- Email: [email protected]
Role of intravascular ultrasonography in detecting intravascular tumor thrombi: A preliminary report
Role of intravascular ultrasonography in detecting intravascular tumor thrombi: A preliminary report Tetsuya Kaneko, MD, Akimasa Nakao, MD, Soichiro Inoue, MD, Hiroomi Funahashi, MD, Akio Harada, MD, Toshiaki Nonami, MD, and Hiroshi Takagi, MD, Nagoya, Japan
Background. We evaluated the role of intravascular ultrasonography in the diagnosis of intravascular tumor thrornbi. Methods. During the past 2 years intracaval endovascular ultrasonography was performed in 26 selected patients to diagnose inferior vena cava invasion. Results of positive intracaval endovascular ultrasonogram were correlated with the pathologic findings of resected specimens and autopsy and with other imaging technologies such as computed tornography and angiography. Results. Six patients had positive studies of intracaval tumor thrornbus. In all cases detailed horizontal images perpendicular to the inferior vena cava axis were studied. Five of the six patients underwent resection. Intravascular ultrasonography correctly predicted the extent of the tumor thrombus, the degree of tumor adherence to the vessel wall, and the intralurninal movement of the tumor thrombus. Floating thrombi were visualized as an intralurninal to-and-fro movement. Thrornbus adhesion to the vessel wall appeared as an absence of space between the tumor and the wall, with no respiratory movement of the thrornbus. Conclusions. Intravascular ultrasonography ~as useful for the accurate diagnosis of intravascular tumor thrornbi and aided in formulating the operative strategy. (SuR(;ERY 199,5;117:538-dd.)
From the Department of Surgery H, Faculty of Medicine, Nagoya University, Nagoya, Japan
INTRAVENOUS
PATIENTS AND METHODS
Accepted for publication Sept. 26, 1994. Reprint requests: Tetsuya Kaneko, MI), Department of Surgery II, Faculty of Medicine, Nagoya University,65 Tsurumai-cho,Showaku, Nagoya 466, Japan. Copyright ~) 1995 by Mosby-Year Book, Inc. 0039-6060/95/$3.00 + 0 11/56/60847
ICEUS was performed in 26 selected patients with possible invasion of the inferior vena cava (IVC) wall and intracaval tumor thrombi on the basis of the results of computed tomography (CT) or conventional ultrasonography from February 1992 to July 1994. Of the 26 patients 11 had hepatocellular carcinomas, five had metastatic liver carcinomas, two had cholangiocellular carcinomas, two had renal cell carcinomas, two had adrenal carcinomas, two had pancreatic carcinomas, and one patient each had biliary cystadenocarcinoma and duodenal cancer. A surgical procedure was performed on 21 patients. Seventeen patients were resected, including three patients who underwent partial resection of the IVC. The IVUS catheter (Aloka Inc., Tokyo, Japan) used in this study consisted of a rotating radial 15 M H z transducer attached to the terminus of a drive shaft, housed and bathed in water for coupling. The diameter of the catheter was 8F. The ultrasonographic beam is projected from the transducer to create a 360 degree real-time image that is perpendicular to the catheter. Axial resolution is set at 300 #m, lateral resolution at 150 #m, and the radius of tissue penetration at 20 mm.
S P R E A I ) O F T U M O R S into the v e n a c a v a poses a challenging surgical undertaking) '2 In some tumors such as hepatocellular carcinoma and renal cell carcinoma, intracaval tumor spread is relatively common. To formulate an appropriate operative strategy for resecting these tumors, it is important to accurately diagnose the presence and extent of intravascular tumor spread. Intravascular ultrasonographic (IVUS) imaging has been developed recently to assess arterial structures. 3 This system operates at considerably higher frequencies than current ultrasonographic systems to produce highresolution cross-sectional images in real time. IVUS was also applied to the portal vein for the accurate diagnosis of portal venous invasion by pancreatic cancer. 4 We have applied this technology to the diagnosis of vascular invasion by tumor and report here the results of intracaval endovascular ultrasonography (ICEUS) in the diagnosis of intravascular tumor thrombus.
538
SURGERY
Surgery Volume 117, Number 5
Kaneko et al.
539
2mm
Fig. 1. A, ICEUS shows an intracaval tumor thrombus (small arrows). No space is seen between tumor thrombus and venous wall (large arrows). Tumor thrombus developed from inferior right hepatic vein. B, Tumor thrombus adherent to wall of IVC at junction of inferior right hepatic vein. (Hematoxylin-eosin stain; original magnification X4).
