INSTITUTE OF RENTGENOLOGY, UNIVERSITY MEDICAL CENTER LJUBLJANA, YUGOSLAVIA INSTITUTE FOR CLINICAL AND EXPERIMENTAL MEDICINE, PRAGUE, CZECHOSLOVAKIA SCIENTIFIC CENTRE OF SURGERY, ACADEMY OF MEDICAL SCIENCES, MOSCOW, SOVIET UNION MUNICIPAL HOSPITAL FRIEDRICHSHEIN, DEPARTMENT OF CARDIOVASCULAR DIAGNOSIS, BERLIN, GERMAN DEMOCRATIC REPUBLIC GLINIC OF RADIOLOGY, PESC, HUNGARY INSTITUTE OF CARDIOVASCULAR SURGERY OF A. N. BAKULEV, MOSCOW, SOVIET UNION PERCUTANEOUS TRANSLUMINAL RENAL ANGIOPLASTY - A MULTICENTRE STUDY OF THE LONG TERM RESULTS Belan A, Pavčnik D, Petrosijan A, Horvath L, Rabkin 1, Muller AH, šurlan M, Klančar J, Knific J, Berden P, Vidmar D, Cesar R, Kocijančič 1, Pavlovčič-Kaplan S Abstract - To assess the long-term clinical effect of percutaneous transluminal angioplasty of the renal artery (PTRA), patients with clinical examinations and laboratory tests performed before PTRA and within a minimum of 3 months following the investigation were considered eligible for inclusion. Patients with fibromuscular disease of the main and/or branch renal arteries were the most suitable candidates for PTRA, because two thirds of them showed a blood pressure benefit at 5-year follow up. These results are similar to those achieved in the group of patients with atheromatous disease. Authors discuss the clinical and laboratory characteristics and radiočlogic aspects of PTRA, the technical standard of the procedure, complications and number of redilatations. UDC: 616.136.7.272-089.844 Key words: renal artery obstruction, angioplasty transluminal Orig sci paper Radiol lugosl 1990; 24: 137-45. Introduction - Percutaneous renal angioplasty (PTRA) has become an established interventional method used in the treatment of renal artery stenosis. While in stenoses caused by fibromuscular dysplasia, it is the method of first choice, there has not been unanimity redgarding its superiority in cases of stenoses of atherosclerotic origin. Substantial improvement of renovascular hypertension following PTRA has been reported in fibromuscular stenoses rather than in atherosclerotic stenoses. Doubtless, any further piece of experience, especially that gained in a large group of patinets followed up for a rather long period of time after PTRA, helps to build the body of hard evidence available. It was for this reason that we decided to conduct a multicentre study designed to assess retrospectively the long-term effect of PTRA on blood pressure in atherosclerotic, fibromuscu-lar and other types of lesions, and to suggest whether the long-term effect can be predicted on the basis of the angiographic finding obtained immediately after PTRA. Patients and methods - The project was joined by the following centres: 1. Department of Radiology; Institute for Clinical and Experimental Medicine (A. Selan), Prague, Czechoslovakia 2. All-Union Scientific Centre of Surgery; Academy of Medical Sciences (l. Kh. Rabkin), Moscow, Soviet Union 3. Institute of Rentgenology University Medical Center (D. Pavcnik), Ljubljana, Yugoslavia 4. Municipal Hospital Friedrichshein, Department of Cardiovascular Diagnosis (J. H. A. Muller), Berlin, German Democratic Republic 5. Clinic of Radiology (L. Horvath), Pecs, Hungary 6. Institute of Cardiovascular Surgery of A. N. Bakulev (l. Petrosyan), Moscow, Soviet Union The study was coordinated by the Department of Radiology of the Institute for Clinical and Experimental medicine in Prague where a questionnaire for retrospective data collection was drawn and distributed to each participating centre. The questionnaire was to be filled for each patient undergoing PTRA before 31 December, 1987. The contribution of each centre to the basic group of patients is shown in Table 1. Twenty-three patients after PTA of the renal graft artery were excluded from the study in order to be assesed separately. The remaining Received: March 3, 1990 - Accepted: April 1O, 1990 137 Belan A. et al. Percutaneous transluminal renal angioplasty - A multicentre study of the long term r11sults Table 1 - Contribution