Prolonged Desflurane Administration for Refractory Status Epilepticus

A 71-yr-old man was transferred to our hospital with generalized myoclonic status epilepticus. He was normal neurologically 1 month prior to admission when focal myoclonic seizures occurred in the right upper limb, which proceeded to be generalized. There were no prodromal signs or symptoms of infection. At his local hospital, initial treatment of the seizures with phenytoin, phenobarbital, and midazolam was ineffective. He was intubated, mechanically ventilated, and transported to our institution in status epilepticus. A propofol infusion was initiated upon admission, and an electroencephalogram at that time demonstrated very active bursts of epileptiform activity in the anterior head (fig. 1). Over the next 24 h, the patient continued to have seizures despite propofol, clonazepam, phenytoin, and valproate, and he subsequently required continuous paralysis with cisatracurium to facilitate ventilatory compliance. Desflurane anesthesia was initiated (1% end-tidal), which immediately suppressed epileptiform activity, and the propofol and cisatracurium infusions were discontinued. Adequate burst suppression was subsequently achieved with an end-tidal concentration of 2% (fig. 2). However, the patient exhibited significant central nervous system (CNS) irritability since whenever he was physically stimulated (e.g ., oral suctioning or changing of position), this caused short bursts of myoclonic seizure activity. During the next 24 h, the frequency of seizures increased, which necessitated desflurane to be increased to an end-tidal concentration of 4%. During the next 13 days, adequate burst suppression was achieved with desflurane with end-tidal concentrations ranging between 3 and 4%. Breakthrough seizures were treated with lorazepam or midazolam boluses. For comparison of effectiveness, desflurane was discontinued on day 13, and isoflurane anesthesia was initiated for an additional 13 days. Isoflurane also provided adequate burst suppression with end-tidal concentrations ranging from 0.10% to 0.66%.

During the initial propofol infusion, dopamine was required (up to 10 μg · kg⁻¹· min⁻¹) to maintain a mean blood pressure greater than 60 mmHg. However, during inhalational anesthesia administration, with both desflurane and isoflurane, a phenylephrine infusion (up to 720 μg/min) was also required to maintain adequate blood pressure. The patient was weaned off both medications following discontinuation of inhalational anesthesia.

The patient was earlier diagnosed with adenocarcinoma of the prostate and was treated with radiotherapy and surgery. He also had been taking colloidal silver and was using higher-than-recommended doses for approximately 1 month prior to his onset of seizures.

Laboratory investigations, including international normalized ratio, hepatic enzymes, blood urea nitrogen, and creatinine did not indicate any significant abnormalities throughout the patient's hospital course (fig. 3). Computerized axial tomography and MRI scans of the head demonstrated no structural abnormalities. Several cerebrospinal fluid samples revealed no organisms or an elevated leukocyte count, no neoplastic cells, a normal glucose concentration, and mildly to moderately elevated protein; all subsequent cultures were negative. All other laboratory investigations to exclude CNS infection, Hashimoto thyroiditis, vasculitis, porphyria, or paraneoplastic syndrome were negative. Since all diagnostic tests were negative, CNS silver toxicity was considered due to the patient's recent ingestion of large amounts of colloidal silver. Silver concentrations were found to be elevated in plasma (41.7; normal, 1.0–2.3 nm), erythrocytes (48.2; normal, 4.3–10 nm), cerebrospinal fluid (2.1 nm), and urine (47.28; normal, 0–0.46 nm). Plasmapheresis was therefore initiated; however, despite a significant reduction in serum silver, the patient's neurologic condition did not improve.

In total, during 26 days of inhalational anesthesia, 238 and 39 MAC hours of desflurane and isoflurane, respectively, were administered to this patient, during which time several parenteral anticonvulsive therapies were administered and dosages were adjusted according to serum drug concentrations and control of the patient's seizure activity. On a daily basis, the dose of inhalational anesthetic was reduced to determine whether seizure activity resumed. Figure 4is a qualitative summary of the patient's anticonvulsant therapy during his hospital stay. Eventually, seizures were controlled with clonazepam and valproate, allowing isoflurane to be gradually withdrawn. However, the patient remained in a persistent vegetative state, presumably from a toxic encephalopathy, the etiology of which was not yet determined. A tracheostomy was performed, and the patient was eventually weaned from the ventilator. His discharge electroencephalogram demonstrated α-theta coma patterns but no epileptiform discharges and no definite electrographic reactivity (fig. 5). The patient remained in the intensive care unit for 50 days and was eventually transferred to his home hospital 76 days following his admission. The patient expired on day 153. An autopsy examination of the brain did not reveal any definitive cause of this patient's encephalopathy, and there was no cerebral edema.