In I C E U S the IVUS catheter was passed through the right femoral vein to the suprahepatic IVC under fluoroscopic guidance. From there the catheter was gradually withdrawn, while cross-sectional ultrasonographic images of the area under investigation were recorded on videotape for later analysis and individual still frames were recorded with Polaroid film (Fuji Film Inc., Tokyo, Japan). Patients with a positive I C E U S study underwent abdominal C T (including a dynamic study) and inferior cavography. The results of the I C E U S studies were compared with those of the two other imaging procedures. The findings on I C E U S were compared with the pathology specimens of those patients with positive studies who underwent resection and autopsy.
RESULTS I C E U S was performed in all patients without complications. Most studies lasted 20 minutes. Detailed horizontal images perpendicular to the IVC axis were obtained. Six of the 26 IVUS studies were positive for tumor thrombi on the basis of the criterion that the echogenic mass was observed in the vascular lumen. T h e patients with positive studies had the following tumors. In intracaval tumor thrombus three hepatocellular carcinomas, two renal cell carcinomas, and one adrenal carcinoma were present. The patients with positive studies were four men and two women, with a mean age of 58.3 years (range, 41 to 74 years). Among these six patients were five in whom postcontrast C T detected intravascular tumor thrombi as an intravascular defect
540
Kaneko et al.
Surgery May 7995
Fig. 2. A, ICEUS shows respiratory movement of tumor thrombus (arrow,S. Tumor thrombus draws back into left renal vein during expiration. B, ICEUS shows respiratory movement of tumor thrombus (arrows). Tumor thrombus enters IVC during inspiration.
and five in whom angiography detected intravascular tumor thrombi as an intravascular filling defect. The study as a whole indicated the presence of thrombi in five of the six patients. Five of the six patients underwent surgical resection of their tumors. The remaining one did not undergo operation because of poor liver function. Autopsy was performed in this patient. In all six cases intravascular tumor thrombi were confirmed by pathologic examination (Table). IVUS indicated specific pathologic findings, visualizing the extent of the tumor thrombus, the degree of tumor adhesion to the vessel wall, and the intraluminal movement of the thrombus. The extent of the intracaval tumor thrombus was determined by observing the most cephalad portion of the tumor thrombus, and the posi-
tion of the IVUS catheter tip was simultaneously checked under fluoroscopy. Tumor adhesion to the vessel was evaluated by observing the separation between the tumor thrombus and the vessel wall. An intraluminal space observed between the tumor thrombus and the vessel indicated that part of the tumor thrombus was adhering to the vessel wall. In addition, real-time respiratory movement of the tumor thrombus could be observed, making it possible to distinguish between floating and nonfloating thrombi. Four (cases 1, 2, 5, and 6) of six intracaval tumor thrombi adhered to the IVC at the level of the draining portion of the vein (Fig. 1, A). In all four cases the tumor was excised together with partial resection of the wall of the IVC around the orifice of the draining vein.
Surgery Volume 117, Number 5
Kaneko et al.
Fig. 3. A, Intracaval tumor thrombus was visualized in CT IVC in cavogram. On histologic examination tumor adherence to the IVC was confirmed (Fig. 1, B). T w o (cases 3 and 4) of six intracaval tumor thrombi were free-floating thrombi that were recognized by ICEUS. To-and-fro respiratory movement of the tumor thrombus was observed. At expiration the tumor thrombus withdrew from the draining vein, and at inspiration it entered the IVC (Fig. 2). C T scan showed the tumor thrombus in the lumen of the IVC (Fig. 3, A). Cavogram also showed the intracaval tumor thrombus (Fig. 3, B). But these two techniques could not determine whether an intracaval tumor thrombus was floating or nonfloat-
scan
541
(arrow); B, Tumor thrombus developed into
ing. According to the finding of C T and cavography, the IVC should be clamped above and below the intracaval tumor thrombus for removal of the tumor. In case 3 the tumor thrombus was withdrawn into the left renal vein so the root of this vein could be ligated. T h e I C E U S finding was confirmed by the resected specimen (Fig. 4). In one case (case 6) the cephalad extent of the intracaval tumor thrombus was different between cavography and ICEUS. In cavography the cephalad extent of the intracaval tumor thrombus was diagnosed below the diaphragm (Fig. 5, A). I C E U S revealed that the intracaval tumor thrombus developed at a point just into the
542
Kaneko et al.
Surgery May 1995
Fig. 4. Free-floating thrombus without adhesion to vessel wall was confirmed by resected specimen.