of each centre to the basic group No. of all pts after PTA Percentage of the No. of pts after PTA of the renal artery entire group of the renal graft artery 1. Prague 134 30.5 18 2. Moscow R. 118 26.9 1 3. Ljubljana 91 20.7 3 4. Berlin 34 7.7 1 5. Pecs 32 7.3 o 6. Moscow P. 30 6.8 o To t a l 439 23 416 patients were divided into three groups by the etiology of the stenosis: I. ATHERO (atherosclerosis), n = 261, II. FMD (fibromuscular dysplasia), n = 109, III. OTHERS (mostly vascular lesions in arteritis and other systemic diseases), n = 46. In some cases, the etiology of stenosis was established by histological examination of the artery after its surgical reconstruction, nephrectomy, orat autopsy. The clinical and laboratory characteristics of the patient group before pTrA are shown in Table 2. Table 2 - Clinico-laboratory characteristics of the group of patients before PTRA l. ATHERO II.FMD III. OTHERS n = 261 n=109 n = 46 Mean age n = 260 n= 107 n= 45 (years) 50.4±7.5 34.5±9.9 32.7±11.3 Sex male 196 75.1% 38 34.9% 28 60.9% female 65 24.9% 71 65.1% 18 39.1% Extrarenal manifestations of atherosclerosis 131 50% 10 10% 5 11% Primary renal disease 38 14.5% 19 18% 7 15.5% Systemic disease (incl. diabetes mellitus) 19 7% 2 2% 5 11% WHO class of hypertension 1 27 10.8% 32 29.6% 15 32% II 192 77.1% 68 63% 29 64% III 30 12% 8 7.4% 2 4% Plasma creatinine n=199 n= 94 n= 27 (umol/I) 127.2±75.6 91.8±26.0 110.6±20.1 (40-764) (46 -198) (88-190) Blood pressure n = 255 n= 108 n= 45 (mmHg) systole 197±31 182±30 190±29 diastole 114±17 113±15 114±16 mean 142±20 136±19 140±18 Antihypertensive therapy -none 11 4.4% 4 3.8% 5 11.1% -1-3 hypotensives 202 80.8% 92 86.8% 36 80.0% - > 3 hypotensives 37 14.8% 10 9.4% 4 8.9% lndications for PIRA - hypertension 223 87.7% 99 94.3% 41 89.1% - hypertension with 31 12.2% 6 5.7% 5 10.9% deteriorated function 138 Radiol lugosl 1990; 24: 137-45. Belan A. et al. Percutaneous transluminal renal angioplasty - A multicentre study of the long term results As expected, the mean age was markedly higher in Group 1 (ATHERO), with men prevailing and extrarenal complications (ischemic heart disease, atherosclerosis of the lower extremities, stroke) present more often than in the other groups. In Group II (FMD), women prevailed and the mean age was lower. There was no difference between the groups as to other parameters (presence of primary renal disease, systemic disease, WHO classification of hypertension, plasma creatinine level, blood pressure before PTRA and the number of hypotensive drugs used). The higher incidence of systemic diseases in Group I was due to the more frequent incidence of diabetes. The radiologic characteristics are given in Table 3. Complications requiring surgery were found in 12 patients. Nephrectomy had to be performed in three cases and aortorenal bypass in nine. The angiographic finding of the renal artery immediately after PTA was assessed as »normalized« in disappeared stenoses, »improved« in cases of stenoses smaller than before the procedure, and »not improved« in persisting stenoses of the same extent. Fram the basic group (n = 416), a total of 154 patients (37%) undergoing successful PTRA without redilatation, with clinical examination and laboratory tests done before PTRA and followed up for a minimum of 90 days since dilatation, were selected to evaluate the effect of PTRA on blood pressure and renal function. The following criteria were chosen: Table 3 - Radiologic characteristics of the group l. ATHERO II. FMD III.OTHERS n = 261 n = 109 n = 46 Side of stenosis * right * left * right + left Number of dilated arteries one two three Technical failure Complications of PTRA Angiographic finding after PTRA * normalization * improved * unchanged Number of redilations 90 125 46 16 23 127 119 14 27 34.4 47.9 17.6 207 79.6 50 19.2 3 1.2 6.1 8.8 48.8 45.8 5.4 10.3 61 55.9 37 33.9 11 10.1 96 10 1 9 17 42 56 11 12 89.7 9.3 1.0 8.2 15.6 38.5 51.4 10.1 11.0 22 12 12 32 14 0 10 5 21 18 6 47.8 26.1 26.1 69.6 30.4 21.7 10.9 46.7 40.0 13.3 2.2 Patients