- Email: [email protected]
Oligohydrosis and fever in pediatric patients treated with zonisamide
Oligohydrosis and Fever in Pediatric Patients Treated with Zonisamide James F. Knudsen, PhD, MD*, Lopa R. Thambi, PharmD*, Leonard P. Kapcala, MD‡, and Judith A. Racoosin, MD, MPH* Zonisamide is an antiepileptic drug developed and first marketed in Japan in 1989. Cases of oligohydrosis, characterized by deficient production and secretion of sweat, were reported in children treated with zonisamide in Japan during development and in the postmarketing period. Zonisamide was approved in the United States in March 2000 for adjunctive treatment of partial seizures in adults. Searching the Food and Drug Administration’s Adverse Events Reporting System, we identified six domestic cases of zonisamideassociated oligohydrosis and/or fever, all in patients < 18 years of age. The calculated reporting rate was 13 cases per 10,000 pediatric-years of exposure, approximately 10 –fold the reporting rate in Japan. A possible risk factor for the development of oligohydrosis in these cases was pediatric age, leading to exposure to elevated zonisamide blood levels relative to patient size. Although the mechanism for zonisamide-associated oligohydrosis has not been fully elucidated, the drug may mediate its effect on eccrine sweat glands by inhibiting carbonic anhydrase, thereby influencing pH dynamics, hydrogen ion concentration, and available calcium transients. Awareness of zonisamide-associated oligohydrosis may prevent morbidity, especially in the pediatric population. © 2003 by Elsevier Inc. All rights reserved. Knudsen JF, Thambi LR, Kapcala LP, Racoosin JA. Oligohydrosis and fever in pediatric patients treated with zonisamide. Pediatr Neurol 2003;28:184-189.
seizure disorders in adult and pediatric patients. Zonisamide was approved for use in the United States in March 2000. The drug is indicated for adjunctive therapy in the treatment of partial epilepsy in adults only. Zonisamide is chemically classified as a sulfonamide and is characterized in vitro as having multiple mechanisms of action, including the ability to inhibit carbonic anhydrase [1,2]. Carbonic anhydrase inhibitors, such as acetazolamide, have been associated with such diverse systemic adverse events as metabolic acidosis, nephrolithiasis, hyperammonemia, and increased renal excretion of sodium bicarbonate [3,4]. Oligohydrosis, a potentially serious adverse event characterized by deficient formation and secretion of sweat, was initially reported in association with zonisamide when Japanese children treated with this drug manifested this condition during the exceptionally hot summer of 1994 [5-8]. To date, there have been no reports in the medical literature describing oligohydrosis in U.S. patients treated with zonisamide. The objectives of our study include the following: (1) documentation of U.S. reports of oligohydrosis in patients treated with zonisamide and identification of risk factors for this adverse event; (2) calculation of the U.S. reporting rate of oligohydrosis for patients treated with zonisamide; (3) systematic review of the medical literature for reports of oligohydrosis associated with exposure to any drug; and (4) review of potential mechanisms for drug-induced hidropoietic dysfunction.
Methods Patient data collection and tabulation
Introduction Zonisamide is an antiepileptic drug (AED) developed in Japan in the late 1970s, and first marketed there in 1989. The drug is used in clinical practice in Japan to treat
From the *Center for Drug Evaluation and Research, Division of Neuropharmacological Drug Products and the†Center for Drug Evaluation and Research, Office of Drug Safety, Food and Drug Administration Rockville, Maryland.
184
PEDIATRIC NEUROLOGY
Vol. 28 No. 3
Through the MedWatch program, the U.S. Food and Drug Administration (FDA) receives adverse drug event (ADE) reports and product quality concerns for drugs, biologic devices, or medical devices used by humans. Reports from manufacturers associated with their product(s) are required by law, whereas reports from health-care professionals and
Communications should be addressed to: Dr. Knudsen; Food and Drug Administration; Center for Drug Evaluation and Research; Division of Neuropharmacological Drug Products; 5600 Fishers Lane, HFD-120; Rockville, MD, 20857 Received June 6, 2002; accepted August 21, 2002.
© 2003 by Elsevier Inc. All rights reserved. doi:10.1016/S0887-8994(02)00511-8 ● 0887-8994/03/$—see front matter
consumers are voluntary. These reports are collected into an electronic database, referred to as the Adverse Events Reporting System (AERS). We searched AERS for United States reports of oligohydrosis with or without fever reported in association with zonisamide use from the time of approval in the United States (March 2000) through December 31, 2001 [9]. We used a broad search strategy to identify any report potentially related to oligohydrosis with or without fever. The search was followed by a manual review of each report to exclude duplicate and unrelated reports. In some cases that lacked specific detail, one of us obtained additional information through direct contact with the reporter. We summarized patient demographics, concomitant medications, zonisamide dose, and duration of therapy from the identified cases.
spiration. Bibliographies of review articles were searched for additional information.
Results Our search of AERS identified six cases of oligohydrosis and/or fever associated with zonisamide use. The clinical diagnosis was established by the health care provider who submitted the adverse event report.
Demographics Reporting rate Within the current passive surveillance system for reporting suspected drug-related adverse events to the FDA, it is not possible to calculate the absolute incidence of a drug-related side effect because we do not know the actual number of patients who experienced the adverse event (numerator) nor the actual number of patients exposed to the drug and duration of use (denominator). However, we are able to estimate the incidence with a reporting rate. A reporting rate is calculated using the number of cases of an adverse event reported to FDA in the numerator, and dividing by an estimate of drug use represented as person-years in the denominator. One person-year is equal to one patient taking a drug for one-year, or two people each taking the drug for six months, or four people each taking the drug for three months, etc. To approximate the extent of zonisamide use, we obtained an estimate of the total number of prescriptions dispensed in the United States for zonisamide from IMS Health’s National Prescription Audit Plus™ (NPA Plus™) [10]. This computerized database provides estimates of the total number of prescriptions dispensed by retail pharmacies. In order to estimate the proportion of zonisamide users in the pediatric age group (0-16 years old), we obtained demographic data from IMS Health’s National Disease & Therapeutic Index Audit™ (NDTI), an ongoing survey of treatment patterns, reported as appearances, at office-based medical practices. An appearance from the NDTI Audit is the mention of a drug in a single patient visit. A drug appearance can be associated with more than one diagnosis in the same patient visit. For example, if a doctor treats two conditions in one patient with the same product, this counts as two drug uses but only one drug appearance. Therefore, drug appearances may be equal to or less than uses. Information retrieved from NDTI included sex and age of patients, as well as mean duration of prescriptions. The average prescription length was 30 days. The method for calculating the reporting rate for oligohydrosis follows below: 1.
