SEARCH

SEARCH BY CITATION

Keywords:

  • Complex-partial status epilepticus;
  • Outcome;
  • Serum level

Summary

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

Oral antiepileptic drugs (AEDs) represent possible add-on options in refractory status epilepticus (SE). In this setting, pregabalin (PGB) has not been reported before. Over the last 42 months, we identified 11 SE episodes (10 patients) treated with PGB in our hospital. Its use was prompted by the favorable pharmacokinetic profile, devoid of drug-drug interactions. The patients mostly had refractory, partial SE. Only two patients were managed in the intensive care unit (ICU). We found a definite electroclinical response in 5 of 11, already evident 24 h after PGB introduction, and a possible response (concomitantly with other AEDs) in 3 of 11 of the episodes; 3/11 did not respond. The treatment was well tolerated. Partial SE appeared to better respond than generalized convulsive SE. PGB appears to be an interesting option as add-on treatment in refractory partial SE.

Status epilepticus (SE) represents a serious condition, with considerable morbidity and mortality. Its emergency treatment usually consists of intravenous administration of benzodiazepines followed by antiepileptic drugs (AEDs) (mostly phenytoin, valproate, or phenobarbital) (Shorvon, 2001). Although in SE refractory to this approach anesthetics agents are often advocated, other AEDs might be used instead, in order to prevent mechanical intubation and related complications. Among AEDs of the new generation, only topiramate (Towne et al., 2003), levetiracetam (Rossetti & Bromfield, 2006; Knake et al., 2008), and, more recently, lacosamide (Kellinghaus et al., 2009) have been described in this context. Although the latter two are available in intravenous formulations, topiramate can be administered only through the nasogastric tube.

Pregabalin (PGB) has been on the market since late 2006. Despite being available only as an oral formulation, this compound represents a potentially interesting option for critically ill patients, as it can be titrated quickly (Ramsay et al., 2009), has good oral bioavailability, and is devoid of hepatic interactions (Ben-Menachem, 2004). We describe here our experience regarding PGB use in subjects with SE.

Methods

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

In our prospective database of patients treated for SE in our center (a tertiary referral hospital) between April 2006 and September 2009, we searched for episodes in which PGB was prescribed. The database, which includes patients older than 16 and received full approval from our ethics commission, has been described previously (Novy et al., 2010). Briefly, SE was defined as the clinical occurrence of ongoing or repeated epileptic seizures without full recovery in between for more than 30 min. If a patient with simple partial seizures presented two or more seizures over at least 30 min without clinical recovery in between, this was considered SE. The type of SE was determined according to the type of the worst seizure (categorized as simple partial, complex partial, or generalized convulsive). All patients underwent repeated electroencephalography (EEG) or EEG monitoring; brain imaging [computed tomography (CT) or magnetic resonance imaging (MRI)], laboratory workup, and cerebrospinal fluid (CSF) analysis were performed as needed.

Demographic and clinical data were retrieved from the database and the computerized patients charts of our hospital, including information about cardiovascular or other possible side effects of medication (particularly, vigilance worsening). SE refractoriness was defined as resistance to benzodiazepines and the first AED in adequate doses. SE termination was defined as resolution of clinical ictal signs along with disappearance of continuous or repetitive epileptiform discharges on the EEG, assessed at least daily. SE response to PGB was categorized as successful (occurring within 24 h following PGB introduction, without modification of concomitant AED), possible (SE termination associated with introduction/increase of PGB concomitantly with other medications), or unsuccessful (no response to PGB treatment). Outcome was measured at hospital discharge, according to the Glasgow–Pittsburgh Cerebral Performance Categories (CPC) (Booth et al., 2004), with CPC 1–2 defined as good outcome.