Table. Comparison of IVUS, CT, and angiographic findings in six patients with tumor thrombi Case no.
Age/ Gender
1
54/M
HCC (Sv, vm)
2
49/M
HCC (Svll)
3
41/F
Left adrenal cancer
4
74/M
HCC (Slv, v, vm)
5
70/M
Right renal cancer
6
62/F
Left renal cancer
Diagnosis
IVUS
CT
Angiography
Nonfloating thrombus; adhering to orifice of RHV Nonfloating thrombus; adhering to orifice of IRHV Floating thrombus; to-and-fro movement Floating thrombus; to-and-fro movement Nonfloating thrombus; adhering to orifice of RRV Nonfloating thrombus; adhering to orifice of LRV; extension to RA
IVC defect
IVC defect
IVC defect
IVC defect
IVC defect
IVC defect
NF
NF
IVC defect
IVC defect
IVC defect below RA
IVC defect below RA
Operation Hepatic right lobectomy + IVC partial resection Hepatic right lobectomy + IVC partial resection Tumor resection + left nephrectomy Not performed (autopsy) Right nephrectomy + IVC partial resection Left nephrectomy + IVC partial resection
HCC, Hepatocellular carcinoma; RttV, right hepatic vein; IRtlV, inferior right hepatic vein; NF, no unusual finding; RA, right atrium.
right atrium (Fig. 5, B). In I C E U S imaging the middle portion of the tumor thrombus was floating (Fig. 5, C). At the draining portion of the left renal vein no space was noted between the tumor thrombus and the IVC wall in I C E U S (Fig. 5, D). At this level, adhesion of the tumor thrombus to the IVC wall could be diagnosed.
DISCUSSION Hepatocellular carcinoma has a tendency to venous invasion. Because the incidence of hepatocellular carcinoma in J a p a n is about 10 times that in Western
countries, s such tumor extension is encountered more frequently. There is a 10% incidence of tumor thrombus in the IVC. Renal and adrenal carcinomas also may invade the renal vein and can extend into the IVC.6, 7 For hepatic tumors with intracaval spread an aggressive radical approach, including resection of the intracaval tumor thrombus, has been advocated, s A recent report 9 has suggested that for renal cell carcinoma with vena cava involvement an aggressive surgical approach is warranted in the absence of identifiable metastases.
Surgery Volume 717, Number 5
Kaneko et al.
543
Fig. 5. A, Inferior cavogram shows that tip of intracaval tumor thrombus was below the diaphragm. B, ICEUS shows that tip of intracaval tumor thrombus (arrow) was visualized at right atrium; C, At draining portion of hepatic vein (small arrow) intracaval tumor thrombus was not adhered to IVC wall; Large arrow, diaphragm; D, At draining portion of left renal vein, the intracaval tumor thrombus adhered to IVC wall (arrows).
To formulate an operative strategy for resecting these tumors, it is essential to determine the extent of the tumor in the vena cava and the degree of adherence to the wall of these vessels. A tumor thrombus that extends into the draining vein near the surgical resection margin or into the IVC is at risk of being dislodged during resection. A tumor thrombus that is confined to the draining vein proximal to the point of surgical ligation is generally not at risk of being so dislodged. With real-time observation of the tumor thrombus by a high-resolution IVUS catheter, the spatial relation between the tumor thrombus and the vessel wall was clearly delineated in our patients. T h e distinction between a free-floating thrombus and a non-free-floating thrombus is impor-
tant. A free-floating thrombus that does not adhere to the margin of the draining vein can be removed by pulling the main tumor, and resection can be performed at the draining portion to the IVC without resection of the IVC wall. In our study respiratory to-and-fro movement of an intracaval tumor thrombus seen on I C E U S indicated a free-floating thrombus. In nonfloating thrombi the intracaval thrombus cannot be drawn back into the draining vein, and partial resection of the IVC wall is necessary. The extent of intracaval tumor thrombi can be diagnosed by conventional imaging techniques such as C T and cavography but not the degree of tumor adherence to the IVC wall. The transverse plane images obtained
544
Kaneko et al.
