Lam0 trig he-induced Severe Cutaneous Adverse Reactions

Transcription

1 Epilepsia, 39(Suppl. 7):S22-S26, 1998 Lippincott Williams & Wilkins, Philadelphia 0 International League Against Epilepsy Lam0 trig he-induced Severe Cutaneous Adverse Reactions *Raymond G. Schlienger, *?$Lori E. Shapiro, and *?$Neil H. Shear Divisions of *Clinical Pharmacology and?dermatology, Departments of?$medicine, $Pharmacology, and $Pharmacy, and the f$glaro Wellcome-Sunnybrook Drug Safety Clinic, Sunnybrook Health Science Centre, University of Toronto, Toronto, Ontario, Canada Summary: Purpose: We systematically reviewed and analyzed published and unpublished cases of Stevens-Johnson syndrome (SJS), or toxic epidermal necrolysis () associated with lamotrigine (LTG) therapy to identify characteristics of these reactions. Methods: We performed a MEDLINE search (January 1985 to April 1998) and citation tracking for published reports. In addition, reports were requested from the Uppsala Monitoring Centre of the World Health Organization (WHO). Published and WHO cases of LTG-associated SJS or were included if the causal relationship was assessed as either possible, probable, or definite. Results: We identified a total of 57 cases (43 cases of SJS, cases of ), of which 13 (23%) were published. Cases in the SJS group were significantly younger-than in the group (21 years vs. 31 years). The median time to onset (17 days for SJS and ) and the median dosage at onset (50 mg vs mg) for SJS and did not differ significantly. Concomitant use of valproate (VPA) was reported in 74% of the SJS cases and 64% of the cases. In three cases, was the cutaneous manifestation of the antiepileptic drug hypersensitivity syndrome (AHS). Conclusions: The main features of severe cutaneous drug reactions, such as dosage, onset, and concomitant VPA use, do not differ in patients with LTG-induced SJS or. SJS or may also be the cutaneous manifestations of LTG-induced AHS. Further epidemiologic studies are needed to identify the incidence of severe LTG-induced cutaneous adverse reactions and the relative risk compared with other AEDs. Key Words: Lamotrigine-Antiepileptic drugs-stevens-johnson syndrome-toxic epidermal necrolysis-hypersensitivity. Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (, Lye11 syndrome) are severe albeit rare adverse drug reactions (ADRs) to several antiepileptic drugs (AEDs) such as phenytoin (PHT), carbamazepine (CBZ), phenobarbital (PB), and valproate (VPA) (1-6). A recent cohort study indicated that the risks for severe cutaneous reactions in first-time users of PHT and CBZ are 9110,000 and 6.2/10,000, respectively (7). Severe cutaneous adverse reactions (SCARS) such as erythema multiforme (EM), SJS, or may also represent the cutaneous manifestation of the antiepileptic drug hypersensitivity syndrome (AHS), which is defined by the triad of fever, a cutaneous reaction, and internal organ involvement (3,8). Lamotrigine (LTG), a phenyltriazine, is chemically unrelated to existing AEDs (9). It is extensively metabolized, predominantly by N-glucuronidation, whereas only minor fractions undergo N-oxidation and N-methylation Address correspondence and reprint requests to Dr. N. H. Shear at Division of Clinical Pharmacology, Room E-240, Sunnybrook Health Science Centre, 2075 Bayview Avenue, Toronto, ON, Canada M4N 3M5. (10,ll). It has a wide range of efficacy for partial ind generalized seizures (12) and is as effective as CBZ and PHT when used as monotherapy in newly diagnosed epilepsy (13,). LTG is being investigated for variety of additional indications, such as bipolar disorders (19, cocaine abuse (16). trigeminal neuralgia (17), and postoperative analgesia (1 8). As with other AEDs, most of the ADRs attributed to LTG therapy are related to the central nervous system and include headache, nausea, vomiting, diplopia, and ataxia (9,19). Rash of any type occurred in 10% of subjects receiving LTG in controlled trials (9). The incidence of rash appears to increase with the magnitude of the initial dose and the subsequent rate of dose escalation (19), and in patients receiving concomitant therapy with VPA (19,20). Occasionally the cutaneous reaction is more severe and may progress to desquamation with mucous membrane involvement (1 0). The product information for lamotrigine states that the incidence of severe rashes (i.e., SJS or ) is approximately 1/1,000 in adults and /100 in children (cited in ref. 21). This study systematically reviewed and analyzed spontaneously reported cases (published and unpub- s22

