| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorial |
1 From the Department of Radiology, Charité, Campus Virchow-Klinikum, Klinik für Strahlenheilkunde, Augustenburger Platz 1, Berlin 13353, Germany. Received May 16, 2002; revision requested June 17; revision received June 19; accepted July 1. Address correspondence to the author (e-mail: hans-joachim.wagner@charite.de).
Index terms: Bile ducts, interventional procedures, 766.1229, 766.1267, 774.1269 • Bile ducts, stenosis or obstruction, 766.1229, 766.1267, 774.1269 • Editorials • Pancreas, interventional procedures, 774.1229, 774.1267, 774.1269 • Pancreas, therapeutic radiology, 774.1229, 774.1267, 774.1269
For nearly 2 decades, the primary treatment option for palliation of malignant jaundice due to incurable or inoperable obstruction of the bile ducts has been endoscopic retrograde relief with insertion of a biliary stent. This paradigm was created during a prospective randomized trial in which the endoscopic route for bile duct decompression was compared with the minimally invasive alternative of percutaneous transhepatic access to the biliary system to drain the bile. This study, published in The Lancet in 1987 (1), demonstrated superior technical and clinical success in terms of decrease in bilirubin level, significantly fewer complications, and a lower 30-day mortality rate. Since that time, both interventional radiologists and gastroenterologists have agreed that every patient with incurable or inoperable malignant bile duct obstruction should first undergo an endoscopic attempt at bile duct decompression. This widely accepted and unquestioned policy did not change with the development of self-expanding metal stents, which became part of the clinical routine for palliation of malignant obstructive jaundice in a short time (2).
However, the results of a new prospective randomized trial published in this issue of Radiology (3) might lead to a paradigm shift. Piñol et al (3) initiated a trial in which percutaneous transhepatic access was compared with the endoscopic retrograde approach for decompressing the biliary system. Different from the study performed in the 1980s (1), their study involved self-expanding metal stents for percutaneous biliary drainage. Endoscopic drainage, however, was carried out in the conventional manner by using plastic stents instead of self-expanding metal stents. Of the 54 patients, 28 were randomly allocated to percutaneous insertion of a metal stent and 26 to endoscopic placement of an plastic stent. The two groups were compared with respect to technical and clinical success, morbidity, and mortality. Technical success did not differ between the groups; clinical success was significantly higher in the percutaneous group (71% vs 42%). Major complications were more frequent in patients who received percutaneous treatment (61% vs 35%; P = .09), but 30-day mortality rates did not differ significantly (36% vs 42%). Median survival was significantly higher in patients with percutaneous metal stent insertion (3.7 vs 2.0 months), and the percutaneously placed metal stent was the only independent variable of survival in a regression analysis. Piñol et al (3) concluded that percutaneous metal stent insertion is an alternative to endoscopic polyethylene stent placement for relief of malignant obstructive jaundice.
This study by the group from the University of Barcelona had an excellent design, in which percutaneous and endoscopic access to the biliary system were compared (3). The choice of plastic stents to endoscopically decompress the bile ducts may be one of the few weak points of the trial and can be criticized. The superiority of self-expanding metal stents over plastic prostheses has been demonstrated in a variety of prospective randomized trials (2,4–6). All of these studies unanimously demonstrated that metal stents offer better long-term patency results. Therefore, self-expanding metal stents require fewer reinterventions to treat recurrent jaundice, which compensates for the initial higher costs associated with their placement, making them cost-effective in the long term (2,4–6). This effect was demonstrated for the endoscopic route (2,4,6), as well as for the percutaneous transhepatic approach (5,7), although in most of these trials, endoscopic access to the biliary system was always attempted first if patients did not have hilar lesions requiring multisegmental drainage procedures. It may be argued that the significantly higher clinical success rate (decrease in bilirubin level) reported in the study by Piñol et al (3) might have been due to the mere fact that the Poiseuille law was unchallenged: Bigger is better; that is, a 30-F self-expanding metal stent achieves superior drainage, as compared with a 12-F plastic tube.
The study design chosen by the authors reflects the clinical reality in most hospitals around the world. After an initial preference for the use of self-expanding metal stents in the treatment of malignant biliary stenoses, most gastroenterologists returned to inserting plastic stents. To avoid plastic stent obstruction, which most often leads to major complications including cholangitis and sepsis, regular exchange of plastic stents every 3 months was performed in most institutions (8). A prospective randomized trial (9) demonstrated that complication-free survival rates did not differ between patients undergoing metal stent placement and a patient group randomly assigned to plastic stent placement with elective exchange every 3 months. However, cumulated hospital days were significantly longer in the plastic stent group, and cost analysis showed that metal stents were advantageous in patients surviving longer than 6 months (9). A decision analysis with Markov modeling, in which the cost-effectiveness of plastic stents was compared with that of metal stents in patients with unresectable pancreatic carcinoma, demonstrated that initial endoscopic placement of a metal stent is a cost-saving strategy (10). This study (10) showed that initial plastic stent placement was associated with 1.799 quality-adjusted life-months; initial metal stent placement conferred 1.832 quality-adjusted life-months.
The change of the gastroenterologic policy from metal stent insertion to regularly exchanged plastic stents has occurred for many reasons. Plastic stents are less expensive, easier to place, and, in case of misplacement, easily extracted (11). On the contrary, self-expanding metal stents became the device of choice for percutaneous transhepatic procedures (12). Apart from one prospective randomized trial in which plastic and metal stents were compared (5), to our knowledge, virtually no studies were published in the past decade in which the placement of plastic stents by way of the transhepatic route was addressed.