by C T are only m o m e n t a r y . T h e t u m o r t h r o m b u s is seen as a low attenuation filling defect in the vena cava. D e spite its accuracy in identifying t u m o r t h r o m b i , C T is u n r e l i a b l e in delineating the c e p h a l a d extent of a t h r o m b u s t~ because that depends on the injection speed and the total v o l u m e of contrast m e d i u m . A l t h o u g h C T scans in our study w e r e limited to axial views, three-dim e n s i o n a l helical C T scans are possible and m a y offer i m p r o v e d results. I n t r a v a s c u l a r t u m o r thrombi are detected as filling defects of the vascular l u m e n by a n g i o g r a p h y . H o w ever, a single a n t e g r a d e cavogram is often insufficient to delineate the cephalad extent of a t h r o m b u s , particularly in patients w i t h vena cava o c c l u s i o n ) 1 W h e n the area of the transverse section of the intravascular t u m o r t h r o m b u s is smaller t h a n one half the a r e a of the vascular lumen, the t h r o m b u s m a y not be visualized as a filling defect. If the t u m o r t h r o m b u s is small and does not exist in the lateral side of the l u m i nal profile, it might be overlooked. A d d i t i o n a l relative disadvantages of a n g i o g r a p h y include the need for administration of potentially n e p h r o t o x i c contrast medium. In our study we experienced a case in w h i c h the intracaval t u m o r t h r o m b u s developed in a m o r e cephalad portion t h a n the position diagnosed by inferior cavography. I C E U S could d e t e r m i n e the c e p h a l a d extent accurately, along with the exact position of the I V C clamp. I C E U S provides sequential, r e a l - t i m e i m a g i n g of intravascular t u m o r thrombi. It can accurately detect tum o r thrombi, the degree of t u m o r adhesion to the vessel wall, and the length of the t u m o r thrombus. T h u s it m a y provide m o r e v a l u a b l e i n f o r m a t i o n about intravascular t u m o r thrombi t h a n C T or angiogr a p h y and can be of assistance in the selection of the ap-
Surgery May 1995
p r o p r i a t e o p e r a t i o n for t u m o r s w i t h intracaval t u m o r thrombi. REFERENCES 1. Kumada K, Shimahara Y, Fukui K, et al. Extended right hepatic lobectomy: combined resection of inferior vena cava and its reconstruction by EPTFE. Acta Chir Stand 1988;154: 481-3. 2. Marshall FF, Reiz BA. Technique for removal of renal cell carcinoma with suprahepatic vena caval tumor thrombus. Urol Clin North Am 1986;13:551-7. 3. Tobis JM, Mallery J, Mahon D, et al. Intravascular ultrasound imaging of human coronary arteries in vivo: analysis of tissue characterizations with comparison to in vitro histological specimens. Circulation 1991;83:913-26. 4. Kaneko T, Nakao A, Harada A, et al. Intraportal endovascular ultrasonography in pancreatic cancer: a new technique for the diagnosis of portal invasion-a preliminary report. SURGERY 1994;115:438-44. 5. Kojiro M, Nakashima T. Pathology of hepatocellular carcinoma. In: Okuda K, Ishak K, eds. Neoplasms of the liver. 1st ed. Tokyo: Springer-Verlag, 1987:81-104. 6. Schefft P, Novick AC, Straffon RA, et al. Extension of renal cell carcinoma into the vena cava: rationale for aggressive surgical management. J Urol 1978;120:28-31. 7. Levine E, de Varies P, Wetzel LH. MR imaging of inferior vena caval recurrence of extraadrenal pheochromocytoma. J Comput Assist Tomogr 1987;11:717-8. 8. Fujisaki M, Kurihara E, Kikuchi K, et al. Hepatocellular carcinoma with tumor thrombus extending into the right atrium: report of a successful resection with the use of cardiopulmonary bypass. SURGERY1990;109:214-9. 9. Waters WB, Richie JP. Aggressive surgical approach to renal cell carcinoma: review of 130 cases. J Urol 1979;122:306-9. 10. Goldbarle DA, Novick AC, Lorig R, et al. Magnetic resonance imaging for assessment of vena caval tumor thrombi: a comparative study with venacavography and computerized tomography scanning. J Urol 1990;144:1100-4. I 1. Siminovitch JMP, Montie JE, Straffon RA. Inferior venacavography in the preoperative assessment of renal cell adenocarcinoma. J Urol 1982;128:908-9.
B O U N D VOLUMES AVAILABLE T O SUBSCRIBERS
Bound volumes of SURGERY are available to subscribers (only) for the 1995 issues from the publisher at a cost of $74.00 for domestic, $100.58 for Canadian, and $94.00 for international subscribers for Vol. 117 (January to June) and Vol. 118 (July to December). Shipping charges are included. Each bound volume contains a subject index and an author index, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular journal subscription. Payment must accompany all orders. Contact Mosby-Year Book, Inc., Subscription Services, 11830 Westline Industrial Dr., St. Louis, M O 63146-3318; phone (800)3254177, ext. 4351, or (314)453-4351.
