Correspondence
Antithrombotic Therapy after Coronary-Artery Stenting
N Engl J Med 1999; 340:1365-1368April 29, 1999
- Article
To the Editor:
The study by Leon et al. (Dec. 3 issue)1 comparing three antithrombotic-drug regimens — aspirin alone, aspirin and warfarin, and aspirin and ticlopidine — after coronary stenting showed a lower rate of stent thrombosis with aspirin and ticlopidine, although there were more hemorrhagic complications than with aspirin alone. By contrast, there were no significant differences in the incidence of neutropenia or thrombocytopenia among the three regimens (overall incidence, 0.3 percent).
Despite the absence of severe hematologic dyscrasias in the study, the authors refer to reports of the risk of serious hematologic complications, such as thrombotic thrombocytopenic purpura, among patients who received ticlopidine after coronary stent placement. However, they did not mention the risk of aplastic anemia, a life-threatening condition that may be induced by ticlopidine. Recently, Yeh et al. reviewed the cases of ticlopidine-induced aplastic anemia published in the English literature,2 and four other cases have been reported3,4 (including one case reported by us). Ten of 23 patients died (43 percent): 7 because of infection (candidal and gram-negative sepsis) and 1 from massive gastrointestinal bleeding. The only factor significantly correlated with survival was the patients' age. Other factors, such as sex, duration of ticlopidine use, and the presence of agranulocytosis at onset, had no prognostic value.2 Although the pathogenesis of marrow toxicity related to this drug is still unclear, both direct toxicity and immunologic effects have been proposed.2 The therapeutic use of hematopoietic growth factors has been suggested,4 but this treatment seems to have little effect on survival.2,3
In summary, although it is an infrequent event, ticlopidine-induced marrow aplasia is possible, and patients and doctors should be aware of it. However, aplastic anemia, like other hematologic dyscrasias induced by ticlopidine, appears after 2 weeks of treatment,2-4 whereas stent thrombosis generally occurs within the first 14 days after coronary stenting (a mean of 0.7 day in the study by Leon et al.).1 So, we agree with the suggestion made by Leon et al. to shorten the duration of ticlopidine therapy after stenting to two weeks.
4 ReferencesFrancisca Ferrer, M.D.
Jose María Moraleda, M.D., Ph.D.
Vicente Vicente, M.D., Ph.D.
Hospital General Universitario, 30003 Murcia, Spain1
Leon MB ,Baim DS ,Popma JJ , et al. A clinical trial comparing three antithrombotic-drug regimens after coronary-artery stenting. N Engl J Med 1998;339:1665-1671
Full Text | Web of Science | Medline2
Yeh SP ,Hsueh EJ ,Wu H ,Wang YC . Ticlopidine-associated aplastic anemia: a case report and review of literature. Ann Hematol 1998;76:87-90
CrossRef | Web of Science | Medline3
Kao T-W ,Hung C-C ,Chen Y-C ,Tien H-F . Ticlopidine-induced aplastic anemia: report of three Chinese patients and review of the literature. Acta Haematol 1997;98:211-213
CrossRef | Web of Science | Medline4
Ferrer F ,De Arriba F ,Moraleda JM ,Vicente V . Aplastic anaemia induced by ticlopidine after placement of coronary artery stents: therapeutic role of granulocyte colony-stimulating factor. Clin Drug Invest 1998;15:261-262
CrossRef | Web of Science
To the Editor:
Erbel et al. (Dec. 3 issue)1 conducted a randomized trial in which coronary-artery stenting was compared with balloon angioplasty for restenosis after initial balloon angioplasty. They conclude that stenting was effective and that it can be recommended. However, their data are open to the opposite interpretation, especially, as noted by Topol in the accompanying editorial,2 if you consider the “hard” end points of myocardial infarction and death. The following are the consequences of stenting: a higher incidence of death, Q-wave infarction, or both (14 of 178 patients, as compared with 5 of 176 patients in the angioplasty group; P=0.048); a longer hospital stay (2.4 days longer) and greater expense, conservatively estimated as $854,400 ($2,000 × 2.4 × 178), including the costs related to the higher incidence of bleeding complications, additional surgical procedures, as well as myocardial infarction and death; additional costs related to the higher incidence of subsequent emergency and elective bypass surgery (10 of 178 patients vs. 3 of 176 patients; P=0.059); the added costs of the stents ($1,500 × 178, or $267,000); and the higher rates of morbidity and mortality (as compared with the angioplasty group, the stent group had 8 more Q-wave myocardial infarctions, 1 more death, 7 more bypass procedures, 9 more arterial repairs, and 10 more patients who required a transfusion).
