The spectrum of epileptic syndromes with fixation off sensitivity persisting in adult life




The term “fixation off sensitivity” (FOS) was proposed by Panayiotopoulos to describe epilepsy/electroencephalography (EEG) changes evoked by the suppression of central vision and fixation. The EEG pattern usually consists of spike/polyspike and waves localized in occipital regions. FOS occurs mainly in children with idiopathic occipital partial epilepsies and rarely in adults. In this retrospective study we evaluated the clinical data, EEG, and magnetic resonance imaging (MRI) findings of patients with epilepsy and FOS persisting in adult life to better define the spectrum of syndromes.


We selected 15 consecutive patients (12 female/3 male; age range 19–59 years). The main inclusion criterion was the diagnosis of epilepsy with FOS persisting in adult life. We retrospectively analyzed clinical EEG and neuroimaging data.

Key Findings

We observed a female prevalence (F/M = 12/3). Eight patients presented both simple and complex partial seizures, whereas seven had only complex partial seizures. Partial seizures evolved into generalized seizures/hemiconvulsions in nine cases. The FOS pattern consisted of spike-and-wave and slow-wave abnormalities with posterior localization (bilateral in eight/monolateral in seven). We recorded seizures in 10/15 patients. All showed a posterior onset (bilateral in 2/left in 2/right in 6). FOS was prevalent in symptomatic epilepsy (cortical malformations in 7; celiac disease in 3; calcified vascular malformation in 1). One patient presented cryptogenic epilepsy and only three idiopathic epilepsy (Gastaut syndrome).


FOS can be observed in adult life in idiopathic epilepsy, representing the “prolongation” of the same phenomenon arisen during childhood. Nevertheless, it often represents the EEG expression of symptomatic epilepsies (cortical malformations/celiac disease).

The term “fixation off sensitivity” (FOS) was first proposed by Panayiotopoulos to describe epilepsy and electroencephalography (EEG) changes, or both, evoked by the suppression of central vision and fixation (Panayiotopoulos, 1998). FOS is a phenomenon elicited by conditions that eliminate central vision, such as closed eyes, complete darkness, modified Ganzfeld stimulation (using a large white surface without visual cues), and Frenzel lenses (Panayiotopoulos, 1998, 2007). The FOS EEG pattern usually consists of spikes/polyspikes and waves localized in occipital regions (bilateral or unilateral) or generalized discharges. It is usually inhibited by intermittent photic stimulation (IPS). FOS occurs mainly in children who have idiopathic partial epilepsies with occipital paroxysms such as Gastaut syndrome and, less frequently, in Panayiotopoulos syndrome (Panayiotopoulos, 2007). It is also described in patients with eyelid myoclonia with absences and atypical features (i.e., cognitive impairment, absence status epilepticus) and in idiopathic generalized epilepsy with photosensitivity (Ogura et al., 2005; Nicolai et al., 2008). In addition, FOS has been documented in nonepileptic patients with posterior paroxysms. Lastly, this pattern has, on rare occasions, been observed in symptomatic or cryptogenic epilepsies (Koutroumanidis et al., 2009; Hassan et al., 2011). In all the aforementioned cases, the FOS phenomenon is typical of infancy or childhood, with reports in adult epileptic patients being an exception (Kurth et al., 2001; Ferlazzo et al., 2010). In this retrospective study, we evaluated the clinical data, EEG, and magnetic resonance imaging (MRI) findings of 15 epileptic patients with FOS persisting in adult life, to better define the spectrum of syndromes that may be associated with this particular phenomenon.

Patients and Methods

We selected, from approximately 3,000 patients attending our epilepsy unit between 2000 and 2010 inclusive, 15 consecutive adult patients (12 female and 3 male; mean age 36.5 years, range 19–59 years) affected by various epileptic syndromes. The main inclusion criteria were the following: (1) a definite and well-documented diagnosis of epilepsy with onset in childhood (including general characteristics, seizure type, and syndrome classification); (2) evidence of a FOS phenomenon persisting in adult life and confirmed over time (for all patients—identified by a careful research on EEG database—at least three video-EEG recordings per year were available). We retrospectively analyzed the clinical data of these patients in order to investigate a possible family history of epilepsy and febrile seizures, age of onset, seizure semiology and frequency, response to therapy, and possible comorbidities. All of the patients underwent a careful EEG and neuroimaging study (MRI scan and, in some cases, computed tomography [CT] scan) as well as a routine neurophysiologic evaluation, including visual evoked potentials (VEPs) and visual field test (VFT).

