Table 1.
Characteristics of the cohorts in the included studies.
| Author, Year | DRE 1 Definition | Participants Included | Inclusion and Exclusion Criteria | Factors Associated with DRE | Factors Not Associated with DRE |
|---|---|---|---|---|---|
| Benova et al., 2018 [22] | Authors did not provide DRE definition. However, the following variables were considered markers of DRE:
|
22 | Inclusion: pre/perinatal diagnosis of cardiac rhabdomyomas | Higher number of areas with FCD-like 3 features (uncorrected p < 0.0001, FDR 4 < 0.01) ID 5 (uncorrected p < 0.001, FDR < 0.05) TSC2 (uncorrected p < 0.01, FDR < 0.05) |
- |
| Chu-Shore et al., 2009 [2] | uncontrolled seizures after more than three AED (not including treatment for infantile spasms) | 173 (2 months to 73 years, median 13 years) | - | At least one cyst-like cortical tuber (p = 0.0007) FCD |
- |
| Chu-Shore et al., 2010 [23] | uncontrolled seizures after at least three first-line AED trials | 291 | - | Infantile spasms (p < 0.0001) | TSC2 vs. TSC1, TSC2 vs. NMI 6 (p = 0.169) |
| Hulshof et al., 2021 [14] | ILAE, 2010 7, at 2 years | 41 | Inclusion: Fetal MRI of sufficient quality and available neurologic outcome data at the age of 2 years Exclusion: Epilepsy surgery before the age of 2 years. |
- | Fetal (sub)cortical lesion sum score—4.89 vs. 4.41 in DRE and non-refractory epilepsy, respectively (p = 0.62) |
| Jeong et al., 2017 [12] | ILAE, 2010 | 1546 (9.6 to 25.5 years, median 16.0 years) 21.4%—TSC1 67.9%—TSC2 10.7%—NMI |
Exclusion: if date fields were missing and age of onset and symptom duration could not be calculated | Onset of focal seizures prior to 1 year of age (p < 0.001) TSC2 (TSC2 vs. TSC1 (p < 0.001)) Infantile spasms (p < 0.001) Drug-resistant infantile spasms (p < 0.001) ASD 8 (p < 0.001) Mild to moderate intellectual disability - ID (p < 0.001) and severe to profound ID (p < 0.001) ADHD 9 (p < 0.001) Anxiety (p = 0.02) Periungual fibromas (p = 0.02)—lower odds of DRE |
Male vs. female (p = 0.94) Race (p = 0.40) TSC2 vs. NMI (p = 0.84) TSC1 vs. NMI (p = 0.12) Duration of infantile spasms (p = 0.90) Depression (p = 0.08) SEGA 10, SEN 11, cortical tubers, cerebral white matter migration lines Anxiety after adjusting for TSC mutation (p = 0.69) |
| Jozwiak et al., 2011 [30] | two or more seizures per month despite the use of two or more AED | 45—total 35—standard treatment (AEDs within a week after the onset of seizures), 14—preventive treatment (AEDs within a week after appearance of active epileptic discharges on consecutive EEG, but before clinical seizures) |
Inclusion: Diagnosis of TSC until the end of second month of life, follow-up till the end of 24 month of life Exclusion: children presenting with seizures |
Standard treatment vs. preventive treatment (p = 0.021) | - |
| Jóźwiak et al., 2019 [29] | two or more seizures a month despite the use of two or more antiepileptic therapies, including AEDs, ketogenic diet, vagus nerve stimulation, and epilepsy surgery | 39—total 25—standard treatment (vigabatrin within a week after first clinical seizures), 14-preventive treatment (vigabatrin introduced within a week after epileptiform discharges, before clinical seizure). |
Inclusion: Diagnosis of TSC until the end of second month of life, follow-up till the end of 24 month of life Exclusion: children presenting with seizures |
- | Standard treatment vs. preventive treatment (p = 0.5) |
| Kotulska et al., 2014 [10] | ILAE, 2010 | 21 | Inclusion: Epilepsy onset within 4 weeks of life. | Presence of FCD | - |
| Kotulska et al., 2021 [24] | ILAE, 2010 | 94 (both groups underwent careful EEG surveillance) | Inclusion: TSC diagnosis within first 4 months of life, no history of clinical seizures or epileptiform abnormalities in EEG. | Lower odds of DRE if preventive treatment (p = 0.047) | - |
