Hironobu Nakamura, MD

Interventional radiology has developed tremendously since the term was first coined some 30 years ago.

 Its minimally invasive nature has made it suitable for use in emergencies and in cases when even the surgeons do not know where to begin. Radiological procedures rival surgical operations in the treatment of aneurysms (Figure 1).

Advances in interventional radiology have been helped by improvements in medical devices, and the trend of interventional radiologists snatching patients away from surgeons has slowed as development of new instruments has settled down. Exposure to radiation, a major shortcoming of interventional radiology, has begun to attract attention as well.

Transcatheter chemoembolization is the most common interventional procedure performed in Japan and other East Asian countries. The technique plays an important role in treating patients with hepatocellular carcinoma (HCC). The cancer cannot be resected easily in most patients because they also have liver cirrhosis.

The treatment strategy for HCC changed following clinical introduction of percutaneous microwave or radio-frequency tumor ablation.

 Many minimally invasive therapeutic modalities are available to treat small (up to 5 cm in diameter) single or multiple (up to three) localized HCC tumors. Choosing the best one is not easy, but modalities can be combined. Localized HCC lesions less than 5 cm in diameter are likely to be well controlled with either one modality or a combination of several, which is good news to patients who cannot tolerate partial hepatectomy.

Surgical resection is an option if liver function is good enough to tolerate surgery. We always perform angiography at our center for accurate diagnosis. CT arterial portography is carried out by placing a catheter in the superior mesenteric artery, and CT hepatic arteriography is performed as well. It is not uncommon to insert a microcatheter through the catheter placed in the hepatic artery to view the subsegment or subsubsegment lesion site. It is even possible to insert a microcatheter into a vessel feeding the tumor. Chemoembolization should then be performed in such a situation, aiming at complete necrosis of the tumor. The patient can be treated with minimal invasion if this is successful (Figure 2).

Systems that combine CT with digital subtraction angiography (DSA) have proved extremely useful.

 Local control has improved considerably following introduction of the angio/CT machine.

 A CT scan taken during or just after chemoembolization may confirm whether the tumor is completely filled with the anticancer drug, suspended in Lipiodol. The presence of a tumor-feeding artery may be investigated if Lipiodol is not sufficiently retained in the tumor, so that complete filling is possible. Poor response to treatment or insufficient retention of Lipiodol in the tumor indicates that chemoembolization should be supplemented or replaced with other treatments. A single lesion located around segment 1 (S1) of the liver, for instance, and fed by many tumor vessels may not respond to chemoembolization but can be surgically resected. If resection proves difficult, an ethanol injection might be a good option when the lesion is less than 2 cm in diameter. Microwave or RF ablation may be preferred if the tumor is up to 5 cm in diameter. Ethanol injection can be added to achieve complete necrosis of the tumor if postchemoembolization CT shows that Lipiodol is not retained in any part of the tumor.

We block blood flow percutaneously when using percutaneous microwave coagulation therapy or ablation to extend the range of coagulation.

 Portal blood flow into the region stops if the hepatic vein is occluded with a balloon catheter in the region of the tumor. Total blood flow in the region containing the tumor is also markedly decreased if the hepatic artery is occluded with a balloon catheter or gelatin sponge. Our experience has shown that the range of coagulation can be expanded from 3 cm to 5 cm with use of the blood flow blockade (Figure 3).

Surgical resection, ethanol injection, and microwave or RF ablation are unlikely to be used to treat advanced HCC with multiple intrahepatic metastases or portal invasion. Many of these patients should be treated with chemoembolization. Advanced HCC without portal invasion can be treated with strong chemoembolization if liver function is not impaired. High doses of Lipiodol or gelatin sponge can be used so that most of the tumor becomes necrotic. It is not uncommon for prolonged survival to follow this strategy. If hepatic function is impaired, however, Lipiodol or gelatin sponges should be used in lower doses. No clearcut evidence exists that chemoembolization given in this way can prolong life. We simply hope that a decrease in tumor size following the treatment will allow an increase in survival time.