Results

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

Among 230 SE episodes that occurred over 42 months, we identified 15 in which PGB was prescribed. Four patients were already treated with PGB before the index SE, and the dosage was not modified; they were, therefore, excluded. Clinical data of the remaining 11 episodes (10 patients, episodes six and eight occurred in the same subject, the second one following PGB withdrawal) are summarized in Table 1. We observed a female predominance (10 of 11), with a median age of 57 years (range 20–80 years). Most patients had partial SE (9 of 11), and SE was most often refractory (9 of 11). We diagnosed an acute symptomatic etiology in four cases, and three patients had de novo SE.

Table 1.   Summary of patient’s data
NGenderAgePrevious seizuresEtiologySeizure typeConsciousnessOutcome (CPC)RefractoryIntubated for treatmentOther treatmentsTime to PGB introductionPGB titration timePGB maximal dosage (mg/day)PGB success?Time between introduction and responseSerum levels (μmol/L)
  1. Seizure types are categorized as generalized convulsive (GC), complex partial (CP), or simple partial (SP). Outcome was assessed according to Glasgow–Pittsburgh Cerebral Performance Categories (Booth et al., 2004).

  2. CLZ, clonazepam; LEV, levetiracetam; PHT, phenytoin; VPA, valproate; PB, phenobarbital; MDZ, midazolam; TPM, topiramate; PRO, propofol; F, female; M, male; WHO, World Health Organization (grade); NMDA. N-methyl-d-aspartate.

  3. aPatient 3 presented initially with generalized convulsive SE that evolved into partial complex SE; he was treated in this second phase with PGB.

  4. bAmong patients with simple partial SE, two patients had motor and aphasic symptoms; the last patient had motor symptoms and was already somnolent before the SE because of brain hemorrhage.

  5. cGiven through the nasogastric tube.

 1F74NoAcute hemorrhageSPSomnolentb2YesNoCLZ, LEV  5 days 1 day150Yes24 h
 2F64YesGlioma WHO IVSPAlertb2YesNoCLZ, LEV  1 day 3 days600Yes24 h
 3M57YesPoor complianceCPaStuporous1YesNoCLZ, LEV  5 days 2 days300cNo
 4F50YesGlioma WHO IVSPAlertb1YesNoCLZ, LEV  0 day 1 day150Possible48 h (with LEV increase)
 5F80NoMeningiomaCPConfused3YesNoCLZ, LEV, PHT  2 days 5 days450PossibleProgressive improvement during titration50
 6F63YesOld hemorrhageCPConfused1YesNoCLZ, LEV, VPA  6 days 2 days450cPossible72 h (with VPA increase)30
 7F34YesCryptogenicCPConfused1YesNoCLZ, PHT  4 days 3 days600Yes24 h32
 8F64YesPoor complianceCPConfused1NoNoCLZ  0 day 1 day600Yes24 h
 9F57YesGlioma WHO IIICPStuporous2NoNoVPA  5 days 2 days300Yes24 h
10F25YesAnti-NMDA encephalitisGCStuporous2YesYesPB, PHT, MDZ 31 days 2 days600cNo19
11F20NoCryptogenicGCStuporous3YesYesMDZ, PHT, LEV, PRO, TPM105 days12 days600cNo

In all episodes, PGB was used as an add-on after failure of other compounds to control SE. The treatment was administered orally in patients with simple-partial SE and through the nasogastric tube in patients who were deeply somnolent or stuporous. In these cases, the capsule was opened and the powder was admixed with water and injected. The median latency to PGB introduction after SE diagnosis was 5 days (range 0–105 days). The median daily target dose was 450 mg (range 150–600), prescribed in two fractioned administrations; this was achieved after a median of 2 days (range 1–13 days). PGB residual serum level was measured in episodes 5, 6, 7, and 11, with results within the therapeutic range (median 31 μmol/L; range 19–50 μmol/L; therapeutic range in our lab 5–55 μmol/L).

