Status epilepticus is a common neurological emergency, causing a wide spectrum of clinical symptoms associated with significant morbidity and mortality.
J Neurol (1998) 245 : 640–646 © Springer-Verlag 1998
Ayse Sagduyu Sultan Tarlaci Hadiye Sirin
Received: 9 December 1997 Received in revised form: 23 February 1998 Accepted: 30 March 1998
A. Sagduyu (Y) · S. Tarlaci · H. Sirin Ege University, Medical School Hospital, Neurology Department, TR-35100 Bornova – Izmir, Turkey e-mail: tarlaci @ med.ege.edu.tr Tel./Fax: +90-232-388-09-80
O R I G I N A L C O M M U N I C AT I O N
Generalized tonic-clonic status epilepticus: causes, treatment, complications and predictors of case fatality
Abstract We retrospectively reviewed the clinical course of 66 patients treated for generalized tonicclonic status epilepticus at the Ege University neurological intensive care unit from 1988 to 1997. Seventy-two per cent of the study group had a pre-existing seizure disorder, and antiepileptic drug withdrawal was the most prominent cause of status epilepticus. The other causes included drug toxicity, central nervous system infection, cerebrovascular disease, tumour and trauma. Seventy-three per cent of all patients responded to the first-line therapy (diazepam and/or phenytoin), and the
Introduction Status epilepticus is a common neurological emergency, causing a wide spectrum of clinical symptoms associated with significant morbidity and mortality. It has a highly variable pathophysiological, anatomical and aetiological basis. Despite improvements in the management of patients with status epilepticus in advanced neurological intensive care units (ICUs), mortality is still high, ranging from 8 to 32% [2, 3, 8, 14, 18], indicating that there still is a substantial need to improve measures for both the prevention and effective management of this syndrome. Although a new classification with basic subdivisions based on age has already been proposed [19], status epilepticus is traditionally classified with reference to seizure types [7], and the most life-threatening pattern and that requiring the most urgent treatment is the generalized tonic-clonic form.
remainder were considered to have refractory status epilepticus and required pentobarbital anaesthesia. Overall case fatality was 21%, but death could be attributed directly to status epilepticus and/or treatment complication in 10% of the study group. Major determinants of fatal outcomes were: increasing age, longer duration of status epilepticus before initiation of therapy and central nervous system infection as a causal factor. Key words Status epilepticus · Causes · Treatment · Case fatality
Prognosis depends on the time allowed to elapse between onset of status epilepticus and initiation of effective treatment as well as on age, seizure type, aetiology and underlying neurological and systemic condition [3, 8, 10, 11, 14, 18, 21]. In the present retrospective hospital-based study, we describe 66 patients treated for generalized status epilepticus at the Ege University neurological ICU of the Neurology Department. We evaluated the clinical features, demographic data, aetiological and prognostic factors, complications and the response to treatment.
Patients and methods We identified the hospital records of all patients admitted with the diagnosis of status epilepticus to the neurological ICU of the Neurology Department, Ege University Medical School Hospital, from 1988 to 1997. The majority of the patients were direct referrals to Ege University Hospital and the rest were referred from the community hospitals around the city. They were all white Turkish peo-
Statistical methods If normally distributed, continuous variables were compared using the independent sample t test and chi-square test; if not, MannWhitney U test. A P value of less than 0.05 was considered significant. Odds ratio (OR) and significant levels were calculated along with a 95% confidence interval (CI). Analyses were performed with SPSS software for Windows (release 6.1) packages.
Results Sixty-six patients (28 women and 38 men) with a mean age of 33.9 years (range 6–77) comprised the study group. Since our neurological ICU serves mainly adult patients, all patients were older than 16 years except 5 children of school age. Nine (14%) patients had more than one episode of status epilepticus. All patients with generalized status epilepticus included in the study had only tonic-clonic seizure type; 47 (71%) of them had primary, 19 (29%) had
Trauma
Tumour
Idiopathic
Metabolic
Intoxication
Tumour
Trauma
CVD
CNSI
The treatment protocol of generalized status epilepticus was mainly based upon a commonly used published regime [4]. After standard initial evaluation, all patients received diazepam (DZP) 10 mg intravenous (IV) bolus or 2 mg/min infusion up to 20–30 mg cumulative doses. If the seizures persisted 30 min after the first injection, phenytoin (DPH) was infused at a dose of 18 mg/kg at a rate of 50 mg/min. If the patients continued to convulse after delivery of these drugs, pentobarbital bolus 5 mg/kg was administered at a rate of 25 mg/min, and the maintenance dose of 2 mg/kg per hour was infused and that dose adjusted according to the patient’s response.
