AAN/AES Guideline Update: Efficacy & Tolerability of New AEDs I — New-Onset Epilepsy (2018)
This topic summarizes the 2018 AAN/AES practice guideline update on the efficacy and tolerability of second- and third-generation antiepileptic drugs for new-onset epilepsy, published in Epilepsy Currents (Kanner AM et al., 2018).
🔹 Bottom Line
- Guideline: 2018 AAN/AES practice guideline update — updates the 2004 guideline on 8 second-generation + 6 third-generation AEDs for new-onset epilepsy
- Focal epilepsy monotherapy: LTG (Level B) and LEV (Level C) may be considered; ZNS (Level C) also an option — all compared against CBZ-CR/CBZ-IR
- Elderly (≥60 yr): LTG (Level B) and GBP (Level C) should be considered for new-onset focal epilepsy
- Children — focal seizures: TPM 400 mg/d is possibly more effective than 50 mg/d (1 Class II) but no clinical recommendation can be made
- Childhood absence epilepsy: ETS or VPA should be used before LTG unless compelling AE-related concerns (Level B); LTG is probably equally effective as ETS or VPA
- No data exist for new AED efficacy in new-onset GE with GTC seizures, JME, or juvenile absence epilepsy
- Third-generation AEDs: No high-quality studies for CLB, ESL, EZG, FBM, GBP, LTG, LEV, LCM, OXC, PER, PGB, RFN, TGB, TPM, VGB, or ZNS in new-onset epilepsy
- FDA extrapolation: ESL and LCM (add-on or monotherapy, ≥4 yr) and PER (monotherapy) approved for focal epilepsy via extrapolation strategy
- OXC remains the only AED with Level A evidence for new-onset focal epilepsy (unchanged from 2004)
Guideline Overview
Source & Scope
- Organizations: AAN Guideline Development, Dissemination, and Implementation Subcommittee + AES
- Published: Epilepsy Currents, Vol. 18, No. 4, July/August 2018, pp. 260–268
- Authors: Kanner AM, Ashman E, Gloss D, Harden C, Bourgeois B, Bautista JF, et al.
- Objective: Update the 2004 AAN guideline for treating new-onset focal or generalized epilepsy with second- and third-generation AEDs
- Literature review: January 2003 to November 2015; 2004 AAN criteria used for classification
- Companion guideline: Part II covers treatment-resistant epilepsy (separate publication)
AEDs Covered
| Generation | Drugs |
|---|
| Second-generation (8) | GBP, LTG, TPM, TGB, OXC, LEV, ZNS, VGB |
| Third-generation (6) | CLB, ESL, EZG, LCM, PER, PGB, RFN |
| Older AEDs (2, newly approved in US) | CLB, VGB |
Key Question
- For adults and children with newly diagnosed epilepsy, how do CLB, ESL, EZG, FBM, GBP, LTG, LEV, LCM, OXC, PER, PGB, RFN, TGB, TPM, VGB, and ZNS compare in efficacy and tolerability with older AEDs?
Evidence Classification Changes from 2004
- 3 LTG studies: downgraded from Class I → Class II
- 1 GBP study: downgraded from Class I → Class III
- 2 TPM studies: downgraded from Class I → Class III
- 4 OXC studies: remain Class I (unchanged)
- Net effect: GBP and TPM → possibly effective (Level C); LTG → probably effective (Level B); OXC → unchanged (Level A)
2004 Guideline Summary (Level A/B Recommendations)
| AED | Monotherapy Focal / Mixed (focal + IGE) | Childhood Absence Epilepsy |
|---|
| Gabapentin | Yes | No |
| Lamotrigine | Yes | Yes |
| Topiramate | Yes | No |
| Tiagabine | No | No |
| Oxcarbazepine | Yes | No |
| Levetiracetam | No | No |
| Zonisamide | No | No |
🔹 Clinical Pearl
OXC is the ONLY AED with Level A evidence for new-onset focal epilepsy (4 Class I studies, unchanged in the 2018 update). LTG is Level B. GBP and TPM were downgraded from their 2004 evidence levels. For boards, remember: OXC > LTG > GBP/TPM in evidence strength for new-onset focal epilepsy.
