The 5-HT3 antagonist d·evelo.ed Navohan.tropisetron) Always once a day. Always 5 mg. & SANDO.Z Sandoz Pharma Ltd, Pharma Basle Operations, Marketing & Sales, CH 4002 Basle/Switzerland RADIOLOGY AND ONCOLOGY Established in 1964 as Radiologia Iugoslavica in Ljubljana, Slovenia. Radiology and Oncology is a journal devoted to publication of original contributions in diagnostic and interventional radiology, computerized tomography, ultrasound, magnetic resonance, nuclear medicine, radiotherapy, clinical and experimental oncology, radiophysics and radiation protection. Editor in chief Tomaž Benulic Ljubljana, Slovenia Associate editors Gregor Serša Ljubljana, Slovenia Viljem Kovac Ljubljana, Slovenia Editorial board Bela Fomet Budapest, Hungary Marija Auersperg Tullio Giraldi Ljubljana, Slovenia Udine, ltaly Andrija Hebrang Matija Bistrovic Zagreb, Croatia Zagreb, Croatia Durila Horvat Haris Boko Zagreb, Croatia Zagreb, Croatia Berta Jereb Malte Clausen Ljubljana, Slovenia Kiel, Germany Vladimir Jevtic Ljubljana, Slovenia Christoph Clemm Munchen, Germany H. Dieter Kogelnik Salzburg, Austria Mario Corsi Udine, Italy Ivan Lovasic Rijeka, Croatia Christian Dittrich Marijan Lovrencic Vienna, Austria Zagreb, Croatia Ivan Drinkovic Luka Milas Zagreb, Croatia Houston, USA Gillian Duchesne Maja Osmak London, Great Britain Zagreb, Croatia Branko Palcic Vancouver, Canada ]urica Papa Zagreb, Croatia Dušan Pavcnik Ljubljana, Slovenia Stojan Plesnicar Ljubljana, Slovenia Ervin B. Podgoršak Montreal, Canada Miran Porenta Ljubljana, Slovenia Jan C. Roos Amsterdam, The Netherlands Horst Sack Essen, Germany Slavko Šimunic Zagreb, Croatia Lojze Šmid Ljubljana, Slovenia Andrea V eronesi Gorizia, Italy Publishers Slovenian Medica/ Society -Section oj Radiology, Croatian Medica[ Association -Croatian Society oj Radiology Affiliated with Societas Radiologorum Hungarorum Friuli-Venezia Giulia regional groups of S.l.R.M. (Italian Society of Medica[ Radiology) Correspondence address Radiology and Oncology Institute of Oncology Vrazov trg 4 61000 Ljubljana Slovenia Phone: + 386 61 1320 068 Fax: +386611314180 Reader far English Olga Shrestha Design Monika Fink-Serša Key words und UDC Eva Klemencic Secretaries Milica Harisc/1 Betka Savski Printed by Tiskarna Tone Tomšic, Ljubljana, Slovenia Published quarterly Bank account number 50101678 48454 Foreign currency account number 50100-620-133-27620-5130/6 LB -Ljubljanska banka d. d. -Ljubljana Subscription fee far institutions 100 USD, individuals 50 USD. Single issue far institutions 30 USD, individuals 20 USD. According to the opinion of the Government of the Republic of Slovenia, Public Relation and Media Office, the journal RADIOLOGY AND ONCOLOGY is a publication of informative value, and as such subject to taxation by 5 % sales tax . Indexed and abstracted by BIOMEDICINA SLOVENICA CHEMICAL ABSTRACTS . EXCERPTA MEDICAIELECTRONIC PUBLISHING DIV/SION· CONTENTS DIAGNOSTIC AND INTERVENTIONAL RADIOLOGY Persistent primitive trigeminal artery associated with intracranial aneurysms Klanfar Z, Lovrencic M, Kalousek M 89 Interventional ultrasound in renal transplantation Fuckar Ž, Mozetic V, Maricic A, Dimec D, Šustic A, Miletic D 94 Transcatheter embolization of congenital arteriovenous malformations of the limbs K6nya A, Vigvdry Z, Tasnddi G 98 ULTRASOUND AND COMPUTERISED TOMOGRAPHY CT and US in endocrine orbitopathy Barta M, Miletits E 106 Accuracy of preoperative CT scanning in staging of gastric carcinoma Roic G, Marolti M, Zovak M, Klaric R, Krolo I, Roic D 114 The role of ultrasound in conservative treatment of blunt hepatic injures Miletic D, Fuckar Ž, Uravic M, Dujmovic M, Šustic A, Mozetic V 119 ONCOLOGY Determinants of long-term survival in stage I non-small celi lung carcinoma (NSCLC) Gebitekin C, Olga<;; G, Satur CMR, Martin PG, Saunders NR, Walker DR 124 Results of nonoperative treatment for esophageal cancer Debevec M 129 Breast tumor aspiration biopsy with a multihole needle Vlaisavljevic V, Paja-Perušic D Stomach tumours following vagotomy and pyloroplasty Dujmovic M, Halaji-Laaby A, Uravic M, Lovasic F 138 Primary leiomyosarcoma of the liver Kraus I, Švalba-Novak V, Rubinic M, Ivaniš N, Uravic M, Škarpa A, Kovac D 141 REPORT Repo rt from the joint meeting of European society for radiation biology and European society for hyperthomic oncology Osmak M NOTICES 145 149 The publication of the journal is subsidized by the Ministry of Science and Technology of the Republic Slovenia. Inštitut za diagnosticno in intervencijsko radiologijo, KC Ljubljana; Klinika za otorinolaringologijo in maskilofacialno kirurgijo, KC Ljubljana; Klinicki zavod za dijagnosticku interventnu radiologiju, KBC Rebro, Zagreb; Onkološki inštitut, Ljubljana Radio/ Oncol 1994; 28: 89-93. Persistent primitive trigeminal artery associated with intracranial aneurysms Zoran Klanfar, Marijan Lovrencic, Miljenko Kal o usek Sestre Milosrdnice University Hospital Zagreb, Croatia Institute of Diagnostic and lnterventional Radiology Persistent embryological carotid-basilar anastomoses are rarely encountered in cerebral angiograms. Intracranial bleeding caused by an arteriovenous malformation or a ruptured aneurysm is the most frequent coincidental finding asociated with persistent carotid-basilar anastomoses. A case of a persistent primitive trigeminal artery associated with aneurysms of both posterior communicating arteries is presented in this study. Other embryonal anastomotic remnants, e. g. primitive otic, hypoglossal, proatlantal intersegmental and stapedial arteries, are briefly described. Key words: carotid arteries-abnormalities; basilar artery-abnormalities; cerebral arteriovenous malformations; cerebral angiography Introduction The pathoanatomic and especially angiographic evidence of persistent carotid-basilar anastomo­ses (CBA) appear to be rather rare. These anastomotic channels are the intersegmental arteries connecting longitudinal neural arteries (primitive basilar artery) with primitive interna! carotid artery (ICA) in a 4-mm stage embryo. They follow the course of the fifth, eighth and twelth cranial nerves and are therefore called primitive trigeminal (PTA), primitive otic/acou­stic (POA) and primitive hypoglossal arteries (PHA). After the vertebral-basilar arterial sy­stem and posterior communicating arteries Correspondence to: Prim. Zoran Klanfar, M. D., M. Se., "Sisters of Mercy" University Hospital, Depart­ment of Radiology and Oncology, 41000 Zagreb, Croatia. UDC: 616.133.33-007.24-073.75 (PCoA) have been formed at a 7 to 17-mm stage embryo, the CBA-s are getting gradually atrophic: first the otic followed by hypoglossal and then by trigeminal artery as last. l, 2• 3, 4 In the case of persistent function of these channels in an adult person, it appears to be a conse­quence of non-obliteration of these primitive anastomoses. The CBA-s are often associate with the intra­ 56 cranial haemorrhage.2• , This observation was mentioned in one of our previously published articles concerning four cases of PTA-s but cerebral angiography failed to demonstrate the cause of bleeding.2 This is our fifth case of PTA and second with subarachnoidal haemorrhage (SAH) in the last 23 years. Four-vessel cerebral angiography revealed two berry aneurysms of both PCoA-s. Case report T. B., a 33-year old woman complained of a sudden onset of unconsciousness. After reco­ Klanfar Z et al. very she experienced weakness, headache and nausea with vomiting. A blood stained cerebro­spinal fluid was found on admission, and CT demonstrated SAH. Cerebral angiography, per­formed with a catheter by femoral route de­monstrated an immediate filling of the posterior circulation from the right ICA. Its precavernous segment was connected with the 3 cm from primary tumor or along branches of celiac axis N3 lnvolvement of other abdominal nodes: paraaortic, hepatoduodenal, retropancreatic, mesenteric DIST ANT MET ASTASES (M) MO No evidence of distant metastases Ml Distant metastases present Source -modified from references 12 and 13 Table 2. TNM staging system. Stage TNM stage I Tl, NO, MO II T2 or T3, NO, MO III Tl-T3, Nl-N2, MO T4, NO-N2, MO IV Tl-T3, N3, MO, T4, any N, MO, any T, any N, Ml Source -modified from references 12 and 13 Sixteen cases of nodal involvement on the basis of lymph node enlargement were detected by CT. Due to a high rate of false-positive and false-negative result the statistical values, i. e. sensitivity (67 % ) and specificity (86 % ), were relatively low. CT showed pancreatic invasion in seven of 16 patients with pancreatic invasion present on surgery or pathologic examination, resulting in 116 Roic G et al. Discussion Table 3. Results of CT staging cornpared with surgical/ pathologic findings. Our results are not suggesting that computed SURGICAL STAGING CT I II III IV Tota! tomography should be a primary diagnostic I 4* 2" o o 6 examination in gastric carcinoma staging. 6* 5" 16 CT stage and laparotomy stage were in ac­ III o 2' 3* 6" 11 IV o 2' 2' 8* 12 Tota! 5 12 10 18 45 * CT data correctly staged disease in 21 patients (47%) " CT data incorrectly understaged disease in 17 pa­tients (38 % ) ' CT data incorrectly overstaged disease in 7 patients (15 %) a sensitivity of only 44 % and a specificity of 93%. Omental invasion was correctly predicted by CT in 2 patients, whereas in 9 patients CT failed to demonstrate direct omental invasion found on surgery resulting in a sensitivity of 18%. Out of 6 patients who had ascites diagnosed by CT scanning, two had peritoneal carcinosis during laparotomy. In one patient CT demons­trated infiltration of the aorta by the primary tumor, but it was not confirmed on subsequent surgical procedure. Table 5 shows the sensitivi­ty, specificity, positive and negative predictive values, and accuracy of CT staging parameters used in gastric carcinoma staging. cord in 21 patients ( 47 % ) . In 24 patients CT did not accurately assess the true extent of disease: in 17 patients (38 % ) it was underesti­mated and in 7 patients (15 % ) overestimated. In this series the accuracy of CT staging compared to surgical/pathologic findings was the best in TNM stage IV. Since gastric cancer is usually diagnosed as advanced 14 and, unfortu­nate!Jy, unresectable disease in approximately 50 % of patients,15 laparotomy is often unneces­sary, and preoperative CT scanning in these patients could be the most valuable (Figure 1). Most of these superfluovs exploratory laparoto­mies jeopardizing the quality of patients' lite could be avoided. CT also enables a surgeon to plan operative strategy since extensive gastric resection require great experience and knowledge. 