Epilepsia, 45(11):1416Ð1428, 2004 Blackwell Publishing, Inc. C 2004 International League Against Epilepsy

A Proposal for Case Definitions and Outcome Measures in Studies of Infantile Spasms and West Syndrome: Consensus Statement of the West Delphi Group

Andrew L. Lux and John P. Osborne

Bath Unit for Research in Paediatrics, Royal United Hospital, Bath, and School for Health, University of Bath, Bath, United Kingdom for The West Delphi Group

Summary: Purpose: To reach a broad consensus on case def- Primary electroclinical outcome denotes cessation of spasms initions, outcomes, and outcome measures that will ease future with resolution of hypsarrhythmia. West syndrome should be a study design and facilitate comparison of data from different defined subset of the syndrome of infantile spasms. An infan- studies of infantile spasms and West syndrome. tile spasms single-spasm variant should be recognized. Ways Methods: Persons who had recently presented or published are suggested of handling subtle spasms in the context of clin- first-author original research in this field were invited to partici- ical studies. It proposes a standard for reporting modifying and pate in an e-mail Delphi process and to invite other investigators atypical features of hypsarrhythmia, a minimal set of baseline or clinicians who they thought might participate. characteristics and outcomes that should be reported in trials of Results: The process consisted of six rounds, anonymous ex- infantile spasms, and suggests a standard definition of relapse. cept to the facilitator. In total, responses were received from 46 Consensus was not reached on a definition of hypsarrhythmia. participants. The final statement was approved by 31 participants Conclusions: We reached a clear consensus on many aspects from 15 countries. It concluded that the primary clinical out- of study design for the investigation of infantile spasms, although come, cessation of spasms, should denote absence of witnessed incomplete consensus was found on how to define EEG criteria. spasms from within 14 days of commencement of treatment, Key Words: Epilepsy—Infantile spasms—West syndrome— and for ≥28 consecutive days from the last witnessed spasm. Consensus—Clinical trials.

Infantile spasms (West syndrome) is a form of epilepsy that is associated with many underlying conditions and Accepted June 20, 2004. often a poor developmental outcome. Its main clinical Address correspondence and reprint requests to Dr. J.P. Osborne at features are spasms that usually occur in clusters and Bath Unit for Research in Paediatrics, Royal United Hospital, Combe onset that almost always occurs during the first 2 years Park, Bath, BA1 3NG, U.K. E-mail: [email protected] of life. It is characteristically associated with an EEG The West Delphi Group pattern called hypsarrhythmia, although hypsarrhythmia Andrew L. Lux and John P. Osborne, Bath Unit for Research in Paediatrics, Bath, England, and School for Health, University of may not be present throughout the course of the spasms Bath, Bath, U.K.; Giuliano Avanzini and Tiziana Granata, Istituto and is not regarded as essential for the diagnosis of in- Neurologico “C. Besta,” Milan, Italy; Tallie Z. Baram, University of fantile spasms. Hypsarrhythmia is an interictal pattern California at Irvine, Irvine, CA, U.S.A.; Roberto Caraballo, Hospital Nacional de Pediatria “Juan P. Garrahan,” Buenos Aires, Ar- that usually changes during clinical attacks to lower- gentina; Kevin Farrell, University of British Columbia, Vancouver, amplitude slow waves or to a sudden flattening that is British Columbia, Canada; Ellie Hancock, Great Ormond Street Hospi- tal, London, U.K.; Masatoshi Ito, Shiga Medical Center for Children, Moriyama, Japan; Jun Kohyama, Tokyo Medical and Dental Univer- William Soler Children’s Hospital, Havana, Cuba; Dietz Rating, sity, Tokyo, Japan; Roshan Koul, Sultan Qaboosh Hospital, Alkhad, University of Heidelberg, Heidelberg, Germany; Bernhard Schmitt, Oman; Wei-Ling Lee, National Neuroscience Institute, Singapore; University Children’s Hospital, Zurich, Switzerland; Malinee A. Jianxiang Liao, Shenzhen Children’s Hospital, Shenzhen, China; Thambyayah, Hospital Selayang, Kuala Lumpur, Malaysia; Federico Marissa B. Lukban and Aida R. Mendoza-Salonga, University of Vigevano, Ospedale Pediatrico Bambinu Gesu, Rome, Italy; Anannit the Philippines College of Medicine, Manila, Philippines; Anthony Visudtibhan, Mahidol University, Bangkok, Thailand; William White- G. Marson, University of Liverpool, Liverpool, U.K.; Brian Neville, house, University of Nottingham, Nottingham, U.K.; Virginia Wong, Institute of Child Health, London, U.K.; Harry O. Nomayo, Klinikum The University of Hong Kong, Hong Kong SAR, China; Hideo Ya- Weiden Kinderklinik, Weiden, Germany; Hirokazu Oguni. Tokyo manouchi Dokkyo University School of Medicine, Tochigi, Japan; Women’s Medical University, Tokyo, Japan; Shunsuke Ohtahara, Chainllie Young, Washington University School of Medicine, St. Louis, Okayama University Medical School, Okayama, Japan; Hian-Tat Ong, Mo, U.S.A.; and Zhongshou Zhou, China-Japan Friendship Hospital, National University Hospital, Singapore; Desiderio Pozo Lauzan, Beijing, China.

