Prevalence of Nocturnal Frontal Lobe Epilepsy in the Adult Population of Bologna and Modena, Emilia-Romagna Region, Italy

Abstract

Study Objectives:

To estimate the prevalence of nocturnal frontal lobe epilepsy (NFLE) in the adults of two areas of the Emilia-Romagna region (northeast Italy) and to describe the clinical features from a population-based perspective.

Design:

Population-based retrospective cohort study including adults with NFLE.

Setting:

Two areas of the Emilia-Romagna region: the city of Bologna (330,901 adult residents) and five districts of the province of Modena (424,007). Prevalence day: December 31, 2010.

Participants:

Patients with NFLE collected from multiple databases of neurologic hub centers of the districts involved. Diagnostic criteria: clinical history of sleep related bizarre motor attacks and videopolysomnographic recording confirming the typical features of NFLE. Inclusion criteria for prevalence calculation: residence in one of the two geographic areas on the prevalence day and an “active” or “in remission with treatment” form of NFLE.

Measurements and Results:

Six subjects from Bologna and eight from Modena were included. Crude prevalence (per 100,000 residents) was 1.8 (95% confidence interval 0.7–4.0) in Bologna and 1.9 (0.8–3.7) in Modena. Similarly, the main clinical features were consistent: onset during adolescence (median age 11–13 y), mainly hyperkinetic seizures, nonlesional form in more than two-thirds of cases, an active form of epilepsy in more than two-thirds of cases. A family history of epilepsy was reported only for two patients.

Conclusions:

This epidemiologic study establishes that nocturnal frontal lobe epilepsy is a rare epileptic condition, fulfilling the definition for rare disease. Because of methodological limitations of our case ascertainment, the estimates we disclose must be considered the minimum prevalence.

Citation:

Vignatelli L, Bisulli F, Giovannini G, Licchetta L, Naldi I, Mostacci B, Rubboli G, Provini F, Tinuper P, Meletti S. Prevalence of nocturnal frontal lobe epilepsy in the adult population of Bologna and Modena, Emilia-Romagna Region, Italy. SLEEP 2015;38(3):479–485.

INTRODUCTION

Nocturnal frontal lobe epilepsy (NFLE) is a peculiar form of focal epilepsy in which seizures occur almost exclusively during sleep. Ictal manifestations are characterized by bizarre motor behavior or sustained asymmetric dystonic posture suggesting frontal lobe involvement.¹ NFLE is reported to be a rare disorder but no epidemiologic data are available. The nonlesional form of NFLE seems to be predominant,² and some familial cases (autosomal dominant NFLE, ADNFLE)^3,4 are linked to mutations in the genes coding for the subunits of the neuronal acetylcholine receptors (nAChR), and mutations in new genes (KCNT1, DEPDC).^5,6 The ratio between sporadic and familial forms is unknown.

This study aimed to (1) estimate the prevalence of NLFE in adults of two areas of the Emilia-Romagna region (the city of Bologna and the province of Modena)—with similar population size but different levels of urbanization and health care organization—to test the consistency of results; and (2) to describe the clinical features of NFLE from a population-based perspective.

METHODS

This was a retrospective cohort study, part of a larger study designed by the Istituto di Ricovero e Cura a Carattere Scientifico (IRCSS) of Neurological Sciences, Bologna, to investigate the features of NFLE, and conducted after approval by the local ethical committee (September 16, 2010 Approval Code 10077). This research is reported following the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.⁷

Setting

The study refers to two areas in the Emilia-Romagna region, northeast Italy (Table 1). The first area comprises five health districts of the province of Modena: Castelfranco Emilia, the city of Modena, Pavullo nel Frignano, Sassuolo, and Vignola; the second area corresponds to the health district of the city of Bologna. All subjects living in one of the two areas on December 31, 2010 (prevalence day) were identified from the Emilia-Romagna population register administered at “ER statistica” (http://statistica.regione.emilia-romagna.it/servizi-online/statistica-self-service). The populations at risk in this study were adult (18 y of age and older) residents from the aforementioned two areas on the prevalence day (Bologna: 330,901 adults, 178,583 women, 152,318 men; Modena: 424,007 adults, 218,553 women, 205,454 men).