Post-stroke
Fig. 1 Aetiologies of primary seizure disorders in patients with a history of epilepsy
AEDW
Treatment protocol
Idiopathic
Number of patients
ple. Patients’ hospital records were reviewed initially to determine whether they fulfilled the international classification of epileptic seizures criteria for status epilepticus [7]. Status epilepticus was defined as focal or generalized seizures (determined by clinical observation) that were continuous for at least 30 min, or repeated generalized seizures without recovery of consciousness between attacks and continuing for at least 30 min. Sixty-six patients’ medical records contained information consistent with the diagnosis of status epilepticus. In cases where a patient had multiple hospital admissions related to status epilepticus, only the latest admission was taken into consideration. Seizure types in status epilepticus were defined as partial or generalized status epilepticus based on the International Classification of Seizure Types and types of status epilepticus as defined by the International League Against Epilepsy [7]. Classification of seizure types was based on a review of the patient’s total medical records. The patients who lacked detailed initial admission history and clear identification of the details of the seizure presentation in their medical charts were excluded from the study. All patients’ medical records were evaluated in detail to obtain the following information: demographic data, details of history of seizures or epilepsy, previous or coexisting medical problems, likely aetiologies of status [i.e. alcohol withdrawal, alcohol or drug toxicity, systemic and/or central nervous system (CNS) infections, antiepileptic drug withdrawal (AEDW), cerebrovascular disease (CVD), metabolic disorders, tumour, trauma] in both previously epileptic and non-epileptic groups, seizure type according to the status epilepticus causation, clinical response to antiepileptic drug regimes for status control, need for intubation and mechanical ventilation, the relation between the duration of status epilepticus before initiation of treatment and the outcome, complications during status control, predictors of case fatality and prognostic factors.
Number of patients
641
Fig. 2 Primary aetiologies of status epilepticus in patients with (black bar) and without (white bar) history of epilepsy
secondary generalization. There were only 3 patients with simple partial status epilepticus, so they were excluded from the study, since this group was not large enough for statistical analysis. Unfortunately, no patient met the criteria for complex partial status epilepticus. There was no significant difference between primary or secondary generalized tonic-clonic status epilepticus in terms of demography or causation. Aetiology of status and pre-exisiting seizure disorders Forty-eight (73%) of 66 patients had a history of epilepsy, and the most common cause of status epilepticus was an-
642
Table 1 General features of patients according to antiepileptic drug regimes for status control (AEDW antiepileptic drug withdrawal, CNSI central nervous system infection, CVD cerebrovascular disease, PriGTC primary generalized tonic-clonic convulsion, SecGTC secondary generalized tonic-clonic convulsion, DZP diazepam, DPH phenytoin, CLO chloralhydrate, PB pentobarbital)
DZP
DZP + DPH
DZP + DPH + PB
DZP + DPH + CLO
n
35 (53.0%)
20 (30.3%)
10 (15.2%)
1 (1.5%)
Sex Female Male Mean age (years) Previously epileptic First seizure
19 16 32.4 30 5
14 16 37.1 12 8
5 5 35.6 5 5
1 6 1
3 3
2 6
5 2 1 1 2 1 1 7
8 27
5 15
5 5
2
6
10
3 32
7 13
4 6
Aetiology of status epilepticus AEDW CNSI CVD Trauma Tumour Metabolic Drug intoxication Idiopathic Type of status epilepticus SecGTC PriGTC Need for ventilation Prognosis Death Survivor