16 In our study, CT did poorly in depicting adenopathy, with sensivity of 67 % (Figures 2 and 3). Some investigators have had better results with reported sensitivity of 97 % , 1 but also others had worse (Cook et al.) -49 % .4 A high rate of false-negative results is attribut- Table 4. Assessrnent of gastric carcinoma extent by CT scan (n = 45). SIGN OF DISEASE True positive True negative False negative False positive Undefined Serosal infiltration 15 20 8 2 o Lymph node enlargement 16 18 8 3 o Liver metastases 6 18 1 1 o Pancreatic infiltration 7 25 9 2 2 Omental invasion 2 31 9 o 3 Table S. CT staging versus surgical/pathologic staging (n = 45). STATISTICAL VALUES (%) SIGNS OF DISEASE Positive pre-Negative pre- Sensitivity Specificity Accuracy dictive values dictive values o Serosal infiltration 65 91 88 71 78 Lymph node enlargement 86 84 Liver metastases 86 97 86 76 96 Pancreatic infiltration 78 74 Ornental invasion 18 100 100 78 79 Accuracy of preoperative CT scanning in staging of gastric carcinoma able to the incapability of CT to detect micro metastases in normal-sized nodes. Some authors report on particular difficulties in the detection of lymph node metastases in some noble groups: nodes adjacent to the primary tumor mass, those in gastrohepatic ligament and the greater curvature lymph node chain. L? Possible causes of false-positive CT findings, as reported by Sussman et al., 12 are reactive hyperplasia or benigo lymph oode enlargement due to an adjacent inflammation or infection. In 6 patients of our series, CT examination proved !iver metastases with only one false-positive and one false-negative result (sensitivity of the method 86 % , specificity 97 % ) . Sussman et al. 12 repor­ted worse results, but it is important that in their investigations they did not use i. v. con­strast medi um, which would, presumably, re­duce the false-negative rate markedly. We would also like to emphasize the fact that in ali patients in our study the size of liver meta­stases was over 2cm. CT scan is of relatively low sensitivity in the diagnosis of direct invasion of neighboring organs by gastric cancer: pan­creas 44 % , omentum 18 % (Figure 4). In summary, our experience reported here indicates that CT based staging of gastric carci­noma can not be a substitute for surgical and pathologic staging. CT plays the main role in the preoperative diagnosis of advanced disease (TNM stage IV), i. e. in inoperable cases, when CT findings can be used to avoid unnecessary surgery in patients with minimal symptoms. Figure 4. Metastases in !iver parenchyma. Enlarged lymph nodes along the upper mesenterial artery. Thickened stomach wall. 118 Roic G et al. References l. Moss AA, Schnyder P, Marks W, Margulis AR. Gastric adenocarcinoma comparison of the accu­racy and economics of staging by computed tomo­graphy and surgery. Gastroenterology 1981; 80: 45-50. 2. Dehn TCB, Reznek RH, Nockler IH, White FE. The preoperative assessment of advanced gastric cancer by computed tomography. Br J Surg 1984; 71: 413-7. 3. Mason RC, Rankin S, Taylor PR, Rowe PH, Linsell J, Owen WJ, Jourdan MH, McColl l. Computerised tomographic scanning and staging of gastric carcinoma (letter). Lancet 1987; 1: 108. 4. Cook AO, Levine BA, Sirinek KR, Gaskill HV. Evaluation of gastric adenocarcinoma. Arch Surg 1986; 121: 603-6. 5. Fraser I, Nash R, James DC. Computed tomo­graphy in gastric cancer. Br J Surg 1985; 72: 249. 6. Komaki S, Toyoshima S. CT'capability in detect­ing advanced gastric cancer. Gastrointest Radio/ 1983; 8: 307-13. 7. McFee AS, Aust JB. Gastric carcinoma and the CAT scan. Gastroenterology 1981; 80: 196-8. 8. Terrier F, Schapira C, Fuchs WA. CT assessment of operability in carcinoma of the oesophagoga­stric junction. Eur J Radio/ 1984; 4: 114-7. 9. Gossios KJ, Tsianos EV, Demou LL, Tatsis CK, Papakostas VP, Masalas CN, Merkoulopulos MC, Kontogiannis DS. Use of water or air as oral contrast media for computed tomografic study of the gastric wall: comparison of the two techniques. Gastrointest Radiol 1991; 16(4): 293-7. 10. Beart AL, Roex L, Marchal G, Hermans P, Dewilde D, Wilms G. Computed tomography of the stomach with water as an oral contrast agent: technique and preliminary results. J Comput Assist Tomogr 1989; 13(4): 633-6. 11. Wegener OH. The )iver. In: Wegener OH ed. Whole Body Computed Tomography. Boston, London: Blackwell Scientific Publications, 1993: 244-8. 12. Sussman SK, Halvorsen RA, Illescac FF, Cohan RH, et al. Gastric adenocarcinoma. CT versus surgical staging. Radiology 1988; 167: 335-40. 13. Beahrs OH, Myers MH. Manual for staging of 2nd cancer. ed. Philadelphia: Lippincot, 1983. 14. Fielding JWL, Roginski C, Ellis DJ et al. Clinico­pathological staging of gastric cancer. Br J Surg 1984; 71: 677--80. 15. Leinster SJ, Hughes LE. The role of resection in advanced gastric carcinoma. Ciin. Oncol 1980; 6: 55-61. 16. Andaker L, Morales O, Hojer H, Backatrsrd B, Borch K, Larsson J. Evaluation of preoperative computed tomography in gastric malignancy. Sur­gery 1991; 109: 132-5. 17. Meigbow AJ, Balthazar EJ. Computed tomo­graphy of the gastrointestinal tract. St Louis: Mos­by, 1986. Radio/ Oncol 1994; 28: 119-23. The role of ultrasound in conservative treatment of blunt hepatic injuries Damir Miletic, 1 Željko Fuckar, 2 Miljenko Uravic, 2 Milivoj Dujmovic, 1 Alan Šustic, 3 Vladimir Mozetic4 Clinical Hospital Center Rijeka, 1 Clinical Institute of Radiology, 2Clinic for Surgery, 3 Department of Anesthesiology, 4 Ultrasound Diagnostic Unit, Croatia In the presented JO year period altogether 932 patients after blunt abdominal injury were examined and 65 traumatic lesions of the !iver were sonographically diagnosed; of these, 58 were directly visualized whereas in 7 patients only free abdominal fluid was presented. We have sonographicalty diagnosed 7 (21.1 % of directly visualized lesions) shaltow hepatic lacerations, 14 (24.1 %) subcap­sular hematomas, 26 (44.8 %) paripheral parenchymal hematomas and 11 (19 %) deep hepatic ruptures. Using ultrasound examination immediately after the injury as welt as sonographic foltow-up, we have conservatively cured alt shaltow hepatic lacerations, 86 % of subcapsular hematomas, 80 % of peripheral parenchymal !iver hematomas, while alt deep hepatic ruptures were laparotomized. Ultrasound yielded high sensitivity (87. 9 % ), specificity (100 % ) and accuracy (98. 9 % ) in visualiza­tion of blunt hepatic injuries. Key words: liver-ultrasonography; wounds, nonpenetrating; abdominal injuries; !iver, blunt injury, ultrasonography Introduction Because of its size, location and fixation, the !iver is frequently subjected to trauma. Next to the brain and spleen, it is the organ most commonly hit by blunt violence.1 Although the spleen is very often involved with a blunt abdo­minal trauma,2 !iver injuries have a greater clinical significance due to the vita) importance of this organ and very limited operative possi­bilities. Correspondence to: Damir Miletic, MD, MSc, Klinicki zavod za radiologiju, Klinicki bolnicki centar, Krešimi­rova 42, 51000 Rijeka, Croatia. The blunt injuries lead to ruptures or lacera­tions varying in size and sometimes in number. They are most commonly the result of traffic accidents or falls. Interna! stress or counter­coup effects during a blunt hepatic injury may set up subcapsular or central lacerations or only a subcapsular hematoma in case of mild im­pacts.1 The !iver is easily accessible to ultrasound examination due to its anatomic position. In blunt !iver injury the basic indicator is a hema­toma. 3 In the early stage it is a heterogeneous intraparenchymal zone. During the organisation of a hematoma, interna) echoes increase with retraction and diminish in size of a blood collec­ UDC: 616.36-001.3:534-8 tion. By using ultrasound, we can detect a very Miletic D et al. small amount of blood around the liver and in Douglas pouch.4 The sonographic appearance of blunt hepatic 57 injury has already been reported.3• -This study is aimed to examine the connection bet­ween the prognosis and particular sonographic pattern of hepatic injuries as well as the possi­bilities for conservative treatment of such inju­ries by means of ultrasound. Patients and methods A total of 932 patients with blunt abdominal injury were examined between January 1982, and January 1992, at the Ultrasound Diagnostic Unit of the Clinical Hospital Center Rijeka, Croatia. For the sonographic examination of the liver we used right subcostal and intercostal ap­proach. Other positions of the probe such as suprapubic, paraumbilical, left and right sub­costal as well as intercostal approach, were used for the visualization of peritoneal recesses. Abdominal pansonography was perfonned by means of the following equipment: Fisher Emi­sonic, Bruel&Kjaer 1486, Aloka SSD 260 LS and Hitachi EUB 515 with sector, linear and convex transducers of 3.5 and 5 MHz. Results From the total 932 ultrasonographically exami­ned patients with blunt abdominal injury, 277 had positive finding of a traumatic intraabdomi­nal lesion, while !iver injury was visualized in Table l. Sonography