Intracaval endovascular ultrasonography for large adrenal and retroperitoneal tumors
Toyone Kikumori, MD, Tsuneo Imai, MD, Tetsuya Kaneko, MD, Hiroyuki Sugimoto, MD, Arihiro Shibata, MD, Yatsuka Hibi, MD, and Akimasa Nakao, MD, Nagoya, Japan
Background. An accurate diagnosis of inferior vena cava (IVC) invasion is important in deciding the surgical strategy for a large adrenal tumor. We investigated the diagnostic value of intracaval endovascular ultrasonography (ICEUS) for invasion of the IVC by a large adrenal tumor. Methods. Nine of 163 patients with adrenal and retroperitoneal tumors underwent ICEUS between 1993 and 2002. Intravascular ultrasonography was performed through the right femoral vein with the use of an 8Fr, 20-MHz transducer. The diagnostic criterion for detecting IVC invasion with ICEUS was identification of destruction of a single echogenic layer of the IVC wall or identification of an intracaval tumor mass. The ICEUS finding was confirmed by pathologic examination.
Results. The mean diameter of the tumors in 9 patients undergoing ICEUS and resection was 12.6 cm (range, 8.6-16 cm). Pathologic diagnosis varied: adrenocortical carcinoma, 4; malignant pheochromocytoma, 1; leiomyosarcoma, 1; metastatic lung cancer, 1; paraganglioma, 1; and neurilemmoma, 1. Vascular invasion was identified in 2 patients by ICEUS and confirmed by examination of resected specimens. The sensitivity, specificity, and positive predictive values of ICEUS for the diagnosis of the IVC invasion were 100%, 100%, and 100%, respectively. However, these values for computed tomography were 100%, 14%, and 25%, respectively; and for cavography, 100%, 57%, and 40%, respectively.
Conclusions. ICEUS provides confirmatory information regarding tumor invasion of the IVC. This modality also can assist in formulating an operative strategy for large adrenal or retroperitoneal tumors. (Surgery 2003;134:989-94.)
From the Department of Surgery II, Nagoya University School of Medicine, Nagoya, Japan
INVASION OF THE INFERIOR VENA CAVA (IVC) by adrenal neoplasms has a marked impact on the surgical strategy, including preparation of vascular recon- struction. Large adrenal or retroperitoneal tumors tend to invade the IVC because of the anatomic position of the adrenal glands, especially the right gland.1 Thus, a preoperative evaluation of the IVC is a prerequisite for adrenal neoplasms. Conven- tional imaging techniques (e.g., computed tomog- raphy [CT], cavography, and extracorporeal ultrasonography) have not provided reliable con- firmatory information regarding IVC invasion, espe- cially in the presence of compression of the IVC.2,3
Presented at the 24th Annual Meeting of the American Association of Endocrine Surgeons, San Diego, California, May 11-14, 2003.
Reprint requests: Tsuneo Imai MD, Department of Surgery II, Nagoya University School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya, 466-8550, Japan.
@ 2003, Mosby, Inc. All rights reserved. 0039-6060/2003/$30.00 + 0 doi:10.1016/S0039-6060(03)00484-7
Malignant neoplasms often develop a thrombus in the IVC. It is important to detect the extent of the thrombus and any invasion of the vessel wall. The conventional imaging techniques mentioned above generally cannot provide precise informa- tion on IVC invasion or the real-time observation of a tumor thrombus in the IVC. High-resolution intravascular ultrasonography that allows real-time intraluminal observations has been developed; this technique facilitates evaluations of arterial struc- tures,4 cardiovascular lesions after angioplasty,5 and portal vein invasion by pancreaticobiliary carci- noma.6 It even allows real-time intraluminal obser- vation. In this report, we reviewed the clinical records of patients with large adrenal or retroper- itoneal neoplasms, which were suspected of in- filtrating the IVC, and verified the usefulness of this modality.
SUBJECTS AND METHODS
Nine of 163 patients with adrenal or retroperi- toneal neoplasms underwent intracaval endo- vascular ultrasonography (ICEUS) between 1993
| Case | Sex | Age | Pathologic diagnosis | Laterality | Max diameter (cm) | Diagnosis | Organ(s) resected concomitantly | ||
|---|---|---|---|---|---|---|---|---|---|
| ICEUS | CT | Cavography | |||||||
| 1 | F | 41 | Adrenocortical carcinoma | L | 12 | * | y | y | Left kidney |
| 2 | F | 71 | Adrenocortical carcinoma | R | 16 | Intact | Invasion | Intact | |
| 3 | F | 43 | Retroperitoneal leiomyosarcoma | R | 10 | Invasion | Invasion | Invasion | IVC|| |
| 4 | M | 35 | Adrenal pheochromocytoma | R | 15 | ៛ | ₮ | z | Liver§, IVC|| |
| 5 | M | 70 | Paraganglioma | R | 8.6 | Intact | Invasion | Invasion | Left kidney |
| 6 | F | 32 | Adrenocortical carcinoma | R | 13.5 | Intact | Invasion | Invasion | Liver§ |
| 7 | M | 68 | Neurilemmoma | R | 13 | Intact | Invasion | Intact | |
| 8 | F | 34 | Adrenocortical carcinoma | R | 13 | Intact | Intact | Intact | |
| 9 | M | 73 | Adrenal metastasis of lung carcinoma | R | 12 | Intact | Invasion | Intact | Liver§ |
ICEUS, Intracaval endovascular ultrasonography; CT, computed tomography.
