| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Division of Gastroenterology, Ospedale Casa Sollievo della Sofferenza IRCCS, Viale Cappuccini, I-71013 San Giovanni Rotondo, Foggia, Italy
Editor:
I read with interest the article by Dr Livraghi and colleagues in the March 1999 issue of Radiology (1) about the comparison of radio-frequency (RF) ablation and percutaneous ethanol injection in the treatment of small hepatocellular carcinoma (HCC).
The authors conclude that "RF ablation was more effective and enabled tumor necrosis in fewer treatment sessions, albeit with a higher rate of complications"; thus, they currently prefer to use RF ablation in the majority of patients with small HCC.
I have three concerns about these statements. The first pertains to the modalities with which percutaneous ethanol injection is performed; the second, to the evaluation of the results of RF ablation; and the third, to the comparison of costs of the two therapeutic modalities.
Dr Livraghi and colleagues emphasize the considerably fewer treatment sessions needed to achieve complete tumor necrosis with RF (1.2 sessions per tumor) compared with percutaneous ethanol injection (4.8 sessions per tumor). In their study, the authors injected ethanol in doses of 1–4 mL per session, an amount that is extremely low and increases remarkably the number of sessions required to perfuse HCC nodules completely. When percutaneous ethanol injection was initially introduced into clinical practice, the volume of ethanol injected was limited to 1–8 mL per session (2–4), with the investigators preferring to repeat the procedure two to three times per week until the whole tumor was completely perfused with ethanol.
Shiina et al (4) first tried to establish prior to the procedure the total amount of ethanol to be injected according to the lesion size; nevertheless, many authors (1,5) continued to terminate treatment exclusively on the basis of their impression of having achieved perfusion of the whole lesion. The main argument supportive of such a procedure is that the distribution of ethanol within the tumor is not a predictable event, and sometimes inhomogeneous diffusion occurs inside the lesion because of hypervascularization or fibrosis. Such a phenomenon, however, does not justify the use of numerous, low-volume, percutaneous ethanol injection sessions spaced too far apart. The progressive formation of fibrous tissue in the treated areas should rather suggest spacing these sessions closer together and using a higher volume of alcohol for each puncture.
A formula based on the volume of a sphere has been proposed (6) to establish the amount of ethanol to be injected: V = 4/3
(r + 0.5)3, where V is the volume of ethanol and r is the radius of the lesion measured in millimeters. The addition of 0.5 to the radius is designed to achieve the necrosis of a rim of nontumorous tissue around the tumor. Thus, a dose of 33.5 mL of ethanol is required for a nodule 3 cm in diameter (ie, the largest nodule treated with percutaneous ethanol injection in the study by Dr Livraghi and colleagues [1]).
It has been demonstrated that volumes up to 40 mL per session can be used safely (7,8). Dr Livraghi and colleagues (5,9) and other authors (10) have also proposed a single percutaneous ethanol injection session with a general anesthetic in patients with large or multiple HCCs. Total volumes of up to 210 mL per lesion have been injected (mean, 49–75 mL) (5,9–11). In my experience, patients can easily tolerate injections of 10–15 mL of ethanol per session without any anesthetic, so a nodule 3 cm in diameter can be treated in three sessions at the most. Nodules up to 2 cm in diameter (requiring a volume of ethanol of 14.1 mL, according to the formula by Farmer et al [6]) can be easily treated in a single session. During each session, the needle can be repositioned to ensure complete perfusion of the lesion.
The numbers given by Dr Livraghi and colleagues (4.8 percutaneous ethanol injection sessions per tumor; lesion diameter range, 11–30 mm; mean diameter, 25 mm) reflect unacceptably low doses of ethanol injected for each session.
The other argument favorable to RF ablation by Dr Livraghi and colleagues is that a higher rate of complete necrosis can be achieved with RF ablation compared with percutaneous ethanol injection. It must be emphasized that such a statement is not supported by statistically significant findings, since the P value is only .127; this value is given only in the discussion section of the article (1).