of blunt hepatic injuries. Type of injury Sonographic appearance 58 patients, i.e. in 20.9 % of ali positive ultra­sound findings. Table 1 presents the frequency of particular sonographic categories of hepatic injury. Shallow hepatic lacerations (Table 1) were presented as perihepatic blood collections with diameter of 2-5 cm and did not show a tendency to progress on ultrasound follow-up. After a physical examination, ultrasound was the only diagnostic and follow-up method during conser­vative treatment. With satisfactory clinical course these patients were discharged from hospital after 10-15 days. Subcapsular hepatic hematomas (Table 1) were mostly located near the visceral surface (Figure 1). The greater diameter of hematomas were less than 9 cm. After the initial volume compensation, all the patients were hemodyna­mically stabile. Hematoperitoneum did not exist and we began with conservative treatment. Du­ring the first 24 hours, in 2 patients (14 % of this group) control ultrasound examination sho­wed hematoma to have increased over 10 cm. 3rd 5th Between the and day after the injury hypotension and threatening hemorrhagic shock developed in both patients, while control sono­graphy performed at that tirne showed free liquid (blood) in Morrison's subphrenic and Douglas pouches as well as disappearance of the previously visible hematoma. The ultra­sound finding indicated secondary liver rupture. Explorative laparotomy was done (Table 2) and patients were cured with liver sutures. The most common sonographically detected blunt hepatic injuries were intraparenchymal hematomas (Table 1, Figure 2). At the tirne of Number % Shallow hepatic Hypoechoic collection near the hepatic contour, without 7 12.1 laceration sonographic evidence of hepatic lesion Subcapsular Hypoechoic blood collection which elevates Glisson's capsule 14 24.1 hematoma without interruption of its continuity Peripheral In fresh stage isoechoic or even hyperechoic, after 24 hours 26 44.8 parenchymal heterogeneous (mostly hypoechoic) focal lesion in the hepatic hematoma parenchyma distant from the hilus Deep hepatic Hypoechoic rupture line in the parenchyma with capsular 11 19.0 rupture interruption. and free intraperitoneal fluid Tota! 58 100.0 The role of ultrasound in conservative treatment of blunt hepatic injuries Table 2. Treatment of a blunt hepatic injury with respect to the sonographic appearance. Type of injury Urgent laparotomy Shallow laceration o Subcapsular hematoma o Peripheral parenchymal hematoma 2(7.7%)due to concomitant injury Deep rupture 11 (100%) Tota! 13 (22.4 % ) Figure l. Intercostal sector scan of subcapsular hepatic hematoma (arrow). H = !iver. hematoma imaging none of these patients pre­sented with free intraperitoneal fluid. Intrapa­renchymal hematoma in 21 patients (80.7 % of this group) measured less than 5 cm in the greatest diameter. In 4 patients (15.4 % ) the diameter was 5-lOcm and in one (3.8%) grea­ter than 10 cm. The patients with hematoma smaller than 5 cm were hemodynamically stabile ( except 2 concomitant splenic ruptures which required surgical treatment) and conservatively cured by regular ultrasound follow-up of hema­toma regression. Two patients with 5-lOcm large hematoma were successfully cured without surgery, using ultrasound follow-up. In other two patients the intrahepatic hematoma pro­gressed, and ultrasonography detected endoab­dominal effusion three and six days after the injury. These two patients as well as the patient with hematoma greater than 10 cm were laparo­tomized and successfully managed by surgery (Table 2). Sec. !iver Conservative rupture treatment o 7 (100%) 2 (14 % ) 12 (86 %) 3(11.5%) 21 (80%) o o 5 (8.6%) 40 (69%) Deep hepatic rupture (Figure 3) was sono­graphically diagnosed in 11 (19 % ) patients (Ta­ble 1). Ali the patients of this group were hemodynamically unstable and required imme­diate volume compensation. As many as 8 patients (72. 7 % of this group) had sustained multiple trauma (mostly a concomitant cranio­cerebral inj ury, thoracal inj ury or lesions of the extremities). In ali patients of this group ultra­sound examination detected abundant intraperi­toneal fluid (hematoperitoneum), interruption of the Glisson's capsule and hepatic parenchy­mal injury. In 4 patients (36% of this group) ultrasound examination presented segments of total destruction of the normal sonographic architecture of !iver parenchyma with hypo­echoic foci representing fresh hemorrhage, and Miletic D et al. m two of them (18 % of this group) even a massive central hematoma. After laparotomy, 5 patients ( 45 .5 % of this group) were treated by liver sutures, debridement or segmentecto­my, and rarely by lobectomy and/or perihepatic "packing", while 6 polytraumatized patients (54.5 % ) have succumbed. Figure 3. Subcostal sector scan of deep hepatic ruptu­re. From the total of 277 patients by whom ultrasound examination detected a traumatic lesion of one or more abdominal organs, in 58 (20.9 % ) !iver injury was directly visualized by ultrasound. In other 7 patients (2.5 % ) only free intraperitoneal fluid was sonographically presented and explorative laparotomy detected a !iver injury. Thus, directly and indirectly, we presented altogether 65 blunt hepatic injuries, which means 23.4 % of ali sonographically de­tected blunt abdominal injuries. Among ali false negative results of abdominal pansonography in blunt abdominal trauma, only one (0.2 % ) related to !iver injury. If we consider direct and indirect sonographic finding of blunt hepatic injury as a real positive result, the sensitivity of the method is 98.5 % , specificity 100 % and accuracy 99 .8 % . In view of exclusively direct sonographic visualization of blunt hepatic injury, sensitivity is 87.9%, specificity 100 % and accuracy 98.9 % . Discussion and conclusions We have diagnosed !iver lesions in 23.4 % of ali sonographically detected blunt abdominal injuries, which is in agreement with the data from literature.2 The relation between the di­rect and indirect ultrasound findings of !iver injury was abouth 8: l. Therefore, we have to insist on proper !iver scanning when free abdo­minal fluid is sonographically evident. Vollmer et al. consider that !iver injuries can be detected only indirectly ,8 which is opposite to our results. Ultrasound is required to determine the post­traumatic !iver status in most of blunt abdomi­nal injuries. Shallow hepatic lacerations with restricted perihepatic hematoma are not frequent hepatic lesions (Table 1). In ali such cases treatment has to be conservative, and the prognosis is favourable. Subcapsular hematomas represen­ted one fourth of ali !iver injuries (Table 1). Most of them can be treated without surgery, especially when the hematoma is smaller than 10 cm in diameter. Ultrasound follow-up is ne­cessary for early diagnosis -of secondary !iver rupture, which is a rare complication (Table 2). Almost a half of ali blunt hepatic injuries were presented as intraparenchymal hematoma. Ge­nerally, shortly after the trauma, sonographic diagnosis of intrahepatic hematomas may be difficult or even impossible, since fresh hemato­mas have the same echogenicity as normal !iver parenchyma. This impression changes after hours or if there has been no diffuse bleeding but an accumulation of fluid.2 In this group of our patients hematomas were situated on the !iver periphery. Most of them were less than 5 cm in diameter; these were cured conservati­vely (Table 2) provided that regression is follo­wed-up sonographically. In about a half of hematomas greater than 5 cm, conservative treatment would be possible on condition of stable hemodynamics. Deep hepatic rupture was found in 19 % of ali hepatic injuries (Table 1). These lesions were severe, and had doubtful prognosis. Severe lacerations or ruptures of the !iver have a high mortality rate. Death early after the trauma occurs due to severe hepatic The role of ultrasound in conservative treatment of blunt hepatic injuries hemorrhage which does not stop for severa! reasons: the walls of the valveless hepatic veins are thin, the !iver is extremely vascular and the diaphragm massages the !iver .1 Following our results, we can conclude that shallow hepatic lacerations, subcapsular hema­tomas and peripheral parenchymal hematomas of the !iver less than 5 cm represent low risk injuries. However, secondary !iver rupture is an exception which requires conservative treat­ment with regular ultrasound control. Intrahe­patic hematomas greater than 5 cm in diameter represent moderate risk injuries. They require an intensive clinical and ultrasound follow-up, while secondary !iver rupture is a real threat in the first two weeks after the injury. Finally, massive hepatic hematom as ( especially central), scaltered of a hepatic segment or lobe and rupture of the great hepatic veins are severe injuries with high mortality risk. Hoshi et al. reported a significant increase in GPT leve! in patients who sustained blunt !iver injury.9 Tanaka et al. found out that the use of ultrasound, significantly decreased the number of laparotomies5 owing to its precise determina­tion of intraperitoneal bleeding extent. A major emphasis should be laid on the exact preoperative diagnosis since unnecessary lapa­rotomies can entail a mortality of 0.25 % 10 and a morbidity of up to 4.9 % .11 We consider ultrasound to be a convenient diagnostic tech­nique for this purpose. High rate of direct sonographic visualization of blunt hepatic inju­ries, simplicity and promptness render ultra­sound a method of first approach to the patient with suspected blunt hepatic injury; this has also been confirmed by the results of other aut­hors.6· 7 References l. Netter FH. The CIBA collection of medica/ illu­strations 3 -digestive system III, 1979: 116-7. 