clinical and eeg features of patients with eeg wicket rhythms misdiagnosed with epilepsy

DOI 10121201WNL000016399197456032005641879-1883 Neurology

G L Krauss A Abdallah R Lesser et al misdiagnosed with epilepsy

Clinical and EEG features of patients with EEG wicket rhythms

This information is current as of June 13 2005

httpwwwneurologyorgcontent64111879fullhtmllocated on the World Wide Web at

The online version of this article along with updated information and services is

0028-3878 Online ISSN 1526-632Xsince 1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN

reg is the official journal of the American Academy of Neurology Published continuouslyNeurology

Clinical and EEG features of patients withEEG wicket rhythms misdiagnosed

with epilepsyGL Krauss MD A Abdallah BA R Lesser MD RE Thompson PhD and E Niedermeyer MD

AbstractmdashBackground EEG wicket rhythms are 6- to 11-Hz medium-to-high voltage bursts that are sometimes misiden-tified as epileptogenic activity The authors determined the clinical and EEG features of patients with wicket rhythmswho had been incorrectly diagnosed with epilepsy Methods Electroencephalographers at an epilepsy center re-read EEGsfor patients referred for epilepsy management and identified patients with wicket rhythms On further evaluation themajority (54 2546) of these patients were found not to have epilepsy The authors compared the clinical and EEGfeatures for the 25 patients with wickets and nonepileptic episodes with those of age- and sex-matched patients withpartial-onset epilepsy using univariate and multivariate analysis Results Several features distinguished patients withEEG wicket patterns and nonepileptic episodes (n 25) from age- and sex-matched patients with epilepsy (n 25)mid-adult age at onset of episodes (mean 384 years vs 198 years) prolonged clinical episodes (mean 155 minutes vs 23minutes) and long duration of EEG wicket patterns (mean 066 seconds vs 011 second spikes) After controlling for otherfactors patients without major confusion during episodes were unlikely to have epilepsy Conclusion Wicket patterns areoften interpreted as epileptogenic This distinctive EEG pattern should be considered in patients with clinical episodesatypical for epilepsy

NEUROLOGY 2005641879ndash1883

Patients with nonepileptic episodes may be misdiag-nosed as having epilepsy when EEG wicket rhythmsare misinterpreted as interictal spikes and sharpwaves1 Wicket rhythms are ldquonormal variantrdquo 6- to11-Hz medium-to-high voltage EEG bursts that haveno epileptogenic potential23 We determined the fea-tures of clinical episodes and EEG that distinguishedpatients with nonepileptic episodes and EEG wicketpatterns from patients with partial-onset epilepsyand epileptogenic activity on EEG Patients mightavoid delayed diagnosis and unnecessary exposure toantiepilepsy drugs if neurologists more accuratelydiscriminate between wicket patterns and epilepto-genic activity

Methods Patients previously diagnosed with epilepsy wereevaluated at the Johns Hopkins outpatient epilepsy clinic Pa-tientsrsquo previous EEG tracings were obtained from other EEGlaboratories and ldquoblindlyrdquo re-read by Johns Hopkins electroen-cephalographers (GLK or RPL) During a 6-year period (Janu-ary 1996 to December 2002) we evaluated 2274 new patients 46patients had previous EEG available and had EEG wicket pat-terns The patients with EEG wicket patterns had been diagnosedwith epilepsy prior to their Johns Hopkins visit Twenty-one of thepatients with wicket patterns also had epileptogenic patterns onthe previously performed EEGs or had seizures confirmed withprolonged video EEG monitoring The remaining 25 (54) pa-tients had EEG wicket patterns but no true spikes or sharpwaves on previous EEGs and had nonepileptic episodes We per-formed one-to-one matching of these 25 ldquowicketrdquo patients withpatients with partial seizures based on age at date of EEG (3years) and sex Partial-seizure disorders were confirmed by theepileptogenic patterns on EEG or video-EEG (interictal spikes and

sharp waves or rhythmic ictal patterns) For both patients andmatches we collected demographic information a description ofpatientsrsquo clinical episodes (precipitants duration presence of pro-dromal symptoms presence of confusion and motor signs) andEEG findings characteristic of wicket and interictal spikes andsharp waves (frequency amplitude duration) Clinical informa-tion such as seizure etiology and MRI was collected via chartreview and an extensive questionnaire The presence of headtrauma was defined as a history of major head injury (eg skullfracture cerebral contusion on imaging) or trauma causing loss ofconsciousness