EEG study

We analyzed the EEG findings of a total 450 video-EEG recordings obtained in the 15 selected patients (a mean of 26 recordings was available for each patient). Video-EEG monitoring was performed using a Telefactor (Conshohocken, PA, U.S.A.) or Micromed (Treviso, Italy) or Extelch System (Oakville, Canada), 21 channels, International 10–20 system. During each recording (mean duration of 40 min) patients were asked to open and close their eyes for 5 seconds for a total period of 5 min; FOS was evaluated according to the technique proposed by Panayiotopoulos (during eye-closed condition or during wakefulness, when central vision was eliminated by a modified Ganzfeld stimulation—by placing a sheet of white paper 20 cm in front of the subject with the eyes open without fixating—or complete darkness with the patient's eyes open, or Frenzel lenses) and by means of IPS (Panayiotopoulos, 1998).

Neuroimaging study

All the patients underwent high-resolution MRI performed with a Philips (Andover, MA, U.S.A.) 1.5 T and/or Siemens (Munich, Germany) 3 T system using a standardized epilepsy protocol that included high-resolution T1-weighted volume acquisition, and T2-weighted and fluid-attenuated inversion recovery (FLAIR) sequences; 11 patients also underwent a CT scan. Some of the patients we studied were evaluated by means of new, more advanced MRI techniques (e.g., EEG/functional MRI [fMRI], magnetic resonance spectroscopy [MRS], diffusion tensor imaging [DTI]).

Other neurophysiologic tests to assess visual functions

We also evaluated VFT and VEP in most of the patients.

The patients’ general characteristics are shown in the Table 1.

Table 1. General characteristics, electroclinical, and neuroimaging findings of patients with epilepsy and FOS persisting in adult life.
PatientSex/AgeSeizure onsetFamily historySeizure TypesFOS and interictal EEG findingsIctal EEG/seizure onsetEpileptic SyndromeNEMRI/CTVisual field/visual evoked potentialCurrent therapySeizure frequencyComorbidity
  1. F, female; M, male; FC, febrile convulsions; SPS, simple partial seizure; CPS, complex partial seizure; SGS, secondary generalized seizure; R, right; L, left; Bil, bilateral; T, temporal; O, occipital; P, parietal; S, spike; S&W, spike-and-wave; SW, slow- wave; IPE, idiopathic partial epilepsy; CPE, cryptogenic partial epilepsy; SPE, symptomatic partial epilepsy; AVM, arteriovenous malformation; NE, neurologic examination, CBZ, carbamazepine; CD, celiac disease, CLL, chronic lymphocytic leukemia; CNZ, clonazepam; LCS, lacosamide; LEV, levetiracetam; LTG, lamotrigine; OXC, oxcarbazepine; PB, phenobarbital; PGB, pregabalin; RFD, rufinamide; TPM, topiramate; VPA, valproate; ZNS, zonisamide.

