1 INTRODUCTION

Epileptic myoclonus or myoclonic seizure is defined as a very short electromyographic burst or silent period (<50 ms for positive myoclonus, up to 500 ms for negative myoclonus) preceded by an epileptiform discharge on the electroencephalogram (EEG).^{1, 2} Myoclonic seizures are a key feature in various pediatric epilepsy syndromes, such as myoclonic epilepsy in infancy (MEI),³ juvenile myoclonic epilepsy (JME), and progressive myoclonus epilepsy (PME).⁴ Myoclonic seizures could also be the predominant seizure type in epilepsy with myoclonic atonic seizures (EMAtS). There are also a large number of patients who have a myoclonic seizure as the predominant seizure type without fulfilling the criteria for any syndrome; they are usually called myoclonic epilepsies (MEs). The most common etiologies are structural, metabolic, and genetic, particularly chromosomal disorders.

The entity of PME, first described by Herman Lundborg in 1903,⁵ is currently defined by the presence of myoclonus, progressive motor and cognitive impairment, sensory and cerebellar signs, and abnormal background slowing on EEG that appear in an individual with prior normal development and cognition.⁶ PME is a rare syndrome encompassing various underlying etiologies, such as neuronal ceroid lipofuscinosis (NCL) and mitochondrial disorders in infants and children.^{6, 7} Due to the progressive nature of these conditions, PMEs might be challenging to diagnose in the early stages.⁸

However, early etiological diagnosis is crucial for prognosis and genetic counseling, and it can influence the choice of antiseizure medication (ASM).⁹ This becomes even more critical when an etiology-specific treatment is available and needs to be initiated without delay, as in NCL type 2 (NCL2).¹⁰ Furthermore, new therapeutic strategies are emerging with advances in gene therapy.¹¹

For pediatric neurologists in charge of a patient with an unidentified epileptic syndrome, MEs are associated with suspicion of PME, including NCL2, or some treatable underlying etiologies, particularly glucose transporter type 1 deficiency syndrome. Various investigations are prescribed in this context, including laboratory tests such as enzymatic activity or glycorrhachia measurements.^{11, 12} However, daily practice does not investigate most patients with JME and EMAtS without alerting signs.^{13, 14}

This retrospective longitudinal study aimed to assess the diagnostic performance of investigations in a cohort of patients with MEs diagnosed from 2009 to 2022 at a tertiary care center for rare epilepsies. Furthermore, we described the definitive syndromic diagnoses and the outcomes of these patients.

2 MATERIALS AND METHODS

3 RESULTS

Fifty-nine patients were identified in our database. Nine patients were excluded because they were referred from another hospital for a second opinion. Nine patients were excluded after reviewing their files because they did not exhibit any epileptic myoclonus. A total of 41 patients were included in the study.

Nine patients (9/41) did not undergo untargeted investigations. One had a family history of ceroid lipofuscinosis type 2 (CLN2), one had consecutive brain MRIs diagnosing Rasmussen syndrome (RS), and one had a history and brain MRI consistent with hypoxic–ischemic encephalopathy. Two patients had prior investigations for other symptoms, another patient with a predominant phenotype of autism spectrum disorder was prescribed only genetic testing, and one had focal seizures and underwent a restrained set of investigations. A patient with multiple seizure types did not undergo additional tests to determine the cause of the ME. One patient had parents who refused the investigation during the admission.

The study population consisted of 41 patients, 24 of whom were male (58.5%). Table 1 presents their characteristics based on data collected during ME investigations.

The first seizure (excluding febrile seizures) occurred at a median age of 2.1 years (interquartile range [IQR] = 1.0–4.0) and was a myoclonic seizure for 25 patients (61%). Eight patients (19.5%) experienced at least one febrile seizure. The median delay from the first seizure to the first neurology evaluation was 7.3 months (IQR = 2.6–22.8). More than half of the patients (56.1%) had other seizure types, such as absences (29.3%) and focal (22%) seizures.

Most patients had no family or neonatal history (90.2%) and were not born from consanguineous unions (82.4%).

Developmental delay was present in 22 patients (53.7%) and regression in two (4.88%). More than half of the patients' clinical examinations were normal (58.5%); the abnormal findings were mainly neurological (nine patients, 22%). The head circumference was in the normal range for 81.3% of the patients.

MRI results showed nonspecific abnormalities, with only 17 (42.5%) having normal MRI.