Estimation of total person-years of exposure to zonisamide: (Total # of zonisamide prescriptions dispensed by retail pharmacy) ⫻ (Average prescription length) ⫼ 365 days. 2. Calculation of the proportion of person-year exposure in the pediatric age group (“pediatric-years exposure”): (Proportion of drug use in children 0-16 from NDTI) ⫻ (Total person-years calculated in step 1). 3. Calculation of reporting rate using the formula: (Number of U.S. cases from AERS) ⫼ (Pediatric-years exposure).
Searching and evaluating the literature A search was performed using MEDLINE, EMBASE, TOXLINE, DERWENT DRUG FILE, PHARM-LINE and BIOSIS through December 2001. The medical subject heading, title, abstract, and key words were searched using the following terms: zonisamide, carbonic anhydrase inhibitor, acetazolamide, benzolamide, dichlorphenamide, ethoxzolamide, methazolamide, topiramate, oligohydrosis, hypohidrosis, dyshidrosis, anhidrosis, hidropoeitic disorders, overheating, body temperature regulation, sweat glands, decreased/reduced sweating/per-
Demographic information and a clinical summary of each of the six cases are shown in Table 1. There were no cases identified in patients older than 18 years of age. Three cases (case numbers 1, 2, and 4) were characterized by both oligohydrosis and fever. In one patient with fever (case number 6) the report did not specify whether the patient had oligohydrosis; however, the health-care provider stated that an evaluation for the etiology of the fever was unrevealing for any cause other than zonisamide. Although oligohydrosis was not reported, we include this case in the series because the patient went on to develop heat stroke requiring intensive care, similar to cases in the Japanese literature [6]. We also note that this patient was taking concomitant topiramate, another AED with reports of associated oligohydrosis. However, the patient had been on a stable dose of topiramate, the heat stroke episode occurred in temporal association with initiation of zonisamide, and the patient recovered following discontinuation of zonisamide with ongoing topiramate therapy. In addition to this case, it was reported in cases 1, 2, and 4 that other etiologies for fever were investigated with no conclusive findings. Five of the six patients were in the pediatric age range, with the sixth being 18 years old. The median age was 4.5 years. All of the patients were taking zonisamide for a seizure disorder. Two patients (case numbers 5 and 6) were taking other antiepileptic drugs in addition to zonisamide. It was not always possible to assign a precise time for the onset of oligohydrosis (with or without fevers) following initiation of zonisamide treatment. In these cases the time to detection was noted to be within a time frame for which the adverse event had been clearly established. Using these data, the time until detection of these adverse events ranged between 7 to 225 days after the start of zonisamide treatment. The median time until detection was 62.5 days. Patients experiencing oligohydrosis with or without fever during zonisamide treatment had received a wide range of doses. The daily doses ranged between l00 – 600 mg/day and corresponded to 6 –22 mg/kg/day. The recommended daily dose in children, according to Japanese literature, is 4 – 8 mg/kg. In three of the six cases (case numbers 1, 2, and 6) the patients with oligohydrosis and fever were reported to reside in southern states. Five of six
Knudsen et al: Oligohydrosis and Zonisamide 185
Table 1.
Clinical summary of U.S. cases of oligohydrosis/fever in patients treated with zonisamide
Case number
1
2
3
4 M 5 years Potassium phosphate, sodium bicarbonate, pyridoxine, calcium, levocarnitine Refractory multifocal onset tonic seizures 300 mg (22) Max 39 °C ⫹ 225 Fevers and oligohydrosis resolved after ZNS decreased to 200mg/day, ZNS was eventually d/c
Sex Age Concomitant medications
M 6 months Ranitidine, pyridoxine
M 10 months None reported
F 4 years None reported
ZNS indication
Infantile spasms
Abnormal EEG
Abnormal EEG
Dose* (mg/kg) Fever Oligohydrosis Onset** (days) Outcome
150 mg (17) Max 41 °C ⫹ 7 Hospitalized; fevers and decreased sweating improved but did not completely recover after ZNS d/c
100 mg (12) Max 41 °C ⫹ 42 Hospitalized; fevers and oligohydrosis resolved 11 and 9 months, respectively, after ZNS d/c
100 mg (6) None reported ⫹ ⱕ11 Oligohydrosis resolved after ZNS d/c
5
6
M 13 years Lamotrigine, levetiracetam, fluticasone
M 18 years Lamotrigine, levetiracetam, topiramate, propranolol, clonidine.
Unspecified seizure disorder
Partial seizure disorder
600 mg (9) None reported ⫹ 83 Oligohydrosis resolved after ZNS decreased, and was eventually d/c
600 mg (7) Max 42 °C none reported 83 Hospitalized; complete recovery; heat stroke syndrome and complications resolved with ICU treatment and ZNS d/c
* Dose reported as total daily dose and (mg/kg) ** Approximate duration of time from ZNS initiation to detection of symptoms (days) Abbreviations: d/c ⫽ Discontinued EEG ⫽ Electroencephalogram F ⫽ Female kg ⫽ Kilograms
M max mg ZNS
⫽ ⫽ ⫽ ⫽
Male Maximum Milligrams Zonisamide
cases (case numbers 1, 3, 4, 5, and 6) of oligohydrosis and fever were reported during summer months. Three patients required hospitalization for their reported adverse event. Five patients were considered to have completely recovered from oligohydrosis after the discontinuation of zonisamide. The remaining patient (case number 1) was considered to reveal only partial recovery after zonisamide therapy was stopped. Subsequently, this patient was diagnosed with a congenital mitochondrial metabolic disorder associated with sweating dysfunction.
younger. The reporting rate for zonisamide-associated oligohydrosis with or without fever is presented in Table 3 for patients 16 years of age and younger, as well as those 18 years of age and younger. The reporting rate for each of these age groups was similar, about 13 cases per 10,000 pediatric-years.