Table 1.

Geographical areas under observation (Bologna and Modena) and patients' origin.

Participants: Definitions and Inclusion Criteria

NFLE was ascertained when patients, during the diagnostic workup period, fulfilled both of the following clinical and videopolysomnographic criteria⁸:

1. A history of motor events arising predominantly from sleep more than 90% of times, suggestive of an involvement of frontal lobe structures.

2. Videopolysomnographic recording of at least one hypermotor episode (either asymmetric tonic-dystonic or hyperkinetic, i.e., bilateral asymmetric tonic/dystonic posturing or other choreoathetoid and ballistic movements of the limbs lasting about 20–30 sec) or at least two minor stereotyped episodes (i.e., paroxysmal arousals).

All patients underwent a comprehensive clinical, neuro-physiological, and neuroradiological evaluation. The final diagnosis was confirmed by at least two experts in sleep medicine and epileptology (in each area: Epilepsy Center of Modena, Epilepsy Center of Bologna) considering, in addition to ictal semiology from videopolysomnographic recordings, all clinical features useful to discriminate NFLE from parasomnias (such as age at onset, duration, frequency and number of events per night, stage of sleep when events arise, recall on awakening, presence of daytime sleepiness, outcome, and response to treatment). All cases with videopolysomnographic recording of stereotyped paroxysmal arousals not associated with hypermotor events were carefully reviewed and discussed collegially also considering the follow-up data. The agreement required for the final diagnosis was 100%; otherwise, these cases were considered doubtful and excluded. Clinical records of subjects identified through sources other than the two principal centers (see flowchart in Figure 1) were reviewed by experts at the Epilepsy Center of Bologna, in order to check the fulfillment of inclusion criteria.

Figure 1.

Flow diagram of eligible and included patients with nocturnal frontal lobe epilepsy (NFLE). Data refer to the prevalence day (December 31, 2010). * From other outpatient epilepsy offices of the city of Bologna.

Following the Guidelines for Epidemiologic Studies on Epilepsy,⁹ patients were classified as having (1) active epilepsy (at least one epileptic seizure in the previous 5 y, regardless of antiepileptic drug [AED] treatment), or (2) epilepsy in remission with treatment (no seizures for ≥ 5 y with AED treatment at the time of ascertainment), or (3) epilepsy in remission without treatment (no seizures for ≥ 5 y without AED treatment at the time of ascertainment). Etiology of NFLE was classified as lesional (structural abnormalities at neuroimaging) or nonlesional (no factor of increased risk of seizures identified). According to family history, cases were classified as familial or sporadic. The familial form was ascertained when at least one first- or second-degree relative had a history of epilepsy (including NFLE). Seizure types in NFLE were classified as follows^1,10: paroxysmal arousal (PA), hyperkinetic (HK), tonic/ dystonic (TD), or epileptic nocturnal wandering (ENW).

Subjects included in the study were (1) adults and residents in one of the two aforementioned specified geographic areas on the prevalence day (December 31, 2010) and (2) with an “active” or “in remission with treatment” NFLE.

Case Finding

We used multiple methods to identify eligible patients. For the area of Modena the main data source was the database of the Epilepsy Center of the Department of Neurosciences of the Nuovo Ospedale Civile S.Agostino-Estense (NOCSAE), recording all consecutive patients with epilepsy since 2007. The NOCSAE is the hub hospital for all neurological disorders of the five health districts of Modena province included in the study. For the area of Bologna the main data source was the databases—recording patients with epilepsy or sleep disorders since 1986—of the Epilepsy Center and the Sleep Center of the IRCSS of Bologna. The IRCSS is the hub center for epileptic and sleep disorders of the district of Bologna and one of the most important centers for studying these disorders in Italy. The clinical files and electrophysiologic (interictal and ictal videopolysomnographic) recordings of both centers were reviewed in detail. All candidates had been followed at the two centers for many years, with seizure diary assessed and clinical evaluation visit at least every 12 mo. Finally, other hospitals hosting neurology wards (none in Modena and two in Bologna) and neurology territorial offices (two in Bologna and two in Modena) were also asked for suspected NFLE patients in both areas.