2 LAMOTRIGINE-INDUCED CUTANEOUS ADRs S23 lished) of LTG-associated SJS or to characterize the typical aspects of these reactions in association with LTG therapy. METHODS A comprehensive MEDLINE search (covering the period from January 1985 through April 1998) of the English and non-english literature was conducted to identify published reports of SCARS (defined as SJS or ) associated with lamotrigine. Search terms included the text word lamotrigine and the MeSH terms triazinedadverse effects, erythema multiforme, Stevens Johnson syndrome, and toxic epidermal necrolysis. The reference lists of all pertinent articles were reviewed to identify any additional cases that might have been missed in the computer search. Published reports were included if the reported reaction was diagnosed as SJS or and if the causality of the reaction was assessed as either definite, probable, or possible. Causality assessment was done by consensus by two rates using the validated ADR probability scale of Naranjo et al. (22). Additional information was requested from the Uppsala Monitoring Centre of the World Health Organization (WHO), Uppsala, Sweden, including all cases spontaneously reported to the WFO by the countries agreeing to the release of data (27 of 49 countries participating in the program). Data included information up to the second quarter of Cases were identified using the search terms Stevens Johnson syndrome and epidermal necrolysis (according to the Adverse Reaction Terminology of the WHO Collaborating Centre for International Drug Monitoring). WHO cases were included if the causal relationship was classified as either certain, probable, or possible according to the WHO causality assessment. In cases for which causality assessment was not available from the WHO data, it was assessed using the ADR probability scale (22) if sufficient information was available. These cases were included if the causality was classified at least as possible. Cases for which the necessary information for causality assessment was lacking were excluded. Possible duplicate reports between published and WHO cases were identified on the basis of demographics and case descriptions. When overlap occurred, the reports were attributed to the published literature. Statistical Analysis Cases were grouped according to their cutaneous diagnosis (SJS vs. ). The Wilcoxon rank-sum test was used for the analysis of quantitative data. Nominal comparisons were made using Fisher s exact test. Data are given as means k SD, medians, and ranges. A two-sided value of p < 0.05 was considered to indicate statistical significance. RESULTS We identified 57 reports that met our inclusion criteria, of which 13 cases (23%) were published (21,23-32). In the majority, the patients diagnosis was SJS [75% (43/57)]. Overall, 43 (75%) of the reactions were classified as possible, 11 (19%) as probable, and three (5%) as definite. Patients in the group were significantly older than those with a diagnosis of SJS (Table 1). Only 7% of the patients were younger than 18 years, whereas more than half of the patients in the SJS group were younger than 18 years. There was no difference in gender distribution, LTG dosage at onset, time to onset, and number of drugs concomitantly used at onset of the reaction (Table 1). The percentage of cases with concomitant valproate therapy was higher in the patients with a diagnosis of SJS (74 and 64%, respectively), but this difference was not significant. Concomitant use of other aromatic AEDs was similar in both groups (Table 1). Published Cases Of the 13 published cases, there were four reports (31%) of patients with a diagnosis of SJS (23-26), and nine (69%) of (21,27-30,32). One SJS cases was published twice (26,33), but we only included the more detailed one (26) for further analysis. Two fatal reactions were reported, one in a patient with and multiple TABLE 1. Comparison of 57 published and unpublished cases with hmotrigine-associated Stevens-Johnson syndrome or toxic epidermal necrolysis Age (years) mean f SD range 48 years, n (%) Female, n (96) Daily LTG dose (mg) at onset Mean f SD Time to onset (days) Mean * SD Total number of drugs at onset Mean f SD Concomitant use of VPA, n (%) Concomitant use of aromatic AEDs, n (%) SJS (n = 43) (n = ) pvalue 21 f (52) 28 (65) 75 f f f (74) 10 (23) 31 * (7) 6 (46) 134f f f (64) 2 () AEDs, antiepileptic drugs; LTG, lamotrigine; SJS, Stevens-Johnson syndrome;, toxic epidermal necrolysis;, not significant; VPA, valproate.