Based on the premise that the study by Piñol et al (3) does reflect current clinical practice in the treatment of malignant biliary obstructions with respect to endoscopic plastic stent placement, it presents interesting results. First, the higher technical success rate, although not statistically significant, of stent insertion in the percutaneous group is different from that in the British study by Speer et al (1). The 58% rate of technical success for endoscopic retrograde stent placement in the study by Piñol et al (3) is disappointing and far lower than the rate reported in previous trials (1,2,4,6,9). In six of eleven patients, stent insertion failed because of difficulties in stent manipulation through the stricture, and in four of 11 patients it failed because of anatomic variants that restricted access to the papilla (gastric outlet obstruction and previous surgery). Inability to reach the papilla was an exclusion criterion in some previous trials (eg, Speer et al [1], Smith et al [13]), which contributes to the higher success rates in these studies. However, from the patient’s viewpoint and the intention-to-treat principle, a study design that does not exclude these patients reflects the clinical situation more closely. Six of the eleven patients with failed endoscopic stent placement were subsequently treated successfully with percutaneous stent placement.
Second, the statistically significant higher clinical success rate in the percutaneous group in the study by Piñol et al (3) is surprising. A combination of two factors might explain this finding. The first factor is the higher technical success rate, and the second factor is the larger diameter of the metal stents, which allows better bile flow and clearance of the obstructed biliary system. Whereas 20 (95%) of the 21 patients with successful metal stent placement had a significant decrease in their elevated bilirubin level, only 11 (73%) of the 15 patients with successfully placed plastic prostheses showed relief of jaundice. The assumption that the low technical success rate with endoscopically placed stents mainly contributed to the inferior clinical success rate is further supported by the finding that the analysis of the data according to the treatment actually applied did not show a statistically significant difference between groups. However, even the analysis according to the actual treatment demonstrates higher success rates for the metal stent (93%), as compared with those for the plastic prosthesis (75%).
Third, the complication rate for percutaneous access was not significantly higher than that for the endoscopic approach, although there was a clear trend toward more percutaneous complications (61% vs 35%). A closer look at the type of complications shows an extremely high rate of cholangitis, which contributed to two-thirds of the overall rate in both groups (12 of 17 patients in the percutaneous group and six of nine in the endoscopic group). To our knowledge, this finding has not been reported elsewhere thus far and was unexpected in light of the fact that patients usually undergo antibiotic prophylaxis during transhepatic biliary drainage. However, the study does not mention antibiotic treatment in either study group.
The 30-day mortality rate in the current study by Piñol et al (3) was similar in both treatment groups (36% in the percutaneous access group and 42% in the endoscopic access group); this represented the most significant difference at comparison with the study by Speer et al (1). The rate in the endoscopic group was significantly higher than in most previous trials (1,2,6,8,9,13), in which 30-day survival was usually achieved in more than 90% of patients. However, often only 30-day mortality results are reported for these patients, in whom the procedure was successful. But in the study by Piñol et al (3), the analysis, according to the treatment actually applied, also reveals an extremely high mortality rate of 44%. The reasons for such a high mortality rate are not clear. The discussion about early mortality leads to the next unexpected result of the study, survival.
Fourth, survival in the study by Piñol et al (3) was significantly higher in the patients with percutaneously placed stents than in the patients with endoscopically inserted stents (median survival, 3.7 vs 2.0 months). Furthermore, percutaneous transhepatic placement of a metallic stent was identified with Cox regression analysis as an independent predictor of survival. Both results are contrary to the findings of Speer et al (1) in the earlier trial. Speer et al (1) could not find significantly different long-term survival times for either of the two groups (median survival time, 119 days in the endoscopic group and 88 days in the percutaneous group), nor was the route of stent placement a prognostic factor. Again, the short median survival time of 2.0 months in the endoscopic treatment group might lead to criticism in light of the longer survival times in other prospective trials (eg, 4.8 months in the study by Prat et al [9]). However, the median survival time in the percutaneous group compares favorably with that in previous studies (eg, 122 days in the study by Lammer et al [5]). I have little doubt that the findings on long-term survival will encourage further discussion of this study.
However, even if the decreased patient survival in the endoscopic treatment group will not stand the test of time, the study by the group from Barcelona (3) will become important, since it challenges the old findings of Speer et al (1). The old question of whether the percutaneous transhepatic access route or the endoscopic retrograde approach to the biliary system is preferred in the treatment of malignant obstructive jaundice is again open to discussion.
Factors can easily be identified for the new results. The percutaneous way to treat malignant obstructive jaundice has changed very much since the study by Speer et al (1). Nearly everything is done differently now. Interventional radiologists use a C-arm system for rotational cholangiography and can select an optimal entrance point of the drainage catheter into the peripheral biliary system, thus reducing the vascular complications of the drainage procedure. The material used for external and internal biliary drainage has changed dramatically. The development of small and flexible catheters, hydrophilic guide wires, and self-expanding metal stents has, indeed, made this procedure safe and eliminated most complications.
I hope that the study by Piñol et al (3) will stimulate research in this field. It seems as if the time has come for a large-scale prospective randomized multicenter trial to assess not only technical and clinical success rates but also complication and short- and long-term survival rates for both routes to the biliary system, as well as quality of life assessment in patients with an incurable disease and a life expectancy of less than 6 months, regardless of the treatment option. I eagerly await a cost-effectiveness analysis comparing the two treatment options with regard to quality-adjusted life-months.
FOOTNOTES
See also the article by Piñol et al in this issue.
REFERENCES
Related Article
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