Table 2. Projected number of total U.S. drug appearances, stratified by age range in pediatric patients treated with zonisamide*
Calculation of U.S. reporting rate for oligohydrosis IMS Health’s NPA Plus™ data for prescription use revealed that approximately 212,000 prescriptions had been dispensed for zonisamide in the United States since approval on March 27, 2000. During the year 2000, there were 28,000 prescriptions written, and during 2001, the number increased to 184,000. Table 2 presents the breakdown of zonisamide prescriptions by age from NDTI™. Although zonisamide is not approved for pediatric use in the U.S., the IMS data showed that 22% of zonisamide appearances in 2001 were for patients aged 16 years and
186
PEDIATRIC NEUROLOGY
Vol. 28 No. 3
Age Range (years) ⬍1-2 3-6 7-9 10-12 13-16 Total
Total (ⴛ 1000)
% of total drug appearances
4 9 3 3 9 28
2.8 7.3 2.3 2.2 7.2 22
* Distributed by calendar year—March 2000 through December 2001. ** Source: IMS Health; National Disease & Therapeutic Index AuditTM
Table 3. Estimated U.S. reporting rates of oligohydrosis and/or fevers associated with zonisamide* U.S. Experience <16 years old Number of cases received by FDA Zonisamide usage (number of pediatric years of exposure) Reporting rate†
U.S. Experience <18 years old
5
6
3799
4565
13 cases/10,000 person-years
13 cases/10,000 person-years
* U.S. usage for March 2000-December 2001, pediatric years of exposure estimated from prescription use assuming each prescription is for 30 days. † Reporting rate: number of U.S. case reports received ⫼ pediatric years of use
Literature search The search strategies identified 314 citations. After a review of their titles, key words, and abstracts, we found no U.S. cases of oligohydrosis and/or fever in association with the use of zonisamide. We identified Japanese literature reports of oligohydrosis, fever, and heat stroke during exposure to zonisamide. In addition, we identified a report by Arcas et al. [11] that consisted of three cases of oligohydrosis and fever in pediatric patients treated with the recently marketed AED topiramate. Discussion We describe the first U.S. reports of oligohydrosis with or without accompanying fever in pediatric patients treated with zonisamide. Health-care practitioners should consider this potentially serious event as being possibly drug-induced. Although elevated ambient temperature is not a risk factor for development of oligohydrosis, it increases the likelihood of detecting the condition, and creates the setting where oligohydrosis can lead to serious sequelae. We calculated a U.S. reporting rate for oligohydrosis with or without fever of 13 cases/10,000 pediatric years of exposure (Table 3). To place this rate in perspective, 38 cases of oligohydrosis with or without fever were reported in Japan during the first 11 years of marketing, for an estimated reporting rate of 1 case per 10,000 pediatricyears of exposure [12]. Despite the 10-fold difference in these rates, both are underestimates of the true incidence because of the widely recognized underreporting of adverse events in a passive surveillance program [13]. One study estimated that between ⬍1% to 10% of serious adverse events are reported to the FDA [13]. If the index of suspicion for oligohydrosis in pediatric patients is low, then correspondingly, recognition and reporting will be low. Recognition may occur only when the patient develops fever for which an infectious or neoplastic source cannot be identified [14]. Mild or moderate decrements in normal sweating may not be recognized. Tanaka et al. [14] characterized the presentation of oligohydrosis in Japanese children treated with zonisamide as subtle and missed
frequently, with recognition occurring when the patient developed hyperthermia. Based on the U.S. cases, as well as those from Japan [5– 8], pediatric age appears to be a risk factor for zonisamide-associated oligohydrosis. The apparent increased risk of oligohydrosis in the pediatric age group may be related to the dose and resulting blood levels of zonisamide and its metabolites. Dosing based on body weight was relatively higher in four of our six patients than the standard dose range (4 – 8 mg/kg/day) used in Japan. In one patient the dose was 22 mg/kg/day, nearly two times the maximal pediatric dose of 12 mg/kg/day recommended in Japan. Further supporting our suspicion that an elevated blood level of zonisamide may play a role in the development of oligohydrosis is the finding that two of the patients were taking lamotrigine concomitantly with zonisamide. Concomitant therapy of lamotrigine with zonisamide results in substantial elevations of serum levels of zonisamide and its metabolite [15]. Thus, these patients may have been exposed to inappropriately high blood levels of zonisamide. Additionally, Tanaka et al. [14] documented that dose levels of zonisamide were high for pediatric patients in all reported cases of dyshidrosis. In their study, perspiration returned after lowering the dose of zonisamide or stopping the drug. One more factor that may contribute to elevated zonisamide serum levels in pediatric patients is the alteration of renal drug clearance by zonisamide itself. According to the drug labeling, clinical trials in adults revealed that zonisamide is eliminated from the body by renal excretion and is associated with a statistically significant increase in serum creatinine and blood urea nitrogen compared with placebo-treated patients. The increase persists over time and is interpreted as an effect on glomerular filtration rate (GFR). The GFR returns to baseline within 2–3 weeks of drug discontinuation. Infants may be particularly sensitive to this effect of the drug given that they have a markedly reduced GFR, about 30% of adult normals when expressed on a surface area basis [16]. Of interest, we note that males comprised five of our six cases and three of three U.S. cases of oligohydrosis/fever
Knudsen et al: Oligohydrosis and Zonisamide 187
