Aminoglutethimide

EVALUATION OF THE CHARACTERISTICS OF SAFETY WITHDRAWAL OF PRESCRIPTION DRUGS FROM WORLDWIDE PHARMACEUTICAL MARKETS-1960 TO 1999*

A descriptive analysis was conducted to evaluate prescription drugs withdrawn from worldwide pharmaceutical markets over the past four decades due to safety reasons. The list of drugs, including indication, the duration of marketing, and reasons for withdrawal were examined. Among the 121 products identified, 42. 1 % were withdrawn from European markets alone, 5.0% from North America, 3.3% from Asia Pacific, and 49.6% from markets in multiple continents. Distributions of these withdrawals in each decade were: 12.4% from the 1960s, 165% from the 1970s, 39.7% from the 1980s, and 31.4% from the 1990s. Unfortunately, since the denominators (number of drug approvals) were not readily available, an accurate rate of withdrawal could not be reliably calculated. The most common categories of drugs withdrawn were: Non-Steroidal Anti-Inflammatory Drugs (13.2%), nonnarcotic analgesics (8.3%), antidepressants (7.4%), and vasodilators (5.8%). The top five safety reasons for withdrawals were: hepatic (262%), hematologic

(10.5%), cardiovascular (8.7%), dermatologic (6.3%), and carcinogenic (6.3%) issues. Among the 87 products for which the timing of marketing was available, the median time on the market was 5.4 years with about one-third withdrawn within the first two years. It is hoped that the current review will stimulate other future research into this important topic. However due to the intrinsic limitations of the descriptive analysis design, our observations are subject to the availability of data in the public domain. Readers are cautioned to be objective and careful when approaching data of this nature in order not to misinterpret the results due to potential data gaps.

Key Words: Market withdrawal; Market discontinuation; Adverse drug reactions; Postmarketing surveillance

INTRODUCTION

CLINICAL STUDIES HAVE been the cornerstone of new drug development in modem medicine. Through vigorous preclinical and clinical testing, many innovative therapies became available to needy patients in recent decades. However, although clinical studies remain the best possible approach (scientifically and practically) in new drug development, there are some potential limitations. The limitations are due to confined sample sizes, stringent inclusion/exclusion criteria, and rare adverse events (eg, aplastic anemia) as well as other unpredictable conditions after marketing such as off-label use, use in previously untested populations (eg, pediatric, geriatric, pregnant, very sick), and so forth. Thus, the complete safety profile of a drug will not usually be available until a product is exposed to a sufficiently large population in the postmarketing phase.

In addition, there are the potentials of misuse, abuse, and concomitant use in previous untested conditions such as together with herbal medicines, homeopathic treatments, and other alternate therapies (1,2,3). Consequently, newly identified drug reactions or interactions may be observed. This may alter the risk-to-benefit ratio which might lead to a subsequent label change, new precautions or warnings, reduced dosage, restriction, or possible withdrawal of the drug product from the market. Because of these issues, active postmarketing surveillance has long been an important aspect of the life cycle of a drug product (4-17).

Drug withdrawal has been a popular issue in recent years (18-25). In a recent study by Jefferys et al., 59 pharmaceutical products were found to be withdrawn from the British market between 1972 and 1994 from a total of 583 new active substances approved (26). This corresponded to an approximate 10% withdrawal rate. Among the 59 withdrawals, 59% were primarily due to commercial reasons, 39% were due to safety concerns, and 2% were due to efficacy concerns. Thus, in addition to commercial reasons, safety issues were the primary cause of market withdrawal.

The impact of market withdrawal of a drug product is far-reaching. In addition to the manufacturers (27-32), oftentimes, regulators, health care professionals, and patients are also affected. A review of all drugs withdrawn in worldwide markets over the past four decades was, therefore, undertaken to examine the nature, number, and other characteristics of these withdrawals to see whether there were any patterns, trends, and similarities so that we might be able to learn from these experiences. Specifically, the therapeutic classes and indications of the products involved, the primary reasons for the withdrawal, the average duration of marketing, geographical variation in withdrawal, and the potential changes in pattern of withdrawals over the past four decades were explored in this review.

METHODOLOGY

Prescription drug withdrawals from worldwide market over the past four decades (January 1960 to December 1999) were identified by the following major sources:

1. Published literature through MedLine, Health-Star, CancerLit, International Pharmaceutical Abstracts, and EM-Base,

2. Newsletters or other communications released from the manufacturers, professional/trade organizations (eg, PhRMA), and pharmaceutical news agencies (eg, the Pink Sheet, Scrip Report, etc.), and

3. Electronic databases via the Internet and other sources such as the World Health Organization (WHO) database and other major regulator Web sites (English based) such as the United States Food and Drug Administration (FDA), European Medicines Evaluation Agency (EMEA), United Kingdom Medicines Control Agency (MCA), Canadian Health Protection Branch (HPB), Australian Therapeutic Goods Administration (ATGA), Irish Medicines Board (IMB), Malaysia Adverse Drug Reactions Advisory Committee (MADRAC), New Zealand Medicines and Medical Devices Safety Authority (MMDSA), and the National Institute of Health Sciences of the Japanese Ministry of Health and Welfare (NIHS-MHW).

The current review focused only on prescription drugs that were new molecular entities. Therefore, new salts, esters, single purified isomer, and other dosage forms of previously available products were excluded. In addition, over-the-counter medicines, herbal drugs, veterinary products, medical devices, diagnostic agents, biologics, radiopharmaceuticals, and vaccines are covered in separate reports that will be published at later dates due to the large number of, and differing market for, these agents.

The searches conducted were meant to be broad. In addition to "Withdrawal," other related terms such as "Discontinuation," "Suspension," "Removal," "Recall," "Surrender," or "Revocation of License" were also used in identifying all possible reports that were available for review. However, it should be noted that product recall (ie, individual lot or batch) due to temporary departure from good manufacturing practices, impurities, microbial contamination, subpotency, mislabeling, instability, reduced shelf-life, uniformity problem, and so forth were excluded.

In addition, products that fell into one of the following situations were also excluded:

1. Withdrawal due to nonsafety reasons (business, patent, legal, efficacy, manufacturing difficulties, tampering, consumer pressure, political sanctions, etc.),

2. Withdrawal of a generic version of a brand-name product due to bioinequivalence or manufacturing problems,

3. Drugs that were withdrawn initially but reinstated later (eg, alphraprodine, aminoglutethimide, clozapine, felbamate, megestrol, molsidomine, thalidomide, tranylcypromine, etc.) (1,20,33-40), and

4. Withdrawal of only one particular dosage form of a certain product when the other dosage forms) remained on the market (eg, short-acting nifedipine, indomethacin-R, intravenous domeperidone, intravenous naftidrofuryl, oral practolol) (4,27, 34,35,40,41,42).

Data collected from these sources were then compiled and analyzed using the methodology of descriptive statistics. A primary list of all withdrawn drugs was created and any duplicated cases were eliminated. An analysis was then conducted as follows: First, the classification of the types and therapeutic classes of the withdrawn drugs were evaluated and the most common therapeutic categories of the products withdrawn were examined. Second, the alleged safety reasons for all withdrawals were tabulated and the most common reasons were examined. In addition to individual body system toxicities, withdrawal due to abuse and misuse, drug interaction, allergy and anaphylactic reaction, and overdose complications were also explored.

Third, the year of withdrawal and the average duration of marketing before withdrawal were reviewed. Potential changes in regulatory or medical practices over time were explored to see whether there were any differences in the characteristics of drug withdrawal over the years. For example, withdrawals in this decade (the 1990s) were compared to those of the previous decades to examine whether there were any differences in the way drug withdrawal (ie, numbers, reasons, classes of drugs withdrawn, etc.) had changed over time. Since the withdrawal year might have been different for some products that were discontinued in multiple countries, the year when the first withdrawal occurred was used in this analysis for simplicity.

