| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
and
Robert I. White, MD
Department of Radiology, University of Utah, 50 North Medical Drive, Salt Lake City, UT 84132*; Department of Medicine, Washington University, St Louis, Mo; Department of Radiology, Yale University, New Haven, Conn
Editor:
We read with great interest and concern the article by Dr Chavan and colleagues (1) on hepatic embolization in the treatment of hereditary hemorrhagic telangiectasia. The authors describe five patients who underwent embolization without major complications and advocate this as standard therapy for patients with advanced symptoms of high-output congestive heart failure, severe liver discomfort, or intestinal ischemia secondary to a steal phenomenon. While these are recognized manifestations of arteriovenous malformations of the liver in patients with hereditary hemorrhagic telangiectasia, we want to point out to all physicians that catastrophic necrosis and death have occurred in two of our patients who underwent embolization with the use of the same techniques described by the authors.
Dr Chavan and colleagues refer to our radionuclide technique (2), which has been helpful in limiting any pulmonary embolization at the time of the procedure. However, the two recent deaths from hepatic infarction and an additional report (3) of liver necrosis requiring urgent transplantation described at the August 1998 Morbidity and Mortality Conference held by the Society of Cardiovascular and Interventional Radiology should give all of us second thoughts about when and in whom to attempt this procedure. These catastrophic results indicate to us that the outcome is unpredictable.
We have currently placed a moratorium on performing the procedure in our patients until we can better predict who will develop fatal hepatic necrosis and who will benefit from embolotherapy. In those patients with a life-threatening condition, we now believe that a transplant team must be consulted in advance, and the issue of a fatal outcome must be part of the consent if in the judgment of the physician embolization is the only other option. Liver transplantation following the procedure will not be an option for some elderly patients because of age criteria. In the two patients who died, the courses were prolonged (5 and 15 months) and agonizing not only for the patients but also for all those caring for them.
References
,
Martin Caselitz, MD,,
Hans Jürgen Schlitt, MD,,
Klaus Friedrich Gratz, MD,§ and
Michael Manns, MD
Departments of Diagnostic Radiology,* Gastroenterology and Hepatology, Abdominal and Transplantation Surgery, and Nuclear Medicine,§ Hannover Medical School, Carl Neuberg Strasse 1, 30625 Hannover, Germany
Although the two articles mentioned by Dr Jackson (1,2) concern the same set of patients, the contents of the two are different from each other. The article in Gut (2) deals with the clinical outcome in symptomatic patients who have hereditary hemorrhagic telangiectasia with predominant hepatic involvement. The article in Radiology (1) describes how the procedure was performed, its side effects, and its tolerability. The details are of particular importance from the point of view of the radiologist performing the intervention. Because embolization of the hepatic artery in these patients is markedly different from embolization of the hepatic artery performed for other conditions, such as hepatic tumors, we believe it was particularly important to highlight the different steps in detail for the readers of Radiology, especially because, to our knowledge, no uniform method for treating this subgroup of patients can be found in the literature.
Regarding the various points raised by Dr Jackson, our replies are as follows:
1. Before treatment and at the end of follow-up, the measurement of cardiac output was performed by using cardiac catheterization with use of a thermodilution method. For additional screening, echocardiography was performed by experienced investigators from the Department of Cardiology at our institution (3) who were unaware of the results of the treatment and of the cardiac catheterization. Measurements were taken three to five times on a single occasion with good correlation. The range of variation between consecutive measurements at our institution is 10%–20%. There was a good correlation between both techniques. The measurements obtained with cardiac catheterization were taken for the purposes of evaluation.
2. The cardiac output measurements at the end of therapy in the patient with the delayed decrease in cardiac output (patient 3 [2]) were taken when the patient was still experiencing postembolization pain; pain-induced tachycardia and consequently a higher cardiac output reading that decreased after the disappearance of pain is a possible explanation for a further decrease in the readings during follow-up. A marked reduction of arteriovenous shunting in the liver in this patient had been documented with angiography at the last embolization procedure (figs 1, 2 [1]); thus, one need not assume spontaneous resolution of arteriovenous shunting, as mentioned by Dr Jackson. The validity of the last follow-up cardiac output measurement in this patient was supported by additional intermediate measurements, which revealed similar results. It is interesting that a similar delayed decrease in cardiac output was noticed in another patient, as well.
3. The patient with the arterioportal shunts also had arteriovenous shunts, which were markedly less pronounced than the arterioportal shunts. Hence, the malformation was classified as arterioportal. As stated in the Radiology article (1), the primary purpose of testing with the radiolabeled polyvinyl alcohol particles was to rule out shunting of the particles into the pulmonary circulation. We were aware that the method would not reliably depict shunting into the portal circulation. Even then, when we were faced with a patient who required urgent intervention, and the two possible treatment options were hepatic arterial embolization or liver transplantation, a decision was made in favor of embolization because transplantation was not considered a realistic option due to the patient's condition at that stage.