Statistical Analysis

To calculate the crude prevalence, the included patients with NFLE (active or in remission with treatment) were the numerator, and the resident adults in each of the two areas were the denominator. We calculated the prevalence of NFLE (and 95% confidence interval [CI] according to Poisson distribution) for both areas (Bologna, Modena), overall and separately by sex. Clinical features were descriptively reported for the two areas and summarized in a single statistic.

RESULTS

The flow diagram of eligible and included patients is shown in Figure 1. Seven adults (three women, four men) from Bologna and nine adults (six women, three men) from Modena fulfilled the diagnosis of NFLE. Two female patients (one from both areas) had epilepsy in remission without treatment, therefore six adults (two women, four men) from Bologna and eight adults (five women, three men) from Modena were finally included in the analysis (Table 1).

The overall crude prevalence (per 100,000 adult residents) was 1.8 (95% CI, 0.7–4.0) in Bologna and 1.9 (95% CI, 0.8–3.7) in Modena. The prevalence for women was 1.1 (95% CI, 0.1–4.0) in Bologna and 2.3 (95% CI, 0.7–5.3) in Modena; for men it was 2.6 (95% CI, 0.7–6.7) in Bologna and 1.5 (95% CI, 0.3–4.3) in Modena.

The main details of the included subjects are reported in Tables 2 and 3. The median age at the prevalence date was 48 y (range, 35–71) in Bologna and 33 y (range, 18–47) in Modena. Median age at onset was 11 y (range, 6–38) in Bologna and 13 y (range, 1–33) in Modena. Epilepsy began with multiple seizures per night in 66% of patients from both areas. The semiology of the first complaint at onset differed from the patient's lifetime predominant motor events in four cases (29%). In particular, three patients presented with nonlateralizing generalized tonic-clonic seizure in sleep, and one presented with diurnal loss of contact with drop attack. Lifetime predominant seizures were in majority HK (86% of patients from both areas, with/without PA, ENW, TD); predominant TD seizures were reported in two patients (14%). A subjective sensation preceding motor manifestations was reported in four patients (28%). These sensations were described as somato-sensory (limb paresthesias) or viscera-sensory aura (choking sensation/pharyngeal dysesthesia because of a foreign body in the throat). Ictal electroencephalogram (EEG) was positive in 50% of cases, showing epileptiform/paroxysmal abnormalities or EEG flattering. Interictal epileptiform abnormalities were recorded in eight cases (57%) and were clearly lateralized in five (36%). The history of two patients was positive for development delay; one of them also had an intellectual disability (IQ = 62). Nonrapid eye movement parasomnias (pavor nocturnus, bruxism) were present in three patients (21%).

Table 2.

Clinical features of each NFLE patient on the prevalence day (December 31, 2010), ordered by geographical area, gender, and age.

Table 3.

Clinical and electro-clinical features of prevalent patients.

Lesional neuroimaging findings were detected in four patients (28%): one patient from Bologna (focal heterotopy), and in three patients from Modena (two focal cortical dysplasias, one dysembrioplastic neuroepithelial tumor). A family history of epilepsy was reported by two patients in Modena (none in Bologna); no patient had family members with a possible history of NFLE. Neither of the two patients from Bologna performing genetic tests had known mutations in genes coding nAchRs.

Five patients from Bologna had active NFLE (two in treatment) and one was in remission with treatment. In Modena, eight patients had active NFLE (7 in treatment). The 5-y remission ratio is one of 14 on the whole. Including the two patients excluded from the prevalence estimate, as in the condition of 5-y remission without AED treatment (see Table 2), the 5-y remission ratio remains equally very low (3 of 16, about 19%).