ticonvulsant drug withdrawal (n = 32) in these patients. The other causes included drug toxicity or induction (n = 3), CNS infection (n = 3), tumour (n = 2), trauma (n = 1), idiopathic (n = 7). Drugs implicated as a cause of status epilepticus included isoniazid (n = 1) in toxic dosage and clomipramine (n = 1), penicillin (n = 1) in therapeutic doses. The causes of the pre-existing epilepsy in these 48 patients were idiopathic in 26 (39%), post-stroke in 7, trauma in 8 and tumour in 7, as shown in Fig. 1. In patients without a history of epilepsy (n = 18), the primary causes of status epilepticus were CNS infection (n = 4), CVD (n = 2), tumour (n = 2), trauma (n = 1), clomipramine overdose because of suicidal attempt (n = 1), metabolic disorder (n = 1), idiopathic (n = 7). Figure 2 shows the primary cause of status epilepticus in both groups of patients with and without pre-existing seizure disorder. In comparing the likely cause of status epilepticus and the primary seizure disorder in both groups of patients, anticonvulsant drug withdrawal was seen to be the main cause of status epilepticus in patients with pre-existing idiopathic epilepsy. Response to the treatment We evaluated the patients’ charts according to the antiepileptic drug regimes for status control. All of the patients
23 2 1 1
1
1 1 1 1 1
1
received first-line therapy of IV DZP (bolus or infusion): 32 (48%) had status control and survived except for 3 patients of whom 1 died of pneumonia after status control and 2 died without status control owing to herniation resulting from CVD and cardiovascular disease (acute myocardial infarction). There were 20 patients in the second step of therapy and seizure control was sustained after administration of adequate doses of IV DPH in addition to DZP in 16 (80%) patients, which was 24% of the study group. Of these, 3 patients died after seizure control because of underlying disease (tumour, CNS infection, unidentified). Four patients (20%) in this group died without status control because of cardiac arrest before completing the loading dose of DPH. The rest survived without irreversible neurological sequelae. Of 11 patients who were refractory to standard therapy, seizures ended after IV pentobarbital administration in 8 (12% of the study group) and after rectal chloralhydrate administration in one case. Two patients had a fatal outcome in a short time without status control in the pentobarbital group owing to pulmonary oedema in 1, and secondary to severe hypotension in the other. Of the 8 patients treated with pentobarbital with sustained seizure control, 2 died because of unidentified cause in one case and cardiac arrest in the other. Table 1 illustrates the general features of patients according to different antiepilep-
643
Table 2 Profiles of patients who died during their treatment for status epilepticus
Table 3 Risk factors for case fatality (OR odds ratio, CI confidence interval)
Patient
Age (years) sex
PreAetiology existing of status epileptic epilepticus
Seizure duration before treatment (h)
Control of status epilepticus
Need for Cause of death ventilation
1 2 3 4 5 6 7 8 9 10 11 12 13 14
71/F 39/M 72/M 20/M 19/M 61/M 47/M 34/F 19/F 17/M 77/M 34/F 60/M 75/M
– + + + – + – – + + + + – +
4 3 0.5 3 2 5 4 4 3 2 5 9 5 3
+ + + – – – – – – – + – + +
+ + – + + + – + – – + – – –
CNSI Tumour AEDW CVD CNSI AEDW Idiopathic CNSI AEDW AEDW AEDW AEDW Idiopathic AEDW
Factor
Case fatality n
Age 45 years
[%] 7 7
12.2 77.7
Sex M F
4 10
14.2 26.6
Seizure duration before treatment 1h
1 13
3.0 39.4
8 6
16.7 33.3
12 2
25.5 10.5
Aetiology of status epilepticus AEDW CNSI Stroke Tumour Idiopathic
5 5 1 1 2
15.6 71.4 50.0 25.0 16.6
Hyperpyrexia < 38° C > 38° C
8 6
16.0 37.5
Need for ventilation No Yes
6 8
12.5 44.4
Pre-existing epilepsy Yes No Type of seizure PriGTC SecGTC