We implemented the procedure with the awareness that some degree of peripheral embolization of the portal venous branches might occur. Hence, only the left liver lobe was embolized in the first stage to avoid endangering the entire liver. The patency of portal venous branches up to the fourth division even in the previously embolized regions was confirmed at every subsequent angiographic examination. In addition, the patency of the portal vein and its main branches was documented with Doppler ultrasonography following every embolization.
4. Clinically, the patient exhibited two features: a very high cardiac output of 17.3 L/min and portal hypertension presumably because of the arterioportal shunts. The cardiac output decreased after three embolizations to 5.9 L/min, but the portal hypertension presumably recurred. We do not know why this decrease in cardiac output should be excluded from the statistical evaluation. As we have already stated in our article, the patient returned with variceal bleeding presumably due to recurrence of portal hypertension.
Because the patient was admitted in a critical state—her varices had bled while she was in another hospital for 2 days—she underwent emergency surgery to create a portocaval shunt without having undergone angiography or measurement of cardiac output; the last cardiac output measurement had been taken under stable conditions. The failed surgical attempt due to excessive intraoperative bleeding and subsequent death underline the problem of surgery in such patients. Because the patient's critical state after emergency surgery did not allow further angiographic evaluation, the definitive cause for recurrence of variceal bleeding remains unclear.
5. The patient referred to in figure 2 (1) underwent five embolization procedures. Thereafter, the patient's refractory abdominal angina (requiring opiate derivatives) disappeared. Oral food intake returned practically to normal, and she showed a weight gain of more than 10 kg. In view of the complete disappearance of symptoms, further therapy was discontinued; hence, the left gastric artery was not embolized. The treatment was intended not necessarily to obliterate all arteriovenous malformations that can be seen at angiography but to achieve clinical improvement. As is obvious from the sequence of images, the embolization of the larger branches of the right hepatic artery was performed in the last session only after repeated particle embolization of the periphery had ensured satisfactory occlusion of the peripheral shunts. Furthermore, the stable follow-up course, 60 months after treatment, confirms the effectiveness of the embolization.
We add that, with use of the same technique, we have treated a sixth adult patient (aged 30 years) who presented with intractable abdominal angina, a weight loss of more than 10 kg, and a cardiac output of more than 11 L/min. After three embolizations, the patient's abdominal pain resolved completely and her body weight increased by 6 kg because of her ability to eat again. The cardiac output decreased to below 9 L/min. As in our first patient, she developed hypoattenuating areas in the liver—which presumably represented areas of hepatic necrosis—following the third embolization that were without any clinical sequelae. At 15 months follow-up, she was in good health and without any abdominal pain; we will elaborate on her in an upcoming article. In addition, as emphasized in our article in Radiology (1), we advocate treatment only in symptomatic patients in whom all forms of medical therapy to relieve symptoms have failed.
Regarding the points raised by Dr Miller and colleagues, we state that in the six patients we have treated so far, we have observed areas of hepatic necrosis in two of them, both without lasting clinical sequelae. Both these patients were asymptomatic and in good health at 60 and 15 months follow-up. Dr Miller and colleagues reported procedure-related fatal hepatic necrosis in various patients, which is an important observation worthy of mention. It would be informative to know the details on all the patients with hereditary hemorrhagic telangiectasia who were treated by Dr Miller and colleagues.
As stated in our article in Radiology (1), the decision to embolize was reached by a consensus of gastroenterologists, radiologists, and transplant surgeons. Our transplant surgeons are of the opinion that transplantation should be considered in those patients in whom embolization is not possible or as a second step after unsuccessful embolization. Since we have started hepatic embolization for hereditary hemorrhagic telangiectasia at our institution, none of the referred patients with hereditary hemorrhagic telangiectasia and predominant and symptomatic hepatic involvement have undergone liver transplantation for this reason. None of our patients in whom embolization has been performed have had a protracted, agonizing course after the embolotherapy. However, we concur with Dr Miller that conducting a very careful selection of patients prior to performing this invasive procedure and close monitoring between successive procedures are obligatory.
References
This article has been cited by other articles:
|
|
 |
|
  J. S. Wu, S. Saluja, G. Garcia-Tsao, A. Chong, K. J. Henderson, and R. I. White Jr. Liver Involvement in Hereditary Hemorrhagic Telangiectasia: CT and Clinical Findings Do Not Correlate in Symptomatic Patients Am. J. Roentgenol., October 1, 2006; 187(4): W399 - W405. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Saluja and R. I. White Hereditary Hemorrhagic Telangiectasia of the Liver: Hyperperfusion with Relative Ischemia--Poverty Amidst Plenty Radiology, January 1, 2004; 230(1): 25 - 27. [Full Text] [PDF]
|
|
|
|
|
|
A. A. S. Ianora, M. Memeo, C. Sabba, A. Cirulli, A. Rotondo, and G. Angelelli Hereditary Hemorrhagic Telangiectasia: Multi-Detector Row Helical CT Assessment of Hepatic Involvement Radiology, January 1, 2004; 230(1): 250 - 259. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M E Begbie, G M F Wallace, and C L Shovlin Hereditary haemorrhagic telangiectasia (Osler-Weber-Rendu syndrome): a view from the 21st century Postgrad. Med. J., January 1, 2003; 79(927): 18 - 24. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