3 S24 R. G. SCHLIENGER ET AL. organ involvement (27) and one in another patient with a diagnosis of (32). Three of the published cases showed the features of the AHS with as the cutaneous reaction, plus fever and single or multisystem internal organ involvement (21,27,28). The liver was affected in all three cases. Multiorgan involvement with additional renal and toxic or reactive hematologic abnormalities (eosinophilia, neutropenia, thrombocytopenia) was reported in two (27,28). A detailed overview of the published cases is shown in Table 2. WHO Cases We received the information for a total of 79 cases from the WHO (63 with SJS, with, two cases with overlapping SJS/) with a severe cutaneous reaction associated with lamotrigine. Of these, 44 (56%) met our inclusion criteria (39 cases with SJS, five cases with ). Thirty-one cases were excluded either because the causal relationship was classified as unlikely according to the WHO classification or because the information was insufficient to assess causality using the ADR probability scale. Four of the WHO reports were published and were included in the series of published cases. Among the cases for which information for out- come was available, there was one fatal reaction in a patient with SJS. DISCUSSION In this series of 57 cases with LTG-induced SCARS, we found a significant difference in the age distribution. Patients in the SJS group were significantly younger and the percentage of patients who were <18 years of age was significantly higher in the SJS group. Whether this merely reflects a reporting bias or whether in fact there is an age-related difference in the risk for developing LTGinduced SJS or remains unclear. In a survey of the epidemiologic data for and SJS over a 5-year period done by Schopf et al. (34), the authors found an even more pronounced age difference in patients presenting with vs. patients with SJS (mean age 63 years and 25 years, respectively) (34). However, it must be taken into consideration that the study by Schopf et al. (34) was not focused on drug-induced or SJS (almost onethird of the SJS cases in their survey had no drug involvement). In addition, the survey was not looking at a population with a specific underlying disease, such as seizure disorders, as in our study. The typical interval TABLE 2. Characteristics of 12 published cases with Stevens-Johnson syndrome or toxic epidemzal necrolysis with lamotrigine therapy LTG dose Days to Cutaneous Reference AgeJsex at onset onset Concomitant drugs manifestation 40M 25 mg every 2nd day initally, 21 Valproate (2,000 mg/day), SJS then 25 mg/day clonazepam (2 mg/day) 30M 25 mg every 2nd day, 35 Valproate (2,500 mg/day) SJS increased to 50 mg/day 12/F Initially 50 mglday, increased 23 None SJS to 300 mg/day over the next 3 weeks 18/F Initially 25 mg every 2nd day, 30 Valproate (1,500 mg/day), SJS increased to 100 mg/day clonazepam (6 mg/day) over the next 4 weeks 24/F Initially 50 mglday, after None days 200 mglday 29/F 25 mg every 2nd day, then Valproate (as the cutaneous mglday manifestations of AHS) Duval et al (25) Sachs et al (26) Dooley et al. (24) Campistol et al (23) Fogh and Mai 1997 (30) Sullivan and Watson 1996 (28) Chaffin and Davis 1997 (21) Wadelius et al. Wadelius et al. Wadelius, et al. Sterker et al (27) Vukelic et al (31) Page, et al (32) 74M 22E 21/F 22M 56M 2M 54M Initially 50 mg/day for 3 days, then 100 mg/day NR 25 mg every other day 75 mg/day 200 mg/day Initially 1 mg/kg/day, after 10 days increased to 2 mg/kg/d Initially 50 mg/day, increased within 1 week to 100 mg/day 12 3 months 28 Carbamazepine Valproate (1,300 mg/day ) Valproate (720 mglday), levothyroxine, medroxyprogesterone Valproate (2.400 mglday) None Valproate Valproate (1,500 mglday), allopurinol, captopril (as the cutaneous manifestation of AHS) (as the cutaneous manifestation of AHS) AHS, antiepileptic drug hypersensitivity syndrome: LTG, lamotrigine: SJS, Stevens-Johnson syndrome:, toxic epidermal necrolysis. Epilepsia, Vol. 39, Suppl. 7, 1998

4 LAMOTRIGINE-INDUCED CUTANEOUS ADRs s25 from beginning of a drug therapy to the onset of reaction in patients with SJS or is usually 1-3 weeks (23). For PHT, CBZ, PB, and VPA, the risk for development of SJS or is greatest in the first 2 months of treatment, although some increased risk can persist among long-term users of PB and VPA (4). The median time to onset of 17 days, as found in our study, is in accordance with that. In two of the WHO cases with a diagnosis of SJS, the onset was relatively short (2 and 4 days). In addition, there was one WHO case with a diagnosis of, who had a very long time to onset of 130 days. However, it is unknown on the basis of the available data whether in the two SJS cases a previous exposure and sensitization to LTG occurred, which might explain such a short interval to onset. On the other hand, we cannot determine from our data whether some increased risk for SCARs may also persist among long-term users of