Herniation Herniation Cardiac arrest Herniation ? Pulmonary oedema Cardiac arrest Cardiac arrest Cardiac arrest Cardiac arrest Cardiac failure Pulmonary oedema ? Cardiac failure
OR
P-value
95% CI
2.880
0.011
1.3084–6.3784
2.410
0.0003
1.6622–1.6945
13.881
0.014
1.9444–4.0038
1.575
0.004
1.0279–2.4134
644
tic drug regimes. The majority of patients responsive to DZP had a history of epilepsy, and AEDW was the main cause of status epilepticus. Complications Complications seen during the management of status epilepticus were evaluated in two groups in terms of severity. Mild events comprising the first group were: gastrointestinal haemorrhage (2 patients), uraemia (2 patients), mild elevation of hepatic enzymes (11 patients), pneumonia (4 patients), hypotension owing to DZP/pentobarbital infusion (4 patients), tachycardia (25 patients), hyperpyrexia (14 patients), leucocytosis (25 patients). Severe events comprising the second group were: aspiration pneumonia and fatal pulmonary failure (2 patients), significant hypotension (systolic blood pressure < 70 mmHg) owing to pentobarbital infusion (1 patient), hepatic failure owing to DPH (initially infusion, then oral administration in therapeutic doses) (1 patient), bradycardia owing to cardiovascular insufficiency (pulse rate < 50) (1 patient), malign hyperpyrexia owing to underlying disease (CNS infection and CVD) (2 patients), prolonged apnoea owing to DZP infusion (2 patients), significant tachycardia (pulse rate > 180) (5 patients). Of the 5 patients with significant tachycardia, 4 developed cardiac arrest during DPH infusion before termination of status epilepticus. The aetiology of status epilepticus was AEDW in 2 and CNS infection or idiopathic in the rest. Although they were not autopsied, the cause of death was supposed to be owing to DPH infusion in 3 cases. Case fatality Fourteen patients (21%) died during their treatment in our neurological ICU, and status control could not be sustained in 8 (12%) of these patients. Death could be attributed directly to status epilepticus and/or treatment complications in 6 (10%) of them (Table 2), although neuropathological examination was lacking. We performed multiple regression analysis to determine the potential predictive indicators associated with fatality (Table 3). The mean age was significantly higher in the fatal group than that of survivors (46.0 and 30.6 years, respectively) (P = 0.004). Case fatality for male and female patients was 32% and 35%, respectively, and this difference was not statistically significant. Case fatality rates did not show statistically significant differences in patients with and without pre-existing seizure disorder or in patients with primary or secondary type of generalized status epilepticus. We also investigated the effect of duration of status on outcome before initiating antiepileptic therapy. To deter-
mine whether a difference in fatality rate existed between patients with prolonged versus non-prolonged seizure duration, the data sets were divided into two groups: the patients whose treatment for status epilepticus started within 1 h (non-prolonged) or later than 1 h (prolonged). Our results show a clear relation between seizure duration before initiating therapy and case fatality (OR 2.410; 95% CI, 1.6622–1.6945; P = 0.0003). Among the aetiologies of status epilepticus in patients with fatal outcome, only CNS infection was associated with significant case fatality in multivariate analysis (OR 13.881; 95% CI, 1.9444–4.0038; P = 0.014). No specific cause was identified from these patients’ cerebrospinal fluid (CSF) examinations, so they were all supposed to be viral in origin. Need to intubate was also significantly related to poor outcome (OR 1.575; 95% CI, 1.0279– 2.4134; P = 0.004).