For the estimation of the duration of marketing before withdrawal, the timing was estimated to the nearest one year if the exact point in time of market removal was not available. If the withdrawal occurred in more than one country and the length of marketing before withdrawal was different for the same product in different markets, then an average was used. In addition, if the dates of approval or discontinuation were not readily available for some products, then the approximate time of launch and withdrawal was assessed by reviewing secondary sources such as various editions of the Physicians' Desk References, Facts & Comparisons, Goodman and Gillman's Pharmacological Basis of Therapeutics, the United States Pharmacopoeia, Martindale Pharmacopoeia, and MicroMedex.

Lastly, the geographical locations of the withdrawals were also examined. Since there might have been potential variation in cultural or regulatory tolerance regarding safety concerns (ie, perception of the significance of a safety issue and the difference in tolerance of the risk-to-benefit ratio), the features of a drug withdrawn from one country or area might be different from the others. Thus, a gross examination was conducted to see whether there might have been more products withdrawn from one continent or country than others in terms of the number, classes, reasons, duration of marketing, and so forth. However, we were aware of the limitations that not all products were available in all countries or to the same extent. Thus, comparisons of this nature could be limited and/or potentially biased. Because of that, only a general overview was conducted.

The scope of the current study was no more than to confirm the existence of a safety issue that led to the subsequent market withdrawal of a product by either the manufacturer or the regulator. Due to lack of complete information available as well as the controversial nature regarding the legitimacy of some withdrawals, no attempt was made to evaluate whether these withdrawals were indeed appropriate or optimal.

RESULTS

After excluding duplication and withdrawals secondary to unqualified reasons (as specified earlier), there were 121 drugs identified in the current review. Table 1 displays all these drug products withdrawn from the worldwide markets over the past four decades. For therapeutic categories, central nervous system (CNS)-acting agents (31.4%), musculoskeletal products (16.5%), and cardiovascular agents (14.9%) were the three most common types of drugs withdrawn. When a more detailed subclassification was conducted, Non-Steroidal Anti-Inflammatory Drugs (NSAIDs) (13.2%), other nonnarcotic analgesics (8.3%), antidepressants (7.4%), vasodilators (5.8%), anorexiants (4. 1 %), CNS-stimulants (4. 1 %), and barbiturates (4.1%) were the most frequent classes of medication withdrawn. Figure 1 lists the most common classes of drugs withdrawn.

On the other hand, the most common body system toxicities for drug withdrawals were (in descending order): hepatic (26.2%), hematologic (10.5%), cardiovascular (8.7%), dermatologic (6.3%), and carcinogenic (6.3%) issues. Figure 2 displays the most common reasons for the drugs withdrawn included in this review. Table 2 displays the specific subtypes of reactions seen as the causes of withdrawal for the top three reasons (ie, hepatic, hematologic, and cardiovascular toxicities).

Interestingly, despite much attention, teratogenicity/obstetric complications were rarely the reason for market withdrawal (1.7%). Also, although carcinogenicity remained an issue for products withdrawn in the early years, there were fewer products withdrawn due to carcinogenic concerns in recent years and none in the 1990s after strict preclinical testing began to be required by many regulators. As for nonbody system-related safety issues, drug interaction (4.1%), abusive use (3.7%), overdose (3.3%), and anaphylaxis/ allergy (3.3%) all contributed to an important number of drug product withdrawals. Lastly, it should be noted that at times the reason for withdrawing a drug might not have been due to a single safety concern. There were several products withdrawn due to multiple organ toxicities (see Table 1). Therefore, even if the intensity of an individual safety concern might not have been the most severe, when multiple safety issues arose, the risk-to-benefit ratio would certainly be altered and this would have led to a decision to withdraw the drug from the market by either the regulator or the manufacturer.

Figure 3 lists the chronological distribution of these withdrawals. Overall, about 70% of the withdrawals identified in the current review appeared to have occurred over the past two decades: 12.4% (15/121) of the withdrawals were from the 1960s, 16.5% (20/ 121) from the 1970s, 39.7% (48/121) from the 1980s, and 31.4% (38/121) from the 1990s. However, it should be noted that the absolute number of withdrawals could be misleading if the corresponding number of new drugs approved in the same period was not taken into consideration. Unfortunately, the census was not readily available for an accurate rate calculation. Thus, it remained inconclusive whether this might reflect any increase in market withdrawal activity in recent years.

When compared to earlier decades, products withdrawn in the 1990s appeared to occur more frequently in multiple markets rather than in single markets. Overall, about two-thirds of the withdrawals in the 1990s occurred in multiple markets, compared to about half in the 1980s and 1970s, and only about one-fifth in the 1960s. On the other hand, there did not seem to be any major difference regarding the reasons or classes of drugs withdrawn in each of these four decades.

Unfortunately, the duration of marketing before withdrawal for every product was not available. Among the 87 products for which the timing of marketing was known, the median duration of marketing was 5.4 years. However, about 31 % of these products were withdrawn within the first two years and up to approximately 50% were withdrawn within the first five years. In addition, the median duration of marketing before withdrawal also appeared to be shorter in recent years. Except for the 1960s when most data were not readily retrievable, the median duration of marketing before withdrawal was 12.3 years for the 1970s, 6.2 years for the 1980s, and 2.6 years for the 1990s.

Lastly, although about half of the products were withdrawn from multiple markets, there were many single market withdrawals which occurred in Europe with France (13.2%), Germany (13.2%), and the United Kingdom (10.7%), contributing to most of these withdrawals. On the other hand, the United States contributed to only 5% of the single market withdrawals and Asia-Pacific countries contributed to only 3.3%. Figure 4 lists the geographical distribution of these withdrawals by countries. Unfortunately, except for the numeric number of withdrawals from each country or continent, when subgroup analyses were conducted to examine the reasons, therapeutic classes, year of withdrawal, and median duration of marketing before withdrawal, the numbers in the subgroups were too small to make a meaningful comparison.

DISCUSSION

This is the first of a series of reviews in our effort to explore the reasons and features of withdrawal of various pharmaceutical products from worldwide markets. Other research analyses regarding the withdrawal of medical devices, diagnostic products, radiopharmaceuticals, biologics, over-the-counter drugs, veterinary products, and herbal and health supplements are in progress and will be published as separate reports at later dates. However, despite our intensive effort to conduct an exhaustive search and review of all possible public domain data that were available, there are obviously some limitations.

First, most of the data that were accessible were from indirect sources such as the published literature. Therefore, the data available might be incomplete and potential errors are possible. This was especially true for products that were withdrawn in the early years (ie, 1960s and early 1970s) where verifications were more difficult.

Second, the source of references were primarily English literature. Thus, although there was occasional mention of regulatory actions from other countries, the majority of the data were limited to North America and Europe (especially the FDA, MCA, and EMEA). Withdrawal information from many non-English speaking countries such as Japan and other nations was very limited. As such, it was very difficult to examine the issue of pharmacogenetics that might have resulted in different expressions of adverse drug reactions which might have led to isolated drug withdrawals in specific markets (ie, differences in ethnicity and local medical practices such as the use of herbs, homeopathic therapies, etc.).

Third, a descriptive analysis was used as the methodology in the current review. However, due to the intrinsic limitation of this design, our observations would be restricted by the availability of the data in the public domain. Thus, one should be very careful and objective when approaching data of this nature in order not to misinterpret the results due to potential data gaps.