Monotherapy in Adults: New-Onset Focal Epilepsy
New Studies Since 2004
- 2 Class I, 5 Class II, and 2 Class III studies published since 2004
- 1 study in patients aged ≥60 years; 1 study in patients aged ≥65 years
GBP vs LTG vs CBZ-IR (Class II)
- Doses: GBP ≤1,500 mg/d; LTG ≤150 mg/d; CBZ-IR ≤600 mg/d
- Population: Patients ≥60 years
- Primary outcome: Retention in trial for 12 months (seizure recurrence or AEs despite dose adjustments)
- Drug discontinuation less frequent with LTG or GBP than with CBZ-IR → better LTG tolerability
- Common AEs: GBP → higher weight gain, water retention; LTG → rash; CBZ → hyponatremia
- All 3 AEDs did not differ in neurologic AEs
LTG vs CBZ-CR (Class I)
- Doses: LTG 100–500 mg/d; CBZ-CR 400–2,000 mg/d
- Population: Focal epilepsy, patients ≥65 years
- Primary outcome: Retention in trial at final 20 weeks
- Seizure-free rates: LTG 52% vs CBZ-CR 57% (same)
- AE-related withdrawal higher for CBZ-CR (14% LTG vs 25% CBZ-CR) — did not reach statistical significance
- Common AEs: LTG → dizziness, headache, fatigue; CBZ-CR → rash, headache, dizziness, somnolence, fatigue
LEV vs CBZ-CR (Class II)
- Doses: LEV 1,000–3,000 mg/d; CBZ-CR 400–1,200 mg/d
- Seizure-free rates: Almost identical for LEV and CBZ-CR at 6 months and 1 year
- AEs more common with LEV: Depression, insomnia, back pain, weight gain
- AEs more common with CBZ-CR: Headache, fatigue, somnolence, dizziness
ZNS vs CBZ-CR (Class II)
- Doses: ZNS 300–500 mg/d; CBZ-CR 600–1,200 mg/d
- Population: 74% focal-onset, 26% unknown epilepsy
- Primary outcome: Percentage seizure-free for 26 weeks
- Results: Seizure-free rates nearly identical for ZNS and CBZ-CR
- AEs more common with ZNS: Decreased appetite, weight loss
- AEs more common with CBZ-CR: Dizziness
- Headache, dizziness, and somnolence most frequent in both groups
LTG vs GBP vs TPM vs OXC vs CBZ (SANAD-Arm A, Class III)
- Design: Randomized, unblinded trial in children and adults
- Population: 89% focal epilepsy
- Primary outcomes: Time to treatment failure; time to 12-month remission
- Treatment failure: LTG outperformed CBZ, GBP, and TPM (nonsignificant advantage over OXC)
- 12-month remission: CBZ outperformed GBP (nonsignificant advantage over LTG, OXC, and TPM)
- 2-year and 4-year remission (secondary): AE intolerability less frequent with GBP and LTG than OXC and TPM
- Rash: More frequent with CBZ and OXC
- Weight gain, dizziness, ataxia: More frequent with GBP
- Psychiatric symptoms, weight loss, paresthesia: More frequent with TPM
- LTG was noninferior to CBZ for 12-month remission at 2 and 4 years
VGB vs CBZ-IR (1 Class I + 1 Class III)
- Primary outcome (Class I): Time to withdrawal due to lack of efficacy or AEs
- Secondary outcomes: Time to 6-month remission, time to first seizure after initial target dose (≤600 mg/d CBZ-IR and ≤2 g/d VGB)
- No differences in time to withdrawal due to lack of efficacy
- Time to 6-month remission significantly shorter for CBZ-IR than VGB
- Time to first seizure significantly longer for CBZ-IR than VGB
- VGB AEs: More psychiatric symptoms, weight gain
- CBZ-IR AEs: More rash
- In Class III study: significantly more seizure-free patients on CBZ-IR than VGB; serious rash occurred with CBZ-IR; VGB associated with scintillating visual disturbances and myoclonic jerks
PGB vs LTG (Class II)
- Doses: Majority received PGB 150 mg/d or LTG 100 mg/d
- Primary outcome: Proportion seizure-free for 6 continuous months during efficacy phase (first 24 weeks, adjustable doses, then fixed)
- Seizure freedom: Significantly more patients on LTG than PGB achieved seizure freedom
- LTG showed comparatively greater reduction in secondarily generalized tonic–clonic (GTC) seizures
- Weight gain: More common with PGB
- AE-related withdrawal: PGB 8% vs LTG 7% (similar)
- Common AEs in both: headaches, dizziness, somnolence, fatigue
TPM vs PHT (Class II)
- Doses: TPM 100 mg/d followed by PHT 1,000 mg/d load then 300 mg/d maintenance
- Duration: 28 days
- Primary outcome: Time to recurrence of first focal seizure or GTC by day 28
- Seizure recurrence: TPM 18.9% vs PHT 9.7% — could not establish noninferiority
