cm re­sulting in 63 transverse slices with a slice thickness of 2.5 mm started between 70-90 minutes after injection. The transmission scans were obtained with 68-Ge rod sources (4 min acquisition time per bed position) al­ternating with the emission scans of 8 min each. The transmission data were recon­structed with filtered back projection, the emission data were corrected for random events, dead time, scatter and attenuation, and were reconstructed with ordered subset iterative reconstruction (OSEM; 2 iterations, 8 subsets). The resulting in-plane image reso­lution of transaxial images was about 4 to 5 mm fu!l with at half maximum. FOG-PET was done as whole body imaging method in all pa­tients. Cervical CT was done in all patients, too. Thoracic CT was performed only if pathological findings on FOG-PET scans re­quired it. Results FOG-PET: Intense localised pathological FOG-uptake indicating the primary was ob- Radio/ Oncol 2002; 36(4): 327-9. served in 25/25 patients. Intense pathological FOG-uptake in ipsilateral lymph node sites was seen in 8/25 patients, in ipsi-and con­tralateral lymph node sites in 1/ 25 patient. Lung-metastases were found in 2/25 patients. Cervical CT: The primary was recognised in all patients. Multiple artefacts caused by metallic crowns and dental implants did not allow visualising the extension of tumour in all patients. Ipsilateral lymph node sites were seen in 7/25 patients (size: 0.9-1.6 cm), ipsi­and contralateral lymph node sites were demonstrated in 7/25 patients (size: 0.8-1.8 cm). Radiologically unsuspected lymph nodes < 1 cm could be seen highly active on the FOG-PET study in 2/25 patients. On the other hand, lymph nodes > 1 cm were inac­tive on FOG-PET in 2/ 25 patients. Thoracic CT was done after FOG-PET in 2/25 patients: the lung metastases recognised with FOG were visualised with CT in both pa­tients, too. Discussion The majority of malignant ora-maxillofacial tumours are squamous cell carcinomas origi­nating in mucosal structures.1 The incidence of ora-maxillofacial malignancies is still in­creasing. The age of the patients is still de- Figure 1. F18-FDG-PET scan of a 69 year old woman with squamous cell carcinoma of the left maxilla, in­filtration of the left sinus maxillaris, left cervical lymph nodes, left supracervical lymph node bulk. creasing at time of clinical presentation. The earliest possible diagnosis and subsequent treatment planning are essential parameters for better prognosis of these patients. Management and prognosis depend on the age of the patient, local tumour invasion, presence of local lymph nodes (ipsi-and con­tralateral cervical lymph nodes) and distant metastases at time of clinical presentation. Preoperative staging is essential in order to choose the individual therapeutic regime.2·5 We analysed the diagnostic value of FDG­PET as an alternative or additional method for preoperative evaluation of oral-maxillofa­cial malignant tumours knowing some data from the scientific literature.6-8 All in all, we found FDG-PET to be a quite sensitive, non-invasive diagnostic modality for preoperative staging of patients suffering from ore-maxillofacial malignancies. FDG­PET proved to be favourable for the visualisa­tion of active, i.e. of suspicious malignant lymph nodes and for the detection of distant metastases in these patients. It is known that distant metastases are a rare condition at the initial staging of ore-maxillofacial malignan­cies. Nevertheless they were demonstrated in 8 % of our patients on the whole body PET-in­vestigations. On the other hand, the malignant primary is usually identified by means of clinical find­ings and confirmed to be malignant by biop­sy. Therefore the usefulness of FDG-PET for recognition of the malignant primary itself and for establishing the diagnosis is limited. The radiological imaging modalities monitor the malignant tumours and their metastases by the size and structural changes and not by metabolic activities. And there is of course the well-established preoperative importance of CT for determining the algorithms needed for the intraoperative reconstruction of the facial structures. The diagnostic impact of FDG-PET is of extraordinary diagnostic im­portance in patients with artefacts caused by metallic crowns and dental implants. References 1. Arnold WJ, Laissue JA, Friedman A, Naumann HH. Diseases of the head and neck, an atlas of histopathology. Stuttgart: Georg Thieme Verlag; 1987. 2. Rassekh CH. Tobacco cancer of the oral cavity and pharynx. W V Med] 2001; 97(1): 8-12. 3. Macluskey M, Ogden GR. An overview of the pre­vention of oral cancer and diagnostic markers of malignant change: 2. markers of value in tumour diagnosis. Dent Update 2000; 27(3): 148-52. 4. Goldenberg 0, Ardekian L, Rachmiel A, Peled M, Joachims HZ, Laufer D. Carcinoma of the dorsum of the tongue. Head Neck 2000; 22(2): 190-4. 