Fourth, some analyses in the current review implied an equal access of all pharmaceutical products worldwide which, in reality, cannot be true. Differences in regulatory requirements, medical practices, marketing and promotional schemes of the industry, pricing and reimbursement structures, cultural mores, and so forth would all affect drug availability and usage in different countries. Because of that, there were certain withdrawals of locally available products that never reached other markets. Therefore, comparison between countries or regions might not have been totally correct. Also, even if some products were available in multiple markets, the extent of market penetration and market share would also have had an impact on whether the products had been used sufficiently among patients in each individual country.

Despite these limitations, we decided to proceed with this review because our objective was not to provide a comprehensive list of worldwide prescription drug withdrawals but rather to examine from the available literature and public domain data the general trend and pattern of the major withdrawals over the last few decades to see whether we might be able to learn from these experiences.

Due to the very nature of many adverse reactions, some side effects were not necessarily predictable at the time of discovery or during clinical testing. Adverse reactions might arise from the molecular structure of the drug itself, its metabolites, or at times from an interaction with other products or conditions. Also, although the nature of many adverse reactions was due to intrinsic cytotoxicity, immunogenicity, or allergenicity, many times the mechanism might be idiopathic or unknown. Thus, not only would a reaction seen in a mouse not necessarily be repeated in other species, sometimes, even though a drug had been tested extensively in clinical studies, certain rare reactions would still be missed.

For example, aplastic anemia has an occurrence of only a few cases per million (143,144). It would not be practical to test a drug in a million patients before it could be marketed. Not only would the cost incurred be prohibitive, but also the practice would prevent many useful compounds from reaching needy patients in a timely manner. In addition to the issues of species differences and sample size limitation, there were other reasons why adverse drug reactions might not be detected until the postmarketing period. Off-label use, misuse, abuse, drug interactions, and use in untested populations (pediatric, geriatric, pregnant) are a few good examples (1,2,3).

In this report, we reviewed the available sources of information in the public domain and there were several findings worth further discussion. For example, CNS-acting agents, musculoskeletal products, and cardiovascular agents were found to be the three most common types of drugs withdrawn. With further subclassification, NSAIDs (13.2%), other nonnarcotic analgesics (8.3%), antidepressants (7.4%), and vasodilators (5.8%), appeared to be the most common individual classes of drugs removed from the market. Interestingly, in the 1995 FDA Annual Adverse Drug Experience Reports (145), the top three broad categories of agents with the highest number of adverse events reported were CNS acting agents (28%), hormones and synthetic substitutes (14%), and cardiovascular agents (10%) with NSAIDs and oral contraceptives as the two most common groups of drugs that had the highest number of adverse events reported. Therefore, with the exception of oral contraceptives, the frequency of adverse event reports might have some correlation with the types of drugs withdrawn.

In alignment with general knowledge, hepato-toxicity was the single most common reason for withdrawing a drug from the market. This was not surprising as many drugs were metabolized by the liver and there were many common medical disorders (eg, alcoholism, viral hepatitis, etc.) that would also compromise the patients' underlying liver functions. Also, drug interactions due to common metabolic pathways through similar Cytochrome P-450 isoenzymes (eg, 3A4, 2A6, etc.) were also a frequent concern (146,147). Of the list of drugs identified in the current review, several drugs were withdrawn due to drug interaction problems (eg, astemizole, mebanazine, mibefradil, nialamide, phenoxypropazine, and terfenadine).

In addition to liver toxicity, hematologic toxicity and cardiotoxicity were the other more common reasons that led to market withdrawal. Bone marrow suppression, agranulocytosis, hemolytic anemia, and derangement of the homeostasis system that led to either thromboembolism or bleeding were frequently observed hematologic reactions. On the other hand, valvular heart diseases, cardiac arrhythmias, and the consequent excess mortality also led to a significant number of drugs being removed from the market.

As for nonbody system-related safety issues, abusive use, drug interaction, overdose, and anaphylaxis/allergy all contributed to a considerable amount of withdrawals. In addition, there were many products withdrawn due to multiple organ toxicities instead of a single reason. On the other hand, although carcinogenicity was another frequent concern, there were fewer drugs withdrawn due to carcinogenic concerns in recent years (eg, none occurred in the 1990s). All drugs withdrawn for this reason were from experimental animals and most took place in earlier decades. With strict regulations on preclinical carcinogenicity testing in recent years, this was less of a concern at the present time.

Similarly, the thalidomide story also led to strict testing on congenital abnormalities and, therefore, this had also been under strict scrutiny with rare cases of problems surfacing in recent years (148,149,150). On the other hand, with a strict warning provided to patients with child-bearing potential, sometimes even a drug with this concern might still be available. For example, isotretinoin is a drug on the market with known teratogenic potential that is still available for patients with severe acne upon a strict warning being placed on its label (151).

While the absolute rate of withdrawals remained unknown (due to unavailability of the denominator), there were some factors that might help to account for the apparent observation that more products might have been withdrawn in recent decades. A general increase in public awareness of drug safety and the availability of many new drug products might have contributed partly to this observation (18-23). In addition, modern computerized safety signal detection systems, an increase in pharmacovigilance manpower in the regulators and the pharmaceutical industry, and more collaboration among regulators with faster communication might also have had some impact (152). Lastly, the vast number of new products which became available in recent years and the popularity of alternative medicines (eg, herbs, health supplements, homeopathic therapies) would undoubtedly have changed the dynamic of this observation further (3,153). Older drugs might interact with newer compounds and/or alternative medicines causing safety concerns that might not have been previously recognized.

Many drug withdrawals appeared to have occurred within the first few years of marketing. In fact, the median duration of marketing before withdrawal was 5.4 years and about one-third of the products were withdrawn within the first two years. This was understandable because the well-known "Weber-- Effect" of decreased adverse event reporting after the first few years might have played a role as attention to safety of a new drug might dwindle over time (154,155,156). On the other hand, it would take some time for a drug to be marketed long enough to have a sufficiently large and heterogeneous population of patients exposed to it to identify a problem.

It was observed that products withdrawn in recent years occurred more in multiple markets and the median duration of marketing before withdrawal also appeared to be shorter. Obviously, simultaneous launches in multiple markets (This is especially common after the European Medicines Evaluation Agency adopted the centralized approval process for its member countries in the 1990s.) for business reasons from many pharmaceutical manufacturers (thus, multiple withdrawals) might have partly accounted for this observation. In addition, rapid dissemination of safety information in the medical literature and news media, and frequent communication among regulators worldwide might have also contributed to this occurrence.

As for geographic distribution of these withdrawals, there appeared to be more single market withdrawals from Europe than from other continents. In addition to possible differences in regulatory and/or medical practices, one other explanation was that quite a few prescription drugs were first available in Europe before they reached the United States and other markets, especially in earlier years (43,124,153,157). Significant safety problems might have emerged in the first few markets where the product was launched that led to a quick withdrawal before it ever reached the markets in other countries.

Unfortunately, a limitation of the current review was that withdrawals from the Japanese market were not readily available in the English-based medical literature. Many other single market withdrawals might have occurred in Japan although our review could only identify a limited number of these products. Future research should look into more direct communication with the Japanese regulator and possibly translating and reviewing more published Japanese medical literature.

As stated before, the objective of the current review was to see whether we might be able to learn from these experiences. One commonly asked question was that if a drug had major clinical benefits to some patients, could it be reinstated despite obvious safety concerns in other patient populations. A good but controversial example was thalidomide (148). While the tragedy of congenital abnormalities still frightened many people, thalidomide was reinstated and used as an immunosuppressive agent for leprosy, graft-versushost disease, rheumatoid arthritis, oral ulcers, and cutaneous inflammatory disorders (148, 149,150).