Background. We evaluated the role of intravascular ultrasonography in the diagnosis of intravascular tumor thrornbi. Methods. During the past 2 years intracaval endovascular ultrasonography was performed in 26 selected patients to diagnose inferior vena cava invasion. Results of positive intracaval endovascular ultrasonogram were correlated with the pathologic findings of resected specimens and autopsy and with other imaging technologies such as computed tornography and angiography. Results. Six patients had positive studies of intracaval tumor thrornbus. In all cases detailed horizontal images perpendicular to the inferior vena cava axis were studied. Five of the six patients underwent resection. Intravascular ultrasonography correctly predicted the extent of the tumor thrombus, the degree of tumor adherence to the vessel wall, and the intralurninal movement of the tumor thrombus. Floating thrombi were visualized as an intralurninal to-and-fro movement. Thrornbus adhesion to the vessel wall appeared as an absence of space between the tumor and the wall, with no respiratory movement of the thrornbus. Conclusions. Intravascular ultrasonography ~as useful for the accurate diagnosis of intravascular tumor thrornbi and aided in formulating the operative strategy. (SuR(;ERY 199,5;117:538-dd.)
From the Department of Surgery H, Faculty of Medicine, Nagoya University, Nagoya, Japan
INTRAVENOUS
PATIENTS AND METHODS
Accepted for publication Sept. 26, 1994. Reprint requests: Tetsuya Kaneko, MI), Department of Surgery II, Faculty of Medicine, Nagoya University,65 Tsurumai-cho,Showaku, Nagoya 466, Japan. Copyright ~) 1995 by Mosby-Year Book, Inc. 0039-6060/95/$3.00 + 0 11/56/60847
ICEUS was performed in 26 selected patients with possible invasion of the inferior vena cava (IVC) wall and intracaval tumor thrombi on the basis of the results of computed tomography (CT) or conventional ultrasonography from February 1992 to July 1994. Of the 26 patients 11 had hepatocellular carcinomas, five had metastatic liver carcinomas, two had cholangiocellular carcinomas, two had renal cell carcinomas, two had adrenal carcinomas, two had pancreatic carcinomas, and one patient each had biliary cystadenocarcinoma and duodenal cancer. A surgical procedure was performed on 21 patients. Seventeen patients were resected, including three patients who underwent partial resection of the IVC. The IVUS catheter (Aloka Inc., Tokyo, Japan) used in this study consisted of a rotating radial 15 M H z transducer attached to the terminus of a drive shaft, housed and bathed in water for coupling. The diameter of the catheter was 8F. The ultrasonographic beam is projected from the transducer to create a 360 degree real-time image that is perpendicular to the catheter. Axial resolution is set at 300 #m, lateral resolution at 150 #m, and the radius of tissue penetration at 20 mm.
S P R E A I ) O F T U M O R S into the v e n a c a v a poses a challenging surgical undertaking) '2 In some tumors such as hepatocellular carcinoma and renal cell carcinoma, intracaval tumor spread is relatively common. To formulate an appropriate operative strategy for resecting these tumors, it is important to accurately diagnose the presence and extent of intravascular tumor spread. Intravascular ultrasonographic (IVUS) imaging has been developed recently to assess arterial structures. 3 This system operates at considerably higher frequencies than current ultrasonographic systems to produce highresolution cross-sectional images in real time. IVUS was also applied to the portal vein for the accurate diagnosis of portal venous invasion by pancreatic cancer. 4 We have applied this technology to the diagnosis of vascular invasion by tumor and report here the results of intracaval endovascular ultrasonography (ICEUS) in the diagnosis of intravascular tumor thrombus.
538
SURGERY
Surgery Volume 117, Number 5
Kaneko et al.
539
2mm
Fig. 1. A, ICEUS shows an intracaval tumor thrombus (small arrows). No space is seen between tumor thrombus and venous wall (large arrows). Tumor thrombus developed from inferior right hepatic vein. B, Tumor thrombus adherent to wall of IVC at junction of inferior right hepatic vein. (Hematoxylin-eosin stain; original magnification X4).