All statistical analyses were performed using the STATA 70statistical software package Simple descriptive statistics and uni-variate analyses were performed on the clinical and demographicvariables in the data set Simple t tests with the Bonferroni cor-rection for multiple measures were used to compare mean differ-ences between symptoms and signs for wicket and epilepticpatients for the continuous variables Outcome data that wereskewed such as episode duration were log transformed Becauseof a low sample size Fisher exact test was used to compare cate-gorical and dichotomous outcomes between wicket and epilepticpatients Since the data were clustered by case-control pairs con-ditional logistic regression was used in the multivariate analysisThe conditional logistic regression analysis could not be performedfor models that contained outcomes with no variability withinmatched case-control clusters In these cases the HuberndashWhitesandwich estimator of variance along with clustering of observa-tions within case-control cohorts was used4

Results Twenty-five of 46 patients (54) had wicketrhythms misinterpreted as epileptogenic activity all ofthese 25 patients had nonepileptic clinical episodes The 25patients with wicket patterns and nonepileptic episodestypically developed clinical episodes in mid-life mean 384years (95 CI 325 443) compared to late teen and earlyadult years for matched patients with complex partial sei-

From the Departments of Neurology (Drs Krauss Lesser and Niedermeyer and A Abdallah) and Biostatistics (Dr Thompson) School of Public HealthJohns Hopkins University Baltimore MDReceived December 3 2004 Accepted in final form March 3 2005Address correspondence and reprint requests to Dr Gregory Krauss Meyer 2-147 600 N Wolfe St Baltimore MD 21287 e-mail gkraussjhmiedu

Copyright copy 2005 by AAN Enterprises Inc 1879

zures mean 198 years (136 259) p 00001 (table 1)The clinical episodes for the patients with wicket rhythmswere much longer in duration (mean 1558 minutes) com-pared to patients with complex partial seizures (mean 24minutes log duration p 001) The average frequency ofpatientsrsquo clinical episodes was similar for patients withwickets (mean 21 per year [043 426]) and patients withcomplex partial seizures (mean 237 seizures per year [21452] p 09) The duration of wicket rhythm bursts werelonger (mean 066 seconds [05 08]) compared to interictalepileptogenic patterns (mean 011 seconds [009 012] p 0001) EEG frequencies were similar between wicket andinterictal epileptogenic patterns with a slightly slower fre-quency for wicket rhythms (93 Hz [835 1021]) comparedto interictal epileptogenic patterns (108 Hz [97 12] p 005)

There was considerable heterogeneity in clinical symp-toms across patients in both groups (table 2) Signs andsymptoms that differed between patients with wicket pat-

terns and patients with seizures were as follows sensorysymptoms (mostly limb paresthesias) were uncommon inpatients with epilepsy while patients with partial-onsetseizures were much more likely to have confusion and oralautomatisms Fainting symptoms occurred only in patientswith wicket patterns Both groups frequently reported con-fusion and histories of head trauma Patients who experi-enced confusion however were 10 to 40 times more likelyto have epilepsy after controlling for other covariates Thisresult is significant in models that control for age at symp-tom onset and EEG frequency or EEG frequency aloneand approaches significance when controlling for age atonset alone The majority (52) of patients with epilepsyhad abnormal MRI (most commonly hippocampal atrophylow grade tumor encephalomalacia) compared to 16 ofpatients with wicket rhythms (nonspecific white matterchanges one subarachnoid cyst)

An increase in the age at onset resulted in a decreasedlikelihood of having epilepsy with a predicted OR of ap-

Table 1 Clinical and EEG characteristics for patients with wicket patterns and epilepsy

Wicket patients Epilepsy patients p Value

Female 840 720Head trauma 280 240Log (episode duration) min 236 (122 350) 050 (018 083) 00013Episode frequencyy 2150 (043 4257) 2365 (213 4517) 08835EEG duration s 066 (052 081) 011 (009 012) 00001EEG frequency Hz 928 (835 1021) 1084 (970 1198) 00486MRI