* To and fro movement of tumor thrombus in the confluence of left renal vein.
t Tumor thrombus in the confluence of left renal vein.
# Invasion and tumor thrombus.
§ Segmentectomy.
Reconstructed with artificial vessel.
and 2002 in the Department of Surgery II, Nagoya University School of Medicine (Table I). Informed consent was obtained from each patient. The selected patients were suspected of having direct invasion of the IVC on the basis of CT or extracor- poreal ultrasonographic findings. Our study group consisted of 4 men and 5 women with a mean age of 52 years (range, 32-73 years). Operative findings revealed that the mean tumor diameter was 12.6 cm. Pathologic diagnoses were obtained in all 9 patients (Table I), with malignancies indicated in 7 patients. The ICEUS findings were compared with the patho- logic results of resected specimens or intraoperative findings.
The intravascular ultrasonography catheter used in this study had a rotating tip and consisted of a rotating radial 20-MHz transducer attached to the terminus of a drive shaft that was housed and immersed in water for coupling (Aloka, Tokyo, Japan). The ultrasonography beam is projected from the transducer to create a 360-degree, real- time image perpendicular to the catheter. Axial resolution was set at 230 um, lateral resolution at 120 um, and the radius of tissue penetration at 20 mm. The catheter was passed through the right femoral vein into the right atrium. As the catheter was withdrawn, cross-sectional ultrasonographic images of the area under investigation were re- corded, both on videotape and on individual still frames for subsequent analysis. In this study, ICEUS was performed simultaneously with cavography. Both anteroposterior and lateral planes of view were obtained for cavography to differentiate between the extrinsic compression and invasion of the IVC.
In all 9 patients, a preoperative CT was per- formed with a commercially available scanner-an incremental dynamic CT with 5-mm-thick contig- uous sections. The ICEUS and cavography were reviewed independently by two of the authors. The diagnostic criterion for CT detection of an IVC involvement was an irregular deformity of the IVC or a caval defect. The diagnostic criterion for ICEUS detection of an IVC invasion was identifica- tion of the destruction of a single echogenic layer of the IVC wall or identification of an intraca- val tumor mass. When the echogenic layer of the IVC remained intact and no intracaval tumor was detected by ICEUS, the neoplasm was considered negative for IVC invasion.
RESULTS
Pathologic assessment, preoperative evaluation, and surgical procedure are summarized in Table I. ICEUS was carried out safely in all patients in about 15 minutes. No procedure-related complications developed. Basically, the normal portion of the IVC was visualized as a hyperechoic ring with a thickness of 1 mm, depending on the location of the ultra- sonography catheter (Fig 1). Neoplastic involve- ment of the IVC was demonstrated by ICEUS in 2 of the 9 patients in whom an echogenic band of the IVC was disrupted (Table I, Case 3; Fig 2, A) and a tumor thrombus was visualized (Table I, Case 4; Fig 2, B). Invasion of the IVC was confirmed patho- logically with the use of resected specimens. A rep- resentative case of dissociated findings (Table I, Case 6) is shown in Figure 3. Cavography (Fig 3, A)
and CT (Fig 3, B) suggested invasion of the IVC. However, ICEUS demonstrated that the echogenic band was intact (Fig 3, C). Thus, the patient was diagnosed as negative for IVC invasion, which was later confirmed by surgery. In 1 patient (Table I, Case 1), ICEUS visualized a free-floating tumor thrombus. A to-and-fro respiratory movement of the intracaval thrombus was seen during real-time ultrasonographic observation.7 In this case of a left adrenal carcinoma, the thrombus was drawn back into the left renal vein and then ligated.