What is achieved by this approach? Six months after angioplasty, the average minimal luminal diameter in the stent group was 0.19 mm (1/133 in.) greater than that in the angioplasty group, a value close to the intrinsic error of the imaging system used (0.13 mm). Nonetheless, the investigators found a significantly lower rate of restenosis (stenosis of more than 50 percent of the luminal diameter) in the stent group and a greater “need” for target-vessel revascularization in the angioplasty group.
Given the minuscule differences and the uniform distribution seen in the cumulative frequencies of the minimal luminal diameter in the two groups (shown in Figure 1 of the article), I would like the authors to explain the relatively large difference in the rates of restenosis. Might the presence of a stent influence the measurement or the observer? Can they explain why they performed 42 target-vessel revascularizations among the 32 percent of patients with restenosis after angioplasty alone but only 16 procedures among the 18 percent of patients who had restenosis in the stent group?
There is a large and rapidly progressive late loss in luminal diameter after stenting. If the study end point had been extended beyond six months, would the 1/133-in. advantage provided by stenting have been lost?
Topol makes some of the same arguments and points out the enormous force and influence wielded by an industry worth $1.5 billion in the United States alone.2 Recognition of that influence is necessary to protect the welfare of patients. In this case, it would be valid to conclude that stenting for coronary restenosis is not in the interest of the patient.
2 ReferencesFrank W. LoGerfo, M.D.
Beth Israel Deaconess Medical Center, Boston, MA 022151
Erbel R ,Haude M ,Hopp HW , et al. Coronary-artery stenting compared with balloon angioplasty for restenosis after initial balloon angioplasty. N Engl J Med 1998;339:1672-1678
Full Text | Web of Science | Medline2
Topol EJ . Coronary-artery stents -- gauging, gorging, and gouging. N Engl J Med 1998;339:1702-1704
Full Text | Web of Science | Medline
To the Editor:
In his editorial Topol argues that stents, although an important advance, are both overused and overpriced. We wish to address the issue of overpricing, as we have done elsewhere.1 We argue that the current pricing of stents is justified for three reasons.
First, stent manufacturers sell their products in an environment of brief product life cycles. From their perspective, their return on investment must be captured in just a few years, if not months. By comparison, for example, product life cycles in the pharmaceutical industry can be 5 to 10 years or more. The short product life cycle of stents is driven by rapid technological advances and the ever-building demand by clinicians to use the “latest, greatest stent.” If the demand fueled by small increments in benefit subsided, the price of stents would naturally begin to decline. In this event, less technologically advanced competitors could enter the market, challenging the current oligopoly. However, with the prospect of new materials and strut designs and with the introduction of coated and radioactive stents on the horizon, we are predicting that technological advances are likely to continue. If demand for these products emerges, they will be the next to be priced at a premium.
Second, stents are not as expensive as they may appear. Manufacturers are increasingly “bundling” products, providing steeper discounts for a range of interventional devices. In addition, newer stents offer the promise of reducing the need for other ancillary products, such as angioplasty balloons, and in some cases obviating the need for the placement of more than one stent.
Third, the risk associated with the development of new medical devices has increased in recent years. Manufacturers became accustomed to the possibility that a new intervention may not be clinically efficacious; in recent years, despite clinical success, they have also faced possible market failure related to issues of cost containment and reimbursement. Decisions not to use clinically effective medical devices have frequently been made arbitrarily. The Health Care Financing Administration established a technology-assessment committee in order to determine which medical devices were appropriate for reimbursement. However, because of its poor track record in providing justification for decisions regarding devices that were not appropriate for reimbursement, the committee was disbanded, creating more uncertainty. The effort to establish reasons for reimbursement grows increasingly complex in a world where formerly unmet medical needs are being addressed successfully by manufacturers. Manufacturers unable to navigate the maze may never see a return on investment despite having products with demonstrable clinical benefit.
Ironically, the current stent boom exists because third parties elected to provide reimbursement using the coding for angioplasty procedures. If clinicians had had to wait for a new system of coding to be devised, those interested in using stents might have had no capacity to pay for them.
In conclusion, investors in companies that manufacture stents have been rewarded in recent years. However, this selective analysis excludes companies whose business did not prosper for a variety of reasons, including those stated above. As observers of pricing throughout the medical industry, we believe that stent pricing is relatively normal given the risks associated with the business and the short product life cycles. As long as the demand for technological advances remains high, stent pricing should remain relatively stable. If technological advances lack clinical relevance, pricing will suffer. This is a normal economic cycle. Regarding Dr. Topol's suggestion that there is price gouging in the stent industry, we offer a verdict of not guilty. Despite his fine editorial, we don't expect Topol to topple pricing.