1F/435 ySPS, CPS, vomitingR temporal-parietooccipital spike-&-wave and polyspike-&-waveLow amplitude fast activity and rhythmic polyspikes on R temporal-parietooccipital regionsIPENormalNormalNormalVPA + LEV + CBZAnnually
2F/347 ySPS, CPS, vomitingBil temporo-occipital spikes and slow wavesLow amplitude fast activity and rhythmic polyspikes on L temporal-parietooccipital regionsIPENormalNormalAbnormalCBZ + PB + CNZAnnually
3F/376 y+FSCPSBil temporo-occipital spike-&-wave and slow wavesSPENormalR parieto-occipital AVM and bil calcificationsNALEV + TPM + RFDWeekly
4F/3715 y+CPS, SGSR temporal-parietooccipital spikes and spike-&-waveRhythmic sharp waves and irregular spike-and-waves activity on R temporo-occipital regions with spreadingIPENormalNormalAbnormalLEV + CBZ + ZNSMonthly
5F/1915 yCPSR temporal spikes and spike-&-waveSPENormalBil occipital calcificationsNACBZ + LTG + VPAAnnuallyCLL CD
6F/279 ySPS, CPS, SGSBil temporo-occipital spikes and spike-&-waveSPENormalNormalNormalOXC + LEV + ZNSWeeklyCD
7M/272 yCPS, SGSBil temporo-occipital spike-&-waveFast activity and rhythmic polyspikes or rhythmic theta waves on bilateral posterior regionsSPEMental retardationDouble cortexNACBZ + VPA + LCS + RFDDaily
8F/436 mCPS, SGSL temporo-occipital spikes and spike-&-waveFast activity and rhythmic polyspikes on bilateral posterior regionsSPEMental retardationDouble cortexNAVPA + CBZ + LEV + CNZDaily
9M/443 y+SPS, CPS, SGSR temporal-parietooccipital spike-&-wave and slow wavesFast activity and rhythmic polyspikes on R temporal-parietooccipital regionsSPENormalR parieto-occipital cortical dysplasiaAbnormalVPA + CBZ + PB + CNZDaily
10F/313 ySPS, CPS, SGSBil temporoparietal spikes and spike-&-waveFast activity and rhythmic polyspikes of L posterior region; seizure triggered by readingSPENormalDouble cortexNAVPA + LCS + CBZMonthly
11F/3212 ySPS, CPSBil temporal-parietooccipital spikes and spike-&-waveFast activity and rhythmic polyspikes on R posterior regionsSPENormalNormalAbnormalCBZ + PGB + PBWeeklyCD
12M/385 yCPS, SGSL tempora-parietooccipital spikes and spike-&-waveSPEMental retardationL parieto-occipital dysplasiaNACBZ + VPA + PBAnnually
13F/427 mCPS, SGSR temporal-parietooccipital slow waves and spike-&-waveFast activity and rhythmic polyspikes on R posterior regionsSPEStrabismusR parietal dysplasiaNormalVPA + PGB + RFDMonthly
14F/5937 ySPS, CPSBil temporo-occipital spikes and spike-&-waveProbably SPENormalNormalNormalCBZ + PB + CNZWeeklyPrevious diagnosis of CD (not confirmed)
15F/3310 ySPS, CPS, SGSBil temporo-occipital spike-&-waveFast activity and rhythmic spike-and-waves on R temporo-occipital regionsSPEStrabismusDouble cortexNormalVPA + LEV + LCSMonthly


General clinical findings

In this population, we observed a female prevalence (F/M = 12/3). Three patients had a positive family history for epilepsy; one of them also had a positive family history for febrile seizures. The mean age at seizure onset was 8.5 years (range 6 months to 15 years). The neurologic examination was abnormal in five patients (5/15), with strabismus being detected in two patients and mild cognitive impairment in three.

Seizure type and ictal manifestations

Eight patients presented simple partial seizures, prevalently of the visual type. Elementary visual hallucinations were the most common ictal manifestations; typically, they were multicolor, varied in number and size, and were flashing or static. They were usually stereotyped. These patients also displayed complex visual hallucinations (figures or faces). Objective symptoms were rarely observed (tonic eye deviation, nystagmus, eyelid fluttering); two patients also presented vomiting. The average duration was 20 s. All the patients presented complex partial seizures. Simple or complex partial seizures evolved to hemiconvulsions or generalized tonic–clonic seizures in nine cases.


A diagnosis of celiac disease was made in three of our patients. This disease had been diagnosed many years before the onset of epileptic seizures. Calcifications in the cortical posterior regions were detected in one of these three patients at the neuroimaging study, whereas the other two did not present any evident lesions or calcifications.