2. Wening JV. Evaluation of ultrasound, lavage and computed tomography in blunt abdominal trauma. Surg Endosc 1989; 3: 152-8. 3. Bilic A. Atlas ultrazvuka u klinickoj gastroentero­logiji. Zagreb: Medicinska knjiga -Zagreb, 1989: 1-14. 4. Weil F. L'ultrasonographie en patologie digestive. Paris: Vigot 1983; 187-208. 5. Tanaka H, Sugimoto H, Yoshioka T, Sugimoto T. Study of blunt hepatic injury by ultrasonogra­phy. Nippon-Gekka-Gakkai-Zashi 1989; 90(6): 855-{i2. 6. Fusco G, Bufalari G, Lio G, Ricci R, Carli S, Tassi A. Our ultrasonic experience in the diagnosis of traumatic lesions of the !iver and spleen. G-Chir 1989; 10(9): 510-2. 7. Markiert R, Gryszkiewicz M, Dabrowski J, Stud­niarek M. The value of ultrasonography in the diagnosis of traumatic !iver injuries. Wiad-Lek 1989; 42(5): 327-9. 8. Vollmer K, Vollmer S, Almendiger G, Ulrich C, Schmidt E. Blunt abdominal trauma -sonographic findings. Schweiz-Med-Prax 1990; 79(4): 64-{i. 9. Hoshi H, Yuki Y, Kawano S, Kusuhura T, Ta­kechi Y, lwamoto I, Nakamura K. Computed tomography for the assessment of blunt abdominal trauma. Radio/ Med 1989; 7(5): 232-6. 10. Shah R, Martin H, Lewis M, Flint JR. Negative laparotomy: mortality and morbidity among 100 patients. Am Surg 1978; 3: 150-4. 11. Petersen RS, Sheldon FG. Morbidity of a negative finding at laparotomy in abdominal trauma. Surg Gynecol Obstet 1978; 148: 23-6. Radio! Oncol 1994; 28: 124-8. Determinants of long-term survival in stage I non-small cell lung carcinoma (NSCLC) Cengiz Gebitekin, Giiven Olgac;, Christopher MR Satur, Paul G Martin, Nigel R Saunders, Duncan R Walker Department of Cardiothoracic Surgery, Killingbeck Hospital, Leeds, V.K. It is well recognised that surgical resection currently provides superior survival in patients with stage 1 non-small cel! lung carcinoma (NSCLC). The determinants [cell type, type of operation, age, sex and tumour (T) characteristics J of long term survival were analyzed in 297 ( 40 % ) out of 735 patients with Stage I NSCLC. There were 222 (75 % ) males and 75 (25 % ) females with a ratio 3: 1. Age range was 40y-81y with a mean 63.2 years. Hospital mortality was 3.3 % (10 patients). The dominant celi type was squamous cell carcinoma [200 (67%) patients, adenocarcinoma in 71 (24%) and large celi carcinoma in 26 (9 %)] and lobectomy (including wedge resection and bilobectomies) was the treatment of choice in 218 (73.5 %) patients [pneumonectomy was performed in 79 (26.5 %)]. Age, sex, type of operation and cell type were not determinants of long term survival in this series of patients. The overall 5-year actuarial survival was 52 % and significantly influenced by tumour characteristics i. e. TI or T2 tumour (TI = 63 % vs T2 = 49 % , p = 0.003). Key words: carcinoma, non-small cell lung carcinoma, NSCLC; surgery; stage I; TNM classification, neoplasm staging; survival analysis Introduction Potentially curative surgical resection is stili the best treatment modality for non-small celi Jung cancer, comparing favourably with radiotherapy and chemotherapy which provide poor progno­sis at ali stages. Superior survival (as high as Correspondence to: PG Martin PhD, Principal Re­search Scientist, Killingbeck Hospital, York Road, Leeds, LS14 6UQ, England. Tel.: + 44.532.648164. Fax: + 44.532.326053. 1st Read at lnternational General Thoracic Surgery Meeting, 10-11 October 1992, Barcelona, Spain. UDC: 616.24-006.6-089-036 70 % )1 in patients with lung carcinoma has been achieved following surgical resection in Stage I disease. 2-4 The aim of the current study was to assess the effect of type of operation, age, sex, cell type and tumour (T) characteristics (Tl-T2 tu­mours) on long-term actuarial survival in pa­tients with stage I non-small cell lung carcinoma where staging was carried out according to the revised TNM classification (1986).5 Material and methods Seven hundred and thirty five patients treated surgically for non-small cell lung carcinoma at Determinants of long-term survival in stage I NSCLC the Regional Cardiothoracic Centre between 1980-1989 were reviewed and ali patients in postoperative stage I regardless of their opera­tion, age, sex and pathologic celi type were included in the study. Survival data were obtai­ned from the case and/or family practitioners records and were later confirmed by the Regio­nal Cancer Registry Office, Cookridge Hospi­tal, Leeds, UK. Hospital mortality was defined as death oc­curring within the unit at any tirne in the postoperative period and as such was included in actuarial life table analysis. For the purpose of review ali bilobectomies and wedge resect­ions were considered as lobectomies because of the small number of patients in these groups. Local and mediastinal lymph nodes were remo­ved and labelled to facilitate a correct postope­rative staging. Staging was carried out according to the revised TNM classification (1986)5 and was ba­sed on histopathologic examination of the resected specimen and sampled lymph nodes. None of the patients received any form of adjuvant radiotherapy or chemotherapy follo­wing Jung resection. Ali patients were followed at 6 weeks and at 3 monthly intervals following surgery for the first 1 year, and 6 monthly thereafter. Patients who died due to non-tu­mour related cause were included in long-term survival analysis. BMDP Statistical Software (BMDP Statistical Software Inc. Los Angeles, LA 90025) was used for ali statistical and actuarial survival analysis (Cutler and Ederer, 1958) and Genera­lised Wilcoxon test was performed for calcula­ting differential analysis between the groups. Chi-square testing was used for analysing diffe­rences in prevalence between groups. The sta­tistical difference between two groups was assu­med to be significant at p < 0.05. Results During the 10-year period (1980-89), 735 pa­tients were operated on for non-small celi lung carcinoma. Pathological staging revealed, 297 ( 40. 5 % ) patients out of the total were in stage I disease. There were 222 (75 % ) males and 75 (25 % ) females with a ratio 3 : 1 whereas accor­ding to tumour (T) characteristics it was 1.5 : 1 in the group of patients with Tl tumour and 3: 1 in the others (T2). Age range at the tirne of diagnosis was 40y to 81y with a mean of 63.2 years. There were 223 (75 % ) patients under the age of 70 and 74 (25 % ) patients 70 years of age or older. Demographic data is shown in Table l. Table l. Demographic and operative characteri­ stics of patients undergoing surgery for NSCLC ­ comparison of Group 1 and Group 2. Age 69 or younger 223 (75)* 70 or older 74 (25) Sex Male 222 (75) Female 75 (25) Ratio M:F Operation Pneumonectomy 79 (26.5) Lobectomy** 218 (73.5) Celi type Squamous Celi 200 (67) Adenocarcinoma 71 (24) Large Celi Carcinoma 26 (9) * N umbers in paranthesis are percentages. ** Including 19 wedge resections, 11 bilobecto­ mies. Standard pneumonectomy was performed in 79 (36 % ) and lobectomy in 188 (54 % ) patients. Wedge resection (19 patients) or bilobectomy (11 patients) was also carried out in 30 (10 % ) patients. However, owing to small number of cases, these patients were included in the lobec­tomy group. Pneumonectomy was the treatment of choice in 77 (34 % ) patients with T2 tumour however, two (2.5 % ) patients with Tl turno ur underwent completion pneumonectomy due to unresolved postoperative atelectasis. Ten (3.3 % ) patients died in the postoperative pe­riod due to various complications during the hospital stay. Following the histopathological examination of the resected specimen, squamous celi carci­noma was the dominant celi type and was diagnosed in 200 (67 % ) patients whereas ade­nocarcinoma was diagnosed in 71 (24 % ) and large celi carcinoma in 26 (9 % ) patients: TNM classification of the tumour was carried out Gebitekin C et al. 100 80 ti' 60­ :f . 40­ 20 -, e,..,_, O .,.,,.._, 5 Ytars Figure l. 5-year actuarial survival according to type of surgery. 100 80 ti' 60 :f . 40 20 SJ, .... ...., ... 'A.AHtt«•n/11a..,. 0 4 Ytars Figure 2. 5-year actuarial survival according to celi type. according to final histopathology report of the resected specimen and Jymph nodes which re­vealed that 73 (24.5 % ) patients were diagnosed having T1 and 224 (75.5 % ) patients T2 Jung tumour. Following discharge, the cause of death was non-tumour related in 11 (3.7 % ) patients and these deaths were included in actuarial life table analysis. Overall 5-year acturial survivaJ was 52 % . Superior survivaJ was achieved follo­wing Jobectomy compared to pneumonectomy but the difference was not statistically signifi­cant, p = 0.14, Figure l. SurvivaJ based on cell type was aJso superior in favour of the patients with squamous cell carcinoma (56 % ) however once more this did not differ significantly from patients with adenocarcinoma (44 % ), Figure 2. SurvivaJ in patients with Jarge cell carcinoma was 47 % however no comparison was carried out due to small number of patients in this group therefore it was not included in Figure 2. In addition, age was associated with impro­ved survivaJ in patients Jess than 70 years of age compared to patients 70 years of age or older ( 56 % vs 42 % ) , however the difference was not significant, p = 0.11. Although the favourable survivaJ (63 % ) was achieved in te­male cohort compared to male gender ( 48 % ) , the difference was not quite significant, p = 0.06. In contrast to these previous findings, significantJy better survival was observed in the group of patients with Tl tumour comparing to patients with T2 tumour ( 63 % vs 49 % ) , p = 0.003, Figure 3. Discussion The effectiveness of pulmonary resection in achieving Jong term survivaJ in the treatment of stage I non-small cell Jung cancer has been reported by severa] authors. 