The balance of the risk-to-benefit ratio also greatly contributed to the decision to withdraw a pharmaceutical product from the market (32,158). Oftentimes, the public would only focus on the safety concern such as excess mortality or hepatotoxicity. However, the actual decision was sometimes based on the availability of an alternative treatment and what impact the withdrawal would have on patients. A good example was the continued availability of isoniazid and sodium valproate despite the known risk of fatal hepatitis. Another recent example was the rapid decision by the FDA to withdraw mibefradil and bromfenac but to keep troglitazone (Troglitazone was voluntarily withdrawn by the manufacturer in the United Kingdom market in December 1997 but remained available in the United States until March 2000.) on the United States market longer (53,55,56,57,90-94,159,160). While the adverse reactions of concern might be very similar, bromfenac was quickly withdrawn and troglitazone was retained in the United States market initially because the latter was viewed as an important therapeutic modality for diabetic patients (159,160). Therefore, although the ultimate decision making process involved the balance of risk-- to-benefit ratio of the drug product itself, whether there were other alternatives available would also affect the dynamics of the decision.

There were other recent products with important safety concerns that had not been removed from the market. Instead, they were allowed to be marketed continuously but with strict warnings and/or restrictions. For example, several protease inhibitors (eg, indinavir, ritonavir, saquinavir) were allowed to remain on the market despite the well-known adverse effects of hyperglycemia and hyperlipidemia which might lead to potential coronary artery disease and other complications in many young AIDS patients (161-164).

Another frequently asked question regarding drug withdrawal was whether the expedited review process by various regulators in recent years had led to an increase in "unsafe" drugs and thus more recent drugs being withdrawn (24,25,30,53). This was a very controversial area. Spriet-Pourra and Auriche examined 131 cases of withdrawal in four countries (United States, United Kingdom, France, and Germany) between 1961 to 1992 and concluded that withdrawal activity had increased from 2.5 products annually in years prior to 1982 to 7.2 products annually from 1982 to 1992 (34). However, as mentioned before, the absolute number of withdrawals could be misleading if not taking into consideration the corresponding number of new drugs approved in the same period.

Friedman et al. of the FDA examined 11 withdrawals by the agency between 1979 and 1998 in four five-year periods (53). When taking into consideration the total number of new molecular entities approved (ie, 95 in 1979-1983, 113 in 1984-1988, 127 in 19891993, and 172 in 1994-1998), the percentage of drugs withdrawn was actually quite stable. In fact, the percentage had slightly decreased in view of more and more new products approved in later years. On the other hand, Bakke et al. had examined 29 drugs withdrawn from three countries (United States, United Kingdom, and Spain) between 1974 and 1993 and found that even taking into consideration the new molecular entity approval, the percentage of drugs withdrawn after 1983 still increased to a rate of about 3% to 4% among all the new drugs introduced compared to only about 2% in the earlier period (51). Therefore, this remained a controversial subject and current data seemed inconclusive. More research in the future would be needed to examine this issue further.

Market withdrawal is a major concern for the health care providers, patients, and the pharmaceutical industry (27-30). From a patient perspective, no one would like to take a medicine that would have a significant safety concern. On the other hand, a manufacturer needs to invest a considerable amount of time and money in laboratory research, animal testing, and human studies before a drug finally becomes available in the market (31,158). To see a product withdrawn due to safety concerns would not only be a great disappointment to the many researchers who had devoted the time and effort to develop the drug, but it also greatly impacts the patients taking and physicians prescribing the products.

Felbamate was a good example (19,37, 38). It was an anticonvulsant that was first available in the early 1990s for Lennox-Gastaut Syndrome and refractory partial seizures. However, within the first year of marketing, it was reported to be associated with many cases of aplastic anemia and hepatotoxicity that led to its subsequent withdrawal in 1994 in the United States market. The sudden withdrawal of felbamate led to many patients suffering from increased seizure activities that were not well managed by other alternate therapies. Thereby, with appeals from the physicians and patients, felbamate was reinstated to the United States market in 1995, although with strict restriction on its use (19).

Unfortunately, there was a common misconception that if a drug had potential safety concerns and was withdrawn from the market, then all the problems would disappear immediately. However, the impact on the patients and the physicians was frequently not fully appreciated. In addition to felbamate, there were several other examples. Brown et al. had raised the concern of potential difficulty in managing pregnant women with eclampsia when the Australian government announced the sudden withdrawal of shortacting oral nifedipine from the market (due to safety concerns in patients with ischemic heart disease and hypertension) (41). When indalpine and nomifensine were withdrawn from the French market in 1985 to 1986, there were a number of patients who committed suicide because other antidepressants were apparently not as effective for them (40).

In addition to the concern of depriving patient subpopulations of important therapeutic options, sometimes the withdrawals might lead to other problems as the substitutions might not necessarily be better or safer. Ekeberg et al. demonstrated that although barbiturate overdoses decreased after the Norwegian government removed four barbiturates from the market, incidences of self-- poisonings with antidepressants and neuroleptics had increased significantly afterward (46). Ross-Degnan et al. examined the impact of zomepirac withdrawal in the early 1980s (165). They found that this withdrawal not only resulted in substitutions for other NSAIDs but also an increased use of other habit-forming analgesics such as barbiturates and narcotics. These were only a few examples to illustrate the complexity of product withdrawals and their great potential impact to the patients and the medical community. It is definitely an important topic that will warrant further research and examination in the future.

With the popularity of the Internet, public attention, and the increase in pharmacovigilance manpower at both the industry and regulator levels, many adverse drug reactions are now actively monitored and quickly identified. The active postmarketing surveillance system and increased awareness of adverse drug reactions among health care practitioners and patients in many countries will hasten the reporting of unfavorable adverse events for many products (4-17). Therefore, any significant adverse reaction should quickly be known and increased public attention might lead to an earlier decision in managing these safety concerns, including potential withdrawal of the drugs) from the market. Also, routine communication among regulators regarding significant drug safety alerts and the establishment of an electronic network among various European Union regulators will certainly contribute to a rapid exchange of information and decision making in the future (152). These activities will set the stage of a modern era of pharmacovigilance and ensure improved management regarding safety concerns of pharmaceutical products.

CONCLUSIONS

Overall, CNS-acting agents, musculoskeletal products, and cardiovascular agents were found in the current review to be the three most common categories of drugs withdrawn with NSAIDs, other nonnarcotic analgesics, vasodilators, and antidepressants as the most frequent subclass of drugs seen. Whereas rare reactions, such as aplastic anemia, required a large patient exposure before they were detected, more frequent reasons for marketing withdrawal of drugs were the more commonly seen reactions of the hepatic, hematologic, and cardiovascular systems. Despite teratogenicity and obstetric complications being touted as important reasons for drug withdrawal, they only accounted for a few of the drugs withdrawn. While the decision leading to market withdrawal was often manifold, the balance of risk-to-benefit ratio and the availability of alternatives both played a key role in the process.

On the other hand, although different rates of market uptake might have affected the time when unpredictable reactions were detected, about one-third of the drugs withdrawn for safety reasons occurred within the first two years and almost half occurred within the first five years of marketing. In addition, products withdrawn in recent years appeared to have occurred more frequently in multiple markets and the median duration of marketing also seemed shorter. Despite the fact that more products identified in the current review appeared to have been withdrawn in recent decades, it remained inconclusive whether this might reflect any increase in market withdrawal activity as the corresponding number of new drugs approved in the same period was not readily available. Further research will certainly be needed to address this issue in the future.

Lastly, the complexity of market withdrawals does not end with merely removing the products from the market. The ramification and impact of these withdrawals to the patients and the medical community will be an important topic in future research.

ADDENDUM

In March 2000, two other agents were withdrawn from the United States market. Cisapride was withdrawn because of a concern of cardiac arrhythmia and troglitazone for hepato-toxicity.

