Nonconvulsive status epilepticus (NCSE) is an epileptic condition lasting ≥30 min, clinically manifested by an altered mental state and associated with continuous or recurrent epileptiform activity on electroencephalogram (EEG). All of these symptoms exist without overt tonic, clonic, or tonic–clonic movements. In general, an immediate or short-term clinical improvement will be achieved with antiepileptic treatment. Although diverse pathologic processes such as cancer, ring chromosome 20, neurosyphilis, subarachnoid hemorrhage, or Creutzfeldt–Jakob disease have been related to this epileptic entity, the diagnosis of drug-induced NCSE has notably increased on recent years.
A 76-year-old woman with antecedents of chronic alcoholism and neutropenia secondary to the use of antibiotics was admitted to our hospital with the diagnosis of gangrenous pyoderma infected with Pseudomonas aeruginosa. She was receiving treatment with teicoplanin (400 mg/24 h) and fluconazole (100 mg/24 h). No history of seizures or other neurologic disorders was known. Cefepime (2 g/8 h) was added to her antibiotic regimen to broaden P. aeruginosa coverage. Over the following days, she developed a progressive alteration of consciousness level and agitation. Lumbar puncture and computed tomography (CT) scan of the brain were unremarkable. Laboratory studies were normal except for a creatinine level of 3 mg/dl and blood urea nitrogen of 95 mg/dl. An EEG revealed continuous rhythmic generalized bi–triphasic 2.0- to 3.0-Hz sharp waves with frontal maximum. The administration of 5 mg of diazepam (DZP) abolished completely the epileptiform activity, and the patient fell deeply asleep. Cefepime was discontinued, and intravenous treatment with phenytoin (PHT; 100 mg/8h) was started. Thirty-six hours later, the mental status of the patient was considered normal.
A 38-year-old woman with antecedents of hypernephroma and allogenic bone marrow transplantation was admitted to our hospital because of neutropenia and fever. She was receiving treatment with cyclosporine (75 mg/24 h). Blood cultures were positive for Serratia marcescens. Intravenous antibiotic therapy with cefepime (2 g/8 h) was started. Five days later, she became agitated and disorientated. No focal neurologic signs or meningism was seen. Laboratory studies showed a creatinine level of 3.2 mg/dl and blood urea nitrogen of 96 mg/dl. Examination of the cerebrospinal fluid and magnetic resonance imaging of the brain were normal. An emergency EEG revealed continuous generalized rhythmic sharp-and-slow wave complexes and bi–triphasic sharp waves at a rate of 2 to 2.5 Hz (Fig. 1). The status was rapidly terminated by intravenous infusion of 5 mg DZP. Thereafter, the patient became alert and was able to answer simple questions. Intravenous cefepime was discontinued, and antiepileptic treatment with PHT (100 mg/8 h) was initiated. Because the episodes of confusion recurred, NCSE was completely controlled with intravenous benzodiazepines (BZDs) and PHT.
A 43-year-old man with a history of congenital megacolon and chronic alcoholism was admitted to our hospital with the diagnosis of abdominal sepsis. He was receiving treatment with amikacin (750 mg/24 h), vancomycin (1 g/24 h), and fluconazole (200 mg/12 h). Cefepime (2 g/12 h) was added to his antibiotic therapy. A picture of abnormal behavior progressed to a state of severe mutism. No focal neurologic signs or meningism appeared. Initially, a presumptive diagnosis of profound depression was considered. Laboratory tests revealed a creatinine level of 1.7 mg/dl and blood urea nitrogen of 71 mg/dl. An EEG showed continuous or almost continuous diffuse spike–wave and sharp-and-slow wave complexes and biphasic moderate high-amplitude sharp waves with frontal emphasis. Intravenous BZDs were not administrated at the EEG laboratory because of severe weight loss and deterioration of his general clinical state. Treatment with PHT (100 mg/8 h) was started, and cefepime therapy was stopped. Twenty-hours later, he experienced a dramatic clinical improvement, and his mental function was considered normal.
We describe the clinical and EEG features observed in three severely ill subjects receiving treatment with cefepime and variable degrees of renal failure in whom a sudden picture of mental state alteration was considered compatible with the diagnosis of generalized NCSE.
Certainly the boundaries between NCSE and encephalopathy may be imprecise. Thus relevant authors have used the term “allied ictal states” for defining a category of nonconvulsive ictal states in which borderline NCSE versus triphasic wave toxic encephalopathies have been similarly categorized (1). It is not surprising that analogous controversy exists in relation to the clinical picture described here. Although most researches agree about the classification of this condition as NCSE (2–6), others view it as an encephalopathy in which spikes and rhythmic bi–triphasic sharp waves reflect cerebral disturbances from severe toxic–metabolic cause (7,8). Our patient 2 is a clear example that confusional episodes associated with the use of cefepime correspond with an acute epileptic disorder compatible with the definition of generalized NCSE. In this case, the administration of intravenous DZP resulted in an immediate clinical and EEG resolution of confusional state. It is obvious that a positive clinical and electrophysiologic result leads to a definite diagnosis of NCSE. Moreover, the clinical recovery observed by others authors despite the persistence of renal dysfunction strongly supports our view (4,6). However, a problem arises when the abolition of epileptiform anomalies after the injection of BZDs is not accompanied by a significant and immediate clinical improvement (as in patient 1). Unfortunately, this ambiguous result is quite frequent. In these cases, the distinction between NCSE and metabolic encephalopathy may be difficult (9). Response to AEDs in situations of genuine NCSE may be delayed. Furthermore, triphasic waves and other EEG alterations of metabolic origin also may be suppressed by intravenous BZDs. A high level of suspicion, broad metabolic screening, and a precise knowledge of all drugs associated with episodes of NCSE can be helpful (10). In cases of possible NCSE, a therapeutic trial of medications can be useful, despite the persistence of the metabolic derangements.
Continuous generalized rhythmic bi–triphasic sensitive BZD sharp waves have been also described in NCSE precipitated by others drugs (11,12). Interestingly, the increased concentration on the central nervous system of all these drugs seems to reduce the seizure threshold by decreasing brain inhibition mediated by γ-aminobutyric acid (11,12). This consideration has important therapeutic implications because it is likely that both the use of AEDs and the interruption of cephalosporin therapy achieve a shorter and faster clinical improvement, reducing complications and possible brain damage. It is also evident that the existence of an impaired renal function plays an important role in the pathophysiology of cephalosporin-induced NCSE and, therefore, the change of the antibiotic regimen also should be accompanied by measures for normalizing renal clearance.
In summary, the diagnosis of NCSE should be considered in all those patients with unexpected changes in mental state and on treatment with intravenous cephalosporins, particularly when these subjects have some degree of renal dysfunction. A high level of clinical suspicion and an emergency EEG are essential to obtain a prompt and accurate diagnosis.