Abnormal 160 520 0025daggerNormal 600 360No MRI 240 120

Values are or mean (95 CI) p Values from the t test

The natural log was included since episode duration was highly skeweddagger p Value from Fisher exact test

Table 2 Symptoms and signs for patients with wicket activity and epilepsy

Variable Wicket patients n 25 Epilepsy patients n 25 p Value

Motor activity 00 200 0050Confusion 560 960 0001Psychic 200 40 0189Sensory 560 40 0001Oral automatisms 00 400 0001Nausea 160 00 0110Staring 40 240 0098Fainting 320 00 0004Shaking 200 00 0050Weakness 200 00 0050Speech arrest 00 120 0235Vomiting 80 00 0490Aura prior 120 400 0051Other Rapid breathing decreased

ability to concentrate burningsensation headache

Collapse activity arrest olfactoryaura wandering

Group differences are evaluated by Fisher exact test Values are

Significant with Bonferroni correction for multiple measures

1880 NEUROLOGY 64 June (1 of 2) 2005

proximately 09 for every year increase in the age at onset(figure 1) This is significant (p 0001) after controllingfor confusion and EEG frequency

Patients often had single discharges or brief bursts ofwicket activity that appeared morphologically similar tointerictal spikes and sharp waves (figure 2 A and B) Allpatients with wicket activity however also had long focalruns of semirhythmic 6- to 11-Hz activity that containedwickets and distinguished their EEG from interictal spikesand sharp waves and ictal EEG patterns (figure 3 A andB) EEG waveform durations greater than 026 secondsindicated a wicket pattern with a corresponding 100 sen-sitivity and specificity for this clinical group

Discussion Epilepsy was incorrectly diagnosed ina group of patients with clinical episodes that wereatypical for seizures the misinterpretation of EEGwicket patterns contributed to patientsrsquo misdiagno-sis While this is a retrospective study the case-control findings provide clinical clues that may helpclinicians distinguish patients with nonepileptic epi-

sodes and wicket patterns from patients withepilepsy

EEG wicket patterns are usually benign and donot suggest the presence of epilepsy5 Within a singleEEG recording wickets may range in appearancefrom single sporadic spikes to trains of arciform dis-charges6 Trains of wicket activity typically have acrescendo-decrescendo envelope and can often befound bilaterally over temporal regions though notnecessarily on both sides at the same time Singlewicket spikes are those that are commonly misinter-preted as temporal spikes or sharp waves due totheir similarity in appearance6 Interictal spikes andsharp waves often have a following slow wave thatdistinguishes them from isolated wickets In our se-ries patients with brief wicket patterns that mightbe mistaken for interictal spikes and sharp wavesalso had long runs of 6- to 11-Hz semirhythmic activ-ity over the same temporal regions Interictal spikesand sharp waves were all brief in duration for ourpatient sample (026 seconds) a factor that distin-guished patients with partial seizures from thosewith wickets Occasionally patients with partial sei-zures may have longer duration bursts of focal epi-leptogenic activity (spikes sharp waves andparoxysmal slowing)

Wicket patterns are most common in people overage 33 years7 this is in accordance with the mid-adult age at onset of episodes seen in the nonepilep-tic group in this study It has been reported thatwicket activity may be associated with cerebrovascu-lar disease in some patients8 however this associa-tion has not been confirmed in controlled studies andis most commonly reported in patients with EEGwickets and focal slowing (ldquominor sharp and slowpatternrdquo)3910 Wicket activity has been classified as anormal variant EEG pattern This term is meant to

Figure 1 Distribution of age at onset of clinical episodes(n 25 for each group)

Figure 2 (A) Wicket activity ina patient with episodic dizzi-ness There is rhythmic 6-Hzactivity which occurs in briefbursts and longer trains Thiswicket activity is maximal overthe left temporal area but wasalso present on the right Thepatient has a history of a rightcerebrovascular accident (B)Right temporal sharp wave in apatient with complex partialseizures and MRI evidence ofright mesial temporal sclerosisUnlike the wicket activity thesharp wave occurs in isolationwithout a buildup of arciformwaveforms