1416 WEST DELPHI GROUP CONSENSUS STATEMENT 1417 known as an electrodecremental period. Various EEG fea- plored some controversial areas. Those agreeing to par- tures have traditionally been labeled modified or atypical ticipate were asked to formulate further questions as a hypsarrhythmia (1Ð4). unique contribution to the study. We made provision for Several International League Against Epilepsy (ILAE) participants to enter after the first round because not all in- commissions and workshops have proposed definitions terested persons would necessarily have been contacted in and classifications of infantile spasms, and some elements time to respond to Round 1. We suggested that, to be con- of these are not mutually consistent. For example, propos- sidered members of the West Delphi Group, participants als from the ILAE Commissions on Classification and would need to make contributions to other rounds and to Terminology in 1985 and 1989 suggested that cases of agree the final draft of the proposal. We also suggested infantile spasms should always have onset of symptoms that minority and dissenting views would be clearly rep- before age 12 months, whereas a 1991 workshop of the resented in the final draft, and that we would not censor ILAE Commission on Pediatric Epilepsy suggested that views. infantile spasms “transcends age groups and may occur The content and form of subsequent rounds was deter- in infancy or childhood” (5Ð7). Variation in age at onset mined in part by the suggestions of participants. A bal- as a defining criterion alters epidemiologic characteristics ance was reached between eliciting quantitative informa- such as median age at onset of illness and affects estimates tion about the strength of agreements and eliciting qualita- of age-dependent outcomes. In addition to differences in tive and creative views on how we might approach prob- definitions and criteria proposed by the ILAE, many other lems of definition, classification, and measurement. We variations in definition, classification, and outcome crite- tried to maintain a focus on aspects of definition and out- ria are found in published studies. come and to develop proposals that would clarify and An initial proposal for standardized case definitions, simplify study design without introducing unnecessary outcomes, and outcome measures in infantile spasms constraints. The first round of West Delphi was sent in was presented at the Symposium on West Syndrome and January 2001, and the final round was sent in November other Infantile Epileptic Encephalopathies, Tokyo, Febru- 2003. ary 2001 (8). Many participants were enthusiastic about this proposal, although some thought that, in particular, it RESULTS would be difficult to achieve a consensus on EEG charac- teristics. In Round 1, 31 responses were received from partici- pants in 14 countries, and overall, 42 respondents from 16 countries. In addition, three potential participants stated a METHODS firm intention to participate but made no substantial con- West Delphi followed the format of a Delphi process, tribution to the process. One response was initially on a method for eliciting expert consensus developed by the behalf of members of the Konigsteiner¬ Kreis, a group of RAND Corporation in the early 1950s (9). A Delphi pro- pediatric neurologists that had already started to address cess essentially involves isolated experts giving their judg- the issue of case definitions and outcome measures in in- ments or opinions to a moderator or facilitator, who makes fantile spasms. These participants continued to confer and those opinions anonymous and redistributes them to the send a single response to subsequent rounds. The 10 initial full group of experts (10). Because the experts are able questions covered the following areas: clinical features, to relate their opinions to that of the larger group, this restrictions for age at onset, assessing developmental de- facilitates an iterative process by which they can finally lay, the usefulness of etiologic subgroups, the likelihood reach broad agreement. This method has been used before of reaching consensus on a definition of hypsarrhythmia, in the fields of neurology, pediatrics, and psychiatry (11Ð using aggregated data, defining developmental delay, and 14). In West Delphi, responses were collated by A.L.L., the interpretation of adverse effects. who acted as facilitator of the process. J.P.O. acted as Twenty-seven responses to Round 2 were received. The guarantor of the process but was not aware of the identity first section, consisting of seven questions, sought qualita- of individual respondents and participated as an opinion tive comments and suggestions about the following: a gen- giver. eral definition of infantile spasms, terms for specific clini- Invitations were sent by e-mail to 133 persons who had cal and electroclinical scenarios, suggested classifications published articles as primary authors on infantile spasms for etiologic subgroups, necessary and sufficient EEG fea- during the previous 10 years, or who were known to A.L.L. tures of hypsarrhythmia, arguments for and against a 28- or J.P.O. to have an interest in infantile spasms. Invited day spasm-free period to define the clinical response ces- persons were asked to forward the e-mail to anyone who sation of spasms (an outcome that had been suggested they thought would be interested in participating. The first in Round 1), and a list of appropriate tests to measure round included 10 questions with multiple-choice format developmental outcome. The second section took MCQ responses (MCQs) that acted to stimulate debate and ex- format and covered such areas as: clustering of spasms; ic-

Epilepsia, Vol. 45, No. 11, 2004 1418 A. L. LUX AND J. P. OSBORNE tal units; age-related classifications and misclassification final consensus is printed later. Paragraph numbering is of cases; age correction for gestational age at birth; de- provided so that future investigators may make more spe- velopmental delay at time of diagnosis; use of the terms cific reference to the proposals. symptomatic, cryptogenic, and idiopathic; distinguish- ing spasms from myoclonus; the role of the cortex and West Delphi Proposal subcortex in the etiology of infantile spasms; the potential 1. Introduction to West Delphi. The West Delphi Pro- role of the ILAE-proposed five-axis diagnostic scheme; posal is a consensus statement that proposes a series of clinical and electroclinical outcomes; features that con- definitions and standards for use in studies of infantile stitute relapse; periods of necessary and sufficient ther- spasms. In particular, it aims to provide clearer defini- apeutic intervention; and the most appropriate ages for tions of cases and etiologic subgroups, and a standard for developmental assessment. outcomes and outcome measures. We hope that this will In Round 3, participants were invited to respond to 32 help those who wish to design future studies of infantile statements that represented majority opinions from earlier spasms, and that such studies will provide optimal infor- rounds and to state whether they agreed, were unsure, or mation for those who wish to learn from these studies and would prefer an amendment. They also were asked to state to choose the best approach to treating infantile spasms. whether they thought the statement should be included in the final West Delphi proposal. The first two statements Case definitions included a diagram illustrating putative definitional re- 2. Infantile spasms. We propose using the term infan- lations between epileptic spasms, infantile spasms, WS, tile spasms (ISs) to describe an epilepsy syndrome that clustered and nonclustered spasms, and hypsarrhythmia. rarely has onset in children older than 2 years and usu- Twenty participants responded. An area identified as im- ally has onset in children younger than 1 year. Its main portant in Round 2 was the status of pyridoxine admin- clinical manifestation is clinical spasms that usually oc- istration as a diagnostic tool, and discussion of this issue cur in clusters. Many potential etiologies or associated also formed part of Round 3. conditions exist. The most characteristic EEG finding is Round 4 presented modified statements, requesting hypsarrhythmia. However, hypsarrhythmia is not found in comments on their content and suitability for the proposal. all cases, nor is it found throughout the clinical course of Twenty-two responses were received. Five participants the condition. Hypsarrhythmia is usually interrupted dur- were essentially prepared to accept Round 4 statements ing a clinical attack of epileptic spasms. The spasms are as the consensus proposal. Round 4 proposed new terms, often associated with developmental arrest or regression. predisposed and nonpredisposed, in an attempt to clar- 3. West syndrome. We propose using the term West syn- ify inconsistencies in the use of the terms cryptogenic, drome (WS) to describe the combination of spasms that idiopathic, and symptomatic. Although some participants occur in clusters and hypsarrhythmia on an EEG. We do were enthusiastic about these new terms, they did not meet not require, as have some previous definitions of WS, that with universal approval and were dropped. At the sugges- evidence of delayed development occur before the onset tion of two participants, an opinion was sought from an of spasms. Some participants wished to classify children external expert on the role that might be played by pyri- with single spasms (that is, without spasms in clusters) and doxine. with hypsarrhythmia as having WS, but most preferred to Round 5 consisted of a draft paper for submission as a reserve the term West syndrome for cases with clustered final proposal from the West Delphi Group. It contained spasms. a cover letter outlining areas of incomplete agreement 4. Significance of developmental delay. We see no rea- and contention. Before sending this draft, the facilitator son for including developmental delay at the time of onset (A.L.L.) corresponded independently with several partic- as a defining feature of WS. Indeed, developmental delay ipants to clarify their views as expressed in Round 4. Most at onset of spasms is likely to be assessed unreliably be- participants agreed with the statement as worded at this cause spasms are often subtle, and their true time of onset stage, but 12 participants had substantial comments that is often unrecognized, and also because development is were addressed individually. These were reviewed by the hard to assess in early infancy. However, we believe that factilitator (A.L.L.) and guarantor (J.P.O.) of the study, information about unequivocally normal development at and comments were sent to the individual participants for the time of onset of spasms is worth recording, because clarification or further comment before they were sent this may be associated with better developmental outcome. to the whole group with a final draft of the consensus Apparently abnormal development may be confounded by document. the effect of unrecognized spasms at onset or the epileptic Round 6 consisted of a final approval of the draft pa- encephalopathy. per. The detailed responses from each participant remain 5. Infantile spasms—single-spasm variant. It is less pseudo-anonymous. That is, they are known to the facili- usual for ISs always to occur singly rather than in clus- tator but not to any other participant, including J.P.O. The ters, yet we agree that such cases do occur. We propose