*An abstract of this paper was presented at the 15th International Conference on PharmacoEpidemiology, August 26-29, 1999, Boston, Massachusetts.

REFERENCES

1. Wilford BB, Finch J, Czechowicz DJ, Warren D.

An overview of prescription drug misuse and abuse: defining the problem and seeking solutions. J Law Med Ethics. 1994;22:197-203.

2. Charpak Y, Favre 1, Bloch J, Jauffret M, Maire A. Dependence, abuse and misuse of drugs: knowledge and attitudes of private physicians in the Ile-deFrance. Rev Epidem et Sante Publ. 1999;47:239247.

3. Boyle FM. Adverse interaction of herbal medicine with breast cancer treatment. Med JAustralia. 1997; 167:286.

4. Mann RD. Monitoring the Safety of Medicine. In: Luscombe D, Stonier PD, eds. Clinical Research Manual. Haslemere, England: Euromed Communications; 1994:8.1-8.42.

5. Flectcher AP. Post-Marketing Surveillance. In, Luscombe D, Stonier PD, eds. Clinical Research Manual. Haslemere, England: Euromed Communications; 1998:15.1-15.26.

6. Faich GA, Dreis M, Tomita D. National adverse drug reaction surveillance. Arch Int Med. 1998;148: 785-787.

7. Anello C. Management of ADR reports at the FDA. Drug Inf J. 1985;19:291-294.

8. Gordon AJ, Petrick RJ. Worldwide regulations for manufacturers on clinical safety surveillance of drugs. Drug Inf J. 1992;26:1-15.

9. Griffin JP. Survey of spontaneous adverse drug reaction reporting schemes in fifteen countries. Br J Clin Pharm. 1986;22:83S-1005.

10. Albengres E. Features of the French postmarketing drug surveillance system. Application to cutaneous effects of nonsteroidal antiinflammatory drugs. J Rheumatol. 1988;15:20-23.

11. Sachs RM, Bortnichak EA. An evaluation of spontaneous adverse drug reaction monitoring systems. Am J Med. 1986;81:49-55.

12. Fletcher AP. Spontaneous adverse drug reaction reporting vs. event monitoring: a comparison. J Roy Soc Med. 1991;84:341-344.

13. Mann RD. Managing adverse reaction information. Drug Inf J. 1992;26:583-587.

14. Dangoumau J, Lagier G, Royer RJ. A drug information-ADR collection network: the French system. Drug Inf J. 1986;20:337-346.

15. Talbot JCC. Management of ADR data from different sources. Drug Inf J. 1986;20:297-299.

16. Koch-Weser J. International aspects of spontaneous reporting of adverse drug reactions. Drug Inf J. 1985;19:217-221.

17. Rawlins MD. Spontaneous reporting of adverse drug reactions I-II: the data and uses. Br J Clin Pharmacol. 1988;26:1-11.

18. Weisberg E. Contraceptive choices-turning back the clock. Adv Contraception. 1991; 7:115-127.

19. Welty TE, Privitera M, Shukla R. Increased seizure frequency associated with felbamate withdrawal in adults. Arch Neurol. 1998;55:641-645.

20. Ashworth L. Is my antihistamine safe? Home Care Provider. 1997;2:117-120.

21. Bouvier-Colle MH. Drugs: 10 years of recall from the market. Rev Epidem et Sante Publ. 1994;42: 466-467.

22. Lynn NJ, Ellis JM. Pharmacists' liability into the year 2000. J Am Pharm Assoc. 1998;38:747-752.

23. Burkhart GA, Sevka MJ, Temple R, Honig PK. Temporal decline in filling prescriptions for terfenadine closely in time with those for either ketoconazole or erythromycin. Clin Pharm Therap. 1997; 61:93-96.

24. Kleinke JD, Gottlieb S. Is the FDA approving drugs too fast? Probably not-but drug recalls have sparked debate. Br Med J. 1998;317:899.

25. Miller HI. When is a drug 'safe'? Nature. 1990; 344:698.

26. Jefferys DB, Leakey D, Lewis JA, Payne S, Rawlins MD. New active substances authorized in the United Kingdom between 1972 and 1994. BrJ Clin Pharmacol. 1998;45:151-156.

27. Bakke OM. Drug withdrawals-circumstances and market impact. In: Cato AE, ed. Clinical Drug Trials and Tribulations. New York & Basle: Marcel Dekker Inc.; 1988:377-395.

28. McGhan WF, Block LH. An analysis of drug recalls and court actions in the United States Pharmaceutical Industry. Pharm Med. 1986;1:187-193.

29. Anonymous. 1996 product withdrawals in France. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; July 25, 1997;2252:1.

30. Anonymous. Rate of drug recalls rises sharply-- FDA `fast track' approval called into question. NBC Internet News. June 24, 1999. http://www.msnbc. com/news.

31. Steward F, Wibberley G. Drug innovation- what's slowing it down? Nature. 1980;284:118-120.

32. Goyan JE. The role of evidence in the approval of pharmaceuticals. Ann NY Acad Sci. 1993;703: 275-277.

33. Weintraub M, Northington FK. Drugs that wouldn't die. JAMA. 1986;255:2327-2328.

34. Spriet-Pourra C, Auriche M. Drug Withdrawal from Sale. 2nd Edition. Scrip Reports. Richmond, UK: PJB Publications Ltd.; 1994.

35. Bakke OM, Wardel WM, Lasagna L. Drug discontinuation in the United Kingdom and Untied States, 1964 to 1983: issues of safety. Clin Pharm Therap. 1984;35:559-567.

36. Lindstrom LH. The effect of long-term treatment with clozapine in schizophrenia: in 96 patients treated with clozapine for up to 13 years. Acta Psych Scand. 1988;77:524-529.

37. Bourgeois BR Felbamate. Semin Ped Neurol. 1997; 4:3-8.

38. Blum DE. New drugs for persons with epilepsy. Adv Neurol. 1998;76:57-87.

39. Anonymous. BGA wants molsidomine suspension. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; Oct. 7, 1985;1040:2.

40. Lechat P. Problems raised by the withdrawal from the market of drugs in therapeutic use. Therapie. 1987;42:403-410.

41. Brown MA, McCowan LM, North RA, Walters BN. Withdrawal of nifedipine capsules: jeopardizing the treatment of acute severe hypertension in pregnancy? Australian Society for the Study of Hypertension in Pregnancy. Med J Australia. 1997;166: 640-643.

42. Anonymous. Prescribe reviews French 1996 launches. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; Jan. 14, 1997;2197:1.

43. Hass AE, Portale DB, Grossman RE. New drugs: their market life and safety. Pharm J. 1985;235238.

44. Anonymous. Alpidem Suspended in France. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; 1993;1870:22.

45. Anonymous. Amineptine: marketing authorization suspended-drug abuse and dependence, France. The WHO Pharmaceuticals Newsletter-Regulatory Actions. Jan. & Feb. 1999;1&2. http://www. who.int/dmp/newsletters.

46. Ekeberg 0, Jacobsen D, Flaaten B, Mack A. Effect of regulatory withdrawal of drugs and prescription recommendations on the pattern of self-poisonings in Oslo. Acta Med Scand. 1987;221:483-487.

47. Anonymous. J&J antihistamine withdrawn. Hismanal taken off shelves. ABC Internet News. June 21, 1999. http://abcnews.go.com/sections/living/Daily News.

48. Anonymous. Maker to Discontinue Allergy Drug Hismanal. Washington Post. Online. Jun. 22, 1999. http://washingtonpost.com/wp-srv/health.

49. Anonymous. Hismanal-worldwide withdrawal. Information for Health Professionals. Malaysia Adverse Drug Reaction Advisory Committee Web site. July 1999. http://www3.jaring.my/mad.