in Groups I and III showed more frequent bilateral stenosis. The number of dilated arteries in the ATHERO and OTHERS groups is likewise higher than in that with FMD stenoses. In atherosclerotic stenoses neither information on the type of stenosis nor records on the angiographic finding of the peripheral arterial bed are available. Post-PTRA complications, regardless the etiology of stenosis, were observed in 10.8% of patients. Half of them were minor complications, i.e., renal artery spasm, and complications at the puncture site, with the remaining 50% of complications made up by dissections, embolization, perforation or rupture of the artery, occllusions and an immediate decrease in renal function. Radiol lugasl 1990; 24: 137-45. Renal function was regarded as unchanged if plasma creatinine level had been within normal limits (i.e., up to 125 umol/I) also before PTRA, or when the change, in patients with initial levels over 125 umol/I, did not exceed 20% of the initial value after PTRA. Our definition of functional deterioration included cases with normal ilnitial creatinine levels and follow-up levels exceeding 125 umol/I, or a rise from values over 125 umol/I by more than 20%. An improvement in function was registered if creatinine decreased from levels over 125 umol/I to below 125 umol/I, or by more than 20% from levels initially higher than 125 umol/1. Blood pressure was considered normal if the value of systolic pressure was lower than 165 139 Belan A. et Percutaneous transluminal renal angioplasty - A multicentre study of the long term results mmHg, that of diastolic pressure lower than 95 mmHg, and the value of mean pressure below 11O mmHg. -Our definition of improvement was a decrease of elevated values to normal level, or a decrease of elevated values by at least 15% of the initial value. Results - The values of plasma creatinine and blood pressure before and after a minimum of 90 days following PTRA in our group of 154 patients are shown in Table 4. Changes • in renal function before and after PTRA were assessed by changes in plasma creatinine levels. In Group 1 (ATHERO) improvement and deterioration were noted in five cases each. There was no change in the remaining patients. lmprovement and deterioration were observed in one case each in Group II (FMD) Table 4 - Plasma creatinine and blood pressure values in a group of 154 patients examined before PTRA and after a minimum od 90 days later I. ATHERO II.FMD III. OTHERS n = 97 n = 44 n = 13 before after before after before after n = 64 53 n = 39 35 n = 6 6 Plasma creatinine 121.9 123.3 93.9 91.8 129.1 106.2 (umol/I) ± 52.8 62.9 27.1 25.5 35.0 14.2 Blood pressure n = 94 n = 44 n=11 systole 190.6 151.6 177.8 143.7 177.7 136.1 ± 33.4 19.8 33.9 15.6 38.3 41.4 diastote 110.1 93.3 108.9 91.7 107.5 92.7 ± 16.4 10.1 15.5 9.3 16.0 14.9 mean 136.9 113.0 131.9 109.0 133.0 110.6 ± 20.7 12.0 20.6 10.4 20.2 21.4 Table 5 1. Follow-up > months after PTRA (n = 149) Mean BP ATHERO FMD OTHERS TOTAL n % n % n % n % Improved 67 71.3 28 63.6 9 81.8 104 67.8 Not improved 27 27.7 16 36.4 2 18.2 45 30.2 Total 94 44 11 149 2. Folow up > 12 months after PTRA (n = 99) Mean BP ATHERO FMD OTHERS TOTAL n % n % n % n % Improved 51 75.0 13 52.0 6 100.0 70 70.1 Not improved 17 25.0 12 48.0 o o 29 29.3 Total 68 25 6 99 3. FoHowüp > months after PTRA (n 66) Mean BP ATHERO FMD OTHERS TOTAL n % n % n % n % Improved 30 69.8 9 53.0 6 100.0 45 68 2 Not improved 13 30.2 8 47.0 o o 21 31.8 Total 43 17 6 66 4. FoNowijp > 60 months after PTRA (n = 23) Mean BP ATHERO FMD OTHERS TOTAL n % n % n % n % 1 mproved 14 82.4 3 75.0 2 100.0 19 82.6 Not improved 3 17.6 1 25.0 o o 4 17.4 Total 17 4 2 23 Radiol lugosl 1990; 24: 137-45. Belan A. et al. Percutaneous transluminal renal angioplasty - A multicentre study of the long term results patients. Since plasma creatinine was determined in six patients of Group III (OTHERS) only, the changes were not assessed. The long-term effect of PTRA on blood pressure was evaluated at three months (n = 149), at 12 months (n = 99), at 24 months (n= 66) and at 60 months (n = 23). The mean follow-up period (n = 154) was 31 ± 26.8 months (range, 3.3 - 92.8 months). Improvement of mean blood pressure is shown in Table 5. The cumulative curves of improvement of blood pressure after PTRA according to the etiology of stenosis do not differ statistically over a five-year follow-up period (Fig. 1). 