Epilepsia, Vol. 45, No. 11, 2004 WEST DELPHI GROUP CONSENSUS STATEMENT 1419 that these cases are distinguished in studies of ISs by be- be flexor or extensor, or a mixture of extensor and flexor ing called infantile spasms single-spasm variant (ISSV). movements, and they may be asymmetrical. These move- A child with single spasms and hypsarrhythmia should be ments have a longer duration than the movements of my- classified as having ISSV rather than WS. We do not pro- oclonus, and they are shorter than the movements associ- pose a strict definition of clustering, but, as a general guide ated with a tonic seizure. Thus their duration is ∼1 s. The and until further evidence is available, we suggest that a subtlest movement that might be classifiable as a clini- spasm can be regarded as a single spasm if no other spasms cal spasm is a head nod. Although it has been suggested occur for 1 min beforehand and for 1 min afterward. Any that the term describing this seizure semiology should be definition of clustering should be clearly stated. We wish epileptic spasm (ILAE Glossary of Descriptive Terminol- to emphasize that a child with single spasms and without ogy for Ictal Semiology, 2001: paragraph II 1.1.1.1), we hypsarrhythmia is rare, and such cases are likely to have propose using the term clinical spasms to describe the ictal a diagnosis other than ISSV. Some participants thought phenomenology and reserving the term epileptic spasms that they would never diagnose ISs in such a case. We to describe the seizure type (16). would suggest that authors state clearly, in cases with nei- 10. Subtle spasms. More subtle movements, often de- ther clustering of spasms nor hypsarrhythmia, why they scribed as subtle spasms, may constitute the clinical attack are accepting the diagnosis of ISSV. associated with hypsarrhythmia. Such movements include 6. Hypsarrhythmia without infantile spasms. Some chil- episodes of yawning, gasping, facial grimacing, isolated dren are investigated for symptoms other than clinical eye movements, and transient focal motor activity. These spasms and are found to have hypsarrhythmia. If these patients may be having focal seizures, and the episodes children, even after video-EEG investigation, are found are difficult to count and record. Ideally, a video-EEG not to have any evidence of clinical spasms, we propose would be performed to confirm that clinical spasms are that they be classified as having hypsarrhythmia without not occurring. infantile spasms (HWIS). Future studies will inform us 11. Subtle spasms at onset or at follow-up. We pro- whether these children have a high risk of developing ISs pose that children with these subtle ictal events be clas- at a later date and about their developmental outcome. sified at onset as having hypsarrhythmia without infan- 7. Relation between definitional terms. We propose us- tile spasms (HWIS) because the events are different from ing ISs as an inclusive term for all children with clinical clinical spasms, they are difficult to recognize and count, spasms who have evidence of EEG abnormalities consis- and they are not reliably measured as a primary clinical tent with the clinical syndrome of ISs—typically, although outcome. However, we suggest that investigators should, not necessarily, hypsarrhythmia or modified hypsarrhyth- during follow-up of a case of ISs, report subtle move- mia (see statements 16 and 17)—provided that the EEG ments associated with typical electrodecremental EEG findings do not suggest another specific diagnosis. The changes or interruption in the interictal background, and terms West syndrome and infantile spasms single-spasm that this combination of clinical and neurophysiologic variant describe specific subgroups of ISs. Investigators findings would be sufficient to constitute electroclinical might wish to report studies of children with all forms nonresponse or relapse. of ISs and report the proportions in their study with WS 12. Epileptic spasms. We propose using the term epilep- and ISSV. Or they might wish to perform a study that in- tic spasms to describe the epileptic seizureÐtype of clinical cluded, say, only children with WS and ISSV with hypsar- spasms associated with an epileptiform EEG. This corre- rhythmia. Studies that focus on developmental and EEG sponds to Axis 2 of the proposed ILAE scheme. This term outcomes might wish to include children with HWIS (al- would include conditions such as the periodic spasms of though HWIS is not a subgroup of ISs). Consistent use of Gobbi and ISs (17,18). The term epileptic spasms implies these definitional terms and a clear statement of how many that the EEG is consistent with the diagnosis of an epilepsy, children came from each group will assist those who wish but does not by itself imply hypsarrhythmia or any more to use and interpret study data. We believe that these def- specific EEG pattern. initions clearly distinguish clinical subgroups, encourage 13. Relation to proposed ILAE five-axis system. By us- clarity, and permit flexibility in study design. ing these definitions and classification, clinical spasms, 8. Our proposed relations between ISs, WS, ISSV and epileptic spasms, and ISs correspond with axes 1 (ictal HWIS are illustrated in Figures 1 and 2. These figures phenomenology), 2 (seizure type), and 3 (syndrome), re- also illustrate the relations between these terms and the spectively, of the proposed ILAE five-axis scheme. West five-axis system proposed by the ILAE Task Force on syndrome is a syndrome in its own right and a subgroup Classification and Terminology, 2001 (15). of ISs, as is ISSV. 9. Features of clinical spasms. We propose that the term 14. With respect to Axis 3 (syndrome) of the proposed clinical spasms be used to describe brief and synchronous ILAE diagnostic scheme, we propose that ISs is the syn- movements of the head, trunk, and limbs, or sometimes drome of epileptic spasms with onset generally during the of the head, trunk, or limbs alone. The movements may first 2 years of life, in association with high-voltage, mul-

Epilepsia, Vol. 45, No. 11, 2004 1420 A. L. LUX AND J. P. OSBORNE

Clinical spasms* No Not infantile spasms or West syndrome

Yes

EEG**: Nonepilepsy condition, such as epileptiform No benign myoclonus of infancy Yes

Epileptic spasms

Other syndrome of spasms, Age at onset <2 years*** No such as Gobbi syndrome

Yes

EEG*: interictal EEG consistent with infantile No spasms and not suggesting other syndrome Other epilepsy condition

Infantile spasms

Infantile spasms Spasms occur in clusters No single-spasm variant

EEG*: hypsarrhythmia (with ISSV without No or without atypical features) hypsarrhythmia— Yes a rare diagnosis

ISSV with hypsarrhythmia

EEG**: hypsarrhythmia (with Infantile spasms without No or without atypical features) hypsarrhythmia

Yes

West syndrome

FIG. 1. Classification of infantile spasms for study inclusion.

∗Excludes subtle spasms. ∗∗EEG, standard, sleep and video EEGs are performed, as necessary, before excluding a positive finding. ∗∗∗2 years is not an absolute upper limit, but substantially older cases would be likely to be classified as another syndrome, such as that of periodic spasms. If thought to be infantile spasms, they would be clearly identified as unusual, and the numbers of such patients should be clearly stated. Suggested cross-reference to ILAE multiaxial classification: Axis 1 (ictal phenomenology): clinical spasms Axis 2 (seizure type): epileptic spasms Axis 3 (syndrome): infantile spasms (ISs), West syndrome (WS), and infantile spasms single-spasm variant (ISSV) are syndrome subgroups of ISs. Suggested use for terms: • Studies of infantile spasms would include cases of ISSV and cases of West syndrome, unless exclusions or restrictions are explicitly stated • Studies of West syndrome would exclude cases without clustered spasms and cases without hypsarrhythmia Note that this classification requires clusters of spasms for a diagnosis of West syndrome. Cases with spasms occurring singly and without any witnessed clusters, even when associated with hypsarrhythmia, are not classified as West syndrome.