associated with topiramate treatment [11]. This observation raises the question of a gender predisposition for oligohydrosis/fever related to zonisamide and topiramate. However, our survey of the Japanese literature [17,18] did not reveal any suggestion of male gender as a risk factor for zonisamide-induced oligohydrosis/fever. Ultimately, the question of a gender predisposition would have to addressed in a controlled study, and can not be answered by a case series. Our present understanding of the pathophysiology of drug-induced oligohydrosis in pediatric patients is limited. However, the medical literature may suggest clues to possible mechanisms. Shimizu et al. [6] reported a reduced response to pilocarpine electrophoresis and acetylcholine stimulation in pediatric patients given zonisamide. Okumura et al. [19] used the subcutaneous acetylcholine stimulation test to detect patients at risk of oligohydrosis caused by zonisamide and concluded that dyshidrosis was due to a functional disorder of the sweat glands. Diminished sudomotor responsiveness and hypohidrosis have also been reported to occur in patients treated with topiramate, another recently marketed AED that inhibits carbonic anhydrase [11]. This report described three male pediatric patients who exhibited hypohidrosis/oligohydrosis, heat and exercise intolerance, and fever after 2-3 months of treatment (during summer heat). The assessment of the patients’ sudomotor function with pilocarpine iontophoresis showed reduced sweat production in each of the individuals. It was concluded that hypohidrosis was caused by a dysfunction of sweat glands because of complex interacting mechanisms triggered by topiramate. Carbonic anhydrase isozymes are located in secretory and transporting epithelia, including eccrine sweat glands [20]. Studies in animal models have demonstrated that carbonic anhydrase localized to sweat glands is important for the production of sweat during physical activity and especially at higher ambient temperatures [21]. Physiologic eccrine sweating depends on the presence of extracellular Ca2⫹ and its entry into the secretory cell, leading to stimulus secretion coupling and, ultimately, sweat production [22]. Calcium ionophores, carriers of calcium ions across the cell membrane, stimulate sweat secretion in various species [22]. The cumulative weight of 70 years of research on the enzyme carbonic anhydrase may provide the framework for conceptualizing how modulation of Ca2⫹ levels by fluxes in acid equivalents regulates secretion. Zonisamide inhibition of carbonic anhydrase could modify the acid— base (acid equivalents) and ionic (mainly Ca2⫹ availability) cellular environment of the eccrine sweat gland epithelial cells. The result would be a conformational change (alteration in membrane form) and a diminution in the responsiveness of eccrine sweat gland muscarinic receptors to cholinergic stimulation. In conclusion, our finding of infrequent oligohydrosis/ fever in children taking zonisamide is consistent with the Japanese literature describing this adverse event. Aware-
188
PEDIATRIC NEUROLOGY
Vol. 28 No. 3
ness of this uncommon drug-related adverse event may prevent unnecessary morbidity, and potentially simplify the evaluation and treatment of fever in children who are taking zonisamide. Although the mechanism for zonisamide-associated oligohydrosis/fever is not fully understood, the drug may in part mediate its effect on eccrine sweat glands by inhibiting carbonic anhydrase thereby influencing pH dynamics, H⫹ concentration, and available Ca2⫹ transients at the level of ion channels and receptors.
References [1] Masereel B, Rolin S, Abbate F, Scozzafava Asupuran CT. Carbonic anhydrase inhibitors: anticonvulsant sulfonamides incorporating valproyl and lipophilic moieties. J Med Chem 2002;45:312-20. [2] Nakatsuji K, Matsuno N, Nakamura N, Fujitani B, Ito T, Kadokawa T. General pharmacology of the novel antiepileptic compound zonisamide. Arzneim Forsch 1987;37:1131-6. [3] Maren TH. Carbonic anhydrase: Chemistry, physiology and inhibition. Physiol Rev 1967;47:595-781. [4] Swenson ER. Respiratory and renal roles of carbonic anhydrase in gas exchange and acid-base regulation. In: Chegwidden WR, Carter ND, Edwards TH, eds. The carbonic anhydrase new horizons. Basal: Birkha¨ user Verlag, 2000:281-341. [5] Okumura A, Hayakawa F, Kuno K, Watanabe K. Oligohydrosis caused by zonisamide. No To Hattatsu 1996;28:44-7. [6] Shimizu Y, Togo A, Eto M, et al. Oligohydrosis induced by zonisamide, clinical features of 7 cases and skin histology. J Japan Ped Society 1996;100:350-1. [7] Hosada N, Miura T, Takanashi S, Shirai H, Sunaoshi W, Takei K, Clinical experience of dyshidrosis induced by zonisamide. Kanagawa Seishin Igaku Kaishi 1996;46:121. [8] Hosada N, Miura H, Shirai H, Sunaoshi W, Takei K. Dyshidrosis by zonisamide. Japanese J Ped 1996;49(11):2445-9. [9] Brinker A, Beitz J. Use of a spontaneous adverse drug events database for identification of unanticipated drug benefits. Clin Pharmacol Ther 2002;71:99-102. [10] Rodriguez EM, Staffa JA, Graham DJ. The role of databases in drug postmarketing surveillance. Pharmacoepidemiol Drug Saf 2001;10: 407-10. [11] Arcas J, Ferrer T, Roche M, Bermejo-Martinez A, LopezMartin V. Hypohidrosis related to the administration of topiramate to children. Epilepsia 2001;42:1363-5. [12] Walsh P. Product information (zonisamide). In: Physicians’ Desk Reference. Montvale, NJ: Medical Economics Co; 2002:1307–11. [13] LaGrenade L, Graham D, Nourjah P. Underreporting of hemorrhagic stroke associated with phenylpropanolamine. Letter to the Editor, JAMA 2001;286:3081. [14] Tanaka K, Matsuoka O, Hattori H, Hattori K. Seto T, Murata R: Dyshidrosis caused by zonisamide. The Osaka Journal of Epilepsy Research 1997;8:31-4. [15] McJilton J, ReToledo J, DeCerce J, Huda S, Abubakr A, Ramsay R. Cotherapy of lamotrigine/zonisamide results in significant elevation of zonisamide levels. Epilepsia 1996;37(suppl 5):173. [16] Rudolf A, Hoffman J, Rudolf C, eds. Rudolph’s Pediatrics 20th ed. Stamford, CT: Appleton and Lange, 1996:1321–2. [17] Fujita Y, Okubo O, Noguchi Y, et al. A study of zonisamideinduced hidropoietic disorder. No To Hattatsu 1995;27(suppl):S187. [18] Otani K, Yanagihara K, Imai K. Zonisamide and oligohydrosis. No To Hattatsu 1995;27(Suppl):S127. [19] Okumura A, Ishihara N, Kato T, Hayakawa F, Kuno K, Watanabe K. Predictive value of acetylcholine stimulation testing for oligohydrosis caused by zonisamide. Pediatr Neurol 2000;23:59-61.