The first EEG report leading to the diagnosis of epileptic myoclonus was available for 39 of 41 patients exhibiting normal background (n = 20), with sleep recording for 14 patients (35.9%). An abnormal organization was found for 11 patients (28.2%). Interictal generalized spikes or polyspike-and-wave were seen in one third of patients (13, 33.3%). For four patients, this first EEG showed nonepileptic myoclonus.

3.2 Investigations performed for ME

The percentage of patients who underwent each test composing the ME investigations is shown in Figure 1. Investigations completed at >80% were CAA, measurements of lactate, pyruvate, ketone bodies, glucose, and free fatty acids, fasting and after a meal (redox), and CSF glucose measurement. At least half of the patients underwent CGH array or SNP array, urinary COA and CAA, CSF lactate measurement, and VEP-ERG.

Supplementary Material 1 displays, for each year, the number of patients who underwent ME investigations; 12 patients completed their investigations before 2017 (when a protocol for ME investigations was initiated in our department).

3.4 Syndromic diagnosis

Figure 2 exhibits the proportion of patients classified into each diagnostic group (syndromes and nonsyndromic epilepsies). An equal number of six patients (14.6%) were classified as MEI (detailed characteristics in Supplementary Material 3) and epilepsy with myoclonic absence (EMA; detailed characteristics in Supplementary Material 4). Eleven patients (26.8%) were diagnosed with DEE, and an etiological diagnosis was established for six of them during follow-up.

Ten patients (24.4%) were labeled nonsyndromic ME, as they had generalized epilepsy with predominantly myoclonic seizures, which did not match the diagnostic criteria for a syndrome or DEE. Five patients (12.2%) had unclassified epilepsy with multiple types of seizures not matching the criteria for DEE. One patient had focal epilepsy without lesion, another had RS, and a third had progressive myoclonus epilepsy due to CLN2.

4 DISCUSSION

Conducting investigations to identify etiology in pediatric onset ME results in a limited number of final diagnoses (9/41). However, our study showed that nonsyndromic ME is frequently observed. We show that some features in pediatric onset MEs are linked with unfavorable neurological or epileptic outcomes. Initial developmental delay or regression and initial abnormal clinical examination were associated with a worse neurological outcome, and the initial EEG background was associated with a worse epilepsy outcome. Moreover, initial developmental delay or regression was significantly associated with a higher total number of ASMs and the likelihood of remaining on ASM at the last follow-up point. Some trends were also observed in our cohort, with no patient with an abnormal head circumference or born of a consanguineous union being in the favorable neurological outcome group.

Investigations, even conducted for each patient at diagnosis, resulted in a single diagnosis of PME by untargeted investigations during the 15 years of our study. It was a patient with CLN2, but the diagnosis was made due to the family history. In the meantime, no patient had a syndromic diagnosis of PME with an unknown etiology, suggesting the unlikelihood of undiagnosed disorders responsible for PME. The incidence of PME, mainly very low and with various ages at onset, explained the absence of PME diagnoses in our cohort. In a large cohort of 147 patients with PME from India published in 2010,¹⁶ the mean age at onset of pediatric onset PME was 14.6 ± 5.8 years for ME with ragged-red fibers and 5.9 ± 9.1 years for NCL. Studies have shown that the average delay for etiological diagnosis of PME remains as long as 1.4 years, even for patients treated in the past decade,¹⁷ emphasizing the ongoing need to enhance diagnostic methods and effectiveness. The importance of achieving a rapid etiological diagnosis in pediatric ME may become more pronounced with the upcoming rise of disease-modifying therapies beyond its current role in facilitating genetic counseling.

Genetic investigation has significantly transformed the field of epilepsy, with a growing focus on its clinical utility.¹⁸ In cases of epilepsies and DEE in infancy and early childhood, early targeted whole exome sequencing has already proven to be a cost-effective approach.¹⁹ For PME, next generation sequencing in a large Italian cohort of 204 patients increased the overall genetic diagnostic yield to >80%, underscoring the value of a genetic approach for diagnosis.²⁰ In the most recent report of the ILAE Genetics Commission,²¹ the group recommends early genetic testing as the primary diagnostic tool to diagnose PMEs, given the association of nearly 50 genes with the condition, stating that laboratory tests and tissue biopsies should be used to confirm genetic findings rather than serving as primary diagnostic methods. In our study, genetic tests provided an etiological diagnosis for six patients, resulting in significant changes in the management of three of them.