Epilepsia, Vol. 45, No. 11, 2004 WEST DELPHI GROUP CONSENSUS STATEMENT 1421

Hypsarrhythmic EEG

Hypsarrhythmia without infantile Spasms* No spasms (HWIS) (may be subclassified FIG. 2. Classification of children with hypsarrhythmia without clinical spasms but who may have subtle spasms. according to clinical features) It is well known that hypsarrhythmia can occur in the ab- Yes sence of spasms and sometimes without any evident clin- ical seizure (26). Because hypsarrhythmia is effectively the presenting feature in such cases (at least, with re- Clusters* No Infantile spasms spect to considering a diagnosis of infantile spasms), it is single-spasm variant useful to have a scheme for classification, conditional on with hypsarrhythmia the presence of hypsarrhythmia. Yes This flow diagram suggests a classification for children with hypsarrhythmia identified at younger than 2 years. West syndrome

*If spasms or clusters are not observed clinically, we recommend performing video-EEG for a period of ≥24 h to rule out their occurrence reliably. tifocal, or diffuse interictal EEG abnormalities (17). We 1991 workshop on infantile spasms (7) that study partici- propose that WS be defined by the association of clini- pants with modified hypsarrhythmia not be reported in di- cal spasms, at least some of which occur in clusters, with chotomized groups but that, where appropriate, a descrip- an interictal EEG showing hypsarrhythmia. This may be tion be given of the individual modifying features. The either “classic” hypsarrhythmia, or hypsarrhythmia with presence of modified features may depend on the stage of modified features. Infantile Spasms single-spasm variant ISs at which the EEG is performed; it may depend on treat- (ISs) is another form of ISs and also is regarded as an ment; and as an aggregate variable, it probably has little epilepsy syndrome. practical prognostic significance in randomized studies. 15. Ictal unit of infantile spasms. Disagreement exists 18. Forms of EEG recording. EEG findings provide es- about whether the appropriate ictal unit in studies of ISs sential information in studies of ISs. We recommend a should be the spasm or the cluster of spasms. Most West degree of standardization in the timing of EEG investi- Delphi participants thought that the more appropriate ic- gations, and that investigators limit potential information tal unit was the cluster of spasms. However, we thought bias in studies by making allowance for differences in the that limited value existed in outcomes that rely on mea- sensitivity of different investigations. Increased sensitiv- surement of ictal units, such as time to 50% reduction of ity is likely in the case of video-EEGs, with longer periods spasms or the proportion of study participants having a of recording, and with recordings that include periods of ≥50% reduction in spasms. non-REM sleep. Although performing less-sensitive in- vestigations is reasonable, it is prudent to design studies Electrographic features that use more-sensitive investigations before inferring that 16. Hypsarrhythmia. We propose that the term hyp- EEG abnormalities are genuinely not present. sarrhythmia be used to describe an EEG pattern that is characterised by random, high-voltage spikes and slow Etiology of infantile spasms waves. The most striking features of hypsarrhythmia are 19. Etiologic subgroups of ISs. Several terms have been high-voltage (generally >200 µV) slow waves with vari- used to describe etiologic subgroups of ISs, but these terms able amplitude; spikes and waves from many foci, and have not been used consistently between studies. Incom- varying with time; and a lack of synchrony, with a gener- plete agreement remains about the use of the terms symp- ally “chaotic” appearance. The typical appearance is more tomatic, idiopathic, and cryptogenic, even though such likely to be found in earlier stages of ISs and when onset terms also are used in the classification of other epilepsies. occurs at a younger age. The hypsarrhythmic pattern may The terms idiopathic and cryptogenic have been used in disappear during rapid-eye-movement (REM) sleep, but it some studies as synonyms of each other and as antonyms may be found with greater sensitivity in some other stages of symptomatic. of sleep (19). 20. Idiopathic infantile spasms. The term idiopathic in- 17. Modified and atypical hypsarrhythmia. With re- fantile spasms is used to describe cases in which ISs occur spect to the use of the term modified hypsarrhythmia (1,4) without any identifiable underlying cause, other neuro- and its synonym (or at least near-synonym) atypical hyp- logic signs, or symptoms. Such cases may be associated sarrhythmia (3), we endorse the recommendation of the with a family history. Although we agree that cases in

Epilepsia, Vol. 45, No. 11, 2004 1422 A. L. LUX AND J. P. OSBORNE this group may have good developmental outcomes, we choice for children with specific diagnoses, although we do not approve making eventual normal development part should be aware of the bias that might be introduced by of its definition, because we are reserving its use as part incomplete diagnostic information at the time of treatment of an etiologic classification for use at the time of study decision. We propose that, to make this clear, studies re- enrolment. serve the terms cryptogenic, idiopathic, and symptomatic 21. Cryptogenic infantile spasms. The term cryptogenic to refer to an etiologic classification made at the time of infantile spasms has been used to describe patients sus- randomization or treatment choice, and that they report the pected of being symptomatic but for whom an underlying final diagnostic classification in more-specific diagnostic structural or biochemical cause could not be identified. groups. We suggest a list of specific diagnostic groups in Features suggesting an unidentified underlying cause in- Appendix B. Of course, study-inclusion criteria or strati- clude preceding developmental delay, or other neurologic fied randomization might include more-specific diagnos- features, such as seizures. It has been suggested that the tic groups, such as tuberous sclerosis, but the study report term probably symptomatic is preferred to cryptogenic should make clear whether diagnostic reclassification of (15), although West Delphi participants had a range of cases has occurred by the time of study completion. views on this. Some prefer to retain the term cryptogenic; Reporting baseline characteristics in studies some prefer probably symptomatic; and some suggest that, of infantile spasms if we cannot identify the underlying cause, we ought to 26. Baseline characteristics in interventional studies. use the term nonsymptomatic rather than subjectively to We propose that the following baseline characteristics classify cases further. should be reported for each treatment or intervention 22. Symptomatic infantile spasms (ISs). The term symp- group: (a) etiologic group, if used for prerandomization tomatic ISs has been used to refer to cases in which the stratification; (b) sex; (c) age at onset (median and range, seizures result from an identifiable cause, and often to refer and proportions in each age category; see statement 27); also to cases in which neurologic features or an unequivo- (d) gestation (proportion born preterm, and median gesta- cal developmental delay precedes the onset of spasms. We tion); (e) birthweight (median birthweight and proportion suggest reserving the term symptomatic for cases with an <1,500 g); (f) preceding seizures; (g) concurrent and pre- identified underlying disorder, and classifying cases with vious treatment with AEDs or steroids; (h) lead time from neurologic symptoms, signs, or developmental delay, but apparent onset of spasms to treatment (median and range); no proven cause or etiology, as cryptogenic. (i) presence of hypsarrhythmia; and (j) unequivocally nor- 23. Specific etiologies. However constituted, the eti- mal development at apparent onset of spasms. Attention ologic subgroups discussed earlier will include children should be drawn to large differences that might bias the re- with different underlying causes, which may or may not be sults, even where treatments have been allocated randomly identified, and such aggregation into etiologic subgroups and these differences must have occurred by chance. is of limited value. However, within any one study, it is 27. Classification by age at onset of spasms. ISs typ- possible to obtain precise estimates of effect, and to per- ically have an onset between ages 3 and 12 months. form formal statistical tests for heterogeneity of effect, on Younger and older children are more likely to have other groups with specific diagnoses only if they are relatively conditions, and a greater risk of misclassification exists. commonly associated with ISs. An example of such com- Age at onset of spasms also may influence outcomes. We monly associated diagnoses is tuberous sclerosis. Greater propose that studies of ISs report outcomes for children in homogeneity of study design will help in aggregating data age at onset groups: (a) younger than 3 months (corrected about rare causes of ISs. for preterm delivery); (b) 3 months (corrected age) and up 24. Timing of etiologic classification. It is often unclear to 12 months; (c) 12 months and older. We propose that in study reports whether the classification into etiologic these be referred to by the terms: early onset, classic on- subgroups is based on diagnostic information known at set, and late onset, respectively. Onset would rarely occur the time of treatment decision or randomization, or infor- after age 2 years, but it is biologically implausible to set mation available at the time of study completion. If the an absolute upper age limit. data are to be used to inform future treatment choice, it is rational for the study to report outcomes according the Duration of treatment classification made at the time of treatment decision. This 28. Duration of treatment in interventional studies. We avoids the information bias introduced by reclassification propose that treatments for ISs should be considered to of cases between time of treatment decision and study have had a sufficient therapeutic trial after 14 days. In gen- completion. Initial classification is the natural choice for eral, it is necessary to give a treatment for 14 days before stratified analyses by intention-to-treat. deciding that it has no treatment effect, but discontinua- 25. We ought also to use the more complete and specific tion should be considered sooner if an adverse effect of diagnostic information that is available at the time of study that treatment appears. We suggest that any study that uses completion. This informs us about outcomes and treatment a different period of necessary and sufficient therapeutic