June (1 of 2) 2005 NEUROLOGY 64 1881

help de-emphasize an association between wicket ac-tivity and epilepsy

Following evaluation at our epilepsy clinic studypatients initially diagnosed with epilepsy were foundto have alternative diagnoses most commonly near-syncope psychogenic nonepileptic seizures anxietyhyperventilation migraine or postconcussive syn-dromes Many patients had typical symptoms of syn-cope or near-syncope postural light-headedness orunconsciousness slumping or falling occasionalstiffening or jerking but no full convulsions pale-ness and nausea1112 Many patients also had nonepi-leptic psychogenic seizures with features atypical forseizures such as retained alertness despite bilateralmotor signs and repeated sensory episodes lastingmore than 4 minutes13 These symptoms should sug-gest the possibility of diagnoses other than epilepsywhen they occur

A limitation of this study is the retrospective de-sign and the limited sample size During rereading ofprevious EEGs records may have been classified asshowing wicket activity rather than sharp waves orspikes based on patientsrsquo atypical clinical presenta-tions Wicket activity and epileptogenic activityhowever were confirmed on readings of additionalEEG with the electroencephalographer blinded tothe clinical history and previous EEG findingsMoreover 46 of patients with wicket activity alsohad spikes and sharp waves on their previous andJohns Hopkins University EEGs These patientswere confirmed to have epilepsy Finally the referralpopulation from which the patients for this studywere chosen is a limitation

Neurologists often noted in their consultationsthat they were uncertain of the etiology of patientsrsquo

clinical episodes due to their atypical features nev-ertheless they subsequently relied on incorrect EEGinterpretations in making a diagnosis of epilepsyNonepileptic conditions are frequently misdiagnosedas epilepsy and this error can lead to many prob-lems such as frustration for the patient when themisdiagnosis is realized a delay in appropriate diag-nosis and treatment and inappropriate treatmentwith antiepileptic drugs1114 Several patients wereunnecessarily exposed to antiepileptic drugs for 2 ormore years Antiepileptic drugs are costly and pa-tients without epilepsy need not tolerate the sideeffects of such drugs which are unlikely to controltheir episodes15

Although the EEG is a useful diagnostic test forepilepsy incorrect interpretations may lead to incor-rect diagnoses While spikes and sharp waves arerelatively specific for epilepsy they are sometimesdifficult to identify correctly some patterns includ-ing wickets appear similar in appearance to interic-tal epileptogenic activity Physicians should becautious in diagnosing epilepsy in the presence ofatypical clinical episodes combined with uncertainEEG findings16 A useful clinical technique for deter-mining whether brief arciform transients are actu-ally fragments of a wicket rhythm is to see whethersimilar but prolonged wicket rhythms are presentelsewhere in the EEG (see figure 3 A and B) Somepatients may benefit from additional EEG recordingto see whether brief transients are components of awicket rhythm or conversely to see if true spikesand sharp waves might appear A better understand-ing of normal variant patterns such as wickets iscrucial to the prevention of epilepsy misdiagnosis8

Figure 3 There is a brief burst of wicket activity in the left and right temporal leads (A) Six seconds later there is moreprolonged wicket activity (B) which is recorded maximally over the left temporal lobe The more prolonged wicket activityhelps distinguish the brief bursts of wicket activity from interictal spikes and sharp waves


References1 Benbadis SR Tatum WO Overinterpretation of EEGs and misdiagnosis

of epilepsy J Clin Neurophysiol 20032042ndash442 Westmoreland BF Epileptiform electroencephalographic patterns

Mayo Clin Proc 199671501ndash511 Review3 Asokan G Pareja J Niedermeyer E Temporal minor slow and sharp

EEG activity and cerebrovascular disorder Clin Electroencephalogr198718201ndash210

4 StataCorp 2001 Stata Statistical Software Release 70 Vol 1 CollegeStation TX Stata Corporation 2001209

5 Klass DW Westmoreland BF Nonepileptogenic epileptiform electroen-cephalographic activity Ann Neurol 198518627ndash635

6 Reiher J Lebel M Wicket spikes clinical correlates of a previouslyundescribed EEG pattern Can J Neurol Sci 1977439ndash47

7 Batista MS Coelho CF de Lima MM Silva DF A case-control study ofa benign electroencephalographic variant pattern Arq Neuropsiquiatr199957561

8 Kellaway P Orderly approach to visual analysis elements of the nor-mal EEG and their characteristics in children and adults In Ebersole