50. Anonymous. Astemizole prescription only in UK. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; July 8, 1998;2350:21.

51. Bakke OM, Manocchia M, de Abajo F, Kaitin KI, Lasagna L. Drug safety discontinuations in the United Kingdom, the United States, and Spain from 1974 through 1993: a regulatory perspective. Clin Pharmacol Ther. 1995;58:108-117.

52. Fenner H. Evaluation of the efficacy and safety of NSAIDs. A new methodological approach. Scand J Rheum. 1989;80(suppl):32-39.

53. Friedman MA, Woodcock J, Lumpkin MM, Shuren JE, Hass AE, Thompson LJ. The safety of newly approved medicines. Do recent market removals mean there is a problem? JAMA. 1999;281:17281734.

54. Williams P, Williams P, Currie WJ, VandenBurg

MJ. A double-blind comparison of 'Osmosin', benoxaprofen and placebo in the treatment of osteoarthritis. Curr Med Res Opin. 1983;8(Suppl 2):90-98.

55. Anonymous. Bromfenac sodium capsules (DURACT) withdrawn from the market. Dear Doctor Letter. Philadelphia, PA: Wyeth Ayerst Company; 1998.

56. Anonymous. Bromfenac (DuractR) withdrawn from the market-severe liver failure. The WHO Information Exchange System. Alert No. 71, Jun. 26, 1998. http://www.who.int/dmp/Alerts.

57. Anonymous. Bromfenac withdrawn-hepatotoxicity, Saudi Arabia. The WHO Pharmaceuticals Newsletter-Regulatory Actions. May & June 1999;5&6. http://www.who.int/dmp/newsletters.

58. Anonymous. EC chlormezanone withdrawal confirmed. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; Oct. 10, 1997; 2276:5.

59. Anonymous. Chlormezanone withdrawn because of cutaneous toxicity-update. The WHO Pharmaceuticals Newsletter-Regulatory Actions. Jan. & Feb. 1997;1&2. http:/www.who.int/dmp/newsletters.

60. Anonymous. Chlormezanone marketing authorization withdrawn, Germany. The WHO Pharmaceuticals Newsletter-Regulatory Actions. Nov. & Dec. 1997;11&12. http://www.who.int/dmp/newsletters.

61. Anonymous. Chlormezanone withdrawn, United Arab Emirates. The WHO Pharmaceuticals Newsletter-Regulatory Actions. Mar. & Apr. 1998;3& 4. http://www.who.int/dmp/newsletters.

62. Anonymous. Chlormezanone withdrawn, Zimbabwe. The WHO Pharmaceuticals Newsletter-- Regulatory Actions. Sept. & Oct. 1998;9&10. http:/ www.who.int/dmp/newsletters.

63. Anonymous. Chlormezanone withdrawn, Saudi Arabia. The WHO Pharmaceuticals Newsletter-- Regulatory Actions. May & Jun. 1999;5&6. http:// www.who.int/dmp/newsletters.

64. Anonymous. Cardiac valvulopathy associated with exposure to fenfluramine or dexfenfluramine: U.S. Department of Health and Human Services interim public health recommendations, Nov. 1997. MMWR. 1997;46:1061-1066.

65. Anonymous. Warning not to use products containing fenfluramine (Ponderal, Ponderax) or dexfenfluramine (Redux). Health Canada Web site. Sept. 15, 1997. http://www.hc-sc.gc.ca.

66. Anonymous. Withdrawal of fenfluramine (Ponderax) and dexfenfluramine (Adifax). The Medicines Control Agency Web site. http://www.open. gov.uk/mca.

67. Anonymous. Fenfluramine and dexfenfluramine. The European Agency for the Evaluation of Medicinal Products Press Release. The EMEA Web site. Sept. 9, 1999. http://www.eudra.org/emea.html.

68. Anonymous. Fenfluramine and dexfenfluramine. Information for Health Professionals. Malaysia Adverse Drug Reaction Advisory Committee Web site. http://www3.jaring.my/mad.

69. Anonymous. Fenfluramine and dexfenfluramine withdrawn-cardiotoxicity (USA). The WHO Information Exchange System. Alert No. 63, Sept. 16, 1997. http://www.who.int/dmp/Alerts.

70. Anonymous. Fenfluramine and dexfenfluramine withdrawn, India. The WHO Pharmaceuticals Newsletter-Regulatory Actions. May & Jun. 1998; 5&6. http://www.who.int/dmp/newsletters.

71. Anonymous. Fenfluramine and dexfenfluramine: withdrawn: Saudi Arabia. The WHO Pharmaceuticals Newsletter-Regulatory Actions. May & Jun. 1999;5&6. http://www.who.int/dmp/newsletters.

72. Anonymous. Diet drugs pulled off the market. The Wall Street Journal Internet News. Sept. 15, 1997. http:/interactive.wsj.com/dailyedition.

73. Anonymous. Withdrawal of oxyphenisatin acetate, diacetoxydiphenolisatin and triacetyldiphenolisatin from the Australian market. Med J Australia. 1972; 1:1051-1053.

74. Mihajlovic G. Recall of metamizole from the market in countries with modern pharmacotherapy. Srpski Arhiv Za Celokupno Lekarstvo. 1996;124: 44-46.

75. Anonymous. Metamizole sodium-prohibited, Nepal. The WHO Pharmaceuticals Newsletter-Regulatory Actions. Nov. & Dec. 1997;11&12. http:H www.who.int/dmp/newsletters.

76. Anonymous. Metamizole sodium-withdrawn due to agranulocytosis, Zimbabwe. The WHO Pharmaceuticals Newsletter-Regulatory Actions. Sept. & Oct. 1998;9&10. http://www.who.int/dmp/ newsletters.

77. Anonymous. Metamizole sodium-withdrawn due to agranulocytosis, Yemen. The WHO Pharmaceuticals Newsletter-Regulatory Actions. Jan. & Feb. 1999;1&2. http://www.who.int/dmp/newsletters.

78. Anonymous. Metamizole sodium-marketing authorization suspended, Sweden. The WHO Pharmaceuticals Newsletter-Regulatory Actions. May & Jun. 1999;5&6. http://www.who.int(dmp/newsletters.

79. Anonymous. Withdrawal of medicinal products containing ebrotidine-liver toxicity. The WHO Information Exchange System. Alert No. 72, Jul. 29, 1998. http://www.who.int/dmp/Alerts.

80. Anonymous. Bristol-Myers Squibb (BMS) Enkaid withdrawn in US. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; Sep. 25, 1991;1654:20.

81. Anonymous. Fenfluramine-withdrawn. The WHO Pharmaceuticals Newsletter-Regulatory Actions. Mar. & Apr. 1998;3&4. http://www.who.int/dmp/ newsletters.

82. Beasley R, Pearce N, Crane J, Burgess C. Withdrawal of fenoterol and the end of the New Zealand asthma mortality epidemic. Int Arch Allerg Immun. 1995;107:325-327.

83. Crane J, Pearce N, Flatt A, et al. Prescribed fenoterol and death from asthma in New Zealand, 198183: case-control study. Lancet. 1989;1:917-922.

84. Anonymous. Glaxo pulls antibiotic on safety fears.

Reuters Internet News. Oct. 27, 1999. http://www. reuters.com/news.

85. Henry JA, Martin AJ. The risk-benefit assessment of antidepressant drugs. Med Toxic Ad Drug Experience. 1987;2:445-462.

86. Anonymous. Ketorolac withdrawn in France. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; Jan. 7th/llth, 1994;1886/87: 27.

87. Anonymous. Ketorolac post marketing surveillance (PMS) study published. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; Feb. 16, 1996;2103:1.