Epilepsia, Vol. 45, No. 11, 2004 WEST DELPHI GROUP CONSENSUS STATEMENT 1423 trial should state a rationale or empirical evidence for that with no witnessed spasms but continuing hypsarrhythmia decision. If adjunct treatments are used, the study protocol should be classified as primary responders. They suggest should state clearly how these might affect outcome. that a primary responder should have both cessation of spasms and resolution of hypsarrhythmia. We propose the Outcomes term primary electroclinical outcome to describe this com- 29. Primary clinical outcome. We propose that the pri- bination of outcomes, and we propose that it be regarded mary clinical outcome of studies of ISs is cessation of as an essential outcome (20). spasms, and we suggest a standard definition for this out- 34. Studies that report both primary clinical outcome come. Without qualification, the term cessation of spasms and primary electroclinical outcome provide important should denote that no clinical spasms have been witnessed information about how often the clinical response is as- from a time commencing within 14 days of treatment, and sociated with resolution of hypsarrhythmia. Such studies for a period of ≥28 consecutive days from the time of would permit a degree of extrapolation of data from stud- the last witnessed spasm. Studies of ISs should report the ies performed in places where access to neurophysiologic number and proportion of children in each treatment group equipment is more restricted and where the primary elec- who meet these criteria. troclinical outcome cannot be reliably determined. 30. Timing of EEG investigations. Although we do not have evidence for the clinical importance of cessation of Adverse events spasms and resolution of abnormal EEG findings, in terms 35. Deaths during study period. Studies should report of prognosis for development and function, we propose all deaths occurring during the study period, even when that it is essential to report the EEG status of children they are not regarded as attributable to treatment. Study re- at baseline and at follow-up. We suggest that an EEG be ports should state which adverse events were investigated, performed at least: how they were sought, and how they were classified. • before treatment is allocated; Outcome measures and analysis of data • between study days 14 and 21 (that is, within 7 days 36. We propose that at least the following outcomes and of the end of the period of sufficient therapeutic trial); measures, stratified by intervention group, be reported in and studies of ISs: • between study days 42 and 49 (that is, after the period • primary clinical response during which 28 days of freedom from spasms should • primary electroclinical response have occurred). • relapse-free primary responses (clinical and electro- 31. EEG classification. Resolution of hypsarrhythmia clinical, numbers and proportions) is a more appropriate EEG outcome than normalization of • continuing subtle spasms where clinical spasms have EEG, because with many underlying etiologies, we cannot ceased expect the background EEG to normalize even if epilepti- • distribution of time to relapse (from time of ran- form changes resolve. Progression to other seizure types domization, perhaps best represented graphically as a also may occur without any evidence of ISs or hypsar- variable proportion with relapse-free remission over rhythmia. time) • development at age 2 years (medians and ranges for Electroclinical outcomes tests chosen) 32. If less-sensitive investigations do not show hypsar- • deaths and other serious adverse events (numbers and rhythmia, we recommend that a sleep EEG be performed descriptions of events) to provide stronger evidence that the hypsarrhythmia has • presence of and progression to other seizure types resolved. Ideally, a video-EEG recording would include (numbers, proportions, and seizure types) sleep. No evidence suggests recommending a specific du- • nonserious adverse events (numbers, proportions, ration for video-EEG recording, but a 24-h period is typ- and description of events) ical, and some participants thought6hwould be as sen- sitive as 24 h. Although it might be inferred that the EEG 37. Analyzing different etiologic subgroups. We suggest remains abnormal in cases in which continuing spasms are that most outcomes are influenced more strongly by un- found, it also is useful to have information from EEGs in derlying disorders than by treatment. We recommend that children with continuing spasms. We regard this informa- investigators exercise caution in performing multiple tests tion as essential because it will help to inform us that the on such outcomes and attributing significant differences seizure type has not changed, about the prognostic impor- to treatments because, in addition to there being biases tance of EEG findings, and also about treatment effects introduced by multiple testing, the conclusions may lack on the EEG. biologic plausibility. However, one aim of treatment is to 33. Primary electroclinical response. Some West Del- improve development and function by limiting any dam- phi participants are uneasy with the idea that children age that might be caused by repeated spasms.