Discussion The intent of our study was to evaluate our patients with generalized status epilepticus, whether primary or secondary, in terms of demographics, aetiology, therapeutic and prognostic factors. Demographic data and the aetiology of status epilepticus did not show a statistical significance in seizure type subgroups, either primary or secondary generalized tonic-clonic convulsion, as has been indicated in other reports [8, 10], even though partial with secondary generalization was reported as the major seizure type in the elderly by De Lorenzo et al. [3]. In our study, 73% of our patients had pre-existing seizure disorder, which was also discordant with the results of the latest population-based survey of status epilepticus [3]. AEDW was the main cause of status epilepticus in 66% of these, which was 48% of the whole study group. In 27% of the patients, status was the first epileptic event, and the cause of status epilepticus was unidentified in 39% of them. Our results are similar to other reports [1, 2, 5, 6, 10], emphasizing the point that most of the patients presenting with status epilepticus in the medical emergency wards have pre-existing epilepsy, which is a consequence of antiepileptic drug discontinuation in more than half of them. It is also notable that isoniazid, clomipramine and penicillin might induce status epilepticus in patients with epilepsy, and in about 21% of all patients, the cause for status epilepticus could not be identified. Although it is commonly reported in the literature [1, 3, 8, 14, 15], there was no patient with alcohol-related seizures among the other likely aetiologies of status epilepticus in our study group. Almost half of the patients in our study group responded succesfully to only IV DZP therapy. Only 3 patients died, but death could not be attributed directly to status epilepticus in them because the cause of death was clinically clear in 2 non-responders, and the 3rd one died
645
of pneumonia after status control. Of the 31 patients who did not respond to IV DZP, 16 responded to a loading dose of IV DPH. The results show that 68% of our patients with status epilepticus responded successfully to initial 10–30 mg IV DZP, followed if necessary by 18 mg/kg of DPH. We assume that this therapeutic approach is probably representative of the standard practice of most neurologists [4, 9, 17, 18, 20]. Patients who failed to respond to the first two drugs were considered to have refractory status epilepticus, and pentobarbital was administered to all except one patient who received rectal chloralhydrate even though it was not routinely used. Pentobarbital is most widely accepted as a drug of choice in patients with refractory status epilepticus [4, 12, 13, 15, 18, 22], but the high mortality rate in some series [16, 22] may discourage its use. Four patients died in our study group who received pentobarbital: two during anaesthesia without effective status control, which might be attributed directly to prolonged status or pentobarbital anaesthesia complication; the other two patients’ deaths were unexpected, because they occurred after status control and cessation of treatment. Although our study group was not large enough to provide a clear idea of the efficacy and safety of pentobarbital anaesthesia in status control, we suggest that 50% case fatality in this group was associated not only with drug complications, but also with underlying disease and status duration. Complications seen during the management of status epilepticus in our study group were similar to those in other reports [1, 8, 14, 18, 21, 22], and most of them were not serious and were successfully managed in our ICU except for patients in whom the cause of death could not be
identified by autopsy. However, severe complications were also seen during the treatment in our study group, and four patients’ deaths might be attributed to treatment complications. We analysed several determinants of fatal outcome in status epilepticus. Our results show statistically significant association of age, duration of status epilepticus and aetiology with case fatality in status epilepticus, and these findings are consistent with certain previous observations [1, 3, 8, 21]. We found that mean age was significantly higher in the fatal group, and a poor outcome of status epilepticus was more likely as its duration before initiation of therapy increased. On the other hand, CNS infection was the only aetiology associated with significant case fatality in multiple regression analysis in our study, whereas it was reported to have a comparatively better prognosis by De Lorenzo and Lowenstein [3, 10]. Towne et al. reported that aetiologies including CNS infection, anoxia and haemorrhage were associated with high mortality, but in multivariate analysis only anoxia was associated with significant mortality [21], as also indicated by De Lorenzo et al. [3]. In conclusion, our results indicate that in generalized status epilepticus, AEDW is the main cause and seizures can successfully be controlled with first-line therapy (DZP and DPH) in about two-thirds of the patients. We also think that pentobarbital infusion is very effective in suppressing seizures in refractory status epilepticus. However, mortality remains unacceptably high, and age, duration of status and underlying disease, especially CNS infection, are the major determinants of fatal outcome.