Isoflurane anesthesia has become our agent of choice for refractory status epilepticus 6and has also been recommended by several practitioners. 2–5Although halothane was initially recommended for this disorder, the potential organ toxicity associated with its toxic metabolites is concerning, particularly during prolonged administration. 7Isoflurane, on the other hand, undergoes significantly less metabolism, and its potential to produce organ toxicity is substantially reduced. Although significant increases in plasma organic fluoride have been demonstrated in patients receiving isoflurane for prolonged periods, there was no renal toxicity associated with this observation. 8As well, the published experience with isoflurane anesthesia for refractory status epilepticus and our own experience in seven patients over the past 5 yr have indicated no organ toxicity associated with prolonged exposure to isoflurane. 2–6 

Desflurane is a third-generation inhalational anesthetic agent with physiochemical properties comparable to isoflurane. Desflurane undergoes biotransformation by cytochrome P-450, similar to isoflurane; however, the formation of its toxic metabolites was shown to be even less than isoflurane. 9This apparent resistance to biotransformation may be an advantage of desflurane over isoflurane, particularly during prolonged administration. As well, this patient also received a course of phenobarbital, which is a known inducer of cytochrome P-450 activity. However, despite receiving a combined total dose of 277 MAC hours of desflurane and isoflurane, there was no indication of renal, hepatic, or other organ toxicity (fig. 5).

The effects of desflurane on depression of electroencephalographic activity, including burst suppression, is dose related and comparable to the effects seen with equipotent doses of isoflurane. The effects of desflurane and isoflurane on cerebral blood flow, intracranial pressure, and cerebral metabolic rate are also comparable. 10Although the possibility exists that the toxic metabolites of these agents contributed to the underlying encephalopathy, there is no published evidence of CNS toxicity with either desflurane or isoflurane. 7 

During isoflurane administration, there is a dose-dependent reduction in systemic vascular resistance due to peripheral vasodilation. Indeed, all of our seven patients with refractory status epilepticus treated with isoflurane or desflurane required an inotrope and/or a vasopressor during its administration, despite adequate fluid resuscitation. 6Desflurane also has similar effects to isoflurane on systemic coronary, renal, hepatic, and cerebral hemodynamics. However, its ability to cause systemic vasodilation is less than isoflurane and therefore should theoretically be less prone to cause hypotension negating the need for vasopressor therapy. 10Despite this, dopamine and phenylephrine were required during both desflurane and isoflurane administration.

The blood–gas partition coefficient of desflurane is lower than isoflurane and implies a more rapid onset of action and elimination once discontinued. 10However, this theoretical advantage will likely not be appreciated in the treatment of refractory status epilepticus since experience has demonstrated administration of isoflurane results in a prompt burst-suppression response on the electroencephalogram. Therefore, a quicker response to desflurane would likely not be appreciated clinically. In terms of recovery, however, the elimination of isoflurane can be rather prolonged, particularly when it has been administered for many days in the intensive care unit. Whether long-term administration of desflurane would translate into a more rapid recovery and whether this would result in any clinical benefit (e.g ., an earlier extubation time) are unknown.

In summary, desflurane is a newer inhalational agent with pharmacokinetic and pharmacodynamic properties comparable to isoflurane. As demonstrated in this case report, it was as effective as isoflurane in producing burst suppression and was easily titratable over a prolonged period of time in response to the electroencephalogram. Although the theoretical advantages of desflurane over isoflurane include a more rapid onset of action and elimination, as well as perhaps fewer requirements for inotropic and/or vasopressor therapy for hypotension and reduced potential for organ toxicity due to its relative resistance to biotransformation, these potential advantages need to be evaluated in future studies. Inhalational anesthesia with either isoflurane or desflurane is a suitable therapy and is recommended for patients with refractory status epilepticus.