| Mert et al., 2019 [26] | seizures once a month or more for at least 1 year, while using at least two AED at the appropriate dose | 83 | Inclusion: At least 1 year follow-up. | Seizures in the neonatal period Age of onset of seizure less than 2 years of age ASD Status epilepticus Infantile spasms Generalization of EEG finding Tuber count of more than 3 (p < 0.001) IQ < 70 |
Sex Consanguinity Family history of TSC Attention-deficit and hyperactivity disorder SEN SEGA White matter dysplasia (p > 0.05) |
| Monteiro et al., 2014 [27] | ILAE, 2010 | 35 | - | TSC2 mutation | - |
| Ogórek et al., 2020 [28] | ILAE, 2010 | 94 | Inclusion: Age ≤ 4 months, no prior seizures, no clinical seizures on baseline video EEG Exclusion: any condition considered by the investigator to hinder participation in the study or affect primary outcome. |
TSC2 (TSC2 vs. TSC1 mutation (p = 0.0245)) | - |
| Peron et al., 2018 [19] | - | 240 | Inclusion: 0–80 years of age, conventional molecular analysis available for both TSC1 and TSC2, complete clinical and imaging data available and updated to the latest follow-up encounter. Exclusion: (1) Possible clinical diagnosis or (2) Insufficient clinical records. |
- |
TSC1 vs. NMI (p = 1) TSC2 vs. NMI (p = 0.7) |
| Savini et al., 2020 [25] | - | 6 | - | ID Pathogenic variants in the GAP domain of TSC2 (no p-value, just case reports) |
- |
| de Ridder et al., 2021 [33] | ILAE, 2010 | 83—total 51—standard (S; clinical and EEG follow-up and start of vigabatrin after seizure onset) 23—preventive (P; follow-up and introduction of vigabatrin once EEG criteria met—focal IED for >10% of the recording time, multifocal IED, generalized IED, or hypsarrhythmia—and before seizure onset) |
- | S group: Younger age of first IED 12 on EEG (p = 0.019). Multifocal IED on the first EEG compared to focal IED (OR 4.4, 95% CI 1.1–16, p =0.026). |
S group: Younger age of first IED on EEG in a multivariable model (p = 0.429). Multifocal IED on the first EEG compared to focal IED in a multivariable model (p = 0.058). P group: None of the features of the first EEG with epileptiform discharges. |
| Vignoli et al., 2013 [21] | ILAE, 2010 | 160 | Inclusion: At least 1 year follow-up | Cognitive impairment (p < 0.05) TSC2 mutation More than 6 cortical tubers SEN or SEGA Lower educational level Psychiatric disorder Earlier mean age of epilepsy onset (3.3 vs. 5.3 years, p > 0.05) Status epilepticus (p < 0.05) Younger age at TSC diagnosis (7.6 vs. 13.2 years, p < 0.05) |
Infantile spasms (p > 0.05) Epilepsy onset in the first year of life |
| Vignoli et al., 2021 [20] | ILAE, 2010 | 257 (>18 years old) | - | ID (p < 0.001) Psychiatric disorders (p = 0.004) No family history of TSC (p = 0.010) Younger age of seizure (6 vs. 27 months, p = 0.001) Higher rate of spasms (27.1% vs. 48.8%, p = 0.007) Less frequently focal epilepsy (p = 0.029) Lower level of education (p = 0.002) |
Age Sex Mutation Tubers SEN |
| Winterkorn et al., 2007 [31] | one of the following criteria met: more than three AED, epilepsy surgery was performed, or one or more seizures per day continued despite therapy | 208 | - | Family history of TSC—lower odds of DRE (p = 0.003) low IQ/DQ (p < 0.0005) |
- |
| Zhang et al., 2018 [32] | ILAE, 2010 | 108 (3 months to 10 years, mean 2.2 years, median 1.4 years) | Inclusion: Taking rapamycin > 1 year |
Calcification in the cerebral parenchyma (p < 0.006) | Patient’s age (p = 0.745) Seizure type (p = 0.788) Genetic mutation (p = 0.204) Family history (p = 0.927) |
1 DRE—Drug-resistant epilepsy, 2 AED—antiepileptic drugs, 3 FCD—Focal cortical dysplasia, 4 FDR—False Discovery Rate correction from univariate tests, 5 ID—Intellectual disability, 6 NMI—No mutation identified, 7 “Drug-resistant epilepsy is defined as failure of adequate trials of two tolerated, appropriately chosen and used antiepileptic drug schedules (whether as monotherapies or in combination) to achieve sustained seizure freedom”. In [3] 8 ASD—Autism spectrum disoder, 9 ADHD—Attention deficit hyperactivity disorder, 10 SEGA—Subependymal giant cell astrocytomas, 11 SEN—Subependymal nodules, 12 IED—ictal epileptiform discharges.