20 10 o _,_,_, ! J j Î i yan Fig. 1 - The cumulative curves of improvement of blood pressure after PTRA according to the etiology of the stenosis. Three and more months after PTRA, mean blood pressure was improved in 104 patients (67.8%). The proportion of »improvement« in each group by the etiology of stenosis is shown in Table 6. Both in the ATHERO and FMD groups, while the mean pressure of most patients was within normal values, they had to continue receiving hypotensives. The difference was Group II which comprised a substantially higher proportion of Table 6 - Improvement of hypertension after PTRA Blood pressure improvement > months after PTRA ATHERO FMD OTHERS n = 67 n = 28 n = 9 Improvement but mean BP >110 mmHg thereafter 23 34.3% 3 10.7% 2 22.2% Mean BP > mmHg with hypotensives 29 43.3% 14 50.0% 3 33.3% Mean BP > 110 mmHg without hypotensives 15 22.4% 11 39.3% 4 44.5% Radio! lugosl 1990; 24: 137-45. 141 Belan A. et al. Percutaneous transluminal renal angioplasty - A multicentre study of the long term results normotensives not taking hypotensive drugs (39.3% vs. 22.4%) and, on the contrary, the percentage of those remaining hypertensive following PTRA was considerably lower (10.7% vs. 34.3%). Group III could not be evaluated because of the small number of followed patients. Comparison of patients who, while not taking hypotensive drugs, were normotensive at three months after PTRA (n = 30) with other patients on follow-up (n = 124) revealed that all the former had significantly lower mean blood pressure before PTRA (127 ± 10 vs. 137 ± 22 mmHG), and the WHO classification of their hypertension was likewise lower (Stage I hypertension in 57%, Stage III hypertension in 0%). The therapeutic protocol in patients whose pressure remained unchanged three months after PTRA (n = 45, i.e., 30.2%) did not differ before and after PTRA, i. e., they received the same number of hypotensive drugs. The correlation between the post - PTRA angiographic finding and the effect of the procedure is shown in Table 7. The options listed in the questionnaire regarding the post - PTRA angiographic finding on the renal artery comprised »normalized«, »improved« and »unchanged«. The number of patinets with prolonged blood pressure improvement is substantially higher in the group with a »normalized« angiographic finding than in the group whose finding was »improved« only. The group with an »unchanged« angiographic finding was not evaluated owing to the small number of patients. The cumulative curves of blood pressure improvement according to the post - PTRA angio-graphic finding of the artery irrespective of the etiology is shown in Fig. 2. The difference in the effect on blood pressure in normalized vs. improved findings is statistically significant. Table 7 1. Follow-up at 3 months after PTRA (n = 149) Angiographic finding on the Normalized lmproved n = 64 43.0% n = 83 55.7% reanl artery after PTRA Not improved n = 1 0.7% Mean BP improved Mean BP not improved 54 10 84.5 15.5 49 34 59.0 41.0 50.0 50.0 2. Follow-up at > 12 months after PTRA (n = 99) Angiographic finding on the Normalized lmproved n=42 42.4% n = 56 56.6% renal artery after PTRA Not improved n=1 1.0% Mean BP improved Mean BP not improved 35 7 83.5 16.5 34 22 60.7 39.3 100.0 O.O 3. Follow-up at > 24 months after PTRA (n = 66) Angiographic finding on the Normalized lmproved n = 27 40.9% n = 38 57.6% renal artery after PTRA Not improved n=1 1.5% Mean BP improved Mean BP not improved 23 4 85.2 14.8 21 17 55.3 44.7 100.0 0.8 4. Follow-up at > 60 months after PTRA (n = 23) Angiographic finding on the Normalized lmproved n=14 60.9% n = 8 34.8% renal artery after PTRA Not improved