FOS phenomenon

The FOS pattern was observed bilaterally in eight patients and unilaterally in seven patients (five right, two left). Epileptic abnormalities characterized by slow-wave, spike and polyspike discharges were documented in all the patients. Generalized bursts of fast activity were observed during sleep in two patients. Four patients presented slowing of background activity. Temporo-occipital regions were the most common location of EEG abnormalities (7/15), followed by temporal-parietooccipital (6/15), temporal (1/15), and temporoparietal regions (1/15). Paroxysmal photic response (PPR) suppressed FOS activity in three patients. The FOS pattern disappeared during sleep in all the patients. In some patients the EEG recordings, performed over time, revealed other associated independent interictal abnormalities.

Ictal EEG

We recorded epileptic events in 10 of 15 patients. The onset of the seizures occurred in the posterior regions in all cases (bilateral in two patients, left in two patients, and right in six). In one case, left temporoparietal partial seizures were triggered by reading (summarized in the Table).

Neuroimaging findings

Six patients had a normal structural MRI scan and nine cases presented brain alterations. Seven patients had cortical development malformations (MCDs): three had focal dysplasia (two on the right side, one on the left), whereas the remaining four had a double cortex. One patient had a vascular malformation and bilateral calcifications. Occipital calcifications were observed in one of the three cases affected by celiac disease.

Neurophysiologic evaluation

The visual field test and visual evoked potentials, which were performed in nine patients, were altered in four in whom the P100 latency was increased.

Spectrum of epileptic syndromes

In our population, FOS was predominantly observed in symptomatic epileptic syndromes (11/15). The causes, in order of frequency, were the following: cortical malformations (7/11), celiac disease (3/11), and calcified vascular malformations (1/11). In the remaining four patients, epilepsy was cryptogenic in one case and idiopathic in three. All the three patients in whom the cause of FOS was presumed to be idiopathic had Gastaut syndrome.

Electroclinical and neuroimaging findings of three patients with FOS pattern are reported in Figure 1.

Figure 1.

Electroclinical and neuroimaging findings of three patients selected from the FOS population. In all the cases EEG tracings show the FOS phenomenon evoked by eye closure: EEG patterns are characterized by bilateral posterior spike-and-wave activity and slow waves in case 2 (panel A); brief runs of paroxysmal fast activity followed by sharp waves and irregular spike-and-waves located in bilateral posterior regions with left predominance and subsequent spreading in case 10 (panel B); and slow waves and slow spike-and-wave activity located in right temporooccipital areas in case 5 (panel C). In the same patients, neuroimaging findings, obtained by using different approaches, are reported: during eye closed condition, EEG/fMRI study clearly shows BOLD activation clusters, located on bilateral temporooccipital regions (peristriate areas) in case 2; structural MRI scan shows a wide malformation of cortical development consisting of double cortex in case 10; CT scan demonstrates some calcifications in posterior regions with right predominance in case 5.