- PHT AEs: Higher rash incidence leading to discontinuation; dizziness, somnolence
- TPM AEs: Paresthesia; cognitive AEs more common
- Note: Data apply only to 28-day efficacy and cannot be generalized to long-term treatment
Conclusions — Adults with New-Onset Focal Epilepsy
- LTG is probably effective in patients ≥60 yr (1 Class I, 1 Class II study); LTG was better tolerated than CBZ-IR but not CBZ-CR
- GBP is possibly as effective and better tolerated than CBZ-IR in patients ≥60 yr (1 Class II)
- LEV is possibly as effective as CBZ-CR (1 Class II); AEs comparable
- ZNS is possibly as effective as CBZ-CR (1 Class II); comparable AE frequency
- Evidence insufficient to compare GBP, OXC, TPM efficacy with CBZ-IR/CBZ-CR (1 Class III)
- VGB is probably less efficacious than CBZ-IR (1 Class I + 1 Class III); associated with increased risk of serious AEs (visual field defects) → precludes first-line use
- PGB is possibly less effective than LTG at study doses (1 Class II), but PGB doses were lower than typically used
- Cannot determine if TPM is equivalent to PHT for urgent treatment (1 Class II, 28-day study)
- No high-quality studies for CLB, ESL, EZG, FBM, GBP, LCM, LEV, LTG, OXC, PER, PGB, RFN, TGB, TPM, VGB, or ZNS in new-onset epilepsy
- Evidence insufficient to demonstrate AED efficacy in unclassified GTC seizures (no study had enough patients)
Recommendations — Adults with New-Onset Focal Epilepsy or Unclassified GTC Seizures
| # | Recommendation | Level |
|---|
| 1 | LTG use should be considered to decrease seizure frequency | B |
| 2 | LTG (Level B) and GBP (Level C) may be considered in patients ≥60 years | B / C |
| 3 | LEV use may be considered to decrease seizure frequency | C |
| 4 | ZNS use may be considered to decrease seizure frequency | C |
| 5 | VGB appears less efficacious than CBZ-IR; toxicity profile precludes first-line therapy | C |
| 6 | PGB at 150 mg/d is possibly less efficacious than LTG at 100 mg/d | C |
| 7 | Evidence insufficient to consider GBP, OXC, or TPM instead of CBZ | U |
| 8 | Evidence insufficient to consider TPM instead of PHT for urgent treatment | U |
| 9 | Data lacking to support or refute use of third-generation AEDs (CLB, FBM, VGB) in new-onset epilepsy | U |
| 10 | Data lacking to support or refute use of newer AEDs in treating unclassified GTC seizures | U |
🔹 Clinical Pearl
For new-onset focal epilepsy in adults: OXC has Level A evidence (unchanged), LTG has Level B, and LEV/ZNS have Level C. VGB is probably less effective than CBZ-IR AND carries risk of irreversible visual field defects — never first-line. PGB was less effective than LTG at the study doses (150 mg/d vs 100 mg/d).
Monotherapy in Children: New-Onset Focal Epilepsy
High-Dose vs Low-Dose TPM (Class II)
- Study: 1 Class II study in children and adolescents
- High dose: TPM 400 mg/d; Low dose: TPM 50 mg/d
- Kaplan–Meier survival analyses: Time to next seizure favored the higher dose
- Seizure freedom probability:
- At 6 months: high-dose 90% vs low-dose 78%
- At 12 months: high-dose 85% vs low-dose 62%
- Seizure freedom significantly higher in the high-dose group
- AEs at 400 mg/d: Occurred in 4% of children taking 50 mg/d and in 14% of those taking 400 mg/d
- Most frequent AEs: Headache, decreased appetite, weight loss, somnolence, dizziness, paresthesia, problems with concentration or attention
Conclusion
- TPM monotherapy at 400 mg/d is possibly more efficacious than 50 mg/d in children/adolescents with new-onset focal or generalized-onset GTC seizures (1 Class II study)
- However, the higher dose is associated with more AEs and is not typically used in clinical practice
- The study was done for regulatory purposes — no clinical recommendation can be made
Recommendation
- Although data suggest TPM 400 mg/d > 50 mg/d, no recommendations can be made regarding TPM use at these studied doses, particularly in pediatric patients
Monotherapy in Adults & Children: New-Onset GE or Unclassified GTC Seizures
LTG vs TPM vs VPA (SANAD-Arm B, Class III)
- Design: Class III multicenter, randomized, open-label, parallel study