Direct parietal invasion of the IVC wall was diagnosed on the basis of detection by all 3 modalities concordantly in 2 patients (Table I, Cases 3, 4). On the other hand, ICEUS ruled out IVC invasion in 6 patients in whom it had been suggested by CT (Table I, Cases 1, 2, 5, 6, 7, 9), and in 3 in whom it had been similarly suggested by cavography (Table I, Cases 1, 5, 6). ICEUS accu- rately diagnosed tumor compression without ex- ception, despite the “IVC invasion” suggested by CT and/or cavography. The results of tumor involvement of the IVC detected by ICEUS were compared with those of CT and cavography. The sensitivity, specificity, and positive predictive values of ICEUS were 100%, 100%, and 100%, respec- tively, whereas these values were 100%, 14%, and 25%, respectively, for CT; and 100%, 57%, and 40%, respectively, for cavography.
DISCUSSION
Retroperitoneal neoplasms often involve the IVC. Adrenal carcinoma also tends to invade the IVC with an incidence of approximately 20%.1,8 Combined resection of the IVC for adrenal carcinoma has been reported,9 and improvements in surgical techniques expand its indications. Although debulking of nonfunctioning malignant tumors has little effect, palliative surgery plays an important role in the treatment of functioning, hormonally active neoplasms. The endocrinopathy can be better managed if a significant debulking of the secreting neoplastic mass can be accom- plished.1º Designing the operative strategy for such aggressive surgery requires an accurate preopera- tive diagnosis of tumor involvement of the IVC.
The diagnostic limitation of conventional imag- ing techniques is their occasional failure to distin- guish between direct parietal invasion of the vessel wall and extrinsic compression. Cavography of a large tumor adjacent to the IVC commonly shows a spurious obstruction, which is often misinter- preted as an invasion.3 In CT and cavography, irregularity of the IVC wall often suggests a neo-
plastic parietal invasion, but this finding is not reliable. Intravascular ultrasonographic imaging is an innovative technology in which a high-frequency transducer placed on the tip of a catheter provides a high-resolution, cross-sectional image perpendic- ular to the long axis of the vessels. Recently, the intravascular ultrasonography catheter has been applied to the urinary tract,11 pancreatic duct, and bile duct.12 The authors have also reported the value of intraportal endovascular ultrasonography in the diagnosis of portal venous invasion by pancreaticobiliary carcinoma.6 The application of endovascular ultrasonography to the evaluation of an intracaval tumor thrombus has also been reported.7
The endovascular ultrasonography technique visualizes the wall of the IVC as a 1-layer echogenic ring. Conventional imaging by CT or extracorpo- real ultrasonography cannot visualize the IVC wall because the wall is too thin. ICEUS shows the echogenic layer of the wall, which allows differen- tiation between compression of the vessel wall and a tumor invasion. Thus, ICEUS can resolve the limitations of interpretation inherent in the con- ventional images of invasion of the IVC wall. This resolution allows a definite distinction to be made between compression and invasion.
A tumor thrombus in the IVC is another issue. Adherence of a tumor thrombus to the IVC wall requires a segmental or partial resection of the IVC, which usually requires venous bypass surgery. Cavography was previously used to detect an in- tracaval tumor thrombus, which is shown as a fill- ing defect within the vascular lumen. Cavography
A
B
A
B
C
R
0
cavography
CT
ICEUS
antero-posterior view
basically provides a vascular luminal profile of the contrast medium, so that a small tumor thrombus is detected as a filling defect. It should also be noted that cavographic and CT images are momentary; therefore, they make any evaluation of the degree of adherence of the thrombus to the IVC wall difficult because respiratory movement of the tumor thrombus cannot be visualized. Likewise, a false-positive diagnosis may be made because of laminar flow phenomenon and unopacified blood flowing from the adjacent venous branches during peripheral vein infusion. Since the evaluation of
IVC patency depends on the injection speed and total volume of contrast medium, a meticulous technique with a fast scanner is required. The real- time observation provided by ICEUS enables us to evaluate the degree of thrombus adherence to the IVC. Free space in the IVC lumen together with the respiratory to-and-fro movement of the tumor thrombus on ICEUS accurately indicates a free- floating lesion.
Another important feature of the technique is that it allows a precise preoperative diagnosis of the cranial extent of both tumor invasion and
intracaval tumor thrombus, thus facilitating the planning of an optimal surgical approach. ICEUS can determine the cranial extent by confirming the site of the tip of the ultrasonography probe under fluoroscopy along with the exact position of the IVC clamp. The sequential, real-time observation of intracaval ultrasonographic images can clearly reveal the length of both tumor invasion and in- tracaval tumor thrombus without the use of a potentially nephrotoxic contrast medium.
Nevertheless, certain limitations of ICEUS must be acknowledged. For example, it is unsuitable for severe IVC stenosis because a near-field ring-down effect often occurs. Another drawback is the relatively small extent of tissue penetration. Since the evaluation of extraluminal structures such as para-aortic lymph nodes and liver involvement is limited, a tumor may not be visualized in its entirety. There is also a potential risk of dislodging an intracaval tumor thrombus. Although no such problem was experienced in the present study, intracaval ultrasonography must be performed cautiously to prevent it.