1 ReferencesVivian R. Wohl, M.B.A.
Eric Hecht, M.D.
Kelly Shaughnessy, M.B.A.
Merrill Lynch Medical Technology Equity Research, San Francisco, CA 94111Author/Editor Response
The authors reply:
To the Editor: We appreciate the comments of Ferrer et al. regarding the occurrence of aplastic anemia as a result of ticlopidine therapy. Coronary stenting has evolved into a safe and effective revascularization strategy, largely because of the identification of the optimal adjuvant antithrombotic regimen. We found that an antiplatelet regimen that included an arachidonic acid–pathway inhibitor (aspirin) and an adenosine diphosphate-receptor–P2T inhibitor (ticlopidine) was superior in reducing the rate of early adverse clinical events to aspirin alone or aspirin in combination with warfarin. Hematologic dyscrasias were rare, and there were no cases of thrombotic thrombocytopenic purpura or aplastic anemia. However, we acknowledge the importance of these serious side effects associated with ticlopidine therapy. Reports of irreversible1 or even fatal2 aplastic anemia are indeed fearsome, despite their rarity.
Given the growing use of stenting, even the rarest of side effects will inevitably occur. Therefore, it is time to explore safer antiplatelet regimens. A large cardiovascular study found that clopidogrel, an oral platelet P2T–receptor antagonist (like ticlopidine), has an excellent safety profile (unlike ticlopidine), with essentially no risk of hematologic dyscrasias (incidence of thrombocytopenia, 0.26 percent, and of neutropenia, 0.10 percent — rates similar to or less than those for aspirin).3 The emergence of other new antiplatelet agents may lead to the discovery of truly safe antithrombotic regimens after stenting.
Further investigation should establish whether the safety of the gold-standard antiplatelet regimen established in our study can be further improved without affecting the clinical efficacy.
3 ReferencesMartin B. Leon, M.D.
George Dangas, M.D.
Cardiovascular Research Foundation, Washington, DC 200101
Mallet L ,Mallet J . Ticlopidine and fatal aplastic anemia in an elderly woman. Ann Pharmacother 1994;28:1169-1171
Web of Science | Medline2
Ceylan C ,Kirimli O ,Akarsu M ,Undar B ,Guneri S . Early ticlopidine-induced hepatic dysfunction, dermatitis and irreversible aplastic anemia after coronary artery stenting. Am J Hematol 1998;59:260-260
CrossRef | Web of Science | Medline3
CAPRIE Steering Committee
CrossRef | Web of Science | Medline
Author/Editor Response
Dr. LoGerfo uses our data to draw different conclusions related to “hard” end points, which he defined as myocardial infarction and death. We agree that death and myocardial infarction are hard end points, which can be referred to as major and adverse cardiac events. The primary end point of our study was angiographic evidence of restenosis on quantitative coronary angiography, an end point that has been used by others.1 As a primary end point it provides information on the efficacy of treatment. Quantitative coronary angiography was performed by independent investigators using a validated technique.
The decision to perform target-lesion revascularization was left to the operators and was based on a patient's symptoms with objective signs of ischemia (in the case of 14 of 16 patients in the stent group as compared with 38 of 42 patients in the angioplasty group) or without such signs (in the case of 2 of 16 patients and 4 of 42 patients, respectively). The higher morbidity with respect to myocardial infarction in the stent group was clearly related to the incidence of subacute stent thrombosis (3.9 percent), which was very similar to rates in other reports of elective stenting with adjunctive aggressive anticoagulation therapy including phenprocoumon and aspirin.1,2 Bleeding complications were related to the use of this regimen, which has now been replaced by aspirin and ticlopidine.2,3 This therapy reduced the rate of stent thrombosis and reduced the rate of bleeding complications to less than 1 percent.
The issue of cost is, of course, of utmost importance. The longer duration of hospitalization in the stent group was related to the investigators' agreement that oral anticoagulation after stent placement be completed before hospital discharge. The price of stents in Europe has dropped dramatically, so that a premounted stent on a high-pressure balloon now costs $500 to $700, and more than 60 different coronary stents are available. Previous reports showed that no progressive late loss in luminal diameter on serial coronary angiography after stent placement is expected.4 Thus, the incidence of the primary end point is unlikely to be different after one year.
Elective coronary stenting was effective in the treatment of restenosis after balloon angioplasty, because the rate of recurrent stenosis was lower despite a higher incidence of subacute thrombosis.
4 ReferencesRaimund Erbel, M.D.
Michael Haude, M.D.