- Population: 716 outpatient children and adults with seizure disorders (~90% had focal epilepsy in whom clinician regarded VPA as better than CBZ)
- Epilepsy types: GE diagnosed in 63%, focal epilepsy 7.3%, 27% unclassified
- Diagnosis: 87.7% new-onset epilepsy
- Treatments: LTG, TPM, or VPA at clinician-chosen target doses
- Primary outcomes: Time to treatment failure (AED discontinuation because of seizures or AEs, or both); time to 1-year remission
- Treatment failure: VPA outperformed TPM; comparable with LTG (analysis restricted to GE only — VPA superior to LTG)
- 1-year remission: VPA superior to LTG when all patients included or restricted to GE; VPA did not differ from TPM
- Weight gain: Most frequent AE causing treatment failure with VPA
- Rash: Most common AE causing LTG discontinuation (4%)
- Fatigue & psychiatric/cognitive symptoms: Most common with TPM
Conclusion
- Evidence is insufficient to compare efficacy of LTG and TPM with VPA in children and adults with new-onset or relapsing GE (1 Class III study)
LEV vs VPA-ER or CBZ-CR (KOMET Study, Class III)
- Design: Class III multicenter, randomized, open-label parallel study
- Population: 1,688 outpatient adolescents (≥16 yr) and adults with new-onset epilepsy
- Diagnosis: GE in 34.8%; focal epilepsy in 64.7%; 2.1% unclassified
- Treatment: Clinician chose VPA-ER or CBZ-CR as standard treatment → randomized 1:1 to LEV or the chosen standard AED
- Target doses: LEV 1,000 mg/d; VPA-ER 1,000 mg/d; CBZ-CR 600 mg/d (could increase to LEV 3,000, VPA-ER 2,000, CBZ-CR 1,600 mg/d)
- Patients: 65.8% VPA-ER had GE only; 86.5% CBZ-CR had focal epilepsy only
- Primary outcomes: Time to treatment failure; time to treatment withdrawal
- Time to treatment withdrawal similar for LEV and VPA-ER, and for LEV and CBZ-CR
- Nonsignificantly longer time to treatment withdrawal with LEV than CBZ-CR
- AEs & discontinuation: Comparable across the 3 drugs
- Common AEs: Weight gain and tremor with VPA; depression with LEV; rash with CBZ-CR
- Headache, fatigue, and dizziness equally frequent across AEDs
Conclusion
- Evidence insufficient to compare efficacy of CBZ-CR, LEV, and VPA-ER in adolescents and adults with new-onset GE and focal epilepsy (1 Class III study)
🔹 Clinical Pearl
Both SANAD and KOMET were Class III studies and therefore insufficient to generate formal recommendations. However, SANAD-Arm B showed VPA was superior to LTG for 1-year remission in GE — consistent with VPA remaining the first-line drug for generalized epilepsy. Weight gain is the most common AE leading to VPA failure; rash is the most common for LTG.
Childhood Absence Epilepsy
LTG vs ETS vs VPA (Class I)
- Study: Class I study comparing efficacy, tolerability, and neuropsychological effects
- Doses: LTG 12 mg/kg/d; ETS 60 mg/kg/d; VPA 60 mg/kg/d
- Study outcomes: Freedom from treatment failure after 16 weeks (could extend to 20 weeks if necessary); attention disturbances measured by objective tests (e.g., continuous performance test)
- Freedom from treatment failure rates:
- ETS and VPA → comparable, both significantly higher than LTG
- Attention disturbances: Significantly more common with VPA than with ETS
- Seizure control and cognitive AE differences maintained at 12-month follow-up study
Conclusion
- LTG is probably not as effective as ETS or VPA for treating absence seizures in childhood absence epilepsy (1 Class I study)
- Attention disturbances are more common with VPA than with ETS
Clinical Context
- ETS use is limited to patients with childhood absence epilepsy without associated GTC seizures (ETS has no efficacy against GTC)
- If GTC seizures are present → VPA or LTG (not ETS alone)
Recommendation
| Recommendation | Level |
|---|
| Unless there are compelling adverse-effect–related concerns, ETS or VPA should be considered before LTG to decrease seizure frequency in childhood absence epilepsy | B |
🔹 Clinical Pearl
For childhood absence epilepsy (Level B): ETS or VPA first, then LTG. ETS = VPA for seizure freedom, but VPA causes more attention problems. ETS cannot be used alone if GTC seizures are present. This is a high-yield topic — the 2010 NEJM trial (Glauser et al.) is frequently tested on boards.