Possible decrease of both portal blood flow and hepatic reserve should be considered if the portal vein is invaded. CT arterial portography is helpful when judging whether chemoembolization is feasible, providing information on both tumor extension and portal blood flow. Arterial infusion of anticancer drugs through a reservoir is used if chemoembolization cannot be performed, although response to this treatment is not good.

Therapeutic outcome of advanced HCC has not improved regardless of the treatment given. Several randomized controlled trials have been reported from Europe, all with negative results, showing tumor response but no survival benefit.

 It is difficult to judge whether chemoembolization should be chosen for patients with impaired hepatic function. Patients without liver dysfunction, however, appear to benefit greatly from chemoembolization. The lack of patients in Japan willing to receive conservative treatment means that, regretfully, we are unable to conduct randomized controlled studies ourselves.

We are developing stent-graft placement for aortic aneurysms in our department under the guidance of Dr. Munehiro Maeda. Stents are placed through a sheath inserted from a small incision in the inguinal region under local anesthetic. This is much less invasive than surgical procedures, which involve laparotomy or thoracotomy, and so can be used for high-risk patients who may not tolerate surgery. Anatomical limitations for stent-grafting do exist, though. The femoral and iliac arteries, which are the site of stent insertion, must be sufficiently wide and straight for unhindered advancement of a large-sized sheath. The neck of the aneurysm should also be more than 1.5 cm long, to allow the stents to be placed in a stable manner both distally and proximally to the aneurysm.

Basic research and instrument development for stent-graft placement from the late '80s and early '90s enabled clinical application to begin in the '90s. Stent-grafts were at first placed only in patients who were considered difficult to operate on, although they have since been used in operable patients as well. Various types of stent-grafts have been developed in Western countries and placed in more than 10,000 patients with aneurysms. The government in Japan has not approved any stent-grafts for clinical use, however. Each center has to use its own handmade versions. We make ours by covering commercially available Z stents with materials used to make artificial vessels, such as polyester (Dacron), polyurethane, or polytetrafluoroethylene (PTFE).

Radiologists and vascular surgeons at Osaka University Hospital work together in treating aneurysms. Our handmade stent-grafts are made up of spiral zigzag stents covered with 0.1-mm polyester combined with our own 15 to 18-French catheters.

 We have attempted placement of these stent-grafts in 57 patients with thoracic and abdominal aneurysms. The outcome has been good, with 54 of the 57 procedures proving to be successful (Figure 1). Stent insertion was suspended in two cases owing to anatomical limitations. An endoleak, the leakage of contrast agent from the space between the stent-graft and the arterial wall into the aneurysm, caused an aneurysm to rupture in the third case.

Ideally, stent-grafts will be developed in the future that can be placed through a small 10-French sheath, can follow severely kinked vessels, and do not allow endoleaks, and clearcut evidence of the value of stent-grafts should be accumulated from long-term follow-up data. Surgical operations may soon give way to stent-grafting.

Radiation exposure to both patients and staff has become a crucial topic of discussion as interventional radiology has developed. The advent of new techniques means that more patients are undergoing interventional procedures, and the increasing complexity of advanced techniques has lengthened examinations, adding to exposure time.

Reports of skin lesions in patients who had undergone interventional radiology procedures first appeared in 1994, and the number of reports has continued to rise. Lesions vary in severity from transient erythema and alopecia to severe skin necrosis. These symptoms are related to so-called deterministic effects, with severity depending on the radiation doses. All patient exposure is within the accepted range of medical exposure. Because patients are assumed to benefit from interventional radiology, there are no legal limitations on dose. Practitioners should always keep the likely patient exposure in mind during interventional procedures, however, and take every possible step to minimize it.

Our own survey showed the mean exposure of the area dorsal to the liver to be 973 mGy on average (with a maximum of 3543 mGy) in 39 patients who underwent transcatheter chemoembolization.11 No patient developed erythema. Repeated treatments may be given, however, and any exposure exceeding 1 Gy should be recorded in the clinical records.