1-4 In contrast, ra­diotherapy alone or in combination with che­motherapy has not been shown to be as effect­ive as surgicaJ therapy under similar circumstan­ces. Sobue et aJ6 reported 14.3 % 5-year survivaJ in patient with clinical stage I lung cancer treated non-surgically. SimiJarly, SandJer et al. 7 reported 17 % three years survivaJ in patients with stage I lung tumour treated with megavoJ­tage radiotherapy. In addition, Haffty et al. 8 showed 21 % , 5-year survival in patients with IOOT. @TJNO ·•TJNO 90 *OVEJULL 80 -, 70 ­ # 60 .• :r so-, . 40 ,,. 30­ 20""" 10­ O 4 S Y,,rn Figure 3. 5-year actuarial survival according to tumour characteristics. Determinants of long-term survival in stage I NSCLC clinical stage I non-small celi lung carcinoma felt to be surgically resectable but treated with radical radiation therapy either for medica! reasons or due to failure to obtain patient's consent for surgery. Emami et al.9 and Cape­well and Sudlow10 reported 29 % and 17 % , 5-year survival with radical radiotherapy, re­spectively. These results reveal radiotherapy to be inferior to the surgical therapy in patients with Stage I NSCLC reported previously. The high male : temale ratio observed in this group could be explained by the increased incidence of squamous celi carcinoma thought to be associated with smoking habits. However, the ratio observed in patients with Tl tumour was 1.5 : 1 could be explained by the fact most Tl tumours are diagnosed coincidentally on routine chest roentgenogram. In this study overall 5-year actuarial survival (including 11 non-tumour related deaths) was 11 52 % as similarly observed by the others.3• Type of surgery performed was not determinant of the 5-year actuarial survival although survival was generally superior in patients treated by lobectomy as similarly observed by Little et al.4 In contrast to the findings of Williams2 et al., we failed to demonstrate a statistically signifi­cant difference in survival probability between patients of less than 70 years of age and those older than 70 years although the trend was for poorer survival in the latter group of patients (56 % vs 42 % , p = 0.11). Similarly superior survival was observed in female group, however the difference was not statistically significant ( 48 % vs 63 % , p = 0.067) as observed similarly 13 by the others.2• Vincent et al.14 and Naruke et al3 have repor­ted, with the exception of small celi carcinoma, absence of significant difference in survival bet­ween the different celi types. Although patients with adenocarcinoma generally do worse than those with squamous celi carcinoma, considered 11• 12 by stage they have similar prognosis.2• This observation was also confirmed in the present study. Deneffe et al. 15 reported a better 5-year survival for squamous celi carcinoma as compared with adenocarcinoma for any given stage, but unlike with disease stage, the progno­stic significance of histopathologic celi type per se, was not clearly defined. In contrast to the previous findings, the tu­mour characteristics (T-Factor as defined by the new TNM classification) was the singular most important factor affecting 5-year acturial survival in both patient groups TlN0 = 63 % , TINO = 49 % , p = 0.0033. Significant differen­ces in 5-year survival rates according to Tl and T2 has also been reported by Naruke et al.3 (76.4% vs 56.9%), Read et al.11 (73.2% vs 49 .5 % ) and also Pairolero et al.1 (70 % vs 58.2%). In conclusion, this review has confirmed that the Iong term survival in the stage I non-small lung celi carcinoma is not a function of age, sex, type of operation, or the celi type but is more Iikely of the T-characteristics of the tu­mour. On the basis of the observed significant survival difference and the findings of the other authors we would strongly recommend sub-clas­sification of stage I NSCLC. Acknowledgements We thank Dr. Canan Gebitekin, Mrs. Leslie Rider and Mrs. Leslie Sampson for their inva­Iuable help in data retrieval. This work was generously supported by Children's Heart Sur­gery Fund, Killingbeck Hospital, Leeds, UK. References l. Pairolero PC, Williams DE, Bergstralh EJ, Pue­hler JM, Bernatz PE, Payne WS. Postsurgical Stage I Bronchogenic carcinoma: Morbid implica­tions of recurrent disease. Ann Thorac Surg 1984; 38(4): 331-8. 2. Williams DE, Pairolero PC, Davies CS, Bernatz PE, Payne WS, Taylor WF et al. Survival of Patients surgically treated for Stage I Jung Cancer. I Thorac Cardiovasc Surg 1981; 82: 70--6. 3. Naruke T, Goya T, Tsuchiya R, Suemasu K. Prognosis and survival in resected Jung carcinoma based on the new international staging system. J Thorac Cardiovasc Surg 1988; 96: 440-7. 4. Little AG, De Meester TR, Ferguson MK, Skin­ner D, Hoffman PC, Skosey C, et al. Modified stage I (TlN0M0, TIN0M0), non-small celi Jung cancer: Treatment results, recurrence patterns, and adjuvant immunotherapy. Surgery 1986; 100(4): 621-7. Gebitekin C et al. 5. Mountain CF. A new international staging system for lung cancer. Chest 1986; 89 (suppl): 225-33. 6. Sobue T, Suzuki T, Matsuda M, Kuroishi T, Ikeda S, Naruke T. Survival for clinical stage I lung cancer not surgically treated. Cancer 1992; 69(3): 685-91. 7. Sandler H, Curran WJ, Turrisi T. The influence of tumour size and pretreatment staging on out­come following radiation therapy alone for stage I non-small cell lung cancer. lnt J Radiation Onco­logy Biol Phys 1990, 19: 9-13. 8. Haffty BG, Goldberg NB, Gerstley J, Fischer DB, Peschel RE. Results of radical radiation therapy in clinical stage I, technically operable non-small celi Jung cancer. Jnt J Radiat Oncol Biol Phys 1988; 15: 69-73. 9. Emami B, Kirn T, Roper C, Simpson JR, Plepich M, Hederman MA. Postoperative radiation the­rapy in the management of lung cancer. Radiology 1987; 164(1): 251-3. 10. Capewell S, SudJow MF. Performance and prog­nosis in patient with lung cancer. Thorax 1990; 45: 951-6. 11. Read RC, Schaefer R, North N, Walls R. Diame­ter, Celi type, and survival in stage I primary non-small celi Jung cancer. Arch Surg 1988; 123: 446-9. 12. The Jung cancer study group: Postoperative TINO non-small celi lung cancer. J Thorac Cardiovasc Surg 1987; 94: 349-54. 13. Martini N. SurgicaJ treatment of non-small celi lung cancer by stage. Semin Surg Oncol 1990, 6: 248-54. 14. Vincent RG, Takita H, Lane WW, Gutierrez AC, Pickren JW. SurgicaJ therapy of lung cancer. J Thorac Cardiovasc Surg 1976; 71: 581-91. 15. Deneffe G, Lacquet LM, Verbaken E, Vermant G. Surgical treatment of bronchogenic carcinoma: a retrospective study of 720 thoracotomies. Ann Thorac Surg 1988, 45: 380-3. Radio/ Oncol 1994; 28: 129-33. Results of nonoperative treatment for esophageal cancer Miha Debevec Institute of Oncology, Ljubljana, Slovenia Fram 1983 to 1987, 152 patients with esophageal cancer were treated at the Institute of Oncology, Ljubljana, Slovenia. Ninety-eight of these had radiation therapy alone, 36 radio-and chemotherapy, whereas 18 patients were treated symptomatically only. One, two and five-year survival of 69 irradiated patients with TD> 45 Gy was 32 %, 13 %, and 5 % respectively. There was a significant difference in survival according to the tumor dose delivered (i. e. > 45 Gy or < 45 Gy ), length of tumor stenosis, and performance status. Chemotherapy, tumor site, duration of dysphagia, and sex had no influence on the survival. There was no difference in survival between patients treated symptomatically and those irradiated with TD< 45 Gy. The effect of radiotherapy and chemotherapy on dysphagia was poor: in only 1/5 of the patients the improvement lasted more than two months whereas in 2/5 of the patients dysphagia worsened already after two months. It seems reasonable to restrict radiotherapy only to patients with radical intent, these being chiefly the patients in good general condition and with short tumor stenosis. Key words: esophageal neoplasms-therapy; radiotherapy, antineoplastic agents; survial analysis Introduction The prognosis for patients with esophageal can­cer is poor. Irrespective of therapeutic modali­ty, only few patients survive five years after diagnosis. It is important to choose such a therapy, that would influence the survival and diminish dysphagia at minimal hospitalization tirne. Standard therapies for esophageal cancer are surgery and radiotherapy; recently, the use of chemotherapy has been reported as well. The aim of this article is to present the results of nonoperative treatment of 152 patients with Correspondence to: Prof. Miha Debevec, MD, PhD, Institute of Oncology, 61105 Ljubljana, Zaloška 2, Slovenia. UDC: 616.329-006.6-036:615.849 esophageal cancer at the Institute of Oncology, Ljubljana, Slovenia, in the period 1983-1987. Methods and patients Of the 152 patients, 136 were male and 16 female, i. e. the sex ratio 8.5: l. The age of patients ranged from 35 to 86 years; the highest incidence was in the age group 50-65 years. Tumor was microscopically confirmed in 141 patients. There were 126 epidermoid carcino­mas, 7 adenocarcinomas, 6 undifferentiated and 2 small-cell carcinomas. Site of primary tumor was as follows: 10 cervical, 30 upper, 76 middle and 36 lower thoracic region. According to clinical TNM classificiation of 130 Debevec M esophageal carcinoma valid in this tirne period there were 13 (9 % ) Stage I, 50 (33 % ) Stage II, 55 (36 % ) Stage III, and 34 (22 % ) Stage IV tumors. Twelve patients had fistuals, 11 bron­chial or tracheal infiltration, and 2 invasion to the aorta. Palliative surgical procedures were performed in 15 patients: 8 gastrostomies, 5 tubus insertion, and 2 by-pass. Three patients had a naso-gastric tube inserted for the needs of nutrition. Performance status (Karnofsky) was as fol­lows: > 70 in 106 patients, 50-70 in 33 patients and < 50 in 13 patients. The length of esophageal stenosis was estima­ted in 120 patients: up to 5 cm in 26, 5-10 cm in 74, and over 10 cm in 20 patients. The duration of dysphagia was as follows: 1 month in 15, 1-3 months in 61, 3-6 months in 36, and over 6 months in 24 patients. For 16 patients there were no reliable data on the · duration of dysphagia. Radiotherapy was performed by a linear ac­celerator (x ray, 8 or 10 MeV), daily doses ranged between 1.5 and 3 Gy with two opposite or three planned fields, and maximum equiva­lent TD 70 Gy/7 weeks, mostly in split course regimen. Chemotherapy consisted of 5-FU 1.000 mg/m2 h in 24infusion, on days 1-4, and cisplatin 90 mg/m2 on day 1, repeated 1-4 times. Ninety-eight patients were treated by radiothe­ rapy alone, 36 by radiotherapy and chemothe­ rapy, and 18 symptomatically only. 1.0 N = 152 . o.s Y E A R S Figure l. Crude survival of ali treated patients. in "radically" irradiated patients the rate was 32 % , 13 % and 5 % respectively; median 8 months. The difference between "radically" and palliatively irradiated patients was significant (Figure 2). There was no difference in the survival of our patients irradiated with palliative doses and those treated symptomatically (Figure 3). Performance status and length of esophageal stenosis influenced the survival of irradiated patients (Figures 4 and 5). 