In I C E U S the IVUS catheter was passed through the right femoral vein to the suprahepatic IVC under fluoroscopic guidance. From there the catheter was gradually withdrawn, while cross-sectional ultrasonographic images of the area under investigation were recorded on videotape for later analysis and individual still frames were recorded with Polaroid film (Fuji Film Inc., Tokyo, Japan). Patients with a positive I C E U S study underwent abdominal C T (including a dynamic study) and inferior cavography. The results of the I C E U S studies were compared with those of the two other imaging procedures. The findings on I C E U S were compared with the pathology specimens of those patients with positive studies who underwent resection and autopsy.
RESULTS I C E U S was performed in all patients without complications. Most studies lasted 20 minutes. Detailed horizontal images perpendicular to the IVC axis were obtained. Six of the 26 IVUS studies were positive for tumor thrombi on the basis of the criterion that the echogenic mass was observed in the vascular lumen. T h e patients with positive studies had the following tumors. In intracaval tumor thrombus three hepatocellular carcinomas, two renal cell carcinomas, and one adrenal carcinoma were present. The patients with positive studies were four men and two women, with a mean age of 58.3 years (range, 41 to 74 years). Among these six patients were five in whom postcontrast C T detected intravascular tumor thrombi as an intravascular defect
540
Kaneko et al.
Surgery May 7995
Fig. 2. A, ICEUS shows respiratory movement of tumor thrombus (arrow,S. Tumor thrombus draws back into left renal vein during expiration. B, ICEUS shows respiratory movement of tumor thrombus (arrows). Tumor thrombus enters IVC during inspiration.
and five in whom angiography detected intravascular tumor thrombi as an intravascular filling defect. The study as a whole indicated the presence of thrombi in five of the six patients. Five of the six patients underwent surgical resection of their tumors. The remaining one did not undergo operation because of poor liver function. Autopsy was performed in this patient. In all six cases intravascular tumor thrombi were confirmed by pathologic examination (Table). IVUS indicated specific pathologic findings, visualizing the extent of the tumor thrombus, the degree of tumor adhesion to the vessel wall, and the intraluminal movement of the thrombus. The extent of the intracaval tumor thrombus was determined by observing the most cephalad portion of the tumor thrombus, and the posi-
tion of the IVUS catheter tip was simultaneously checked under fluoroscopy. Tumor adhesion to the vessel was evaluated by observing the separation between the tumor thrombus and the vessel wall. An intraluminal space observed between the tumor thrombus and the vessel indicated that part of the tumor thrombus was adhering to the vessel wall. In addition, real-time respiratory movement of the tumor thrombus could be observed, making it possible to distinguish between floating and nonfloating thrombi. Four (cases 1, 2, 5, and 6) of six intracaval tumor thrombi adhered to the IVC at the level of the draining portion of the vein (Fig. 1, A). In all four cases the tumor was excised together with partial resection of the wall of the IVC around the orifice of the draining vein.
Surgery Volume 117, Number 5
Kaneko et al.
Fig. 3. A, Intracaval tumor thrombus was visualized in CT IVC in cavogram. On histologic examination tumor adherence to the IVC was confirmed (Fig. 1, B). T w o (cases 3 and 4) of six intracaval tumor thrombi were free-floating thrombi that were recognized by ICEUS. To-and-fro respiratory movement of the tumor thrombus was observed. At expiration the tumor thrombus withdrew from the draining vein, and at inspiration it entered the IVC (Fig. 2). C T scan showed the tumor thrombus in the lumen of the IVC (Fig. 3, A). Cavogram also showed the intracaval tumor thrombus (Fig. 3, B). But these two techniques could not determine whether an intracaval tumor thrombus was floating or nonfloat-
scan
541
(arrow); B, Tumor thrombus developed into
ing. According to the finding of C T and cavography, the IVC should be clamped above and below the intracaval tumor thrombus for removal of the tumor. In case 3 the tumor thrombus was withdrawn into the left renal vein so the root of this vein could be ligated. T h e I C E U S finding was confirmed by the resected specimen (Fig. 4). In one case (case 6) the cephalad extent of the intracaval tumor thrombus was different between cavography and ICEUS. In cavography the cephalad extent of the intracaval tumor thrombus was diagnosed below the diaphragm (Fig. 5, A). I C E U S revealed that the intracaval tumor thrombus developed at a point just into the
542
Kaneko et al.
Surgery May 1995
Fig. 4. Free-floating thrombus without adhesion to vessel wall was confirmed by resected specimen.
Table. Comparison of IVUS, CT, and angiographic findings in six patients with tumor thrombi Case no.
