Dear Editor,
Cefepime is a fourth-generation cephalosporin that is widely used in hospital settings.1 Since its approval, isolated cases of encephalopathy have been reported in patients with both normal2 and impaired kidney function.3,4 Nonetheless, the information about the clinical manifestations and the prognosis of this adverse reaction is scarce. Therefore, we believe it is important to report seven cases of cefepime-induced encephalopathy in patients with kidney failure. These cases corresponded to 4 males and 3 females with an average age of 63 years. All of the patients had acute or chronic renal failure when cefepime was prescribed. The average value of creatinine at the beginning of treatment was 3.6mg/dl and the initial dose of cefepime was 2.75g/day; in five patients the dose was adjusted for the degree of kidney function. The average time period between beginning of treatment and symptoms was 5.4 days. The most common clinical manifestations were a decreased level of consciousness (71.4%) and myoclonus (71.4%). The EEG was pathological in the six cases where it was carried out, demonstrating a nonconvulsive epileptic status in three, slowed global activity with repetitive paroxysm in two, and diffuse affectation with a predominance of triphasic waves in one. The CT scan and the spinal tap were normal in all cases. After diagnosing the encephalopathy, treatment with cefepime was discontinued. Three of the patients received dialysis. Three patients improved (42.9%), one of whom required haemodialysis. The 4 remaining patients (57.1%) died from the encephalopathy.
The use of cefepime in patients with kidney failure, even at adjusted doses, may cause serious encephalopathy, and thus its administration should be avoided or used with close monitoring. The appearance of alterations in the level of consciousness and the myoclonus should alert us to the appearance of a nonconvulsive status that requires an EEG as it is the most useful diagnostic test. Haemodialysis does not seem to modify the clinical outcome.