[20] Ono M, Inagaki Y, Ono M, Ueki H. Localization of carbonic anhydrase I and II in the normal human skin. J Dermatol Sci 1996;12: 201-6. [21] Dahlborn K, Jansson A, Nyman S, Morgan K, Holm L, Ridderstrale Y. Sweat production and localization of carbonic anhydrase
in the equine sweat gland during exercise. Equine Vet J 1999;30:398403. [22] Sato K, Kang W, Sato F. Eccrine sweat glands. In: Goldsmith LA, ed. Physiology, Biochemistry and Molecular Biology of the Skin, 2nd edition. New York: Oxford University Press, 1991:741-61.
Knudsen et al: Oligohydrosis and Zonisamide 189
seizure disorders in adult and pediatric patients. Zonisamide was approved for use in the United States in March 2000. The drug is indicated for adjunctive therapy in the treatment of partial epilepsy in adults only. Zonisamide is chemically classified as a sulfonamide and is characterized in vitro as having multiple mechanisms of action, including the ability to inhibit carbonic anhydrase [1,2]. Carbonic anhydrase inhibitors, such as acetazolamide, have been associated with such diverse systemic adverse events as metabolic acidosis, nephrolithiasis, hyperammonemia, and increased renal excretion of sodium bicarbonate [3,4]. Oligohydrosis, a potentially serious adverse event characterized by deficient formation and secretion of sweat, was initially reported in association with zonisamide when Japanese children treated with this drug manifested this condition during the exceptionally hot summer of 1994 [5-8]. To date, there have been no reports in the medical literature describing oligohydrosis in U.S. patients treated with zonisamide. The objectives of our study include the following: (1) documentation of U.S. reports of oligohydrosis in patients treated with zonisamide and identification of risk factors for this adverse event; (2) calculation of the U.S. reporting rate of oligohydrosis for patients treated with zonisamide; (3) systematic review of the medical literature for reports of oligohydrosis associated with exposure to any drug; and (4) review of potential mechanisms for drug-induced hidropoietic dysfunction.
Methods Patient data collection and tabulation
Introduction Zonisamide is an antiepileptic drug (AED) developed in Japan in the late 1970s, and first marketed there in 1989. The drug is used in clinical practice in Japan to treat
From the *Center for Drug Evaluation and Research, Division of Neuropharmacological Drug Products and the†Center for Drug Evaluation and Research, Office of Drug Safety, Food and Drug Administration Rockville, Maryland.
184
PEDIATRIC NEUROLOGY
Vol. 28 No. 3
Through the MedWatch program, the U.S. Food and Drug Administration (FDA) receives adverse drug event (ADE) reports and product quality concerns for drugs, biologic devices, or medical devices used by humans. Reports from manufacturers associated with their product(s) are required by law, whereas reports from health-care professionals and
Communications should be addressed to: Dr. Knudsen; Food and Drug Administration; Center for Drug Evaluation and Research; Division of Neuropharmacological Drug Products; 5600 Fishers Lane, HFD-120; Rockville, MD, 20857 Received June 6, 2002; accepted August 21, 2002.
© 2003 by Elsevier Inc. All rights reserved. doi:10.1016/S0887-8994(02)00511-8 ● 0887-8994/03/$—see front matter
consumers are voluntary. These reports are collected into an electronic database, referred to as the Adverse Events Reporting System (AERS). We searched AERS for United States reports of oligohydrosis with or without fever reported in association with zonisamide use from the time of approval in the United States (March 2000) through December 31, 2001 [9]. We used a broad search strategy to identify any report potentially related to oligohydrosis with or without fever. The search was followed by a manual review of each report to exclude duplicate and unrelated reports. In some cases that lacked specific detail, one of us obtained additional information through direct contact with the reporter. We summarized patient demographics, concomitant medications, zonisamide dose, and duration of therapy from the identified cases.
spiration. Bibliographies of review articles were searched for additional information.
Results Our search of AERS identified six cases of oligohydrosis and/or fever associated with zonisamide use. The clinical diagnosis was established by the health care provider who submitted the adverse event report.
Demographics Reporting rate Within the current passive surveillance system for reporting suspected drug-related adverse events to the FDA, it is not possible to calculate the absolute incidence of a drug-related side effect because we do not know the actual number of patients who experienced the adverse event (numerator) nor the actual number of patients exposed to the drug and duration of use (denominator). However, we are able to estimate the incidence with a reporting rate. A reporting rate is calculated using the number of cases of an adverse event reported to FDA in the numerator, and dividing by an estimate of drug use represented as person-years in the denominator. One person-year is equal to one patient taking a drug for one-year, or two people each taking the drug for six months, or four people each taking the drug for three months, etc. To approximate the extent of zonisamide use, we obtained an estimate of the total number of prescriptions dispensed in the United States for zonisamide from IMS Health’s National Prescription Audit Plus™ (NPA Plus™) [10]. This computerized database provides estimates of the total number of prescriptions dispensed by retail pharmacies. In order to estimate the proportion of zonisamide users in the pediatric age group (0-16 years old), we obtained demographic data from IMS Health’s National Disease & Therapeutic Index Audit™ (NDTI), an ongoing survey of treatment patterns, reported as appearances, at office-based medical practices. An appearance from the NDTI Audit is the mention of a drug in a single patient visit. A drug appearance can be associated with more than one diagnosis in the same patient visit. For example, if a doctor treats two conditions in one patient with the same product, this counts as two drug uses but only one drug appearance. Therefore, drug appearances may be equal to or less than uses. Information retrieved from NDTI included sex and age of patients, as well as mean duration of prescriptions. The average prescription length was 30 days. The method for calculating the reporting rate for oligohydrosis follows below: 1.
Estimation of total person-years of exposure to zonisamide: (Total # of zonisamide prescriptions dispensed by retail pharmacy) ⫻ (Average prescription length) ⫼ 365 days. 2. Calculation of the proportion of person-year exposure in the pediatric age group (“pediatric-years exposure”): (Proportion of drug use in children 0-16 from NDTI) ⫻ (Total person-years calculated in step 1). 3. Calculation of reporting rate using the formula: (Number of U.S. cases from AERS) ⫼ (Pediatric-years exposure).