Head-to-Head Comparisons — Key Trial Data
| Comparison | Study Class | Population | Key Finding |
|---|
| GBP vs LTG vs CBZ-IR |
II |
Focal, ≥60 yr |
Discontinuation less with LTG/GBP vs CBZ-IR; GBP → weight gain; CBZ → hyponatremia |
| LTG vs CBZ-CR |
I |
Focal, ≥65 yr |
Seizure-free 52% vs 57% (similar); AE withdrawal 14% LTG vs 25% CBZ-CR (NS) |
| LEV vs CBZ-CR |
II |
Focal |
Seizure-free rates nearly identical at 6 & 12 mo; LEV → depression, insomnia |
| ZNS vs CBZ-CR |
II |
74% focal |
Seizure-free rates identical at 26 wk; ZNS → appetite loss, weight loss |
| LTG vs GBP vs TPM vs OXC vs CBZ |
III (SANAD-A) |
89% focal |
LTG best for treatment failure; CBZ best for 12-mo remission; LTG noninferior to CBZ |
| VGB vs CBZ-IR |
I + III |
Focal |
CBZ-IR superior for 6-mo remission & time to first seizure; VGB → visual disturbances |
| PGB 150 vs LTG 100 mg/d |
II |
Focal |
LTG significantly more seizure-free; LTG better for GTC reduction; PGB → weight gain |
| TPM 100 vs PHT |
II |
Focal (28-day) |
Recurrence: TPM 18.9% vs PHT 9.7%; noninferiority not established; only 28-day data |
| LTG vs TPM vs VPA |
III (SANAD-B) |
GE + unclassified |
VPA superior to LTG for 1-yr remission in GE; VPA → weight gain; LTG → rash (4%) |
| LEV vs VPA-ER/CBZ-CR |
III (KOMET) |
GE + focal |
Similar treatment withdrawal times; LEV → depression; VPA → weight gain, tremor |
| LTG vs ETS vs VPA |
I |
Childhood absence |
ETS = VPA > LTG for seizure freedom; VPA → more attention disturbances than ETS |
| TPM 400 vs 50 mg/d |
II |
Children, focal/GTC |
High dose: 85% seizure-free at 12 mo vs 62%; AEs 14% vs 4% |
Adverse Effect Profiles by Drug
| Drug | Notable Adverse Effects |
|---|
| LTG | Rash (most common cause of discontinuation, 4% in SANAD); dizziness, headache, fatigue |
| GBP | Weight gain, water retention, dizziness, ataxia |
| CBZ / CBZ-IR | Rash, hyponatremia, dizziness, somnolence, fatigue |
| CBZ-CR | Headache, fatigue, somnolence, dizziness (less rash than CBZ-IR) |
| LEV | Depression, insomnia, back pain, weight gain; behavioral/psychiatric AEs |
| ZNS | Decreased appetite, weight loss, dizziness, headache, somnolence |
| TPM | Paresthesia, cognitive AEs, psychiatric symptoms, weight loss, decreased appetite, problems with concentration |
| VPA | Weight gain (#1 cause of treatment failure), tremor, attention disturbances (especially in children) |
| OXC | Rash, tiredness, fatigue (comparable to CBZ) |
| VGB | Psychiatric symptoms, weight gain, irreversible visual field defects (scintillating visual disturbances, myoclonic jerks) |
| PGB | Weight gain, headache, dizziness, somnolence, fatigue |
| PHT | Rash (higher incidence leading to discontinuation), dizziness, somnolence |
| ETS | Generally well-tolerated; fewer attention disturbances than VPA |
Clinical Context & FDA Approvals
Limitations of the Evidence
- Studies limited to comparisons between first- and second-generation AEDs only
- Recommendations apply to treatment of focal epilepsy and cannot be generalized to other AEDs not studied
- The single study with GTC seizures secondary to GE was Class III → no recommendations for second-generation AEDs in GTC
- LTG is probably not as effective in absence epilepsy as the 2004 guideline suggested
- No data at all for third-generation AEDs in new-onset epilepsy
FDA Extrapolation Strategy
- Recent FDA strategy allows extrapolation of efficacy across populations for focal epilepsy
- Eslicarbazepine (ESL): Approved as add-on or monotherapy in persons ≥4 years for focal epilepsy
- Lacosamide (LCM): Approved as add-on or monotherapy in persons ≥4 years for focal epilepsy (oral only for pediatric use)
- Perampanel (PER): Approved as monotherapy for focal epilepsy
- These approvals are based on extrapolation, not on new-onset epilepsy clinical trials
Drugs NOT Recommended for New-Onset Epilepsy
- Felbamate (FBM): Not recommended due to serious AEs (aplastic anemia, hepatotoxicity) — safer agents available
- Vigabatrin (VGB): Not recommended due to serious AEs (irreversible visual field defects) — safer agents available
🔹 Clinical Pearl
Third-generation AED FDA approvals via extrapolation: ESL and LCM are approved as add-on or monotherapy for focal epilepsy in patients ≥4 years old; PER is approved as monotherapy for focal epilepsy. These approvals are not based on new-onset epilepsy trials. FBM and VGB are NOT recommended for new-onset epilepsy due to serious AEs.