The present study demonstrates that ICEUS can accurately resolve questions of neoplastic involve- ment of the IVC. This technique can differentiate extraluminal compression from direct parietal invasion. Moreover, ICEUS is uniquely suited to delineate the degree of tumor adherence within the IVC. When diagnosis of invasion by CT is not confirmatory or the size of tumor is extraordinarily large, use of ICEUS is recommended. Although ICEUS provides important information about cava involvement by a neoplasm, the test itself is invasive, expensive, and not without risk. ICEUS is not recommended as the first step in the evaluation of cava involvement by a tumor; rather, it should be regarded as compensation for the diagnostic limitations of conventional imaging techniques.
REFERENCES
1. Hedican SP, Marshall FF. Adrenocortical carcinoma with intracaval extension. J Urol 1997;158:2056-61.
2. Berdon WE, Baker DH, Santulli TV. Factors producing spurious obstruction of the inferior vena cava in infants and children with abdominal tumors. Radiology 1967;88:111-6.
3. Didier D, Racle A, Etievent JP, Weill F. Tumor thrombus of the inferior vena cava secondary to malignant abdomi- nal neoplasms: US and CT evaluation. Radiology 1987;162: 83-9.
4. Bolz KD, Tveit K, Nordby A, Myhre HO, Gaserud K, Ostrem G, et al. Intravascular ultrasonographic appearance of angiographically normal arteries related to age and the occurrence of vascular disease. Acta Radiol 1992;33:532-7.
5. Tobis JM, Mallery JA, Gessert J, Griffith J, Mahon D, Bessen M, et al. Intravascular ultrasound cross-sectional arterial
imaging before and after balloon angioplasty in vitro. Circulation 1989;80:873-82.
6. Kaneko T, Nakao A, Inoue S, Harada A, Nonami T, Itoh S, et al. Intraportal endovascular ultrasonography in the diagnosis of portal vein invasion by pancreatobiliary carcinoma. Ann Surg 1995;222:711-8.
7. Kaneko T, Nakao A, Funahashi H, Itoh S, Endo T, Harada A, et al. Intracaval ultrasonography in the diagnosis of tumour involvement of the vena cava. Br J Surg 1995;82:1655-9.
8. Tritos NA, Cushing GW, Heatley G, Libertino JA. Clinical features and prognostic factors associated with adrenocor- tical carcinoma: Lahey Clinic Medical Center experience. Am Surg 2000;66:73-9.
9. Huguet C, Ferri M, Gavelli A. Resection of the suprarenal inferior vena cava. The role of prosthetic replacement. Arch Surg 1995;130:793-7.
10. Soreide JA, Brabrand K, Thoresen SO. Adrenal cortical carcinoma in Norway, 1970-1984. World J Surg 1992;16: 663-7.
11. Lee DI, Bagley DH, Liu JB. Experience with endoluminal ultrasonography in the urinary tract. J Endourol 2001;15: 67-74.
12. Furukawa T, Tsukamoto Y, Naitoh Y, Hirooka Y, Hayakawa T. Differential diagnosis between benign and malignant localized stenosis of the main pancreatic duct by intraductal ultrasound of the pancreas. Am J Gastroenterol 1994; 89:2038-41.
DISCUSSION
Dr Thomas J. Musholt (Hannover, Germany). The technique of endovascular ultrasonography that you presented is very sophisticated but also very invasive. From the literature, I know that in adrenal tumors with invasion of the vena cava, the methods used to demonstrate intracaval tumor thrombi are extracorpo- real ultrasound, transesophageal echocardiography, and magnetic resonance imaging. I think that these latter imaging techniques are superior to the cavography that you chose to compare to endovascular ultrasonography. In my personal view, you get no additional information from the intracaval ultrasound compared with the less invasive methods of MRI and echocardiography.
Dr Kikumori. With regard to comparison with the lesser invasive modalities, actually some of our cases underwent MR angiography (MRA). However, resolution of MRA was similar to that of cavography. Therefore, it is very difficult to distinguish between compression and direct invasion by MRA, especially in the presence of compression. The resolution of intracaval ultrasonogra- phy is much higher than that of MRA.
Dr Henning Dralle (Halle, Germany). Very nice study. I have two questions. The first is, “Was there any risk of thromboembolism by the intraluminal device?” Many of the adrenal cancers invading the cava have an intense thrombosis within the cava.
The second question would be, “Does this device give you further information about the critical area of the hepatic veins and the entrance into the heart from the cava?” This would be a critical area in which to decide for a heart/lung machine or for a conventional surgical procedure.