JS Pedley TA eds Current practice of clinical electroencephalography3rd ed Philadelphia Lippincott Williams amp Wilkins 2003127ndash132

9 Bruens JH Gastaut H Giove G Electroencephalographic study of thesigns of chronic vascular insufficiency of the Sylvian region in agedpeople Electroencephalogr Clin Neurophysiol 196012283ndash295

10 Koshino Y et al Temporal minor slow and sharp activity in psychiatricpatients Clin Electroencephalogr 199021225ndash232

11 Smith D et al The misdiagnosis of epilepsy and the management ofrefractory epilepsy in a specialist clinic QJM 19999215ndash23

12 Bergfeldt L Differential diagnosis of cardiogenic syncope and seizuredisorders Heart 200389353ndash358

13 Fleisher W et al Comparative study of trauma-related phenomena insubjects with pseudoseizures and subjects with epilepsy Am J Psychia-try 2002159660ndash663

14 Scheepers B Clough P Pickles C The misdiagnosis of epilepsy find-ings of a population study Seizure 19987403ndash406

15 Chadwick D Smith D The misdiagnosis of epilepsy BMJ 2002324495ndash496

16 Worrell GA Lagerlund TD Buchhalter JR Role and limitations ofroutine and ambulatory scalp electroencephalography in diagnosingand managing seizures Mayo Clin Proc 200277991ndash998 Review

ACTIVATE YOUR ONLINE SUBSCRIPTIONAt wwwneurologyorg subscribers can now access the full text of the current issue of Neurology and back issues to 1999Select the ldquoLogin instructionsrdquo link that is provided on the Help screen Here you will be guided through a step-by-step ac-tivation process

Neurology online offersbull Access to journal content in both Adobe Acrobat PDF or HTML formatsbull Links to PubMedbull Extensive search capabilitiesbull Complete online Information for Authorsbull Examinations on designated articles for CME creditbull Access to in-depth supplementary scientific data



G L Krauss A Abdallah R Lesser et al with epilepsy

Clinical and EEG features of patients with EEG wicket rhythms misdiagnosed


ServicesUpdated Information amp

httpwwwneurologyorgcontent64111879fullhtmlincluding high resolution figures can be found at

References

1httpwwwneurologyorgcontent64111879fullhtmlref-list-at This article cites 14 articles 3 of which you can access for free

Citations

icleshttpwwwneurologyorgcontent64111879fullhtmlotherartThis article has been cited by 8 HighWire-hosted articles

Subspecialty Collections

httpwwwneurologyorgcgicollectioneeg_EEG

httpwwwneurologyorgcgicollectionall_epilepsy_seizuresAll EpilepsySeizuresfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpwwwneurologyorgmiscaboutxhtmlpermissionsor in its entirety can be found online atInformation about reproducing this article in parts (figurestables)

Reprints

httpwwwneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

Clinical and EEG features of patients withEEG wicket rhythms misdiagnosed

with epilepsyGL Krauss MD A Abdallah BA R Lesser MD RE Thompson PhD and E Niedermeyer MD

AbstractmdashBackground EEG wicket rhythms are 6- to 11-Hz medium-to-high voltage bursts that are sometimes misiden-tified as epileptogenic activity The authors determined the clinical and EEG features of patients with wicket rhythmswho had been incorrectly diagnosed with epilepsy Methods Electroencephalographers at an epilepsy center re-read EEGsfor patients referred for epilepsy management and identified patients with wicket rhythms On further evaluation themajority (54 2546) of these patients were found not to have epilepsy The authors compared the clinical and EEGfeatures for the 25 patients with wickets and nonepileptic episodes with those of age- and sex-matched patients withpartial-onset epilepsy using univariate and multivariate analysis Results Several features distinguished patients withEEG wicket patterns and nonepileptic episodes (n 25) from age- and sex-matched patients with epilepsy (n 25)mid-adult age at onset of episodes (mean 384 years vs 198 years) prolonged clinical episodes (mean 155 minutes vs 23minutes) and long duration of EEG wicket patterns (mean 066 seconds vs 011 second spikes) After controlling for otherfactors patients without major confusion during episodes were unlikely to have epilepsy Conclusion Wicket patterns areoften interpreted as epileptogenic This distinctive EEG pattern should be considered in patients with clinical episodesatypical for epilepsy