University of Essen, 45122 Essen, Germanyfor the Restenosis Stent Study Group
1
Fischman DL ,Leon MB ,Baim DS , et al. A randomized comparison of coronary-stent placement and balloon angioplasty in the treatment of coronary artery disease. N Engl J Med 1994;331:496-501
Full Text | Web of Science | Medline2
Schomig A ,Neumann F-J ,Kastrati A , et al. A randomized comparison of antiplatelet and anticoagulant therapy after the placement of coronary-artery stents. N Engl J Med 1996;334:1084-1089
Full Text | Web of Science | Medline3
Leon MB ,Baim DS ,Popma JJ , et al. A clinical trial comparing three antithrombotic-drug regimens after coronary-artery stenting. N Engl J Med 1998;339:1665-1671
Full Text | Web of Science | Medline4
Kimura T ,Yokoi H ,Nakagawa Y , et al. Three-year follow-up after implantation of metallic coronary-artery stents. N Engl J Med 1996;334:561-566
Full Text | Web of Science | Medline
Author/Editor Response
I appreciate the opportunity to expand on my concern about the pricing of coronary stents in the United States. Wohl et al. are top investment analysts in the biomedical industry, and their assertion that “investors in companies that manufacture stents have been rewarded in recent years” is true and greatly understated. The same week that my editorial was published came the announcement that one of the four firms composing the oligopoly of stent manufacturers had been sold to Medtronic for $3.7 billion. This company was part of an initial public offering in 1996 and only began selling stents (their only product) in the United States in late December 1997. Some sales representatives of companies selling coronary stents reported that their income from salary and commissions exceeded $1 million in 1998, further supplemented by an increasing equity stake in the company they represent. Several reports acknowledged that the profit margin for coronary stents is in excess of 90 percent,1 which is considerably higher than that for virtually any other biomedical device marketed in the United States (typically, the profit margin is 60 to 65 percent). The exact stent models that we pay $1,500 to $1,800 for in the United States are available in Canada for less than $650 and in Europe for less than $800. Something appears to be desperately wrong with this picture.
From the opposite perspective of what may be construed as a “predator–prey” relationship, the data supporting wide use of stenting have become remarkably strong. Much of the data that Dr. LoGerfo used for his calculations are derived from an obsolescent era of stenting that required prolonged hospital stays for heparin anticoagulation and led to considerable bleeding complications. Stenting has substantially reduced the rate of target-vessel revascularization, and a recent trial that compared stenting with balloon angioplasty has also demonstrated a significant reduction in mortality at one year of follow-up.2 This comes at a time when 43 million Americans (16.1 percent) are uninsured3 and when medical centers across the country are unable to cope with the shrinking levels of reimbursement, including the $22 billion cut in Medicare resulting from the Balanced Budget Act of 1997.
In 1998, expenditures for coronary stents at the Cleveland Clinic Foundation exceeded $4 million, and stent use at the center is projected to increase by at least 20 percent in 1999. Without question, the stent companies are entitled to make a substantial profit for many of the reasons that Wohl et al. articulate — the only questions are how much and for how long. Gouging refers to an “excessive or exorbitant charge,” and there is no question that the pricing of stents in the United States, relative to the rest of the world and the rest of the biomedical-device industry, meets this definition. Wohl et al. state, “manufacturers are increasingly `bundling' products, providing steeper discounts.” Only in recent weeks have we and other medical centers begun to see this practice, but interestingly, despite intensive negotiation, there has been no reduction in the price of stents, only in accessory equipment. Although it remains entirely impossible for one physician to affect the pricing of a specific industry, recent talks with representatives of the Federal Trade Commission may ultimately be quite helpful.
3 ReferencesEric J. Topol, M.D.
Cleveland Clinic Foundation, Cleveland, OH 441951
Winslow R. Missing a beat: how a breakthrough quickly broke down for Johnson & Johnson. Wall Street Journal. September 18, 1998:1.
2
Topol EJ. EPISTENT 1 year mortality results. Presented to the American Heart Association, Dallas, November 7, 1998.
3
Iglehart JK . The American health care system -- expenditures. N Engl J Med 1999;340:70-76
Full Text | Web of Science | Medline
- Citing Articles (2)
Citing Articles
1
A. Symeonidis, A. Kouraklis-Symeonidis, U. Seimeni, A. Galani, N. Giannakoulas, E. Fragopanagou, M. Tiniakou, P. Matsouka, N. Zoumbos. (2002) Ticlopidine-induced aplastic anemia: Two new case reports, review, and meta-analysis of 55 additional cases. American Journal of Hematology 71:1, 24-32
CrossRef2
M. Gallerani, R. Manfredini, P. Doneg??, F. Lanza, M. Da Busti, G. B. Vigna, R. Fellin. (2000) Adverse Haematological Effects of Ticlopidine. Clinical Drug Investigation 19:3, 231-237
CrossRef