Approximately half of children younger than 3 years of age are classified with epilepsy syndromes.^{3, 22} In our study, seven patients met the diagnostic criteria for an ILAE syndrome comprising myoclonic seizures, namely MEI or PME. We also observed six patients with epilepsy with EMA and one patient with RS. The latter had a particular phenotype that included myoclonic seizures. Based on our inclusion criteria, we found that patients with epilepsy with myoclonic absences were investigated similarly to patients with ME. A group of 11 patients had DEEs that could not be classified into another syndrome. Among them, two patients had SCN1A pathogenic variants with a non-Dravet syndrome as previously described.^{23, 24} Another group consists of the 10 patients with nonsyndromic ME. This group was defined in the first edition of “Guide bleu” (Epileptic Syndromes in Infancy, Childhood, and Adolescence).²⁵ In a more recent study, the term myoclonic encephalopathy in a nonprogressive disorder designates children with myoclonic status who do not have a progressive neurological disorder, regardless of age at onset.²⁶ Despite the meaningful clinical information conveyed by these terms, no such group is recognized in the current ILAE classification.¹⁵

We identified a group of nonsyndromic ME without an identification of an etiology. In our cohort of patients with ME, unfavorable neurological outcome was significantly associated with the initial developmental delay or regression. This abnormal development was significantly associated with the total number of ASMs used and the likelihood of remaining on ASM. These results, although based on univariable analysis, are consistent with the existing literature studying predictors of drug-resistant epilepsy^{27, 28} and recurrence risk after withdrawal.²⁹ Strikingly, a retrospective study of 127 young children with drug-resistant epilepsy demonstrated through multivariable analysis that developmental delay at the time of diagnosis was one of four independent variables.³⁰ Some of the factors with a statistical trend, such as abnormal head circumference or family history of consanguineous union, indirectly point to the genetic origin of ME.

Our study has limitations, including its retrospective real-life design and the inclusion criteria. In real-life studies, routine practice descriptions often face variability in clinician practices, as evidenced by imperfect adherence to local investigation guidelines. The inclusion criteria focused on admission for ME investigations do not allow the identification of all patients with ME but do allow the identification of similar investigations in other syndromes, such as epilepsy with myoclonic absences. The inclusion of patients investigated at daycare centers prevented us from studying patients requiring an initial hospital stay, for example, for myoclonic status epilepticus or neonatal onset epilepsies. The limited number of patients might also have restricted reaching statistical significance in our comparisons by epilepsy. Our findings provide valuable insights, although larger, multicenter studies are needed to confirm these results. In particular, it would be of interest to explore EEG findings as a predictive biomarker of the outcome. Finally, our patients were not all investigated by whole genome sequencing due to the study period. We can hypothesize that a few additional patients might have a final etiological diagnosis.

In conclusion, our study highlights key findings regarding pediatric onset ME. Although our research did not show a high diagnostic rate with early etiological investigation, the treatable nature of some causes of ME makes this evaluation essential and unavoidable. We suggest investigating ME with brain MRI, CSF glucose, targeted tripeptidyl peptidase 1 dosage or molecular biology, and genetic investigations. Currently, genetic testing with whole genome sequencing should be offered early in this context with a possible impact on early diagnosis rate and adequate management. We found some predictive factors on the outcome that could be particularly helpful in informing families on the prognosis in the case of nonsyndromic ME. ME with myoclonic seizures only has a better prognosis than ME with both myoclonic seizures and other seizure types. Our data demonstrate that early developmental abnormalities are significantly linked to a higher total number of ASMs and an increased likelihood of continued ASM use at the end of follow-up. Moreover, patients with abnormal initial EEG patterns, particularly interictal generalized spike or polyspike-and-wave, also tended to have worse outcomes, raising the question of their role as part of epileptic encephalopathy, as recently reported.³¹ Further research with larger cohorts and more standardized diagnostic protocols will be critical to refining the management of ME in children.

FUNDING INFORMATION

This work benefited from a government grant managed by the Agence Nationale de la Recherche under the France 2030 program (ANR-23-IAIIU-0010).

CONFLICT OF INTEREST STATEMENT

S.A. is deputy editor of Epilepsia. He has received personal fees for lectures or advice from Angelini, Biocodex, Eisai, Encoded, GRIN Therapeutics, Jazz Pharmaceuticals, Longboard, Neuraxpharm, Nutricia, Orion, Proveca, Servier, Stoke, UCB Pharma, and Xenon. He has been an investigator for Eisai, Marinus, Proveca, Takeda, and UCB Pharma. B.D.-P. has received personal fees for lectures or advice from Biocodex, Jazz Pharmaceuticals, and UCB Pharma. None of the other authors has any conflict of interest to disclose. 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.