LTG, as was shown among users of PB or VPA. The rate of 64 and 74%, respectively, of concomitant VPA was higher than the rate of LTG users as a whole using concomitant VPA therapy, which is reported to be approximately 40% (28). Concomitant use of VPA with LTG significantly increases the risk for development of adverse cutaneous reactions (20). It is known that VPA interacts with LTG metabolism, leading to a reduced total clearance and therefore to an increased elimination half-life of LTG (35,36) and thuf resulting in higher serum concentrations. On the other hand, VPA itself carries a significant risk for SCARs (4). In addition, the use of other AEDs known to produce severe cutaneous reactions, such as the older aromatic AEDs PHT, PB, primidone (PRM), and CBZ, was as high as % and 23%, respectively. Although a causal relationship with these agents cannot definitely be excluded, in the majority of the cases the temporal course made an association unlikely. In some of the unpublished cases, concomitant use of other drugs, such as certain cephalosporins or AIDs that are well known to induce severe cutaneous reactions (4), was also reported. As with the concomitant use of other AEDs, a causal relationship with these drugs cannot be excluded because relevant temporal information was often missing. Furthermore, because some rel: evant information for the WHO cases was sometimes not available, other possible non-drug-related causes also cannot be excluded with certainty. In at least three cases included in our study, SJS or appeared to be the cutaneous manifestation of the AHS. We suggest that in patients with SCARs to LTG, close clinical and laboratory monitoring and follow-up for other signs common with the AHS, such as abnormal hepatic, renal, hematologic, or thyroid function (8,37, 38), is advisable. It has been shown in vivo and in vitro that aromatic AEDs, such as PHT, PB, and CBZ, have a risk of 70-80% of crossreactivity in patients with hypersensitivity or allergic reactions (8). Whether this risk is comparably high with LTG, which shares an aromatic ring structure but otherwise has no chemical similarity to older aromatic AEDs, is unclear. In vitro testing [e.g., with the lymphocyte toxicity assay (S)] might be useful in clarifying this aspect. Spontaneous reports from postmarketing surveillance may be subject to several limitations. Because denominator data (number of exposed patients or person-time of exposure) are not available, a cumulative incidence or an incidence rate cannot be estimated. In addition, voluntary reporting may underrepresent the true incidence of ADRs for a variety of reasons [e.g., ignorance of reporting requirements, fear of involvement in litigation, guilt due to patient harm, lack of time, lethargy (39)l. Reports may be biased because of the perceived severity of the adverse effect, media bias, or physician awareness of a possible drug association. In our study, this is partly illustrated by the fact that cases with a diagnosis of comprised less than one-quarter of all reports included in our analysis but represented almost two-thirds of the published reports. Furthermore, because precise diagnostic boundaries between SJS and have not been clearly established and it is therefore sometimes difficult to distinguish the two disorders (5), we cannot rule out the possibility that the diagnosis was inappropriate in some cases. In conclusion, we did not find significant differences in terms of LTG dosage and interval to onset of a reaction between LTG-induced SJS or, whereas patients with a diagnosis of SJS were significantly younger. The reactions were associated with a high rate of concomitant VPA. Because severe cutaneous reactions to AEDs &e associated with high morbidity, mortality (3), and also with considerable direct medical costs (40), further epidemiologic studies are needed to estimate the relative risk for severe cutaneous reactions to LTG in comparison with other AEDs, as well as the possibly increased risk with concomitant VPA therapy. In vitro testing, e.g., with the lymphocyte toxicity assay, would be helpful in studying the potential for crossreactivity between LTG and older aromatic AEDs. Acknowledgments: We thank Ms. Helena Fucik of the Uppsala Monitoring Centre of the World Health Organization (WHO), Uppsala, Sweden. The information obtained from the WHO is not homogeneous, at least with respect to origin or likelihood that the pharmaceutical product caused the adverse reaction. The opinions expressed in this article are those of the authors and do not necessarily represent the opinion of the WHO. Raymond G. Schlienger was supported by a grant from the Swiss National Science Foundation. Presented in part at the 6th Canadian Pharmacoepidemiology Forum, April 27, 1998, Ottawa, Ontario, Canada REFERENCES 1. Chan H-L, Stem RS, Arndt KA, et al. the incidence of erythema multiforme, Stevens-Johnsons syndrome, and toxic epidermal necrolysis. Arch Demtol 1990;126:43-7.