n=1 23.0% Mean BP improved Mean BP not improved 13 1 92.9 7.1 62.5 37.5 100.0 o.o 142 Radiol lugosl 1990; 24: 137-45. Belan A. et al. Percutaneous transluminal renal angioplasty - A multicentre study of the long term results group either. The mean values of plasma creatinine before PTRA and at three months after PTRA were within normal limits in all three groups and did not change during follow-up. Today, there is no doubt that PTRA has become an established technique for the treatment of renovascular hypertension and its results are comparable with those of surgical treatment (8, 14, 15, 16). The technique of PTRA and technology are being constantly refined (1, 17). Long term improvement of blood pressure and the percentage of cured patients (normotensives not requiring hypotensive therapy) are reportedly higher in Group II (FMD) (2-4, 6, 7, 8, 12, 14) than in group 1 (ATHERO). Significant imrpovement of systolic, diastolic and mean pressure after PTRA irrespective of the etiology of stenosis was found in 68% of our patients on long-term follow-up. Provided our criterion of clinical effect was a 15% decrease in mean blood pressure, or a decrease in mean blood pressure below 11O mmHg, the percentage of improved patients was higher in Group 1 (ATHERO), but the cumulative curves of improvement did not differ statistically (Table 5, Graph 1). However, the number of Group II (FMD) patients with pressure normalization was twice as high as that in Group I and, compared with Group 1 (ATHERO), only a third of FMD and, compared with Group 1 (ATHERO), only a third of FMD patients remained hypertensive, even though improved if assessed by our criteria (Table 6). We are not the first to make such an observation. Kuhlman et. al. (5) reported improvement in blood pressure at 21.6 months after PTRA in 48% of patients with atherosclerotic stenosis, and in as little as 32% of patients with fibromuscular stenosis, even though the percentage of normalized patients was higher in the FMD than in the ATHERO group (50.0 vs. 29.0%). It is implied that the etiology of stenosis, as established by angiography, is not neccessa-rily the basic factor determining the long-term effect of PTRA. Moreover, we are unable to make any conclusions regarding the duration of hypertension before PTRA, nor any other factors that might possibly play a major role. The results of our multicentric study in the ATHERO group were primarily attributable to the extremely good data obtained from the centres headed by Prof. Rabkin from Moscow and Dr. Horvat from Pecs. The angiographic finding after PTRA has turned out to be a significant factor for the prediction of the long-term clinical effect of the procedure. Whereas in the case without residual stenosis (and a pressure gradient no longer persisted) the finding was assessed as »normalized«, patients with residual stenoses and residual pressure gradient were considered »improved«. Regardless the etiology of stenosis, the group with a »normalized« finding of the renal artery showed an effect »improved« finding, and the effect persisted for a long period of time (Table 7). The cumulative curve of blood pressure improvement is significantly better in the group of »normalized« stenoses (Fig. 2). Since no angiographic follow-up in patients after PTRA has been performed, we are unable to provide data on the incidence of restenoses neither can we assess the potential value of subsequent antiaggregation or anticoagulation therapy. PTRA is an effective method for the treatment of renovascular hypertension. It is associated with a high technical success rate and a low rate of serious complications. While, almost as a rule, the improvement of blood pressure in atherosclerotic stenoses is only partial and it is usually necessary to continue hypotensive therapy, in fibromuscular stenoses, normotension is rather often attained without further drug administration. A decrease in blood pressure found at three months after PTRA suggests permanent improvement in most cases. The etiology