Recommendations for Future Research
- Second-generation AEDs for focal epilepsy: GBP, LEV, LTG, OXC, and ZNS can be considered; change from Class I to Class III for 2 TPM studies suggests TPM may be possibly effective and should be reinvestigated in an RCT with clinical doses
- Third-generation AEDs: No data on TGB or any third-generation AED (CLB included) for new-onset focal epilepsy
- PGB trial: Should be repeated using higher doses to determine if PGB can be considered efficacious
- OXC: Only second-generation AED with Class I evidence for new-onset focal epilepsy; no studies on second-generation AEDs in new-onset GE with GTC seizures in children or adolescents
- Absence epilepsy: No studies on second-generation AEDs with juvenile absence or JME
- Third-generation AEDs in children: Data unavailable; RCTs are needed in pediatric new-onset epilepsy
- Adults with GE: No data on CLB, VGB in new-onset GE with GTC seizures or JME
- Third-generation AEDs found equivalent to LTG (or CBZ-CR/VPA) for focal epilepsy and GE should undergo head-to-head double-blind, controlled, parallel studies for efficacy
Summary of All Recommendations by Evidence Level
Level A
| Recommendation | Population | Notes |
|---|
| OXC is effective for new-onset focal epilepsy | Adults & children | 4 Class I studies; unchanged from 2004 |
Level B
| # | Recommendation | Population |
|---|
| 1 | LTG should be considered to decrease seizure frequency in new-onset focal epilepsy | Adults |
| 2 | LTG should be considered in patients ≥60 yr with new-onset focal epilepsy | Elderly adults |
| 3 | Unless compelling AE-related concerns, ETS or VPA should be considered before LTG for childhood absence epilepsy | Children |
Level C
| # | Recommendation | Population |
|---|
| 1 | GBP may be considered in patients ≥60 yr with new-onset focal epilepsy | Elderly adults |
| 2 | LEV may be considered to decrease seizure frequency | Adults with focal epilepsy |
| 3 | ZNS may be considered to decrease seizure frequency | Adults with focal epilepsy |
| 4 | VGB appears less efficacious than CBZ-IR and may not be offered; toxicity precludes first-line use | Adults with focal epilepsy |
| 5 | PGB at 150 mg/d is possibly less efficacious than LTG at 100 mg/d | Adults with focal epilepsy |
Level U (Insufficient Evidence)
| # | Statement |
|---|
| 1 | Insufficient evidence to consider GBP, OXC, or TPM instead of CBZ |
| 2 | Insufficient evidence to consider TPM instead of PHT for urgent treatment of new-onset or recurrent focal epilepsy, unclassified GTC, or GE with GTC |
| 3 | Data lacking on third-generation AEDs (CLB, FBM, VGB) in new-onset epilepsy |
| 4 | Data lacking on newer AEDs in treating unclassified GTC seizures |
| 5 | No high-quality studies suggest CLB, ESL, EZG, FBM, GBP, LCM, LEV, LTG, OXC, PER, PGB, RFN, TGB, TPM, VGB, or ZNS is effective in treating new-onset epilepsy (because no high-quality studies exist) |
🔹 Clinical Pearl
Evidence hierarchy for new-onset focal epilepsy (memorize this): Level A = OXC (only one); Level B = LTG; Level C = LEV, ZNS, GBP (≥60 yr). For childhood absence: ETS or VPA before LTG (Level B). There are NO Level A or B recommendations for any third-generation AED in new-onset epilepsy. The most commonly tested concept: OXC is the strongest evidence-based choice, but LTG is most commonly used in practice due to broad-spectrum efficacy and tolerability.