5 S26 R. G. SCHLIENGER ET AL. 2. Guillaume J-C, Roujeau J-C, Revuz J, Penso D, Touraine R. The culprit drugs in 87 cases of toxic epidermal necrolysis (Lyell's syndrome). Arch Dematol 1987;123: Chang D, Shear N. Cutaneous reactions to anticonvulsants. Semin Neurol 1992;12: Roujeau J-C, Kelly JP, Naldi L, et al. Medication use and the risk of Stevens Johnson syndrome or toxic epidermal necrolysis. N Engl J Med 1995;333: Roujeau JC, Stern RS. Severe adverse cutaneous reactions to drugs. N Engl J Med 1994;331: Roujeau J-C, Guillaume J-C, Fabre J-P, Penso D, Flichet M-L, Girre J-P. Toxic epidermal necrolysis (Lyell syndrome). Incidence and drug etiology in France, Arch Dematol 1990;126: 17 "0 J IYL. 7 Tennis P, Stem RS. Risk of serious cutaneous disorders after initiation of use of phenytoin, carbamazepine, or sodium valproate: a record linkage study. Neurology 1997;49: Shear NH, Spielberg SP. Anticonvulsant hypersensitivity syndrome: In vitro assessment of risk. J Clin Invest 1988;82:182& Messenheimer JA. Lamotrigine. Epilepsia 1995;36(Suppl 2):S Fitton A, Goa KL. Lamotrigine. An update of its pharmacology and therapeutic use in epilepsy. Drugs 1995;50: Rambeck B, Wolf P. Lamotrigine clinical pharmacokinetics. Clin fhanacokinet 1993;25: Kilpatrick ES, Forrest G, Brodie MJ. Concentration-effect and concentration-toxicity relations with lamotrigine: a prospective study. Epilepsia 1996;37: Brodie MJ, Richens A, Yuen AWC, and the UK Lamotriginel Carbamazepine Monotherapy Trial Group. Double-blind comparison of lamotrigine and carbamazepine in newly diagnosed epilepsy. Lancet 1995;345: Steiner TJ, Silveira C, Yuen AWC, and the North Thames Lamictal Study Group. Comparison of lamotrigine (Lamictal) and phenytoin in newly diagnosed epilepsy. Ep'lepsia 1994;35(suppl 7): Spom J, Sachs G. The anticonvulsant lamotrigine in treatmentresistant manic-depressive illness. J Clin fsychopharmucol 1997; 17: Margolin A, Avants SK, DePhilippis D, Kosten TR. A preliminary investigation of lamotrigine for cocaine abuse in HIV-seropositive patients. Am J Drug Alcohol Abuse 1998;24: Zakrzewska JM, Chaudhry Z, Nurmikko TJ, Patton DW, Mullens EL. Lamotrigine (Lamictal) in refractory trigeminal neuralgia: results from a double-blind placebo controlled crossover trial. fain 1997;73: Bonicalzi V, Canavero S, Cerutti F, Piazza M, Clemente M, Chi0 A. Lamotrigine reduces total postoperative analgesic requirement: a randomized double-blind, placebo-controlled pilot study. Surgery 1997;122: Richens A. Safety of lamotrigine. Epilepsia 1994;35(Suppl 5):S Li LM, Russo M, O'Donoghue MF, Duncan JS, Sander JW. Allergic skin rash with lamotrigine and concomitant valproate therapy: evidence for an increased risk. Arq Neuropsiquiatr 1996; 54: Chaffin JJ, Davis SM. Suspected lamotrigine-induced toxic