88. Anonymous, Ketorolac-extension of suspension. The WHO Pharmaceuticals Newsletter-Regulatory Actions. Jan. & Feb. 1997;1&2. http://www. who.int/dmp/newsletters.

89. Garcia C. Mianserin agranulocytosis followed by oxitriptan eosinophilia. Nervenarzt. 1992;63:303307.

90. Anonymous. Posicor. Irish Medicines Board Press Release. June 8, 1998. http://www.imb.ie/news. 91. Anonymous. Mibefradil (Posicor). Information for

Health Professionals. Malaysia Adverse Drug Reaction Advisory Committee Web site. http://www3. jaring.my/mad.

92. Anonymous. Mibefradil (Posicor)-withdrawn from the market: potential harmful interactions. The WHO Information Exchange System. Alert No. 70, Jun. 15, 1998. http://www.who.int/dmp/Alerts.

93. Anonymous. Roche withdraws hypertension drug from the market. The Wall Street Journal Internet News. Jun. 8, 1998. http://interactive.wsj.com/ dailyedition

94. Anonymous. Mebefradil-suspended, Germany. The WHO Pharmaceuticals Newsletter-Regulatory Actions. Sept. & Oct. 1998;9& 10. http://www. who.int/dmp/newsletters.

95. Anonymous. Muzolimine withdrawn in Europe. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; Jul. 17, 1987;1223:22.

96. Anonymous. Ciba-Geigy implements Coramine withdrawal. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; Oct. 4, 1989; 1452:11.

97. Anonymous. CSM Update. Withdrawal of Nomifensine. Br Med J. 1986;293:41.

98. Martlew VJ. Immune haemolytic anemia and nomifensine treatment in north west England 1984-85: report of six cases. J Clin Path. 1986;39:11471150.

99. Delchier JC, Metreau JM, Levy VG, Opolon P, Dhumeaux D. Oxyphenisatin, a laxative responsible for chronic hepatitis and cirrhosis, still marketed in France. Nouvelle Presse Medicale. 1979;8:29552958.

100. Anonymous. Liver complications result in withdrawal of attention deficit hyperactivity disorder drug cylert. Health Canada Web site. Sept. 22, 1999. http://www.hc-sc.gc.ca.

101. Anonymous. Volital (pemoline) has been withdrawn. The Medicines Control Agency Web site. http://www.open.gov.uk/mca.

102. Anonymous. Pemoline withdrawn-hepatotoxicity, UK. The WHO Pharmaceuticals Newsletter-- Regulatory Actions. Nov. & Dec. 1997;11&12. http://www.who.int/dmp/newsletters.

103. Brunner FP, Selwood NH. End-stage renal failure due to analgesic nephropathy, its changing pattern and cardiovascular mortality. EDTA-ERA Registry Committee. Nephrol Dial Transplant. 1994;9: 1371-1376.

104. Pommer W, Glaeske G, Molzahn M. The analgesic problem in the Federal Republic of Germany: analgesic consumption, frequency of analgesic nephropathy and regional differences. Clin Nephrol. 1986; 26:273-278.

105. McGuinness ME, Talbert RL. Phenformin-induced lactic acidosis: a forgotten adverse drug reaction. Ann Pharmaco. 1993;27:1183-1187.

106. Anonymous. Malaysia withdraws prenylamine registrations. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; Nov. 9, 1988; 1359:19.

107. Anonymous. Proxibarbal: voluntary withdrawal. The WHO Pharmaceuticals Newsletter-Regulatory Actions. May & Jun. 1998;5&6. http://www. who.int/dmp/newsletters.

108. Anonymous. UK suspends Lundbeck's antipsychotic treatment. Health News Daily. Dec. 7, 1998. http://www.medicalnetinc.com/reuters.

109. Anonymous. Notification of suspension of Serdolect (sertindole). Irish Medicines Board. Dear Doctor Letter. Nov. 23, 1998. http://www.imb.ie/news.

110. Anonymous. Suspension of availability of sertindole. The Medicines Control Agency Web site. http:/www.open.gov.uk/mca.

111. Anonymous. Sertindole (Serdolect) suspended-- cardiac arrhythmias and sudden cardiac death. The WHO Information Exchange System. Alert No. 78, Dec. 4, 1998. http://www.who.int/dmp/Alerts.

112. Anonymous. Sorivudine & nevirapine on FDA panel. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; May 21, 1996; 2130:1.

113. Davey P, McDonald T. Postmarketing surveillance of quinolones, 1990 to 1992. Drugs. 1993;45 (Suppl 3):46-53.

114. Anonymous. Temafloxacin-anatomy of a withdrawal. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; 1992;1737:20-21.

115. Anonymous. Hoechst Marion Roussel (HMR) withdraws Seldane in US. SCRIP World Pharmaceutical News. London, UK: PJB Publications Ltd.; Feb. 1998;2298:1.

116. Anonymous. Terfenadine suspension & restricted use. The WHO Pharmaceuticals Newsletter-Regulatory Actions. Mar. & Apr. 1997;3&4. http://www. who.int/dmp/newsletters.

117. Anonymous. Terfenadine prohibited. The WHO

Pharmaceuticals Newsletter-Regulatory Actions. May & Jun. 1998;5&6. http:/www.who.int/dmp/ newsletters.

118. Anonymous. Terfenadine withdrawn-cardiotoxicity, Saudi Arabia. The WHO Pharmaceuticals Newsletter-Regulatory Actions. May & Jun. 1999; 5&6. http://www.who.int/dmp/newsletters.

119. van der Klauw MM, van Rey FJ, Stricker BE Polymorph ventricular tachycardia with torsades de pointes caused by administration of terodiline (Mictrol). Nederlands Tijdschrift voor Geneeskunde. 1992;136:91-93.

120. Edwards RI, Lindquist M, Wiholm BE, Napke E. Quality criteria for early signals of possible adverse drug reactions. Lancet. 1990;336:156-158.

121. Haberle DA, Davis JM, Kawabata M, Metz C, Wapler P, Stachl M. Renal and single-nephron function is comparable in thiobutabarbitone- and thiopentone-anaesthetised rats. Pflugers Archive-Eur J Physiol. 1993;424:224-230.

122. Barclay WR. Ticrynafen's withdrawal from the market. JAMA. 1980;243:771.

123. Finkle MJ. Anatomy of drug withdrawals in the United States. In: Cato AE, ed. Clinical Drug Trials and Tribulations. New York & Basle: Marcel Dekker Inc.; 1988:363-375.

124. Marcus CJ, Griffin JP. New chemical entities 19721982: licensing and subsequent adverse reactions. A UK/US comparison. Pharm Int. 1984;5:146-149.

125. Anonymous. Liver complications reported with anti-Parkinson's drug Tasmar. Health Canada Web site. Nov. 23, 1998. http://www.hc-sc.gc.ca.

126. Anonymous. Notification of withdrawal of Tasmar. Irish Medicines Board. Dear Doctor Letter. Nov. 23, 1998. http://www.imb.ie/news.

127. Anonymous. Withdrawal of Tasmar (tolcapone) from the market. The Medicines Control Agency Web site. http://www.open.gov.uk/mca.

128. Anonymous. Recommendation for the suspension of the marketing authorization for Tasmar. The European Agency for the Evaluation of Medicinal Products Press Release. The EMEA Web site. Nov. 17, 1998. http://www.eudra.org/emea.html.

129. Anonymous. Tolcapone (Tamar)-liver failure: suspension & warnings. The WHO Information Exchange System. Alert No. 75, Nov. 21, 1998. http:fl www.who.int/dmp/Alerts.

130. Anonymous. Tolcapone withdrawn-hepatotoxicity, Australia. The WHO Pharmaceuticals Newsletter-Regulatory Actions. May & Jun. 1999;5&6. http:llwww.who.int/dmp/newsletters.