References 1. Aminoff MJ, Simon RP (1980) Status epilepticus: causes, clinical features and consequences in 98 patients. Am J Med 69 : 657–666 2. Barry E, Hauser WA (1994) Status epilepticus and antiepileptic medication levels. Neurology 44 : 47–50 3. De Lorenzo RJ, Hauser WA, Towne AR, Boggs JE, Pellock JM, Penberthy L, Gardnett L, Fortner CA, Ko D (1996) A prospective, population-based epidemiologic study of status epilepticus in Richmond, Virginia. Neurology 46 : 1029–1035 4. Delgado-Escueta AV, Wasterlain C, Treiman DM, Porter R (1982) Current concepts in neurology: management of status epilepticus. N Engl J Med 306 : 1337–1340
5. Duncan JS, Shorvon SD, Trimble MR (1988) Withdrawal symptoms from phenytoin, carbamazepin and sodium valproate. J Neurol Neurosurg Psychiatry 51 : 924–928 6. Duncan JS, Shorvon SD, Trimble MR (1990) Discontinuation of phenytoin, carbamazepine and valproate in patients with active epilepsia. Epilepsia 31 : 324–333 7. Gastaut H (1983) Classification of status epilepticus. In: Delgado-Escueta AV, Wasterlain CG, Treiman DM, Porter RJ (eds) Status epilepticus: mechanisms of brain damage and treatment. Advances in neurology, vol 34. Raven Press, New York, pp 15–35
8. Hauser WA (1982) Status epilepticus: frequency, etiology and neurological sequelae. In: Delgado-Escueta AV, Westerlain CG, Treiman DM, Porter JM (eds) Status epilepticus: mechanisms of brain damage and treatment. Advances in neurology, vol 34. Raven Press, New York, pp 3–14 9. Leppik IE, Patrick BK and Cranford RE (1983) Treatment of acute seizures and status epilepticus with intravenous phenytoin. In: Delgado-Escueta AV, Wasterlain CG, Treiman DM, Porter JM (eds) Status epilepticus: mechanisms of brain damage and treatment. Advances in neurology, vol 34. Raven Press, New York, pp 447–451 10. Lowenstein DH, Alldredge BK (1993) Status epilepticus at urban public hospital in the 1980s. Neurology 43 : 483– 448
646
11. Lowenstein DH, Aminof MJ (1992) Clinical and EEG features of status epilepticus in comatose patients. Neurology 42 : 100–104 12. Lowenstein DH, Aminoff MJ, Simon RP (1980) Barbiturate anesthesia in the treatment of status epilepticus: clinical experience with 14 patients. Neurology 38 : 395–400 13. Osorio I, Reed RC (1989) Treatment of refractory generalized tonic-clonic status epilepticus with pentobarbital anesthesia after high-dose phenytoin. Epilepsia 30 : 464–471 14. Oxbury JM, Witty CWM (1971) Causes and consequences of SE in adults: a study of 86 cases. Brain 94 : 733–744
15. Pilke A, Partinen M, Kovanen J (1984) Status epilepticus and alcohol abuse; an analysis of 82 status epilepticus admissions. Acta Neurol Scand 70 : 443– 450 16. Rashkin MC, Youngs C, Penovich P (1987) Pentobarbital treatment of refractory status epilepticus. Neurology 37 : 500–503 17. Shaner DM, McCurdy SA, Herring MO, Gabor AJ (1988) Treatment of status epilepticus: a prospective comparison of diazepam and phenytoin versus phenobarbital and optional phenytoin. Neurology 38 : 202–207 18. Shorvon S (1995) The management of status epilepticus. In: Hopkins A, Shorvan S, Cascino G (eds) Epilepsy, 2nd edn. Chapman & Hall, London, pp 331–354 19. Shorvon S (1997) The classification of status epilepticus. J Neurol 244 [Suppl 3] : 53–54
20. Tassinari CA, Daniele O, Michelucci R, Bureau M, Dravet C, Roger R (1983) Benzodiazepines: efficacy in status epilepticus. In: Delgado-Escueta AV, Wasterlain CG, Treiman DM, Porter RJ (eds) Status epilepticus: mechanisms of brain damage and treatment. Advances in neurology, vol 34. Raven Press, New York, pp 465–475 21. Towne AR, Pellock JM, Ko D, De Lorenzo RJ (1994) Determinants of mortality in status epilepticus. Epilepsia 35 : 27–34 22. Yaffe K, Lowenstein DH (1993) Prognostic factors of pentobarbital therapy for refractory generalized status epilepticus. Neurology 43 : 895–900