epidermal necrolysis. Ann fharmacother 1997;31 : Naranjo CA, Busto U, Sellers EM, et al. A method for estimating the probability of adverse drug reactions. Clin fharmucol Ther 1981;30: Campistol J, Geli M, Llistosella E, Molins J, Llobet M. Sindrome de Stevens-Johnson tras las introduccion de lamotrigina. Rev Neurol 1995;23: Dooley J, Camfield P, Gordon K, Camfield C, Wirrell E, Smith E. Lamotrigine-induced rash in children. Neurology 1996;46: Duval X, Chosidow 0, Semah F, Lipsker D, Franc& C, Herson S. Comment on: lamotrigine versus carbamazepine in epilepsy. Lancet 1995;345: Sachs E, Ronnau AC, von Schmiedeberg S, Ruzicka T, Gleichmann E, Schuppe H-C. Lamotrigine-induced Stevens-Johnsons syndrome: demonstration of specific lymphocyte reactivity in vitro. Dermatology 1997;195:6M. 27. Sterker M, Berrouschot J, Schneider D. Fatal course of toxic epidermal necrolysis under treatment with lamotrigine. Int J Clin Pharmacol Ther 1995;33: Sullivan JR, Watson A. Lamotrigine-induced toxic epidermal necrolysis treated with intravenous cyclosporin: a discussion of pathogenesis and immunosuppressive management. Australas J Dermatol 1996;37: Wadelius M, Karlsson T, Wadelius C, Rane A. Lamotrigine and toxic epidermal necrolysis [Letter]. Lancet 1996;348: Fogh K, Mai J. Toxic epidermal necrolysis after treatment with lamotrigine (LamictaF). Seizure 1997;6: Vukelic D, Bozinovic D, Tesovic G, et al. Lamotrigine and toxic epidermai necrolysis. Dermatology 1997; 195: Page RL, O'Neil MG, Yarbrough DR, Conradi S. Fatal toxic epidermal necrolysis related to lamotrigine administration. fharmacotherapy 1998;18: Sachs B, Ronnau AC, Ruzicka T, Gleichmann E, Schuppe H-C. Lamotrigine and toxic epidermal necrolysis [Letter]. Lancet 1996; 348: Schopf E, Stuhmer A, Rzany B, Victor N, Zentgraf R, Kapp JF. Toxic epidermal necrolysis and Stevens-Johnson syndrome. An epidemiologic study from West Germany. Arch Dermatol 1991; 127: Anderson GD, Yau MK, Gidal BE, et al. Bidirectional interaction of valproate and lamotrigine in healthy subjects. Clin fharmaool Ther 1996; Yuen AWC, Land G, Weatherley BC, Peck AW. Sodium valproate acutely inhibits lamotrigine metabolism. Br J Clin fharmacol 1992;33: Vittorio CC, Muglia JJ. Anticonvulsant hypersensitivity syndrome. Arch Intern Med 1995;155: Gupta A, Egg0 MC, Uetrecht JP, et al. Drug-induced hypothyroidism: the thyroid as a target organ in hypersensitivity reactions to anticonvulsants and sulfonamides. Clin Phurmacol Ther 1992;51: 5M Smith Rogers A, Israel E, Smith CR. et al. Physician knowledge, attitudes, and behavior related to reporting adverse drug events. Arch Intern Med 1988;8:159& Schlienger RG, Oh PI, Knowles SR, Shear NH. The economic impact of serious adverse drug reactions (ADRs) to anticonvulsants [Abstract]. Clin fharmacol Ther 1997;61:138.