Age/ Gender
1
54/M
HCC (Sv, vm)
2
49/M
HCC (Svll)
3
41/F
Left adrenal cancer
4
74/M
HCC (Slv, v, vm)
5
70/M
Right renal cancer
6
62/F
Left renal cancer
Diagnosis
IVUS
CT
Angiography
Nonfloating thrombus; adhering to orifice of RHV Nonfloating thrombus; adhering to orifice of IRHV Floating thrombus; to-and-fro movement Floating thrombus; to-and-fro movement Nonfloating thrombus; adhering to orifice of RRV Nonfloating thrombus; adhering to orifice of LRV; extension to RA
IVC defect
IVC defect
IVC defect
IVC defect
IVC defect
IVC defect
NF
NF
IVC defect
IVC defect
IVC defect below RA
IVC defect below RA
Operation Hepatic right lobectomy + IVC partial resection Hepatic right lobectomy + IVC partial resection Tumor resection + left nephrectomy Not performed (autopsy) Right nephrectomy + IVC partial resection Left nephrectomy + IVC partial resection
HCC, Hepatocellular carcinoma; RttV, right hepatic vein; IRtlV, inferior right hepatic vein; NF, no unusual finding; RA, right atrium.
right atrium (Fig. 5, B). In I C E U S imaging the middle portion of the tumor thrombus was floating (Fig. 5, C). At the draining portion of the left renal vein no space was noted between the tumor thrombus and the IVC wall in I C E U S (Fig. 5, D). At this level, adhesion of the tumor thrombus to the IVC wall could be diagnosed.
DISCUSSION Hepatocellular carcinoma has a tendency to venous invasion. Because the incidence of hepatocellular carcinoma in J a p a n is about 10 times that in Western
countries, s such tumor extension is encountered more frequently. There is a 10% incidence of tumor thrombus in the IVC. Renal and adrenal carcinomas also may invade the renal vein and can extend into the IVC.6, 7 For hepatic tumors with intracaval spread an aggressive radical approach, including resection of the intracaval tumor thrombus, has been advocated, s A recent report 9 has suggested that for renal cell carcinoma with vena cava involvement an aggressive surgical approach is warranted in the absence of identifiable metastases.
Surgery Volume 717, Number 5
Kaneko et al.
543
Fig. 5. A, Inferior cavogram shows that tip of intracaval tumor thrombus was below the diaphragm. B, ICEUS shows that tip of intracaval tumor thrombus (arrow) was visualized at right atrium; C, At draining portion of hepatic vein (small arrow) intracaval tumor thrombus was not adhered to IVC wall; Large arrow, diaphragm; D, At draining portion of left renal vein, the intracaval tumor thrombus adhered to IVC wall (arrows).
To formulate an operative strategy for resecting these tumors, it is essential to determine the extent of the tumor in the vena cava and the degree of adherence to the wall of these vessels. A tumor thrombus that extends into the draining vein near the surgical resection margin or into the IVC is at risk of being dislodged during resection. A tumor thrombus that is confined to the draining vein proximal to the point of surgical ligation is generally not at risk of being so dislodged. With real-time observation of the tumor thrombus by a high-resolution IVUS catheter, the spatial relation between the tumor thrombus and the vessel wall was clearly delineated in our patients. T h e distinction between a free-floating thrombus and a non-free-floating thrombus is impor-
tant. A free-floating thrombus that does not adhere to the margin of the draining vein can be removed by pulling the main tumor, and resection can be performed at the draining portion to the IVC without resection of the IVC wall. In our study respiratory to-and-fro movement of an intracaval tumor thrombus seen on I C E U S indicated a free-floating thrombus. In nonfloating thrombi the intracaval thrombus cannot be drawn back into the draining vein, and partial resection of the IVC wall is necessary. The extent of intracaval tumor thrombi can be diagnosed by conventional imaging techniques such as C T and cavography but not the degree of tumor adherence to the IVC wall. The transverse plane images obtained
544
Kaneko et al.