5. Ferlito A, Mannara GM, Rinaldo A, Politi M, Robiony M, Costa F. Is extended selective suprao­mohyoid neck dissection indicated for treatment of oral cancer with clinically negative neck? Acta Otolaryngol 2000; 120(7): 792-5. 6. Kunkel M, Wahlmann U, Grotz KA, Benz P, Kuffner HO, Spitz J, et al. [Value of (F18)-2-fluo­rodeoxyglucose PET scanning in staging mouth cavity carcinoma. Comparative evaluation of PET findings before and after preoperative ra­diochemotherapy with histological and computer­ized tomography findings]. Mund Kiefer Gesichtschir 1998; 2(4): 181-7. [German]. 7. Stuckensen T, Kovacs AF, Adams S, Baum RP. Staging of the neck in patients with oral cavity squamous cell carcinomas: a prospective compari­son of PET, ultrasound, CT and MRI. J Craniomaxillofac Surg 2000; 28(6): 319-24. 8. Braams JW, Pruim J, Nikkels PG, Roodenburg JL, Vaalburg W, Vermey A. Nodal spread of squa­mous cell carcinoma of the oral cavity detected with PET-tyrosine, MRl and Cf. J Nucl Med 1996; 37(6): 897-901. Radial Oncol 2002; 36(4): 327-9.      ! ! ! ! 2 7 7 7 7 7 7 7 7 7 7 7 7 3 7 7 7 7 2 3 7 7 7 7 7 7 7 7 7 7    + +     + + + + + 8 2 8 8 + + + 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 3 8 8 8 2 8 +   + 3 8 8 8 3 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 3 8 + 8 8 8 8 8 8 8 8 8 &       % % % % % % % ) ) ) ) ) % & ) ) ) )  ) )  % ) &  ) )   ) ) ) > : > ) > > > > > > > > > > > > > > > > > > > > > > > > > > >  8 > > > > > > > > : > > : > ) %  %   $ $ 6 < $ < < < 6 < < 6 6 6 < < <  <  < < % %  "     $ $ $ < B B B B B B = 1 =   B B B B B > B 1 B B B # B # B B $ > B B B B B B B B B B , B B  " " "    ! ! & & & & C C C C C C C C C C C C C    C & C C "                                   .                      U                                                                  "  a a a a a a a a a a a a a a a a a a  a a a a a a a a a a a a a a a a  a a a a a a a a a a a a  a a a a a a  a a a a a a a a a a a a a a a a a 9  a a a a a a a a a a a a a a a a a  a a a a a a _  a  a a a a a a a a a a a a a a a a a  a a a a a X a a a a a a a a a 9 9    a a  a a a a a a a a a a a a a a a a a a a a a a a ^ a  " " a a a a a a a a X a a a a a a a a a 9 9     (  . . . . . . . . . . . . . . . . . . . . . . . . T ( . . . . . . . . . . . . . . . T ( . . . . . . . . . . . . . . . . . . . . . . T ( . . . . . . . . . . . . ( . . . . . . . . . . . . . . . T T ( . . . . . . . . . . . . . . . . . . T ( . . . . . . ( ( . . . . . # . . . . . . . ( ( . . . . . . ( . . . T . ( . . . . . . . . . . . . . . . . . . . . . . . . . . T . ( . . . . . . . . . . . . . . . . . . . . . . . . . .   6 * * 6 d d d d d d d d d d d d  * 6  d d d d d d ` d * d d d d d d d d d d d d d a d *    d d d d d d d d d d d d d d d d d = d *   d d d d d d d d d d d d d d d d d d d d * 6 6 d d d d d d d d d d d d d d d d d d d d d d = = 6  d d d d d d d d d d d d d d d d d d * 6  d d d d d d d d d d d d d d d d d d = * d d d d d d d d d 5 2 a d * d d d d d d d d d d d d d d d d d * d d d d d d d d d d d = *  6 6 d d d )   ) ) ) _ _ _ _ _ _ _ _   _ _ _ _ _  ) )  _ _  _ _ _ _ _  _ _ _ _ ?  ] _    ) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _  _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ) )   =   = =  = = = =  = = = =  = =   =  = =    =    =      =    =  =   =   = 6 = W W   W W W W  W W W W  W W W   W n  n n n n  n n n n n n n  n  n n n n n n  n n  n   n n n  n n n  n  n n n n n n n n      G G  G G G G G G     G   G  G G   G  G G G G G G  G  G G G G   G        G G G G G G G G G G  G G G    G  G G G G G G G  G G G G G   G G G G G G  G   G G G  G  G G G G G G G G G  G G G G G G G       G G    G G G G  G  G G G  G G G G G G     G  G G G G G G G G G     6  6 6 6 6 6 6 6 6  6 6 6  6 6 6 6 6   6 6  6 6 6    , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,       G G G G G G G G G G G G G A G G G G G G G G G B G G G G G G G B G B G  B G G G G   , , , , + +   # # + +  # +   #       ,    ,   C C C       a a Z Z   Z Z Z Z Z Z Z Z Z Z Z Z Z E Z Z Z    0 0 0  0              1  1 0 0 0 0  0  0  0 1 1 0 0 , 0 ,   0   0 1 1 0 0 0 1  0 1 0 0 1 1  1  0 0 0 0 1 1  0 0  0 0 1 1 0 0  0 0  0  0  1 0   9 9  ! / 9 9 9 9 9 9 9 9 9 9   9 9                  W   W W W W U W W W W W W W W W W W U U W W U W W W U W W W W W W W W W W W W W W U W W W W W W W W W W W W U W W W  W W W W W W U W  W W W W W W W W W W W W W  U 1 W W W W W 3 W W W W W I  W W W W U W W W W U U W W W W W W W W W W W W W U U W U U W W W W W W W  W W W W W W W W W W W W U W W W W W U U U W U W W W W W W W W W W W W W W U W W W W W W W W W W U W W W W W W W U W W W W W W W W W W W W W  U W H H H H H H H H H H G H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H G H H H H H H H H G H = H H H H H G H H G H H H H H H H H H H H H H H H H G H H H H H G G H H H H H H G H H H H G H H H H H H H H H H H H H H H H H G H H G H H H H H H H H H H H H H H G H H G H H H H H G H H G H H H H H H H H H G H H H H H H H H G H = = H H H H H H H H $ < H # H G H H H H H H H = = = H H : H G H G H G G H H H H H G H H H H H = = = = = = = =