DISCUSSION

To our knowledge, these are the first epidemiological data estimating the adult prevalence of NFLE and classifying NFLE subgroups from a population-based perspective. Two well-defined areas of the same region showed a very similar crude prevalence of NFLE in adults (1.8–1.9 per 100,000), establishing that NFLE is a rare epileptic condition. Projecting the prevalence that we have found, we can infer about 70 prevalent cases out of the whole adult population of the Emilia-Romagna region (3,842,658). If our results are replicated and confirmed in other geographical settings, our article could establish that NFLE is a “rare disease”, according to both the European definition (less than one person per 2,000 affected, available at http://www.orpha.net/consor/cgi-bin/Education_AboutRare-Diseases.php?lng=EN, last access 18 July 2014) and United States definition (patient populations smaller than 200,000 individuals in the United States, Rare Disease Act 2002, available at http://history.nih.gov/research/downloads/PL107-280.pdf, last access 18 July 2014, corresponding roughly to less than one person per 1,600 affected). It is widely recognized that rare diseases suffer from a deficit of medical and scientific knowledge. People with rare diseases all face similar difficul-ties in their quest for a diagnosis, relevant information, and proper direction from qualified professionals (http://www.orpha.net/consor/cgi-bin/Education_AboutRareDiseases.php?lng=EN, last access 18 July 2014). When our results are integrated by reliable prognostic data, they will be useful to define the frame and burden of this condition.

Our population-based data confirm those from the largest case series²: a preponderance (not less than two-thirds of cases) of nonlesional compared to lesional forms; family history of epilepsy in not more than one-fourth of cases; peak onset around 11–13 y of age; seizure-free condition only in a minority of patients; HK seizure as commonest ictal manifestation; high frequency of associated nonrapid eye movement parasomnias^2,11; at least half of patients without interictal epileptiform abnormalities at EEG. Interestingly, we found that complaints at onset differed from lifetime predominant motor events in about one-fourth of patients (generalized tonic-clonic seizure in sleep, diurnal seizures). These features deserve further research.

We did not observe patients pertaining to a suggestive familial NFLE, challenging the common notion that ADNFLE cases make up a large share (up to 43%)^12,13 of the NFLE spectrum. A preliminary systematic review on the geographical distribution of ADNFLE published cases (46 references retrieved on this topic; data not shown) failed to disclose a low awareness of ADFLE in Italy (at least 16 references from Italy). However, no article refers to patients reported to be resident in the areas of our study.

The general perception is that NFLE and ADNFLE typically respond well to treatment,¹³ showing a complete remission around the fourth or fifth decade of life.¹⁴ In spite of this, we observed a very low 5-y remission ratio. This can be partially explained by the fact that we applied the very restrictive definition of remission by the International League Against Epilepsy (absence of seizures for at least 5 y) in order to classify patients' status. Moreover, for the prevalence estimates we excluded (as documented in Table 2) two patients in 5-y remission without AED treatment at the prevalence date. However, also including these two patients, the 5-y remission ratio would be equally very low (about 19%).

Our study has some limitations. As we ascertained cases from the second level of care some patients were certainly missed, so the estimates we disclose must be considered the minimum prevalence. The population we describe is probably shifted toward a more severe phenotype than the hypothetical actual spectrum. Moreover, we included only adults and our observation was limited to care settings for adults; considering that the onset of NFLE is during childhood, this could be another reason some patients were missed. However, because the rates and clinical features we observed from the two areas are similar, the potential selection bias probably acted in the same direction. In addition, the importance of both hospitals involved in this study for the care of sleep and epileptic disorders in the regional and national context makes the migration of patients not captured to distant centers highly unlikely. Another limitation is the small number of included patients, making the estimates imprecise. In particular, we observed inconsistent male/female ratios (4/2 in Bologna, 3/5 in Modena), in turn different from that (7/3) reported by Provini et al.² Because of the low number of patients performing genetic analysis, conclusions cannot be drawn on the frequency of ADNFLE.

Finally, we observed an unusual onset in about one-fourth of patients and an unexpectedly high ratio of active epilepsy. Our study has not the sufficient sample size and the follow-up duration to assess the broad spectrum and long term prognosis of NFLE; thus, cohort studies of consecutive patients should be designed to define these unresolved features of NFLE— namely, the natural history and the remission rate.

DISCLOSURE STATEMENT

This was not an industry supported study. This work was supported by the Telethon Foundation grant GGP 132000 to Dr. Tinuper. Dr. Rubboli has received honoraria from UCB and EISAI. Dr. Provini has received honoraria from MEDAPHARMA. Dr. Meletti has received honoraria from UCB. The other authors have indicated no financial conflicts of interest.