1.0 171 _, 1 i -­ ALL N = 120 :l ············· RT> 45 Gy N = 69 Of 134 irradiated patients, 69 received "radi­ , ----RT< 45 Gy , 1 l 1 ! 1 l : ! cal" TD> 45 Gy/weeks, 51 patients <45 Gy, whereas in 14 patients radiation was started but :., 1 :., - had to be terminated before TD 15 Gy had <% P < 0.001 ! !1 been achieved because of worsening of the . o.s <( 1 „1 1, l patient's condition. So, only 69 "radically" and i g; 51 palliatively irradiated patients could be con­ L 1 :.,.! .. .lL, ... .,'!.., '·· sidered. Results L\L ___ "L.t., .................................. -._ One-, two-and five-year crude survival of all 152 patients was 18 % , 7 % , and 4 % respective­ Y E A R S ly; median survival was 5.5 months (Figure 1). Figure 2. A comparison or irradiated patients by The survival of irradiated patients was better: tumor dose. Results of nonoperative treatment far esophageal cancer Chemotherapy did not improve the survival of irradiated patients, irrespective of tumor 0.1). The influence of radio-and chemotherapy on dysphagia was poor: in only 21 of 101 evaluable patients the improvement lasted two months or more whereas in 39 of 101 patients swallowing ability worsened. --RT < 45 Gy N = 51 ............ Symptomatic TH , N = 18 P > 0.1 = 2 3 V E A R S Figure 3. A comparison of survival by treatment method: radiotherapy with TD <45 Gy and sympto­matic therapy. 1.0 1.0 Stenosis . 5 cm, N = 24 Stenosis > 5 cm , N = 8 6 ; P < 0.025 i.i ;; :.i ·•·,., ..... ,., !. •. ! .... " : .... :.: .................... . 2 3 V E A R S Figure 5. A comparison of irradiated patients by length of esophageal stenosis. 1.0 --RT > 45 Gy, N = 46 ............. RT>45 Gy+ChT,N = 23 P > 0.1 . o.s ;; 0-+-------.---.--.---­ 0 E A R Figure 6. A comparison by treatment method: radio­ . O.S > = = therapy and radiotherapy + chemotherapy. Our 14 selected patients with tumor stenosis up to 5 cm and performance status at least 70 were irradiated with TD> 45 Gy: their one-year survival was 50 % and five-year survival 21 % (Figure 7). The survival of ali patients with complications was very short: with fistulas maximum 10 months, with bronchial (& tracheal) infiltration and invasion of the aorta 18 months. Ali 15 of irradiated Figure 4. A comparison patients by performance status. patients with palliative surgery died within 8 132 Debevec M 1.0 In our patients radiation with TD over 45 Gy Stenosis . 5 cm resulted in significant better survival than with RT > 45 Gy TD bellow 45 Gy. This is in agreement with PS ) 70 N = H the results reported by Albertsson et al.3 Our survival results could be compared with the data of Earlam and Cunha-Melo:4 8489 = 9 cm. There was no relationship between tumor site and survival rate in our patients. This is in accordance with the observations of other aut­hors. 3, 5, 6 Chemotherapy did not improve the survival of our irradiated patients. Also in the case of "radical" radiation doses we could not achieve the results reported in literature.s-11 Hospitalization tirne of our patients treated by chemotherapy was longer because treatment on outpatient basis was not possible. Our treatment results suggest that radiothe­rapy is reasonable only with radical intent in esophageal cancer patients in good general con­dition and with short tumor stenosis; for the tirne being, chemotherapy should be performed only with protocols. References l. UICC. TNM classificiation of malignant tumors. Harmer MH ed. 3ro .• 'le ,, ..,_,._.J ... .A. 1 round lesion in the centre of the cranial part of the Figure 4. A hepatic tumor composed of uniform right !iver lobe. spindleshaped cells arranged in an interlacing pattern. Primary leiomyosarcoma of the !iver More than three years after surgery the pa­tient is feeling well and is free of any complaint. During this period an ultrasound scan of the abdomen was made on severa) occasions, as well as scintigraphy of the !iver and abdominal CT, ali showing through regeneration of the !iver, with no evidence of disease (Figure 5). Figure 5. Ultrasound scan showing regeneration of the !iver. Discussion Leiomyosarcomas of the !iver are usually large, slowly growing tumors found in adults. The main symptoms usually include swelling of the abdomen and pains under the right thoracic arch, accompanied by a loss of body weight. If the tumor is small, pains of some intensity may occur without abdominal swelling. Hepatome­galy may also be observed, sometimes extreme and weighing even up to 11.200 g. 10 Now and then, ascites may also be present. Leiomyo­sarcoma is mostly solitary, but may also be multiple. Ali the cases described so far had no specific symptoms, so that there was a clear-cut discre­pancy between the findings of tumors and the hepatic functional tests. With smaller tumors all laboratory tests are usually normal, whereas with the bigger ones there are alterations of alkaline phosphatase, gammaglutamyl transpep­tidase, transaminases or lactic dehydrogenase, which however are not specific. It is most important to make the diagnosis as early as,-possible, since the prognosis of this tumor is much better than that of other primary malignant hepatic tumors, especially after an adequate surgical treatment. 12 The only sate preoperative diagnostic method of this tumor is histologic analysis. In case there is any doubt as to the histologic finding obtained by means of routine methods, electronic micro­scopy is recommended. In the diagnostics of this tumor the ultra­sound examination of the abdomen, CT of the liver, angiography of the arteria hepatica pro­pria, liver nuclear magnetic resonance (NMR) and laparsocopy with target biopsy of the tumor are also of great help. Of ali these methods, the best results are provided by the CT of the ]iver, ultrasound guided percutaneous biopsy of the tumor and the laparoscopy with target biopsy of the tumor. These are the methods of choice. In particular cases, a percutaneous liver biopsy can also be made using the transjugular approach. This, of course, can be done only if the patient has no coagulation irregularities. 13 The findings of the !iver ultrasound and CT in combination with arteriography will yield a good anatomic image of the tumor mass and local veinal structure, which can be of great assistance when planning surgery. Laparoscopy with target biopsy of the tumor is also recommended.2 This can yield additional information on the size, appearance and any possible spread of the tumor to an extrahepatic area. The diagnosis of this tumor requires a careful search to ascertain that it originates from neigh­bouring structures,14-18 and that it is not meta­stasized from another primary site. The clinical presentation and the results of the examination are not specific for the diagnostic of hepatic leiomyosarcoma. Microscopically it consists of interwoven bundles of spindle cells with a variable amount of cytoplasm. Characteristic are the elongated cigar-like nuclei, as well as their mitotic activity. Morales et ai. 19 present ultrastructural crite­ria of leiomyosarcoma. According to these aut­hors, leiomyosarcoma must meet the following electron microscopic criteria: presence of intra­ Kraus I et al. cytoplasmic myofilaments; dense bodies in both cytoplasm and plasma membrane; pinocytic ve­sicles and invaginations of plasma membranes; remnants of basal lamina or an excessive cell coat. Histologically, hepatic Ieiomyosarcoma has to be distinguished from fibrosarcoma and ma­lignant schwannoma, which are also built of spindle cells. The histologic diagnosis is made on the basis of: (1) proliferation of long spindle­shaped cells in interlacing bundles; (2) presence of cigar-shaped nuclei; (3) presence of intracy­toplasmic thin filaments and marginal dense patches; (4) immunohi'stochemical positivity for vimentin and muscle-specific actin. In response to the muscle-specific actin the antibodies react with both alpha and gamma actins within the skeletal, cardiac and smooth muscle cells, and are not reactive with other mesenchymal and epithelial cells. 