Operators must also remain aware of their own exposure. Two interventional radiologists reported lens abnormalities after performing flouroscopy-guided procedures with an over-the-table x-ray tube.

 Protective glasses could have prevented this, and an under-the-table x-ray tube is recommended for fluoroscopy-guided intervention if possible. We use a C-arm-type fluoroscope, which can convert from over- to under-the-table and vice versa. The over-the-table x-ray tube is used only when absolutely necessary.

The International Commission on Radiological Protection has set limits on the permissible dose of occupational radiation exposure to staff. The mean dose should remain below 20 mSv/year over five years, and the total dose in any year should not exceed 50 mSv. Interventional radiologists should wear two dosimeters (film badges), one inside the protector covering the trunk, and the other on the head or neck. We found the mean exposure of staff performing transcatheter chemoembolization to be 0.01 mSv (0.05 mSv maximum) inside the trunk protector, and 0.04 mSv (0.15 mSv maximum) on the head using an under-the-table x-ray tube. Exposure levels to the head are about 10 times higher if an over-the-table x-ray tube is used.

We hope that vascular interventional radiology will continue to make steady progress in the 21st century and that the problem of radiation exposure will be resolved by the more widespread use of MRI and robotics.

Prof. Nakamura is chair of radiology at the Osaka University School of Medicine in Japan.

1 Margulis AR. Interventional diagnostic radiology-a new subspecialty. AJR 1967;99:761-762.

2 Wallace S. Interventional radiology. Cancer 1976;37:517-531.

3 Livraghi T, Goldberg N, Lazzaroni S, et al. Small hepatocellular carcinoma: treatment with radio-frequency ablation versus ethanol injection. Radiology 1999;210:655-661.

4 Dod G D III, Soulen MC, Kane RA, et al. Minimally invasive treatment of malignant hepatic tumors: at the threshold of a major breakthrough. Radiographics 2000;20:9-27.

5 Nakamura H, Murata M, Oi H, et al. Diagnostic imaging of hepatocellular carcinoma in interventional radiology. Jpn J Cancer Chemother 2000;27(Suppl II):332-335.

6 Murata M, Oi H, Nakamura H, et al. Early result of transcatheter arterial oily chemoembolization with IVR-CT system for hyper vascular hepatocellular carcinoma. Cardiovasc Intervent Radiol 1999;22(Suppl):S59(abstract).

7 Murakami T, Shibata T, Nakamura H, et al. Percutaneous microwave hepatic tumor coagulation with segmental hepatic blood flow occlusion in seven patients. AJR 1999;172:637-640.

8 Shibata T, Murakami T, Ogata N, et al. Percutaneous microwave coagulation therapy for patients with primary and metastatic hepatic tumors during interruption of hepatic blood flow. Cancer 2000;88:302-311.

9 Bruix J, Llovet JM, Castells A, et al. Transarterial embolization versus symptomatic treatment in patients with advanced hepatocellular carcinoma: Results of a randomized, controlled trial in a single institution. Hepatology 1998;27:1578-1583.

10 Maeda M, Timmermans HA, Uchida BT, et al. In vitro comparison of the spiral Z-stent and the Gianturco Z-stent. J Vasc Interv Radiol 1992;3:565-569.

11 Ishiguchi T, Nakamura H, Okazaki M, et al. Radiation exposure to patient and radiologist during transcatheter arterial embolization for hepatocellular carcinoma. Nippn Acta Radiol 2000;60:839-844.

12 Vano E, Gonzalez L, Beneytez F, et al. Lens injuries induced by occupational exposure in non-optimized interventional radiology laboratories. Br J Radiol 1998;71:728-733.

Articles

Interventional radiology has developed tremendously since the term was first coined some 30 years ago.1,2 Its minimally invasive nature has made it suitable for use in emergencies and in cases when even the surgeons do not know where to begin. Radiological procedures rival surgical operations in the treatment of aneurysms (Figure 1).

Hironobu Nakamura, MD

Articles

Interventional Methods Transform Therapy