Considering all 11 episodes, there was no clinical response to PGB in 3, whereas 3others possibly responded. Five episodes were very likely controlled by PGB, and the response occurred within 24 h after PGB introduction. Regarding SE subgroups, 5 of 9 of partial SE episodes versus 0 of 2 of generalized convulsive episodes were responsive to PG; however, PGB was started after more than 1 month in both generalized convulsive cases. SE duration before PGB introduction, as well as PGB titration speed or target dosage, did not show any clear correlation with treatment success, possibly owing to the limited sample size.

All patients were discharged alive, with 9 of 11 achieving a good outcome according to their CPC. PGB administration did not produce any cardiovascular side effects, and there was no reported complication. Particularly, there was no need for mechanical ventilation because of a sedating effect, and the medication was retained until hospital discharge in all episodes.

Discussion

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

We report here our experience concerning patients with mostly refractory SE treated with add-on PGB. This relatively small series, which to the best of our knowledge represents the first description of the use of this compound in SE, suggests that PGB is efficacious and well tolerated in this setting.

PGB showed a good tolerability, and was efficacious at usual dosages (150–600 mg/day, with therapeutic serum levels), with a responder rate (45%) similar to that of previous reports on oral levetiracetam (Rossetti & Bromfield, 2006). The treatment was efficacious even if administered by the oral route, as already described for topiramate (Towne et al., 2003) and levetiracetam (Rossetti & Bromfield, 2006): this corroborates our recent observation that intravenous administration is not always imperative for refractory SE (Novy et al., 2010). The five responders improved their clinical condition within 24 h of PGB administration; the fast action of this compound in SE reflects recent data on patients with refractory epilepsy (Ramsay et al., 2009). Three further patients responded after 48 h, but since they also received increasing doses of other AEDs, it is difficult to judge the specific role of PGB.

The clinical response appeared to be better in partial as compared to generalized convulsive SE; this should, however, be put into the context of the considerable delay of PGB administration in the two patients with generalized convulsive SE, suffering from a severe refractory SE with multiple intensive care unit (ICU) complications that possibly affected intestinal resorption. A long delay before SE treatment is known to represent an important predictor of poor response (Shorvon, 2001; Rossetti & Bromfield, 2006), especially in the first 24–48 h (Drislane et al., 2009). It is interesting to note that the administration through the nasogastric tube (that occurred in the three nonresponders) does not seem to have significantly altered the drug absorption, since the PGB serum level was in the reference interval in two patients. Titration speed or the target dosage of PGB did not appear to determine the clinical response.

From a theoretical perspective, PGB could represent a good pharmacodynamic complement to sodium channel blockers, γ-aminobutyric acid (GABA) agonists or other wide-spectrum AEDs used in SE, owing to its calcium channel modulation (Dooley et al., 2000). Moreover, this compound has an extremely favorable pharmacokinetic profile, with pure renal elimination and no interactions with others drugs, suggesting that PGB is particularly useful in subjects needing other treatments, such as AEDs, antibiotics, immunosuppressives, or cytostatic chemotherapy. The fact that our patients were already receiving polypharmacy and we wanted to avoid pharmacokinetic interactions probably led to a selection of our patients, who had brain tumor or encephalitis in about half of the cases.

This study is limited by its small sample size. Moreover, although our patients mainly had refractory SE, almost half of them had self-limited conditions as self-limited SE etiologies, and most of them also were known for previous seizures, a feature considered to herald a better prognosis than de novo episodes (Holtkamp et al., 2005). The uncontrolled aspect of this study also limits the significance of the results. However, a “regression to the mean,” that is, a spontaneous SE resolution, is unlikely to have occurred in our patients, since refractory SE is rarely self-limited.

In our opinion, since refractory SE may at times be managed outside of the ICU (Novy et al., 2010), the adjunction of PGB could represent an interesting alternative in partial refractory SE before considering coma induction, which has a potential burden of complications.

Acknowledgments

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

Dr. Rossetti received research support from UCB, Pfizer, Astra-Zeneca, and Janssen-Cilag.

Disclosure

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

Dr Novy has nothing 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.

References

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References