Epilepsia, Vol. 45, No. 11, 2004 1424 A. L. LUX AND J. P. OSBORNE

38. Analyzing time to cessation of spasms. Time to ces- assess intrarater and interrater agreement in classification sation of spasms can be represented by actuarial curves of EEGs. and is a clinically useful outcome that might inform fu- ture study designs. We suggest the convention of reporting DISCUSSION time to 28 days of spasm freedom for the primary clinical The rationale behind West Delphi was to propose stan- response. This is a useful and intuitive clinical measure dard case definitions, outcomes, and outcome measures that can be examined by using the statistical techniques of that will ease future study design; that facilitate compar- survival analysis. Measures of time to cessation of spasms ison of data from different studies; and that encourage are more reliable and informative than measures of time to studies to answer questions that are important to clinicians 50% reduction in ictal units. The primary electroclinical and families seeking information about treatment choices. response does not lend itself so readily to such analyses These aims are essential stages on the road to determin- because EEGs are performed in a cross-sectional fashion, ing the best treatments for children with ISs. The process whereas clinical observation is a more continuous mea- dissected many of the issues surrounding case definitions sure. and outcomes. Although substantial areas of consensus 39. Relapse. The following should be considered to con- exist, several areas require further discussion. We wish to stitute a clinical relapse of ISs at any stage after a primary emphasize that the primary aim of the consensus is to aid clinical response has been obtained: (a) any episode of the design of studies of ISs. spasms occurring in clusters; (b) two or more episodes of spasms that occur singly but not in clusters; and (c) subtle Case definitions for studies of infantile spasms spasms (if accompanied by an EEG showing appropriate We aimed to propose definitions that are congruent with changes). A single witnessed spasm would not be reliable previous usage and recent ILAE proposals. For example, and should not be classified as a relapse of clinical spasms, Roger and Dulac (17) used the terms infantile spasms and but investigators might wish to report their occurrence. West syndrome to refer to syndrome diagnoses, and they suggested the term epileptic spasms to refer to a seizure Role of pyridoxine (vitamin B ) 6 type. The West Delphi consensus evolved to have strong 40. Diagnostic role of pyridoxine. Pyridoxine- congruence with these definitions. dependent seizures are rare (21). They even more rarely However, we were prepared to make clear distinctions have the clinical manifestation of infantile spasms, al- and depart from previous practice where we thought that though this is thought to be most likely when other seizure this would be valid and useful. For example, we found types have occurred before the onset of spasms. Giving it valuable to use the term epileptic spasms to describe intravenous pyridoxine is associated with a risk of ap- the seizure type, rather than as a phenomenologic descrip- nea. It requires close monitoring and is not known to be tion, as proposed in the ILAE Glossary of Descriptive associated with rapid resolution of hypsarrhythmia. Oral Terminology, 2001 (16). Where the ILAE Proposed Di- pyridoxine seems to be associated with a median time to agnostic Scheme for People with Epileptic Seizures and response of several days. Thus incorporating a diagnostic with Epilepsy suggests the term spasms to describe the and therapeutic test of pyridoxine into study designs is seizure type, we propose making a distinction between challenging. clinical spasms, which describe seizure phenomenology, 41. Treatment role of pyridoxine. In addition to use in and epileptic spasms, the seizure type. We think that diagnosing potential pyridoxine-dependent seizures, pyri- these terms are more appropriate and that this use of doxine has been used more broadly as a treatment or ad- the modifiers clinical and epileptic has strong intuitive junct treatment for ISs. The Cochrane Review of treatment appeal. of ISs did not identify any evidence of treatment effect in Wegenerally found the proposed five-axis classification trials meeting its inclusion criteria (22,23). If pyridoxine of epilepsy, the overall diagnostic scheme of which was is used, the report should clearly state its timing, dose, proposed by the ILAE General Assembly in May 2001, to duration, and whether it is being used to exclude a diag- be potentially useful, and we described where our classifi- nosis of pyridoxine-dependent seizures or as a treatment cation is congruent with that classification. We wait to see intervention for ISs in its own right. if the details of those proposals will find broad approval, Areas of incomplete consensus and we suggest the changes we outline as modifications 42. Areas of incomplete consensus remain. In the fu- to the scheme (15). ture, we might wish to develop a more detailed struc- It seems that nobody would dispute the status of West ture for reporting adverse events and recommendations syndrome as an epilepsy syndrome, but initial disagree- on which tools to use to assess developmental outcomes ment was found about the status of developmental delay and to classify disability. Finally, we found it difficult to as a defining feature of WS. The consensus was that it provide a detailed description of the limits of definition is often difficult to determine when spasms started and of hypsarrhythmia: those designing studies must consider whether development was normal at that stage. Because the advantages of using several EEG assessors, and how to these are unreliably determined, and misclassification is

Epilepsia, Vol. 45, No. 11, 2004 WEST DELPHI GROUP CONSENSUS STATEMENT 1425 likely, we thought that developmental delay at onset of Although some participants expressed reservations, it spasms ought not to be a defining feature of WS. It was was agreed that cases of ISs might occur without hypsar- suggested that WS is defined as a new triad: spasms, clus- rhythmia. Some studies have reported hypsarrhythmia in ters and hypsarrhythmia. However, a consensus formed fewer than half of cases classified as having ISs at the time that, even though it might make a useful but informal aide of randomized treatment, (24), and it is well recognized memoire,« this triadic definition was too contrived to be that hypsarrhythmia is not found throughout the course of adopted formally. ISs. We debated the status of infantile spasms as a syn- With respect to ISSV without hypsarrhythmia, we drome, and also the status of infantile spasms single-spasm thought that we could not exclude these solely by defini- variant with hypsarrhythmia. An argument was reached tion if we are prepared to admit cases with single spasms for classifying these as seizure types, corresponding to with hypsarrhythmia and cases with clustered spasms Axis 2 of the proposed ILAE diagnostic scheme. How- without hypsarrhythmia. However, we would counsel cau- ever, we did not find this argument compelling and thought tion and clarity when including such cases in studies of ISs, that it is more appropriate to classify ISs and ISSV with and some contributors were certain that they would never hypsarrhythmia as epilepsy syndromes. The features that diagnose ISs in such cases. Spasms that are never observed define WS as a syndrome subgroup of ISs are clustering to cluster and that are not associated with hypsarrhythmia of spasms and hypsarrhythmia. Hypsarrhythmia remains are unlikely to be ISs, and if such cases are included in rather loosely defined and has traditionally been prone studies, we would expect the investigators to state clearly to reclassification when modifying features exist. Even their reasons for regarding these cases as valid. the status of a criterion to define nonclustering of spasms Some contributors thought that unequivocally normal requires empirical clarification, because at least one con- development at the time of onset of spasms is a good prog- tributor thought that spasms separated by longer than 1 nostic feature. They suggested that this should be recorded min might genuinely be part of a single cluster. and used for prognostic factor analysis in studies of ISs. As was predicted by several respondents to the paper It also was suggested that this reframing from apparently presented at the International Symposium on West Syn- delayed development to unequivocally normal develop- drome and Other Infantile Epileptic Encephalopathies, ment is useful when it comes to counseling families about Tokyo, February 2001, the task of defining necessary and prognosis. sufficient features of hypsarrhythmia proved very chal- It was evident from the process of eliciting a consen- lenging. We eventually agreed on a description of hyp- sus that clinicians and investigators have used the terms sarrhythmia rather than a strict definition of limits, and infantile spasms and West syndrome variably, and that evidently an element of Gestalt remains in recognizing great variation has been noted in the interpretation of this EEG pattern. We are not aware of studies of intrarater the terms cryptogenic, idiopathic, symptomatic, and non- and interrater agreement with classification of EEGs as symptomatic. Previous ILAE commissions have used the hypsarrhythmic. Such studies would inform future study terms idiopathic and cryptogenic as antonyms of symp- design and, we hope, would provide reassurance that mis- tomatic (5,6). Participants debated these terms a great deal. classification of EEG patterns is not a substantial prob- The eventual consensus was that information about prog- lem. Meanwhile, those designing studies might consider nosis and treatment response related to specific underlying reducing bias and misclassification by, for example, using diagnoses is probably more useful than classification into EEG assessment by two independent, treatment-blinded etiologic subgroups that are themselves heterogeneous. investigators. Wealso discussed the issue of potential information bias We have followed the proposal made by the 1991 ILAE related to reclassification of cases between enrolment in Workshop in recommending that modifying features be a study and the time of study completion. We believe that listed separately rather than regarded in the aggregate as clearly distinguishing between an etiologic classification forming a distinct group of EEG types (7). Individual made at the time of study enrolment and a diagnostic clas- modifying features are listed in Appendix A. Modified sification made at the time of study completion will allow EEGs may bear more or less resemblance to “typical” those using information from studies to be clearer about or “classic” hypsarrhythmia, and we think that it is irra- what type of diagnostic information tends to be available tional to dichotomize cases on the basis of presence of any at the time of treatment decision. one or more modifying features. EEG modification may Some participants wished to emphasize the limited be related to a missed temporal window of opportunity value of aggregating cases into etiologic subgroups, pre- for catching a “typical” recording, because increased syn- ferring to look instead at outcomes in groups of children chrony and organization is related to brain development with the same specific underlying disorder. For example, or sleep state, or it may relate to underlying causes or trig- tuberous sclerosis seems to have a better response to treat- gers: for example, increased focality where the trigger is ment with vigabatrin do than other cases of symptomatic a cortical malformation or tuber. infantile spasms.