131. Anonymous. Tolrestat withdrawn-hepatic necrosis. The WHO Pharmaceuticals Newsletter-Regulatory Actions. Mar. & Apr. 1997;3&4. http://www. who.int/dmp/newsletters.

132. Meyboom RH. The Halcion affair in 1979, a false alarm? Nederlands Tijdschrift voor Geneeskunde. 1989;133:2185-2190.

133. Meyboom RH. Secondary effects of drugs. The

1979 Halcion experience in the Netherlands, a false alarm? Therapie. 1991;46:373-377.

134. Nazareth 1, Ashworth M, Hammond J, King M. Withdrawal of triazolam's product licence: effect on patients 18 months later. Addiction. 1995;90: 927-934.

135. Brahams D. Medicine and the law. Triazolam suspended. Lancet. 1991;338:938.

136. Anonymous. Trovafloxacin & alatrofloxacin-suspension and warnings: hepatotoxicity. The WHO Information Exchange System. Alert No. 86, Jun. 18, 1999. http://www.who.int/dmp/Alerts.

137. Anonymous. Trovafloxacin & alatrofloxacin-suspension and warnings. Information for Health Professionals. Malaysia Adverse Drug Reaction Advisory Committee Web site. July 1999. http://www3. jaring.my/madrac.

138. Anonymous. Diabetes drug spurs worries about safety. The Wall Street Journal Internet News. Dec. 1, 1997. http://interactive.wsj.com/dailyedition.

139. Anonymous. Troglitazone (Romozin) withdrawn. The Medicines Control Agency Web site. http:// www.open.gov.uk/mca.

140. Goyan JE. Withdrawal of a drug from the market. What should the prescriber do? JAMA. 1993; 270: 1976-1977.

141. Levy DB, Vasilomanolakis EC. Anaphylactic reaction due to zomepirac. Drug Intel Clin Pharm. 1984;18:983-984.

142. Speirs CJ. Comparison of the safety of several nonsteroidal anti-inflammatory drugs currently or formerly marketed in the United Kingdom. J Clin Pharmacol. 1988;28:58-S12.

143. Patton WN, Duffull SB. Idiosyncratic drug-induced haematological abnormalities. Incidence, pathogenesis, management and avoidance. Drug Safety. 1994;11:445-462.

144. Heimpel H. When should the clinician suspect a drug-induced blood dyscrasia, and how should he proceed? Eur J Haematol. 1996;60 (supl):11-15.

145. Knapp DE, Robinson JI, Britt AL. Annual adverse drug experience report: 1995. The FDA Web site. http://www.fda.gov/cder/aers/index.htm.

146. Touw DJ. Clinical implications of genetic polymorphisms and drug interactions mediated by cytochrome P-450 enzymes. Drug Met Drug Int. 1997; 14:55-82.

147. Jefferson JW. Drug and diet interactions: avoiding therapeutic paralysis. J Clin Psych. 1998;59(Suppl 16):31-39.

148. Stirling D, Sherman M, Strauss S. Thalidomide: A surprising recovery. J Am Pharm Assoc. 1997;37: 306-313.

149. Schuler U, Ehninger G. Thalidomide: rationale for renewed use in immunological disorders. Drug Safety. 1995;12:364-369.

150. Tseng S, Pak G, Washenik K, Pomeranz MK, Shupack JL. Rediscovering thalidomide: a review of its mechanism of action, side effects, and potential uses. J Am Acad Derm. 1996;35:969-979.

151. Koren G, Pastuszak A. How to ensure fetal safety when mothers use isotretinoin (Accutane). Can Fam Phy. 1997;43:216-219.

152. Montero D, Moore N, Coulson R, et al. Electronic transmission of rapid alerts in pharmacovigilance: a pilot EC experiment. Pharmacoepi Drug Safety. 1994;3:163-167.

153. Kessler DA, Hass AE, Feiden KL, Lumpkin M, Temple R. Approval of new drugs in the United States. Comparison with the United Kingdom, Germany, and Japan. JAMA. 1996;276:1826-1831.

154. Weber JCP. Limitations of a voluntary reporting system. In: Hawkins DF, ed. Drugs and Pregnancy; Human Teratogenesis and Related Problems. Edinburgh: Churchill Livingstone; 1983:31-40.

155. Clark JA, Klincewicz SL, Turner ME, Arellano FM. Spontaneous reporting sequences and the Weber effect: a new look at an old phenomenon. Pharmacoepidem Drug Safety. 1999;8:S95.

156. Fung M, Kwong K, Mockbee C, Hornbuckle K, Muniz E. Analysis of the occurrence of Weber effect in spontaneous adverse event reporting. Pharmacoepidem Drug Safety. 1999;8:S106.

157. Coppinger PL, Peck CC, Temple RJ. Understanding comparisons of drug introductions between the United States and the United Kingdom. Clin Pharmacol Ther. 1989;46:139-145.

158. Hayes AH Jr. Safety considerations in product development. Drug Safety. 1990;5(Suppl 1):24-26.

159. Anonymous. FDA panel recommends continued use of controversial diabetes drug. CNN Interactive News. Mar. 26, 1999. http://cnn.com/HEALTH.

160. Anonymous. FDA places stricter controls on diabetes drug. CNN Interactive News. June 17, 1999. http://cnn.com/HEALTH.

161. Anonymous. New concerns about side effects of AIDS-fighting protease inhibitors. CNN Interactive News. June 30,1998. http://cnn.com/HEALTH.

162. Anonymous. HIV protease inhibitors-hyperglycemia. The WHO Pharmaceuticals Newsletter-- Regulatory Actions. Nov. & Dec. 1997;11&12. http://www.who.int/dmp/newsletters.

163. Anonymous. HIV protease inhibitors revised data sheet-increased blood sugar. The WHO Pharmaceuticals Newsletter-Regulatory Actions. July & Aug. 1998;7&8. http://www.who.int/dmp/newsletters.

164. Anonymous. HIV protease inhibitors: lipodystrophy, hyperglycaemia, and nephrolithasis. The WHO Pharmaceuticals Newsletter-Drug Surveillance. July & Aug. 1998;7&8. http://www.who.int/dmp/ newsletters.

165. Ross-Degnan D, Soumerai SB, Fortess EE, Gurwitz JH. Examining product risk in context. Market withdrawal of zomepirac as a case study. JAMA. 1993;270:1937-1942.

MAN FUNG, MD, MBA, FACP

Worldwide Pharmacovigilance and Epidemiology, Lilly Research Laboratories, Indianapolis, Indiana, and Eli Lilly Taiwan, Taipei, Taiwan

ANNA THORNTON, PHARMD

Worldwide Pharmacovigilance and Epidemiology, Lilly Research Laboratories, Indianapolis, Indiana

KATHY MYBECK, PHARMD

Global Medical Information, Lilly Research Laboratories, Indianapolis, Indiana

JASMANDA HSIAO-HUI WU, MS, MPH

Worldwide Pharmacovigilance and Epidemiology, Lilly Research Laboratories, Indianapolis, Indiana

KEN HORNBUCKLE, MPH, DVM, PHD

Worldwide Pharmacovigilance and Epidemiology, Lilly Research Laboratories, Indianapolis, Indiana

EDMUNDO MUNZ, MD, PHD

Worldwide Pharmacovigilance and Epidemiology, Lilly Research Laboratories, Indianapolis, Indiana

Reprint address: Man C. Fung, MD, MBA, FACP, Worldwide Pharmacovigilance and Epidemiology, DC-1778, Lilly Research Laboratories, Indianapolis, IN 46285. E-mail: fung_mC@lilly.com.

Copyright Drug Information Association Jan-Mar 2001
Provided by ProQuest Information and Learning Company. All rights Reserved