Searching and evaluating the literature A search was performed using MEDLINE, EMBASE, TOXLINE, DERWENT DRUG FILE, PHARM-LINE and BIOSIS through December 2001. The medical subject heading, title, abstract, and key words were searched using the following terms: zonisamide, carbonic anhydrase inhibitor, acetazolamide, benzolamide, dichlorphenamide, ethoxzolamide, methazolamide, topiramate, oligohydrosis, hypohidrosis, dyshidrosis, anhidrosis, hidropoeitic disorders, overheating, body temperature regulation, sweat glands, decreased/reduced sweating/per-
Demographic information and a clinical summary of each of the six cases are shown in Table 1. There were no cases identified in patients older than 18 years of age. Three cases (case numbers 1, 2, and 4) were characterized by both oligohydrosis and fever. In one patient with fever (case number 6) the report did not specify whether the patient had oligohydrosis; however, the health-care provider stated that an evaluation for the etiology of the fever was unrevealing for any cause other than zonisamide. Although oligohydrosis was not reported, we include this case in the series because the patient went on to develop heat stroke requiring intensive care, similar to cases in the Japanese literature [6]. We also note that this patient was taking concomitant topiramate, another AED with reports of associated oligohydrosis. However, the patient had been on a stable dose of topiramate, the heat stroke episode occurred in temporal association with initiation of zonisamide, and the patient recovered following discontinuation of zonisamide with ongoing topiramate therapy. In addition to this case, it was reported in cases 1, 2, and 4 that other etiologies for fever were investigated with no conclusive findings. Five of the six patients were in the pediatric age range, with the sixth being 18 years old. The median age was 4.5 years. All of the patients were taking zonisamide for a seizure disorder. Two patients (case numbers 5 and 6) were taking other antiepileptic drugs in addition to zonisamide. It was not always possible to assign a precise time for the onset of oligohydrosis (with or without fevers) following initiation of zonisamide treatment. In these cases the time to detection was noted to be within a time frame for which the adverse event had been clearly established. Using these data, the time until detection of these adverse events ranged between 7 to 225 days after the start of zonisamide treatment. The median time until detection was 62.5 days. Patients experiencing oligohydrosis with or without fever during zonisamide treatment had received a wide range of doses. The daily doses ranged between l00 – 600 mg/day and corresponded to 6 –22 mg/kg/day. The recommended daily dose in children, according to Japanese literature, is 4 – 8 mg/kg. In three of the six cases (case numbers 1, 2, and 6) the patients with oligohydrosis and fever were reported to reside in southern states. Five of six
Knudsen et al: Oligohydrosis and Zonisamide 185
Table 1.
Clinical summary of U.S. cases of oligohydrosis/fever in patients treated with zonisamide
Case number
1
2
3
4 M 5 years Potassium phosphate, sodium bicarbonate, pyridoxine, calcium, levocarnitine Refractory multifocal onset tonic seizures 300 mg (22) Max 39 °C ⫹ 225 Fevers and oligohydrosis resolved after ZNS decreased to 200mg/day, ZNS was eventually d/c
Sex Age Concomitant medications
M 6 months Ranitidine, pyridoxine
M 10 months None reported
F 4 years None reported
ZNS indication
Infantile spasms
Abnormal EEG
Abnormal EEG
Dose* (mg/kg) Fever Oligohydrosis Onset** (days) Outcome
150 mg (17) Max 41 °C ⫹ 7 Hospitalized; fevers and decreased sweating improved but did not completely recover after ZNS d/c
100 mg (12) Max 41 °C ⫹ 42 Hospitalized; fevers and oligohydrosis resolved 11 and 9 months, respectively, after ZNS d/c
100 mg (6) None reported ⫹ ⱕ11 Oligohydrosis resolved after ZNS d/c
5
6
M 13 years Lamotrigine, levetiracetam, fluticasone
M 18 years Lamotrigine, levetiracetam, topiramate, propranolol, clonidine.
Unspecified seizure disorder
Partial seizure disorder
600 mg (9) None reported ⫹ 83 Oligohydrosis resolved after ZNS decreased, and was eventually d/c
600 mg (7) Max 42 °C none reported 83 Hospitalized; complete recovery; heat stroke syndrome and complications resolved with ICU treatment and ZNS d/c
* Dose reported as total daily dose and (mg/kg) ** Approximate duration of time from ZNS initiation to detection of symptoms (days) Abbreviations: d/c ⫽ Discontinued EEG ⫽ Electroencephalogram F ⫽ Female kg ⫽ Kilograms
M max mg ZNS
⫽ ⫽ ⫽ ⫽
Male Maximum Milligrams Zonisamide
cases (case numbers 1, 3, 4, 5, and 6) of oligohydrosis and fever were reported during summer months. Three patients required hospitalization for their reported adverse event. Five patients were considered to have completely recovered from oligohydrosis after the discontinuation of zonisamide. The remaining patient (case number 1) was considered to reveal only partial recovery after zonisamide therapy was stopped. Subsequently, this patient was diagnosed with a congenital mitochondrial metabolic disorder associated with sweating dysfunction.
younger. The reporting rate for zonisamide-associated oligohydrosis with or without fever is presented in Table 3 for patients 16 years of age and younger, as well as those 18 years of age and younger. The reporting rate for each of these age groups was similar, about 13 cases per 10,000 pediatric-years.