by C T are only m o m e n t a r y . T h e t u m o r t h r o m b u s is seen as a low attenuation filling defect in the vena cava. D e spite its accuracy in identifying t u m o r t h r o m b i , C T is u n r e l i a b l e in delineating the c e p h a l a d extent of a t h r o m b u s t~ because that depends on the injection speed and the total v o l u m e of contrast m e d i u m . A l t h o u g h C T scans in our study w e r e limited to axial views, three-dim e n s i o n a l helical C T scans are possible and m a y offer i m p r o v e d results. I n t r a v a s c u l a r t u m o r thrombi are detected as filling defects of the vascular l u m e n by a n g i o g r a p h y . H o w ever, a single a n t e g r a d e cavogram is often insufficient to delineate the cephalad extent of a t h r o m b u s , particularly in patients w i t h vena cava o c c l u s i o n ) 1 W h e n the area of the transverse section of the intravascular t u m o r t h r o m b u s is smaller t h a n one half the a r e a of the vascular lumen, the t h r o m b u s m a y not be visualized as a filling defect. If the t u m o r t h r o m b u s is small and does not exist in the lateral side of the l u m i nal profile, it might be overlooked. A d d i t i o n a l relative disadvantages of a n g i o g r a p h y include the need for administration of potentially n e p h r o t o x i c contrast medium. In our study we experienced a case in w h i c h the intracaval t u m o r t h r o m b u s developed in a m o r e cephalad portion t h a n the position diagnosed by inferior cavography. I C E U S could d e t e r m i n e the c e p h a l a d extent accurately, along with the exact position of the I V C clamp. I C E U S provides sequential, r e a l - t i m e i m a g i n g of intravascular t u m o r thrombi. It can accurately detect tum o r thrombi, the degree of t u m o r adhesion to the vessel wall, and the length of the t u m o r thrombus. T h u s it m a y provide m o r e v a l u a b l e i n f o r m a t i o n about intravascular t u m o r thrombi t h a n C T or angiogr a p h y and can be of assistance in the selection of the ap-
Surgery May 1995
p r o p r i a t e o p e r a t i o n for t u m o r s w i t h intracaval t u m o r thrombi. REFERENCES 1. Kumada K, Shimahara Y, Fukui K, et al. Extended right hepatic lobectomy: combined resection of inferior vena cava and its reconstruction by EPTFE. Acta Chir Stand 1988;154: 481-3. 2. Marshall FF, Reiz BA. Technique for removal of renal cell carcinoma with suprahepatic vena caval tumor thrombus. Urol Clin North Am 1986;13:551-7. 3. Tobis JM, Mallery J, Mahon D, et al. Intravascular ultrasound imaging of human coronary arteries in vivo: analysis of tissue characterizations with comparison to in vitro histological specimens. Circulation 1991;83:913-26. 4. Kaneko T, Nakao A, Harada A, et al. Intraportal endovascular ultrasonography in pancreatic cancer: a new technique for the diagnosis of portal invasion-a preliminary report. SURGERY 1994;115:438-44. 5. Kojiro M, Nakashima T. Pathology of hepatocellular carcinoma. In: Okuda K, Ishak K, eds. Neoplasms of the liver. 1st ed. Tokyo: Springer-Verlag, 1987:81-104. 6. Schefft P, Novick AC, Straffon RA, et al. Extension of renal cell carcinoma into the vena cava: rationale for aggressive surgical management. J Urol 1978;120:28-31. 7. Levine E, de Varies P, Wetzel LH. MR imaging of inferior vena caval recurrence of extraadrenal pheochromocytoma. J Comput Assist Tomogr 1987;11:717-8. 8. Fujisaki M, Kurihara E, Kikuchi K, et al. Hepatocellular carcinoma with tumor thrombus extending into the right atrium: report of a successful resection with the use of cardiopulmonary bypass. SURGERY1990;109:214-9. 9. Waters WB, Richie JP. Aggressive surgical approach to renal cell carcinoma: review of 130 cases. J Urol 1979;122:306-9. 10. Goldbarle DA, Novick AC, Lorig R, et al. Magnetic resonance imaging for assessment of vena caval tumor thrombi: a comparative study with venacavography and computerized tomography scanning. J Urol 1990;144:1100-4. I 1. Siminovitch JMP, Montie JE, Straffon RA. Inferior venacavography in the preoperative assessment of renal cell adenocarcinoma. J Urol 1982;128:908-9.
B O U N D VOLUMES AVAILABLE T O SUBSCRIBERS
Bound volumes of SURGERY are available to subscribers (only) for the 1995 issues from the publisher at a cost of $74.00 for domestic, $100.58 for Canadian, and $94.00 for international subscribers for Vol. 117 (January to June) and Vol. 118 (July to December). Shipping charges are included. Each bound volume contains a subject index and an author index, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular journal subscription. Payment must accompany all orders. Contact Mosby-Year Book, Inc., Subscription Services, 11830 Westline Industrial Dr., St. Louis, M O 63146-3318; phone (800)3254177, ext. 4351, or (314)453-4351.