20 This anti­body is a much safer marker for differential diagnosis of mesenchymal malignant tumors than desmin, because of its high sensitivity for poorly differentiated muscle tumors. When leiomyosarcoma of the liver is histolo­gically confirmed, the degree of malignant po­tential has to be ascertained on the histological image. In order to achieve this, usually such criteria as the size of the tumor, cellularity, number of mitotic figures and the presence of necrosis within the tumor are used. Enzinger and Weiss21 tried to establish the gradation of malignancy of leiomyosarcoma. They have found that the retroperitoneal tu­mors of smooth muscle cells with only 5 mitoses/ 10 HPF have to be considered as malignant, while those tumors having between 1 and 4 mitoses/10 HPF should be considered as poten­tially malignant. In oposition to these authors, Wolf et al.22 found tumors with smooth muscle cells of the soft tissue, which had mitotic activity of Iess than 1 mitosis/10 HPF, to be nevertheless metastasized. On the basis of this it can be concluded that we have not yet found a reliable morphological criterion for the gradation of malignancy of the hepatic leiomyosarcoma. The reason for this is the absence of a precise analysis of morpholo­gical criteria and criteria of growth. It is not easy to determine the metastatic potential of this tumor. The mitotic index alone is not a sufficient indication. There are a number of morphological criteria such as e.g. the size of the tumor, degree of differentiation, as well as the characteristics of tumor growth, such as e.g. the mitotic index, frequency of cells in the growth fraction, that need to be taken into account. The therapy of choice for this tumor is an extensive surgical resection. The role of the 9 chemotherapy is not yet clearly understood2• as there is no safe attitude towards this question due to the insignificant number of cases repor­ted so far. The patient could be treated also with the hepatic transplantation.23 References l. Paraskevopoulos JA, Stephenson TJ. Primary Leiomyosarcoma of the liver -case report. HPB Surgery 1991; 4: 157-63. 2. O'Leary MR, Hill RB, Levine RA. Peritoneosco­pic diagnosis of primary leiomyosarcoma of liver. Hum Pathol 1982; 13: 76-8. 3. Wilson SE, Braitman H, Plested WG, Longmire WP. Primary leiomyosarcoma of the liver. Ann Surg 1972; 174: 232-7. 4. Shimo M, Murakami T, Endou H, Oohara T, Yamakawa M. A clinicopathologic study of two cases of leiomyosarcoma of the )iver. Byori To Rinsho 1989; 7: 635-42 (in Japanese). 5. Bloustein PA. Repatic leiomyosarcoma: Ultra­structural study and review of the differential diagnosis. Hum Pathol 1978; 9: 713-5. 6. Masur H, Sussman EB, Molander DW. Primary hepatic leiomyosarcoma: a report of two cases. Gastroenterology 1975; 69: 994-7. 7. Chen KTK. Repatic leiomyosarcoma. J Surg On­col 1983; 24: 325-8. 8. Qifang P, Lunan Y, Ruiti R, Xianying Y. Clinico­pathologic manifestations in 2 cases of primary leiomyosarcoma of the )iver. Hua Hsi I Ko Ta Hsueh Hseoh Pao 1987; 18: 90-2 (in Chinese). 9. Maki HS, Hubert BC, Sajjad SM, Kirchner SP, Kuehner ME. Primary hepatic leiomyosarcoma. Arch Surg 1987; 122: 1193-6. 10. Fong JA, Ruebner BH. Primary leiomyosarcoma of the !iver. Hum Pathol 1974; 5: 115-20. Primary leiomyosarcoma of the /iver 11. Watanabe K, Saito A, Wakabayashi H, Kawa­guchi T, Suzuki T. Two autopsy cases of primary leiomyosarcoma of the !iver. Acta Pathol Jpn 1991; 41: 461-5. 12. Echeverria RA. Hepatic tumors of long duration with metastases. Amer J Ciin Pathol 1978; 69: 624-7. 13. Adam A. Percutaneous techniques in the !iver and biliary system: recent advances. Br J Hosp Med 1989; 42: 102-10. 14. Yamagawa H, Oonishi C, Kobayashi M. Leiomyo­sarcoma of the suspensory ligamentum of the !iver. Mie Mad J 1967; 16: 229-33. 15. Mita! R, Bazaz-Malik G. Leiomyosarcoma of liga­mentum teres of the !iver. Amer J Gastroenterol 1971; 56: 48-51. 16. Taylor RW, Sylwestrowicz T, Kossakowska AE, Urbanski S, Minuk GY. Leiomyosarcoma of the inferior vena cava presenting as Budd -Chiari syndrome. Liver 1987; 7: 201-5. 17. Huguet C, Harb J, Gavelli A, Riberi A. Leiomyo­sarcoma de la veine cave inferieure e'tendu au foi. Resection complete avec reconstruction vei­nense. Gastroenterol Ciin Biol 1992; 16: 714-7. 18. Baur M. Primary leiomyosarcoma of the !iver -a case report. Z Gastroenterol 1993; 31: 20--3. 19. Morales AR. The ultrastructure of smooth muscle tumours with a consideration of the possible rela­tionship of glomangiomas, hemangiopericytoma and cardiac myxomas. Pathol Annu 1975; 10: 65-70. 20. Tsukada T, MacNutt MA, Ros RR, Gown AM. HHF35, a muscle actin specific monoclonal antibo­dy: II. Reactivity of normal, reactive, and neopla­stic human tissue. Amer J Pathol 1987; 127: 389­402. 21. Enzinger FM, Weiss SW. Leiomyosarcoma in soft tissue tumor. 2nd ed., CV Mosby Co., St. Louis, Washington, D. C. and Toronto, Canada, 1988, 402-21. 22. Wolff M, Kaye G, Silva F. Pulmonary metastases (with admixed epithelial elements) from smooth muscle neoplasms: Reports of nine cases, inclu­ding three males. Amer J Surg Pathol 1979; 3: 325-42. 23. Saint-Paul MC, Gugenheim J, Hofman P, Arpurt JP, Fabiani P, Michiels F, Fujita N, Goubeaux B, Loubiere R, Delmont J, Mouier J. Leiomyosar­coma du foie: un cas traite par transplantation. Gastroenterol Ciin Biol 1993; 17: 218-22. Radio/ Oncol 1994; 28: 146-8. Repo rt from the joint meeting of European Society f or Radia ti on Biology and European Society for Hyperthermic Oncology Amsterdam, Jone 1-4, 1994 For the first tirne European Society for Radia­tion Biology (ESRB) held the joint meeting with European Society for Hyperthermic Onco­logy (ESHO). More than 350 participants attended this meeting. Many investigators came from Eastern Europe; for the first tirne, scien­tists from Japan and lndia also presented their work. Only the opening and the closing Iecture (J. Denekamp and D. Bootsma) were given for ali participants. Throughout the meeting, the pro­gram started at 8 a. m. in four parallel sections. It began with refresher courses, followed by workshops, symposia, proffered papers or po­ster viewing. With two coffee breaks and one hour for lunch, it finished at 6 p. m. The presentations were divided in the following sections: clinical hyperthermia, hyperthermia physics and technology, hyperthermia biology, photodynamic therapy and the heat shock pro­teins, X-rays, hyperthermia, apoptosis, tumor radiobiology, radiation and environment carci­nogenesis, EULEP: genetic predisposition of radiosensitivity, cellular radiobiology, DNA and chromosome damage and repair, radiation damage to normal tissue, and biological basis for the clinical application of boron neutron capture therapy. In this report, I will point out some of the most important new data or some general conlusions that have both, basic and clinical implications. Basic science The newest basic data were presented by D. Bootsma (DNA repair: genes and syndromes). With his group (no doubt the best in this field in the world), he has made a tremendous pro­gress since last year. His talk was concentrated on three genes involved in the repair of human genome. In collaboration with Japanese collea­gues, HHR23A and HHR23B genes were clon­ed. These genes are Iocated on human chromo­some 3, and show great homology with yeast RADlO and RAD23 genes. The defect in these genes affected the so-called "slow DNA repair" component, while the fast one was unaffected. (During fast repair, the active genes are repair­ed, while, during slow repair the lesions from rest of DNA are removed.) The defect in HHR23 genes is thought to be responsible for the genotypic and phenotypic characteristics of Xeroderma pigmentosum (XP) C type of pa­tients: photosensitisation, characteristic pig­mantation, proneness to cancer). The other gene mentioned was ERRC-6. This gene was cloned two to three years ago. It shows great homology with RAD 16 and RAD 54 genes. The deffect in this gene is expressed in the Cockney syndrome (the lack of pigmentation characteristic of XP patients, dysmyelinisation, short growth, "bird face", but no proneness to cancer). The product of ERCC-6 gene is thought to be a helycase, which is important in fast DNA repair. In his lecture, Bootsma also dealt with TFIIH(BTF2) gene, the first gene known today to participate in recombinational repair in eukaryotes. The product of this gene form a complex with severa! proteins (p34, p41, p44, p62, p80-from ERCC2 gene, p89-from ERCC3 gene, and two or more unknown pro­teins). This complex exerts helycase and phos­phorylating activities, enabling that RNA poly­merase II to transcribe DNA. Thus the deffect in this gene causes the deffect in DNA trans­cription and manifests itself clinically as "trans­criptional syndrome" (neurodysmyelinisation, disturbance in physical and mental develop­ment). Not surprisingly, it is difficult to find a Repar/ 147 patient with this syndrome, because such a defect is more or less incompatible with life. Other genes which have now been known to participate in the DNA repair include: 13 hu­man, more than 10 rodent and more than 15 yeast genes involved in the repair of lesions induced by UV light or mitomycin C, and 5 human, 10 rodents and 10 yeast genes involved in the repair of X-rays induced lesions. An attept to explain the cause of adaptive response of human keratocytes to N-mehyl-N­nitro-N-nitroso-guanidine (MNNG) was made by H. E. Kleczkowska and F. R. Althaus. Like rodent cells, human keratocyites, do not possess 06-methylguanine DNA methyltransferase, which is otherwise specifically induced in bacte­ria following treatment with MNNG and which is specifically involved in the adaptive response to this drug. In keratocytes, the authors reveal­ed the induction of poly-(ADP-ribose) synthesis after pretreatment with MNNG; this synthesis is thought to play an essential role in adaptive response. (The polymers are very efficient in removing histones from DNA, thus facilitating the DNA repair). Basic-clinic data Many presentations at the meeting dealt with the methods that could give the accurate answer to the question what happens in the cells after irradiation, and with the assays that could pre­dict the celi sensitivity to irradiation. Yariations in the initial leve! of DNA damage have pre­viously been suggested as potential determi­nants of radiation sensitivity in both rodent and human celi lines. These studies were usually performed with the 11011-denaturating filter elu­tion technicques, a method often suggesed to be influenced by factors other than DNA brea­kege. H. H. Kampinga gave a talk about the results obtained by some new