Table 2. Projected number of total U.S. drug appearances, stratified by age range in pediatric patients treated with zonisamide*
Calculation of U.S. reporting rate for oligohydrosis IMS Health’s NPA Plus™ data for prescription use revealed that approximately 212,000 prescriptions had been dispensed for zonisamide in the United States since approval on March 27, 2000. During the year 2000, there were 28,000 prescriptions written, and during 2001, the number increased to 184,000. Table 2 presents the breakdown of zonisamide prescriptions by age from NDTI™. Although zonisamide is not approved for pediatric use in the U.S., the IMS data showed that 22% of zonisamide appearances in 2001 were for patients aged 16 years and
186
PEDIATRIC NEUROLOGY
Vol. 28 No. 3
Age Range (years) ⬍1-2 3-6 7-9 10-12 13-16 Total
Total (ⴛ 1000)
% of total drug appearances
4 9 3 3 9 28
2.8 7.3 2.3 2.2 7.2 22
* Distributed by calendar year—March 2000 through December 2001. ** Source: IMS Health; National Disease & Therapeutic Index AuditTM
Table 3. Estimated U.S. reporting rates of oligohydrosis and/or fevers associated with zonisamide* U.S. Experience <16 years old Number of cases received by FDA Zonisamide usage (number of pediatric years of exposure) Reporting rate†
U.S. Experience <18 years old
5
6
3799
4565
13 cases/10,000 person-years
13 cases/10,000 person-years
* U.S. usage for March 2000-December 2001, pediatric years of exposure estimated from prescription use assuming each prescription is for 30 days. † Reporting rate: number of U.S. case reports received ⫼ pediatric years of use
Literature search The search strategies identified 314 citations. After a review of their titles, key words, and abstracts, we found no U.S. cases of oligohydrosis and/or fever in association with the use of zonisamide. We identified Japanese literature reports of oligohydrosis, fever, and heat stroke during exposure to zonisamide. In addition, we identified a report by Arcas et al. [11] that consisted of three cases of oligohydrosis and fever in pediatric patients treated with the recently marketed AED topiramate. Discussion We describe the first U.S. reports of oligohydrosis with or without accompanying fever in pediatric patients treated with zonisamide. Health-care practitioners should consider this potentially serious event as being possibly drug-induced. Although elevated ambient temperature is not a risk factor for development of oligohydrosis, it increases the likelihood of detecting the condition, and creates the setting where oligohydrosis can lead to serious sequelae. We calculated a U.S. reporting rate for oligohydrosis with or without fever of 13 cases/10,000 pediatric years of exposure (Table 3). To place this rate in perspective, 38 cases of oligohydrosis with or without fever were reported in Japan during the first 11 years of marketing, for an estimated reporting rate of 1 case per 10,000 pediatricyears of exposure [12]. Despite the 10-fold difference in these rates, both are underestimates of the true incidence because of the widely recognized underreporting of adverse events in a passive surveillance program [13]. One study estimated that between ⬍1% to 10% of serious adverse events are reported to the FDA [13]. If the index of suspicion for oligohydrosis in pediatric patients is low, then correspondingly, recognition and reporting will be low. Recognition may occur only when the patient develops fever for which an infectious or neoplastic source cannot be identified [14]. Mild or moderate decrements in normal sweating may not be recognized. Tanaka et al. [14] characterized the presentation of oligohydrosis in Japanese children treated with zonisamide as subtle and missed
frequently, with recognition occurring when the patient developed hyperthermia. Based on the U.S. cases, as well as those from Japan [5– 8], pediatric age appears to be a risk factor for zonisamide-associated oligohydrosis. The apparent increased risk of oligohydrosis in the pediatric age group may be related to the dose and resulting blood levels of zonisamide and its metabolites. Dosing based on body weight was relatively higher in four of our six patients than the standard dose range (4 – 8 mg/kg/day) used in Japan. In one patient the dose was 22 mg/kg/day, nearly two times the maximal pediatric dose of 12 mg/kg/day recommended in Japan. Further supporting our suspicion that an elevated blood level of zonisamide may play a role in the development of oligohydrosis is the finding that two of the patients were taking lamotrigine concomitantly with zonisamide. Concomitant therapy of lamotrigine with zonisamide results in substantial elevations of serum levels of zonisamide and its metabolite [15]. Thus, these patients may have been exposed to inappropriately high blood levels of zonisamide. Additionally, Tanaka et al. [14] documented that dose levels of zonisamide were high for pediatric patients in all reported cases of dyshidrosis. In their study, perspiration returned after lowering the dose of zonisamide or stopping the drug. One more factor that may contribute to elevated zonisamide serum levels in pediatric patients is the alteration of renal drug clearance by zonisamide itself. According to the drug labeling, clinical trials in adults revealed that zonisamide is eliminated from the body by renal excretion and is associated with a statistically significant increase in serum creatinine and blood urea nitrogen compared with placebo-treated patients. The increase persists over time and is interpreted as an effect on glomerular filtration rate (GFR). The GFR returns to baseline within 2–3 weeks of drug discontinuation. Infants may be particularly sensitive to this effect of the drug given that they have a markedly reduced GFR, about 30% of adult normals when expressed on a surface area basis [16]. Of interest, we note that males comprised five of our six cases and three of three U.S. cases of oligohydrosis/fever
Knudsen et al: Oligohydrosis and Zonisamide 187
associated with topiramate treatment [11]. This observation raises the question of a gender predisposition for oligohydrosis/fever related to zonisamide and topiramate. However, our survey of the Japanese literature [17,18] did not reveal any suggestion of male gender as a risk factor for zonisamide-induced oligohydrosis/fever. Ultimately, the question of a gender predisposition would have to addressed in a controlled study, and can not be answered by a case series. Our present understanding of the pathophysiology of drug-induced oligohydrosis in pediatric patients is limited. However, the medical literature may suggest clues to possible mechanisms. Shimizu et al. [6] reported a reduced response to pilocarpine electrophoresis and acetylcholine stimulation in pediatric patients given zonisamide. Okumura et al. [19] used the subcutaneous acetylcholine stimulation test to detect patients at risk of oligohydrosis caused by zonisamide and concluded that dyshidrosis was due to a functional disorder of the sweat glands. Diminished sudomotor responsiveness and hypohidrosis have also been reported to occur in patients treated with topiramate, another recently marketed AED that inhibits carbonic anhydrase [11]. This report described three male pediatric patients who exhibited hypohidrosis/oligohydrosis, heat and exercise intolerance, and fever after 2-3 months of treatment (during summer heat). The assessment of the patients’ sudomotor function with pilocarpine iontophoresis showed reduced sweat production in each of the individuals. It was concluded that hypohidrosis was caused by a dysfunction of sweat glands because of complex interacting mechanisms triggered by topiramate. Carbonic anhydrase isozymes are located in secretory and transporting epithelia, including eccrine sweat glands [20]. Studies in animal models have demonstrated that carbonic anhydrase localized to sweat glands is important for the production of sweat during physical activity and especially at higher ambient temperatures [21]. Physiologic eccrine sweating depends on the presence of extracellular Ca2⫹ and its entry into the secretory cell, leading to stimulus secretion coupling and, ultimately, sweat production [22]. Calcium ionophores, carriers of calcium ions across the cell membrane, stimulate sweat secretion in various species [22]. The cumulative weight of 70 years of research on the enzyme carbonic anhydrase may provide the framework for conceptualizing how modulation of Ca2⫹ levels by fluxes in acid equivalents regulates secretion. Zonisamide inhibition of carbonic anhydrase could modify the acid— base (acid equivalents) and ionic (mainly Ca2⫹ availability) cellular environment of the eccrine sweat gland epithelial cells. The result would be a conformational change (alteration in membrane form) and a diminution in the responsiveness of eccrine sweat gland muscarinic receptors to cholinergic stimulation. In conclusion, our finding of infrequent oligohydrosis/ fever in children taking zonisamide is consistent with the Japanese literature describing this adverse event. Aware-
188
PEDIATRIC NEUROLOGY
Vol. 28 No. 3
ness of this uncommon drug-related adverse event may prevent unnecessary morbidity, and potentially simplify the evaluation and treatment of fever in children who are taking zonisamide. Although the mechanism for zonisamide-associated oligohydrosis/fever is not fully understood, the drug may in part mediate its effect on eccrine sweat glands by inhibiting carbonic anhydrase thereby influencing pH dynamics, H⫹ concentration, and available Ca2⫹ transients at the level of ion channels and receptors.