of the stenosis, as established by angiography is not necessarily the main factor determining the long-term clinical effect. Another important predictor of long-term improvement seems to be the angiographic finding of the renal artery immediately after angioplasty. Povzetek multicentiriCna študija perkutane TRANSLUMINALNE ANGIOPLASTIKE Da bi ovrednotili dolgotrajne klinične učinke PTA renalnih arterij, smo pri bolnikih opravili klinični pregled in laboratorijske preiskave pred pTa in tekom (najmanj) treh mesecev po posegu. Najprimernejši bolniki za PTRA so tisti s fibromuskularno boleznijo glavne in katere od vej renalne arterije, saj je bil pri dveh tretjinah bolnikov, 5 let po posegu, učinek na krvni pritisk dober. Podobno velja tudi za bolnike z arteriosklerotično zožitvijo renalne arterije kot tudi za radiološke vidike PTA, tehnične standarde posega, komplikacije ter število ponovnih dilatacij. References 1. Martin LG, Casarella WJ, Alspaugh JP, Chuang VP. Renal Artery Angioplasty: lncreased Technical Success and Decreased Complications in the Second 100 Patients. Radiology 1986; 159 :631-4. 2. Sos TA, Pickering TG, Sniderman K, Saddekni S, Case DB, Silane MF, Vaugham ED Jr, Laragh JH. Percutaneous Transluminal Renal Angioplasty in Reno- 144 Radiol lugosl 1990; 24: 137-45. Selan A. et al. Percutaneous transluminal renal angioplasty - A multicentre study of the long term results vascular Hypertension Due to Atheroma of Fibromus-cular Dysplasia. N Engl J Med 1983; 309:274-9. 3. Puylaet CBAJ, Klinge J, Mali WPTM, Geyskes GG, Becking WB, Feldberg MAM. Percutaneous Transluminal Renal Angioplasty: Initial and Long-Term Results. Radiology 1989; 171:501-6. 5. Kuhlmann U, Greminger P, Gruentzig A, Schneider E, Pouliadis G, Luescher T, Steurer J, Siegenthaler W, Vetter W. Long-Term Experience in Percutaneous Transluminal Dilatation of Renal Artery Stenosis. Am J Med 1985; 79:692-8. 6. Tegtmeyer CJ, Kellum CD, Ayers C. Percutaneous Transluminal Angioplasty of the Renal Artery. Radiology 1984; 153:77-84. 7. Loehr E, Weihert HC, Hartjes H, Schrivjers A. Percutaneous Transluminal Angioplasty of Renal Arteries: Therapeutic Principle - Case Report of 128 Patients with Renovascular Hypertension in: Dotter CT et al. Percutaneous Transluminal Angioplasty. Berlin; Springer 1983; 281-5. 8. Grim CE, Yune HY, Donohue JP, Weinberg MH, Dilley R, Klatte EC. Renal Vascular Hypertension: Surgery vs. Dilatation Nephron 1986; 44(1):96-100. 9. Sos TA, Saddekini S, Pickering TG, Laragh JH. Technical Aspects of Percutaneous Transluminal Angioplasty in Renovascular Disease. Nephron 1986; 44(1):45-50. 10. Becker GJ, Katzen BT, Dake MD. Noncoronary Angioplasty. Radiology 1989; 170:921-40. 11. Maher F, Triller J, Weidmann P, Nachbur b. Complications in Percutaneous Dilatation of Renal Arteries. Nephron 1986; 44(1):60-3. 12. Obrez 1, Šimunic S, Surlan M, Gurtl R, Klenkar M, Kaplan-Pavlovčič S, Fuduric-Winter l. Percutaneous Transluminal Renal Angioplasty: Clinical and Angiographic Follow-up Results. Radiol Today 1983; (2) :126-30. 13. Martin LG, Casarella WJ, Gaylord GM. Azotemia Caused by Renal Artery Stenosis: Treatment by Percutaneous Angioplasty. AJR 1988: 150:839-44. 14. Leuscher TF, Keller HM, Imhof HM, Greminger P, Kuhlmann U, Largiader F, Schneider J, Vetter W. Fibromuscular Hyperplasia: Extension of the Disease and Therapeutic Outcome. Nephron 1986; 44(1):109-14. 15. Zech P, Finaz de Villaine J, Pozet N, Sassard J, Vincent M, Labeeuw M, Collard M, Had-Aissa A. Surgical Versus Medical Treatment in Renovascular Hypertension. Nephron 1986; 44(1) :105-I8. 16. Dean RH. Comparison of Medical and Surgical Treatment of Renovascular Hypertension. Nephron 1986; 44(1):101-4. 17. Rose BD. Renovascular Hypertension. In: Pat-hophysiology of Renal Disease. 18. Tegtmeyer CJ, Sos TA. Techniques of Renal Angioplasty. Radiology 1986; 161:577-86. Author's address: Doc. dr. Dušan Pavčnik. dr. med., Inštitut za rentgenologijo Univerzitetnega kliničnega centra Zaloška 7, 61000 Ljubljana Radiol lugosl 1990; 24: 137-45. 145