techniques: Co­met assay and pulse field gel electrophoresis (PFGE). Both these methods, however, failed to detect any significant difference in sensitivity of radioresistant and radiosensitive celi lines obtained by clonogenic assay. PFGE determi­ nes the retention of DNA on gel, depending on the number of double strand breaks (the main cause of celi death after irradiation). While smaller fragmets are released, more than 70 % of larger fragments are retained. This could limit the capacity of this assay to distin­guish between radiosensitive and radioresistant cells (J. Dalm-Daphi and E. Dikomey). In the case of Comet assay, the situation is even worse; moreover, it is not known what is really detected by this technicque (E. C. Woudstra). It is speculated that the features of chromatine structure interfere with the detection of damage by some assays. lndeed, it is disappointing, that during so many years of investigations, no sim­ple and sensitive method was found that may replace the clonogenic assay. Obviously, the theoretical basis of new techniques is too simple and do not take ali the relevant facts into consideration, beeing therefore inadequate in the detection of the fina! damage effect at the cellular leve!. Clinical application The results of three reports of ESHO trials, presented by J. Overgaard, will be briefly men­tioned. The assessment of the efficacy of hyper­thermia adjuvant to radiotherapy was determin­ed in the treatment of neck and breast tumour and the metastatic melanoma. For metastatic malignant melanoma, 128 patients were treated from 1987 till now. They received either radia­tion alone (3 x 8-9 Gy) or this same radiation treatment was combined with hyperthermia (3 sessions, 43° C, 1 hour). The initial complete response was significantly higher in patients receiving hyperthermia. This was maintained during the two years period (the study is clos­ed). 77 patients with neck tumours node meta­stasis received irradiation alone (60-70 Gy) or combined with hyperthermia (5-6 sessions, once weekly, 43° C, 1 hour). The preliminary analysis was given showing no difference in this group of patients (the study is stili open). Similar results were obtained with the same treatment schedule involving 155 patients with 148 Report locally advanced breast cancer. However, colla­borative phase III hyperthermia trials (C. C. Vernon) on 300 pacients with breast cancer (started in U. K. at 1989 and because of slow accural rate of this study, la ter also the data from ESHO patients were included and analys­ed together) shows better effects in pacients treated with hyperthermia. The disagreament in the results obtained with the breast cancer patients may depend on the different treatment schedules. In the workshop on clinical hyper­thermia it was concluded that in future hyper­thermia should continue to be used as the adjuvant to radiotherapy. The tumours, candi­dates for such treatments, should be accessible, high grade, pathologically deep (meaning that they are spread to muscles), with metastasis. That are breast, colon and rectal cancer and sarcomas. Also, more attention must be given to the adequate heating of tumour mass, be­cause the rare and incomplate studies done so far indicate, that the expected temperature of 43° C is not reached in all parts of the tumor. Clinical data provide a warning that the po­tential of these assays, and their predictive therapeutic power must be critically evaluated for specific endpoints. It is concluded that a more critical approach to experimental radio­therapy is required to avoid the rejection of potential improvements of the treatment, resulting from the falsy optimism. It was poin­ted out (S. M. Bentzen) that in the case when insufficient number of patients is taken into account, it is impossible to draw accurate con­clusions (e. g., to prove that a new treatment protocol improves the success from 30 to 50 per cent, 124 patients are needed, while for the improvement of 30 to 40 per cent, 467 patients are needed). If it is not possible to have such a number of patients, than doing such clinical trials is not only a waste of tirne, but is also unethical to the patient. Therefore, it is recom­mended to clinicians to joint the ESHO pro­gram rather than to start their own trials with a low number of patients. In this report only some of the presentations are mentioned from many interesting ones. The meeting was very well organized, giving all the participants the opportunity to talk with their colleagues. Finally, I must stress, that a lot of very young people attended the Meeting, thus clearly indicating that radiobiology has again come in the focus of interest. Maja Osmak Ph. D. Ruder Boškovic Institute, Zagreb Radio! Oncol 1994; 28: 149. Notices Notices submitted for publication should contain a mailing address, phone and/or fax number of a contact person or department. Radiology The European seminar (ESDIR) "Radiology of Ga­strointestinal Tract Neoplasms: Diagnosis, Staging and Intervention" will be held in Iraklion, Crete, Septem­ber 22-24, 1994. Contact Public Relations Center LTD -102, Michala­kopoulou str., 11528 Athens, Greece; or call + 30 1 777 1056 / 775 6336 / 771 1673; fax + 30 1 771 1289. Breast cancer The ESO refresher day will be held at October 7, 1994. Contact Miss Gollubics, ESO-Vienna-Office, Arzte­kammer fi.ir Wien, Fortbildungsreferat Weihburggasse 10-12, A-1010 Vienna, Austria; or call + 43 151501293; fax + 43 151501 240. Lung cancer 1st The international -lung cancer conference "Non­small Celi Lung Cancer Management: Open Questions and Controversies" will be offered in Alba, Italy, October 7-8, 1994. Contact Cuneo Lung Cancer Study Group, Via Ro­mita 15, 12011 Borgo S. Dalmazzo, Cuneo, Italy; or call + 39 71 441 770 (hours: 12 a.m.-3 p.m., Monday through Friday); fax +39 171 611597. Papillomavirus The 13th international papillomavirus conference will be offered in Amsterdam, The Netherlands, October 8-12, 1994. Contact Bureau PAOG, Mre. C. M. Schoof / Mr. C. H. Walta, Tafelbergweg 25, 1105 BC Amsterdam, The Netherlands; or call + 3120 5564801; fax + 3120 696 3228. IAEA Scientific meeting The international symposium on spent fuel storage ­safety, engineering and environmental aspects will be Contact lnternational Atomic Energy Agency, P. O. Box 100, Vienna International Centre, A-1400 Yien­na, Austria. IAEA Scientific meeting The seminar on radioactive waste management practi­ces and issues in developing countries will be offered in Beijing, China October 10-14, 1994. Contact lnternational Atomic Energy Agency, P. O. Box 100, Yienna lnternational Centre, A-1400 Yien­na, Austria. Radiology The first congress of the Croatian Society of Radiology will be held in Opatia, Croatia, October 11-15, 1994. Contact Congress secretariat, Mr. Berislav Budiselic or Mr. Stjepan Riman, Clinical Institute of Radiology, Clinical Hospital Center Rijeka, Tome Strižica 3, 51000 Rijeka, Croatia; or call + 385 51 216 899; fax + 38551536. Radiobiology ESTRO teaching course "Basic Clinical Radiobiology" will be held in Prague, Czech Republic, October 16--20, 1994. Contact the ESTRO Secretariat, Radiotherapy Department, University Hospital St Rafael, B-3000 Leuven, Belgium; or call + 32 16 33-64-13; fax +32 16 33 64 28. IAEA Scientific meeting The FAO/IAEA international symposium on nuclear and related techniques in soil/plant studies on sustain­able agriculture and environmental preservation will be offered in Yienna, Austria, October 17-21, 1994. Contact International Atomic Energy Agency, P. O. Box 100, Yienna International Centre, A-1400 Yien­ held in Vienna, Austria, October 10-14, 1994. na, Austria. ESTRO-AER PRE-MEETING TEACHING COURSE THE USE OF MODERN DIAGNOSTIC IMAGING TECHNIQUES IN RADIOTHERAPY PLANNING Granada 22-24 September, 1994 Teaching Staff: M. Brada, J. Britton, M.Goitein, G. Kantor, A. Neal, C. Raybaud, D. Ten Haken, A. Warrington. Course aim: This course is a joint venture of ESTRO together with the European Society for Radiology (AER). In the last year new radiotherapy techniques are becoming more and more important. The common aspect of these new techniques is the aim to apply higher doses to smaller volumes.This can be done both with brachytherapy as with extemal beam conformal radiotherapy. This development creates the need for radiotherapists to know how to use the modem imaging techniques such as ultrasound, computerized tomography and magnetic resonance imaging. In this course experts in the field of modem imaging techniques and radiotherapy will discuss and demonstrate modem treatment planning for different types of malignancies in the central nervous system making an optimal use of the imaging techniques. They will also demonstrate how to organize and structure the cooperation between the radiodiagnostic and radiotherapy departments. Although brain is chosen as a site to demonstrate the use of modem imaging techniques the principles demonstrated are also applicable to other sites. Language: English Programme: Curent Status of Conventional Imaging CT and MRI in the diagnosis of adult brain tumours (differential diagnosis, choice of imaging by sites and tumours, visualisation of the extent of tumour, diagnostic aspects of metastatic disease). CT and MRI in the diagnosis of paediatric brain tumours. Spina! imaging (primary and metastatic spina! tumours, techniques of imaging spina! seeding). Imaging in the follow-up of brain tumour patients ( assessment of response, features of early and late toxicity of chemotherapy, radiotherapy and surgery, late vascular events, hydrocephalus). Angiography and MRA in the diagnosis of AVM's. Imaging and Modem Radiotherapy Planning Fixation and localisation devices for brain imaging. Problems of image distortion. 3D image reconstruction and visualisation and radiotherapy planning. Evaluation of 3D plans with