The therapeutic efficacies of both techniques were assessed with dual-phase spiral computed tomography (CT) 4 months after treatment. It is unlikely that the slow growth of HCC allows small residual areas of neoplastic tissue to be visible after such a short time. Dr Livraghi and colleagues reported that only 39 of 86 patients (45%) underwent follow-up CT at 8 months, so I believe that no conclusion can be drawn about the efficacy of an ablative therapy after so short a follow-up. The best criterion to use to evaluate the efficacy of a therapy can be none other than survival rate, as Dr Livraghi has reported (5,12).
In their article (1), Dr Livraghi and colleagues acknowledge the higher rate of complications of RF compared with percutaneous ethanol injection, reporting one "major" (hemothorax requiring drainage) and four "minor" complications (intraperitoneal bleeding, hemobilia, pleural effusion, and cholecystitis) in the RF group versus no complication in the percutaneous ethanol injection group. The four minor complications were so defined because no treatment was required; although the intraperitoneal bleeding resulted in a 4 g/dL decrease in hemoglobin level, the pleural effusion resolved after 2 months, and the cholecystitis resolved in several days. Thus, it seems reasonable to estimate the rate of major complications in patients treated with RF ablation to be at least 10% (four of 42 patients).
Dr Livraghi and colleagues did not perform a formal cost-effectiveness analysis as part of their study, but they suggest that RF ablation is the preferred therapeutic procedure thanks to "the greater success in achieving a total response coupled with the need for fewer treatment sessions." Currently, in Italy, the cost of an RF ablation session is at least 50 times greater than that of a percutaneous ethanol injection session, and such an obvious difference can be justified only by (a) a reduction in the number of sessions, (b) a demonstrated improvement in survival rates, and (c) a low rate of complications (whose occurrence further increases the costs of the procedure). In my opinion, the data from this article do not ensure, for the present, such outcomes.
In conclusion, although it is surrealistic to defend percutaneous ethanol injection in front of its creator, Dr Livraghi, I believe that this technique is still the best ablative therapy for small HCC (13), and more extended and thorough comparisons with RF ablation are needed before leaving percutaneous ethanol injection aside.
REFERENCES
,
Sergio Lazzaroni, MD
,
Franca Meloni, MD*,
Luigi Solbiati, MD§ and
G. Scott Gazelle, MD, MPH
Department of Radiology, Ospedale Civile, Via Cereda 23, 20059 Vimercate, Milan, Italy*; Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston; Department of Internal Medicine, Ospedale San Biagio, Clusone, Italy; Department of Radiology, Ospedale Generale, Busto Arsizio, Italy§
We congratulate Dr Caturelli on his knowledge of percutaneous ethanol injection. In response to his concerns, however, we offer the following comments.
The first observation concerning the amount of ethanol to inject per session is acceptable. Every operator can use the amount that he or she considers most appropriate. It is obvious that if one chooses to inject a larger quantity, the number of sessions will be decreased. In our article (1), we reported on our standard technique, which is always well accepted by our patients. Indeed, a few additional words relating to this issue could have been said. Ideally, one could conduct a CT examination after two sessions to check the necrosis achieved instead of at the presumed end of the procedure. In this way, several sessions could have been avoided.
The second observation concerns the comparison of the complete necrosis rates (80% with percutaneous ethanol injection and 90% with RF). Irrespective of concerns relating to the timing of follow-up studies, the necrosis rates were nearly the same for both procedures. A comparison of survival rates was beyond the aims of our study; perhaps in the future we will conduct one. About the evaluation of the severity of complications, we have always considered as minor those complications that do not require specific treatment or that are not life threatening even if not cured. Our experience, as reported, probably also reflects a learning curve. In fact, in the subsequent 75 patients with an HCC of less than 3 cm in diameter treated with RF, no serious complications occurred.
Percutaneous ethanol injection is certainly less costly than RF with respect to equipment, but RF is less time-consuming, both for physicians and particularly for patients living far from the hospital.
Regarding the conclusions, in our article (1) the following is written: "Nevertheless, given the well-established efficacy of percutaneous ethanol injection (confirmed in this study), percutaneous ethanol injection should remain the primary treatment option when RF ablation is not available." Therefore, this does not mean the end of percutaneous ethanol injection. However, in a referral center with a large volume of patients with HCC (on the average, one new patient per day), RF allows the majority of them to be treated in a shorter time and with complete ablation of an additional 10% of the volume.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