Epilepsia, Vol. 45, No. 11, 2004 1426 A. L. LUX AND J. P. OSBORNE

One controversial area was that of subtle seizures as- eling time to cessation of spasms. This would preserve a sociated with hypsarrhythmia but without evidence of degree of study standardization that would facilitate com- spasms. Whereas eye or subtle distal limb movements parison and aggregation of data from studies. In the longer may be regarded as evidence of continual seizure ac- term, such an approach would provide better information tivity, they do not provide a reliable outcome for the for clinicians and families. purposes of clinical studies because they are likely to We included three categories of age at onset (or more be missed without careful observation. In the absence correctly apparent onset, because timing can be unreli- of hypsarrhythmia, these subtle movements might be able) not to define further subgroups but rather to facilitate classified as focal seizures. However, it is important to its use as a prognostic factor. recognize cases that have interictal hypsarrhythmia and With respect to secondary outcomes, strong support was typical ictal EEG changes without clinical spasms. Al- found for other recommendations, and in particular, for though we suggest classifying these as HWIS, little doubt developmental assessment at age 5 years rather than the exists that some of them still effectively have the electro- more conservative 2 years stated in the consensus. We have clinical syndrome of ISs without the typical clinical mani- limited the statement to what we consider to be a minimal festation. Again, we think that having a specific definition standard of information. and term to describe such cases will allow their appropri- A good deal of negotiation occurred about the most ap- ate inclusion in studies without confusing the reader. How- propriate primary outcome. Most participants were keen ever, we did think that such subtle movements were suf- on emphasizing the clinical response, but others were ficient to define electroclinical nonresponse—they would sceptical at the idea of suggesting that a response had presumably be associated with an EEG correlate; other- occurred when the EEG still showed hypsarrhythmia. wise no evidence would support the idea that such subtle The final consensus is that two proper primary outcomes movements were epileptic—or electroclinical relapse in are found: the primary clinical outcome and the pri- cases in which spasms had ceased. mary electroclinical outcome, as defined earlier. We think that the approach of reporting both outcomes has many Outcome measures for studies of infantile spasms advantages. The latter part of the consensus statement (sections 26 onward) addresses the issue of standardized outcomes and outcome measures for studies of ISs. Most published stud- Summary ies have provided incomplete information about even es- Although higher estimates have been made, worldwide sential outcomes, such as how they defined clinical or elec- probably ∼20,000 to 50,000 cases of infantile spasms are troclinical response. In the early rounds of West Delphi, seen each year, and very few of these are enrolled in formal it became apparent that most clinicians would not regard studies (25). Reliably and convincingly to evaluate treat- cases relapsing within a few weeks as primary clinical ments and to investigate potential variation in responses responders. However, they agreed that spasms recurring in different diagnostic groups, we should consider more after 28 days of absence are described reasonably as re- collaborative and international studies and aggregate data lapse rather than as failed primary response. from disparate studies. We hope that West Delphi is a use- One question that emerged was that of defining peri- ful contribution to this process. ods that constitute necessary and sufficient durations of The clinicians and investigators who contributed to treatment. The consensus was that 14 days was a suit- West Delphi have an interest in ISs. The primary interest able duration for both—at least, with current treatments. of many is that of making treatment decisions and using In other words, we agreed that a treatment has had suffi- reliable and readily interpretable published information to cient opportunity to prove its effect by 14 days, and that make those decisions. Many of the contributors also have it is necessary to give a treatment for this duration before an interest in designing and contributing to studies of ISs. concluding that it has failed to show any treatment effect. The process involved novel suggestions and lively debate. These limits are arbitrary to some degree, and good A strict interpretation of the guidelines for a Delphi pro- reasons might be found for a new study to use a different cess would not have permitted conferring by West Delphi duration. However, we think it is useful to suggest a stan- contributors who were also members of the Konigsteiner¬ dard, and we would invite investigators to explain their Kreis. The Konigsteiner¬ Kreis is a group of German and reasons for choosing a different period. A danger exists Swiss doctors with an interest in pediatric epilepsy who that an extra degree of subjectivity enters study design and deferred completion of an ongoing consensus statement makes aggregation and use of information more difficult while two of its members contributed to West Delphi. One if, for example, investigators hoped to favor a drug that of those members sought the opinions of other members is reputed to act more quickly than another. We believe of that group, but the response to West Delphi was counted that such questions would be better investigated by other as a single response to prevent any bias being introduced means than altering the primary outcome, such as mod- from coordinated responses.

Epilepsia, Vol. 45, No. 11, 2004 WEST DELPHI GROUP CONSENSUS STATEMENT 1427

These proposals are not intended to be immutable Appendix B: A suggested list of diagnostic categories guidelines for study design in ISs, but rather to be an aid This classification uses the 1996 pediatric adaptation of to encourage the collection of clear, consistent, and reli- ICD-10. able information. We would be pleased if they stimulate research that later refines some or all of these proposals. Brain Neoplasm Some of the proposals rest on arbitrary but agreed lim- its, such as the number of days of spasm freedom that is generally regarded as constituting a primary clinical re- Malignant C71 Benign D33 sponse. Future studies using these proposed case defini- tions and outcomes are likely to have wide acceptability and applicability. Inevitably, some participants have ac- Endocrine/Metabolic cepted, in the cause of consensus, definitions and ideas This grouping specifically excludes transient and spe- that would not be their personal first choices. cific conditions of the newborn that are classified under Studies cannot answer all questions put by all people, perinatal. and we suggest that these proposals are better regarded as a minimal standard than as sufficient for any study. We Hypoglycemia E16 hope that constructive criticism of these proposals will Classic phenylketonuria E70 lead to refinement and improved consensus, and that it Organic acidurias E71 will, in the meantime, provide a framework for the design Aminoacidurias E72 Enzyme deficiencies, including biotinidase, Krabbe, of informative and valuable studies of ISs. Leigh’s, pyridoxine, adrenoleucodystrophy, etc. Acknowledgment: We thank the following: Professor Yukio Fukuyama (Tokyo) for his warm reception of the idea behind West Delphi and the invitation to Professor Osborne to speak Nervous System at the 2001 Tokyo Symposium on Infantile Spasms and West This grouping also excludes conditions mentioned else- Syndrome. We thank the members of the Konigsteiner¬ Kreis [B. where, including perinatal. Schmitt (Zurich),¬ U. Brandl (Jena), R. Korinthenberg (Freiburg), I. Krageloh-Mann¬ (Tubingen),¬ G. Kurlemann (Munster),¬ D. Rat- ing (Heidelberg), M. Wolff (Tubingen)],¬ who had already pre- Meningitis GO1ÐO2 pared a consensus paper in German on a similar theme. Two of Encephalitis GO5 its members made substantial contributions to West Delphi and Cerebral abscess GO7 drew on ideas that had been developed in the Konigsteiner¬ Kreis Cerebral palsy (unknown cause ONLY) G80Ð83 consensus. We thank Dr. Peter Baxter (Sheffield) for informa- Porencephaly G93 tion about the use of pyridoxine; Heather Hill (Bath) for able Other (ONLY if unable to classify elsewhere; assistance with secretarial tasks; Cow & Gate, Ltd., Bishops- e.g., leucomalacia not due to perinatal cause). gate Financial Management, and the Bath Unit for Research in Paediatrics (BURP) for assistance with funding. Cerebrovascular Disease Appendix A: List of EEG features that are regarded This grouping also excludes conditions mentioned else- as modifying features of hypsarrhythmia where, including perinatal. These features are traditionally regarded as defining Cerebral hemorrhage I60Ð62 modified hypsarrhythmia, but we recommend listing indi- Cerebral infarct or stroke I63, 64 vidual features. One also should note that many of these features are matters of degree rather than absolute differ- ences. Perinatal Asymmetry Maternal factors (e.g., drug abuse) P04 Consistent focal discharge Birth trauma, including hypoxicÐischemic Episodes of voltage attenuation (local, regional, or gen- encephalopathy (HIE) due to trauma P10Ð11 eral) Intrauterine asphyxia (HIE) P20Ð21 Infections: meningitis, cytomegalovirus (CMV), Excessive rapidity toxoplasmosis, herpes P35Ð36 Excessive slowing Intracranial nontraumatic hemorrhage Fragmentation Transient endocrine or metabolic diseases of the newborn Increased interhemispheric synchronization (e.g., hypoglycemia) P70Ð74 Other: Periventricular hemorrhage/leucomalacia Increased periodicity from preterm injury P91 Predominant high-voltage, bilaterally asynchronous slow HypoxicÐischemic injury of uncertain cause activity