References [1] Masereel B, Rolin S, Abbate F, Scozzafava Asupuran CT. Carbonic anhydrase inhibitors: anticonvulsant sulfonamides incorporating valproyl and lipophilic moieties. J Med Chem 2002;45:312-20. [2] Nakatsuji K, Matsuno N, Nakamura N, Fujitani B, Ito T, Kadokawa T. General pharmacology of the novel antiepileptic compound zonisamide. Arzneim Forsch 1987;37:1131-6. [3] Maren TH. Carbonic anhydrase: Chemistry, physiology and inhibition. Physiol Rev 1967;47:595-781. [4] Swenson ER. Respiratory and renal roles of carbonic anhydrase in gas exchange and acid-base regulation. In: Chegwidden WR, Carter ND, Edwards TH, eds. The carbonic anhydrase new horizons. Basal: Birkha¨ user Verlag, 2000:281-341. [5] Okumura A, Hayakawa F, Kuno K, Watanabe K. Oligohydrosis caused by zonisamide. No To Hattatsu 1996;28:44-7. [6] Shimizu Y, Togo A, Eto M, et al. Oligohydrosis induced by zonisamide, clinical features of 7 cases and skin histology. J Japan Ped Society 1996;100:350-1. [7] Hosada N, Miura T, Takanashi S, Shirai H, Sunaoshi W, Takei K, Clinical experience of dyshidrosis induced by zonisamide. Kanagawa Seishin Igaku Kaishi 1996;46:121. [8] Hosada N, Miura H, Shirai H, Sunaoshi W, Takei K. Dyshidrosis by zonisamide. Japanese J Ped 1996;49(11):2445-9. [9] Brinker A, Beitz J. Use of a spontaneous adverse drug events database for identification of unanticipated drug benefits. Clin Pharmacol Ther 2002;71:99-102. [10] Rodriguez EM, Staffa JA, Graham DJ. The role of databases in drug postmarketing surveillance. Pharmacoepidemiol Drug Saf 2001;10: 407-10. [11] Arcas J, Ferrer T, Roche M, Bermejo-Martinez A, LopezMartin V. Hypohidrosis related to the administration of topiramate to children. Epilepsia 2001;42:1363-5. [12] Walsh P. Product information (zonisamide). In: Physicians’ Desk Reference. Montvale, NJ: Medical Economics Co; 2002:1307–11. [13] LaGrenade L, Graham D, Nourjah P. Underreporting of hemorrhagic stroke associated with phenylpropanolamine. Letter to the Editor, JAMA 2001;286:3081. [14] Tanaka K, Matsuoka O, Hattori H, Hattori K. Seto T, Murata R: Dyshidrosis caused by zonisamide. The Osaka Journal of Epilepsy Research 1997;8:31-4. [15] McJilton J, ReToledo J, DeCerce J, Huda S, Abubakr A, Ramsay R. Cotherapy of lamotrigine/zonisamide results in significant elevation of zonisamide levels. Epilepsia 1996;37(suppl 5):173. [16] Rudolf A, Hoffman J, Rudolf C, eds. Rudolph’s Pediatrics 20th ed. Stamford, CT: Appleton and Lange, 1996:1321–2. [17] Fujita Y, Okubo O, Noguchi Y, et al. A study of zonisamideinduced hidropoietic disorder. No To Hattatsu 1995;27(suppl):S187. [18] Otani K, Yanagihara K, Imai K. Zonisamide and oligohydrosis. No To Hattatsu 1995;27(Suppl):S127. [19] Okumura A, Ishihara N, Kato T, Hayakawa F, Kuno K, Watanabe K. Predictive value of acetylcholine stimulation testing for oligohydrosis caused by zonisamide. Pediatr Neurol 2000;23:59-61.
[20] Ono M, Inagaki Y, Ono M, Ueki H. Localization of carbonic anhydrase I and II in the normal human skin. J Dermatol Sci 1996;12: 201-6. [21] Dahlborn K, Jansson A, Nyman S, Morgan K, Holm L, Ridderstrale Y. Sweat production and localization of carbonic anhydrase
in the equine sweat gland during exercise. Equine Vet J 1999;30:398403. [22] Sato K, Kang W, Sato F. Eccrine sweat glands. In: Goldsmith LA, ed. Physiology, Biochemistry and Molecular Biology of the Skin, 2nd edition. New York: Oxford University Press, 1991:741-61.
Knudsen et al: Oligohydrosis and Zonisamide 189