NEUROLOGY 2005641879ndash1883

Patients with nonepileptic episodes may be misdiag-nosed as having epilepsy when EEG wicket rhythmsare misinterpreted as interictal spikes and sharpwaves1 Wicket rhythms are ldquonormal variantrdquo 6- to11-Hz medium-to-high voltage EEG bursts that haveno epileptogenic potential23 We determined the fea-tures of clinical episodes and EEG that distinguishedpatients with nonepileptic episodes and EEG wicketpatterns from patients with partial-onset epilepsyand epileptogenic activity on EEG Patients mightavoid delayed diagnosis and unnecessary exposure toantiepilepsy drugs if neurologists more accuratelydiscriminate between wicket patterns and epilepto-genic activity

Methods Patients previously diagnosed with epilepsy wereevaluated at the Johns Hopkins outpatient epilepsy clinic Pa-tientsrsquo previous EEG tracings were obtained from other EEGlaboratories and ldquoblindlyrdquo re-read by Johns Hopkins electroen-cephalographers (GLK or RPL) During a 6-year period (Janu-ary 1996 to December 2002) we evaluated 2274 new patients 46patients had previous EEG available and had EEG wicket pat-terns The patients with EEG wicket patterns had been diagnosedwith epilepsy prior to their Johns Hopkins visit Twenty-one of thepatients with wicket patterns also had epileptogenic patterns onthe previously performed EEGs or had seizures confirmed withprolonged video EEG monitoring The remaining 25 (54) pa-tients had EEG wicket patterns but no true spikes or sharpwaves on previous EEGs and had nonepileptic episodes We per-formed one-to-one matching of these 25 ldquowicketrdquo patients withpatients with partial seizures based on age at date of EEG (3years) and sex Partial-seizure disorders were confirmed by theepileptogenic patterns on EEG or video-EEG (interictal spikes and

sharp waves or rhythmic ictal patterns) For both patients andmatches we collected demographic information a description ofpatientsrsquo clinical episodes (precipitants duration presence of pro-dromal symptoms presence of confusion and motor signs) andEEG findings characteristic of wicket and interictal spikes andsharp waves (frequency amplitude duration) Clinical informa-tion such as seizure etiology and MRI was collected via chartreview and an extensive questionnaire The presence of headtrauma was defined as a history of major head injury (eg skullfracture cerebral contusion on imaging) or trauma causing loss ofconsciousness

All statistical analyses were performed using the STATA 70statistical software package Simple descriptive statistics and uni-variate analyses were performed on the clinical and demographicvariables in the data set Simple t tests with the Bonferroni cor-rection for multiple measures were used to compare mean differ-ences between symptoms and signs for wicket and epilepticpatients for the continuous variables Outcome data that wereskewed such as episode duration were log transformed Becauseof a low sample size Fisher exact test was used to compare cate-gorical and dichotomous outcomes between wicket and epilepticpatients Since the data were clustered by case-control pairs con-ditional logistic regression was used in the multivariate analysisThe conditional logistic regression analysis could not be performedfor models that contained outcomes with no variability withinmatched case-control clusters In these cases the HuberndashWhitesandwich estimator of variance along with clustering of observa-tions within case-control cohorts was used4

Results Twenty-five of 46 patients (54) had wicketrhythms misinterpreted as epileptogenic activity all ofthese 25 patients had nonepileptic clinical episodes The 25patients with wicket patterns and nonepileptic episodestypically developed clinical episodes in mid-life mean 384years (95 CI 325 443) compared to late teen and earlyadult years for matched patients with complex partial sei-

From the Departments of Neurology (Drs Krauss Lesser and Niedermeyer and A Abdallah) and Biostatistics (Dr Thompson) School of Public HealthJohns Hopkins University Baltimore MDReceived December 3 2004 Accepted in final form March 3 2005Address correspondence and reprint requests to Dr Gregory Krauss Meyer 2-147 600 N Wolfe St Baltimore MD 21287 e-mail gkraussjhmiedu

Copyright copy 2005 by AAN Enterprises Inc 1879































































References


Citations







Reprints
































































References


Citations







Reprints


clinical and eeg features of patients with eeg wicket rhythms misdiagnosed with epilepsy

Healthcare