Epilepsia, Vol. 45, No. 11, 2004 1428 A. L. LUX AND J. P. OSBORNE

Congenital als (and epidemiological studies) in West syndrome. Brain Dev 2001;23:677Ð82. 9. Cooke RM. Experts in uncertainty: opinion and subjective proba- bility in science. New York: Oxford University Press, 1991. Malformations Agenesis of the corpus callosum Q 04 10. Meyer MA, Booker JM. Eliciting and analyzing expert judgment. (if not chromosomal) Agyria/polygyria Philadelphia: Society for Industrial and Applied Mathematics, 2001. Cortical dysplasia 11. Vijayaraghavan L, Krishnamoorthy ES, Brown RG, et al. Abulia: a Heterotopia Delphi survey of British neurologists and psychiatrists. Mov Disord Holoprosencephaly 2002;17:1052Ð7. Hydrocephalus 12. Herdman M, Rajmil L, Ravens-Sieberer U, et al. Expert consen- Microcephaly Q 02 sus in the development of a European health-related quality of life Other malformations Incontinentia pigmenti Q 82 measure for children and adolescents: a Delphi study. Acta Paediatr (specific diseases) Neurofibromatosis Q 85 2002;91:1385Ð90. Tuberous sclerosis Q 85 13. Burns T, Fiander M, Audini B. A Delphi approach to characteris- Hypomelanosis of Ito Q 92 ing “relapse” as used in UK clinical practice. Int J Soc Psychiatry Chromosomal Down syndrome Q 90 2000;46:220Ð30. Other: please specify 14. Meijer R, Ihnenfeldt D, Vermeulen M, et al. The use of a modified Delphi procedure for the determination of 26 prognostic factors in the sub-acute stage of stroke. Int J Rehabil Res 2003;26:265Ð 70. Signs and Symptoms not otherwise specified 15. Engel J Jr. A proposed diagnostic scheme for people with This section would include information gained from epileptic seizures and with epilepsy: report of the ILAE Task Force on Classification and Terminology. Epilepsia 2001;42:796Ð imaging that is not in itself diagnostic, for example: 803. Cortical atrophy: diffuse 16. Blume WTC, Luders HO, Mizrahi E, et al. Glossary of descriptive Cortical atrophy: focal terminology for ictal semiology: report of the ILAE Task Force on Classification and Terminology. Epilepsia 2001;42:1212Ð8. Hemimegalencephaly 17. Roger J, Dulac O. West syndrome: history and nosology. In: Dulac O, Chugani HT, Dalla Bernardina B, eds. Infantile spasms and West External Injury Syndrome. London: WB Saunders, 1994:6Ð11. Intracranial injury, for example, due to nonaccidental 18. Gobbi G, Bruno L, Pini A, et al. Periodic spasms: an unclassi- injury or trauma. fied type of epileptic seizure in childhood. Dev Med Child Neurol 1987;29:766Ð75. 19. Dalla Bernardina B, Watanabe K. Interictal EEG: variations and pitfalls. In: Dulac O, Chugani HT, Dalla Bernardina B, eds. Infantile REFERENCES spasms and West Syndrome. London: WB Saunders, 1994:63Ð81. 20. Baram TZ, Mitchell WG, Tournay A, et al. High-dose corticotropin 1. Gibbs FA, Gibbs EL. Infantile spasms. In: Atlas of elec- (ACTH) versus prednisone for infantile spasms: a prospective, ran- troencephalography: epilepsy. Cambridge, MA: Addison-Wesley, domized, blinded study. Pediatrics 1996;97:375Ð9. 1952:24Ð30, 52. 21. Baxter P. Pyridoxine dependent and pyridoxine responsive seizures. 2. Shewmon D. Ictal aspects with emphasis on unusual variants. In: In: Baxter P, ed. Vitamin responsive conditions in paediatric neu- Dulac O, Chugani HT, Dalla Bernardina B, eds. Infantile spasms rology. London: Mac Keith Press, 2001:109Ð65. and West syndrome. London: Saunders, 1994:36Ð51. 22. Hancock E, Osborne JP, Milner P. The treatment of West syndrome: 3. Gastaut H. Clinical and electroencephalographical classification of a Cochrane review of the literature to December 2000. Brain Dev epileptic seizures. Epilepsia 1970;11:102Ð13. 2001;23:624Ð34. 4. Hrachovy RA, Frost JD Jr, Kellaway P. Hypsarrhythmia: variations 23. Hancock E, Osborne JP, Milner P. Treatment of infantile spasms on the theme. Epilepsia 1984;25:317Ð25. (Cochrane Review): The Cochrane Library. Oxford: Update Soft- 5. Commission on Classification and Terminology of the International ware, 2002. League Against Epilepsy. Proposal for classification of epilepsies 24. Chiron C, Dumas C, Jambaque« I, et al. Randomized trial comparing and epileptic syndromes. Epilepsia 1985;26:268Ð78. vigabatrin and hydrocortisone in infantile spasms due to tuberous 6. Commission on Classification and Terminology of the International sclerosis. Epilepsy Res 1997;26:389Ð95. League Against Epilepsy. Proposal for revised classification of 25. Dulac O, Soufflet C, Chiron C, et al. What is West syndrome? In: epilepsies and epileptic syndromes. Epilepsia 1989;30:389Ð99. Schwartzkroin P, Rho JM, eds. Epilepsy, infantile spasms, and devel- 7. Commission on Pediatric Epilepsy of the International League opmental encephalopathy. Amsterdam: Academic Press, 2002:1Ð Against Epilepsy. Workshop on infantile spasms. Epilepsia 22. 1992;33:195. 26. Friedman E, Pampiglione G. Prognostic implications of electroen- 8. Osborne JP, Lux A. Towards an international consensus on def- cephalographic findings of hypsarrhythmia in first year of life. Br initions and standardised outcome measures for therapeutic tri- Med J 1971;4:323Ð5.

Epilepsia, Vol. 45, No. 11, 2004