FOS is a phenomenon elicited by conditions that eliminate central vision in both natural situations, for example, with closed eyes or in complete darkness, and the experimental setting (Panayiotopoulos, 1998). It usually occurs in children with an idiopathic epileptic syndrome, whereas reports in children with cryptogenic and symptomatic epilepsies are rare (Koutroumanidis et al., 2009). Given its high degree of reproducibility, this phenomenon offers, from a pathophysiologic point of view, an intriguing and unique opportunity to study epileptic activity. The mechanisms underlying FOS are not yet fully understood. Ever since the first reports (Panayiotopoulos, 1998), FOS has been considered to be related to alpha rhythm generators, its appearance depending on variables that normally modulate the alpha rhythm. Over time, the FOS pattern has acquired an important role both in clinical practice, by providing clues to recognize specific epileptic conditions (especially in infancy) (Panayiotopoulos, 2000), and in the neurophysiologic setting, by enhancing knowledge of cortical occipital hyperexcitability (Strigaro et al., 2011). An ongoing, intriguing historical debate is on the distances and/or points of contact between FOS and the other forms of occipital hyperexcitability such as PPR. Although the borders between these two conditions are not clearly defined (in rare contexts they may even coexist), the main neurophysiologic differences, and in particular the opposite effects induced by IPS (Gumnit et al., 1965), may be due to the involvement of different visual pathways, that is, the magnocellular system in PPR and the parvocellular system in FOS (Wilkins, 1995). The FOS pattern has more recently been studied by means of various advanced techniques, which have provided new insights into the underlying network, excitability mediators, and metabolic changes of this phenomenon. Of interest, during the eyes-closed condition, recent EEG/functional MRI (fMRI) studies have revealed blood oxygen level–dependent (BOLD) predominant activations in the peristriate areas (Iannetti et al., 2002; Di Bonaventura et al., 2005), whereas MRS has documented a dynamic increase in the glutamate concentration in the same regions (Peca et al., 2010), thereby confirming the occipital hyperexcitability. Lastly, studies based on transcranial magnetic stimulation have recently confirmed an alteration in cortical excitability that is probably due to an imbalance between γ-aminobutyric acid (GABA) and glutamate transmission (Strigaro et al., 2011). Because this peculiar phenomenon is rarely observed in adult patients with epilepsy (Kurth et al., 2001), in this article we decided to focus on its occurrence in adult life in order to better define syndromes associated with it and to try to identify the possible relationship between the FOS pattern and specific etiologies. In accordance with data in the literature (Panayiotopoulos, 1998), in our population the typical FOS pattern, whether bilateral or unilateral, was predominant in the posterior regions (especially the temporooccipital, occipital, temporal, temporoparietal regions), and was rarely associated with generalized discharges, it often being an expression of secondary bilateral synchrony. In 10 of the 15 patients, we had the opportunity to document some ictal events (simple partial with predominant visual symptoms, complex partial, and secondary generalized), all of which arose from the posterior regions. These data are of considerable interest, since seizures in adult life have rarely been reported (Ferlazzo et al., 2010). All the seizures were recorded in patients with drug-resistant epilepsy, which accounted for a large proportion of our cohort (11 of 15). The most curious finding to emerge from our study was the prevalence of symptomatic epileptic syndromes (11/15). Indeed, according to published data, FOS is typically a pattern observed in children with idiopathic epileptic syndromes, with rare reports of symptomatic cases (Koutroumanidis et al., 2009; Hassan et al., 2011). The prevalence of symptomatic syndromes in our population suggests that the persistence of the FOS phenomenon in adulthood might be related to specific etiologic factors. The most frequent cause of FOS in the group of patients with symptomatic epileptic syndromes was malformation of cortical development (MCD) (especially band heterotopia). Another interesting finding was the association of this pattern with celiac disease, the typical neurological complication of which is posterior epilepsy, which is usually related to occipital calcifications. Such calcifications were, however, observed only in one of our three cases. Because other anecdotal reports (Licchetta et al., 2011) have also documented the association of FOS phenomenon and celiac disease in the absence of occipital calcifications, we believe that the diagnostic process in patients with “cryptogenic” posterior epilepsy and a FOS pattern should include an accurate screening for celiac disease. In three of our adult patients affected by an idiopathic epileptic syndrome (Gastaut syndrome), the FOS pattern is likely to represent a “continuation” of the same phenomenon started in childhood.

The present study was limited by a number of factors, including its retrospective design and a selection bias. With regard to the selection bias, we identified a significant number of symptomatic drug-resistant epilepsies, owing to the fact that the patients enrolled were selected in a tertiary epilepsy center. In contrast to what is expected in idiopathic forms, in which both seizures and the FOS phenomenon usually gradually disappear, a symptomatic etiology and drug resistance may explain the persistence of this rare EEG phenomenon in adult life. Consequently, we cannot rule out that patients with persisting FOS associated with idiopathic forms of epilepsy, or other cryptogenic forms, in which the seizures were controlled, escaped us. This explanation, however, remains speculative, since the data available on the long-term evolution of idiopathic epilepsies with FOS phenomenon are currently still somewhat limited.

In conclusion, the data that emerge from this retrospective study highlight certain crucial points. The revision of the electroclinical findings and syndromic characteristics in our population suggests that the FOS phenomenon may be observed in adult life, and that when it does, unlike the childhood form, it often represents the EEG expression of symptomatic epilepsies. FOS may, in some conditions, point to specific etiologies underlying the epileptic process, the most noteworthy being MCD (such as double cortex) and celiac disease. In this regard, the FOS EEG pattern may be considered an important practical clue that should be taken into account in clinical investigations.


The authors have no conflicts of interest. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.