General anesthesia with propofol for ovarian teratoma excision associated with anti-N-methyl-D-aspartate receptor encephalitis

Anti-N-methyl-D-aspartate receptor (NMDAR) encephalitis is an autoimmune neurological disorder caused by production of antibodies to NMDAR and has become the most common autoimmune encephalitis described, since its first description in 2007 [1]. Tumors occur in approximately 38–58% of all cases and 94% of the tumor is ovarian teratoma [2]. Both autoimmune responses against neural antigens expressed ectopically in the underlying tumor and cross-reaction with an unknown infectious agent are thought to be responsible for the pathogenic mechanism of anti-NMDAR encephalitis [1, 2]. The treatment bundle, therefore, includes immunotherapy and surgical tumor resection, if applicable. Patients with early tumor resection along with first-line immunotherapy (corticosteroids, gamma globulin, or plasma exchange) recover earlier, less often require second-line immunotherapy (cyclophosphamide or rituximab, or both) and experience fewer relapses, than those with a tumor that was treated late or without a tumor [3, 4]. To establish an anesthetic plan for patients with anti-NMDAR encephalitis, it is important to understand the pharmacological interaction between the anesthetic agents and the disabled NMDAR, because NMDAR is a major site of action for commonly-used anesthetic agents. Although anti-NMDAR encephalitis has gradually attracted the attention of anesthesiologists, there have been only a few discussions about propofol that have some pharmacological effects on NMDAR.

We describe two cases of young female patients with anti-NMDAR encephalitis who underwent resection of ovarian teratoma under general anesthesia using propofol and discuss the validity of propofol in these patients.

The NMDAR is one of two major receptors associated with the effect site of anesthetic agents, along with the gamma-aminobutyric acid type A (GABA_A) receptor. Ketamine and nitrous oxide minimally affect GABA_A receptor and antagonize the NMDAR by direct action [5, 6]. While there have been several reports indicating that propofol also inhibits NMDAR at clinically relevant concentrations, propofol is considered to elicit its anesthetic effects through enhancing a GABAergic mechanism, not via a NMDAR-dependent mechanism [7, 8]. Nakao et al. reported that propofol inhibits ketamine-induced psychotomimetic activities via the activation of GABA_A receptor activation in the rat posterior cingulate and retrosplenial cortices, which are suggested to be the regions of brain concerned with psychotomimetic activity and neural damage of NMDAR antagonists [9]. These findings suggest that propofol, a GABAergic agent, is not only effective in suppressing psychotomimetic complications induced by ketamine, but is also acceptable for the suppressing NMDAR-related pathological mechanisms of anti-NMDAR encephalitis.

Several case reports have included the anesthetic procedures using propofol in patients with anti-NMDAR encephalitis. Collectively, Liu et al. and Pascual-Ramírez et al. described three patients who presented for resection of ovarian teratoma under total intravenous anesthesia using propofol. No complications or unexpected events occurred perioperatively in these patients and they recovered smoothly [10, 11]. Pryzbylkowski et al. reported the anesthetic management of two patients and demonstrated that propofol, isoflurane, desflurane, hydromorphone, and fentanyl were well tolerated, and no paroxysmal sympathetic hyperactivity was seen intraoperatively in these cases [12]. On the other hand, Lapebie et al. reported a case of a patient whose symptoms deteriorated with increasing dyskinesia and generalized seizure under propofol sedation after general anesthesia with propofol and sevoflurane for ovarian teratoma resection [13]. There is, however, a limitation to distinguishing complications induced by anesthetic agents from deteriorations of this encephalitis.

In our report, total intravenous anesthesia, including propofol, remifentanil, and fentanyl, was utilized to induce and maintain anesthesia for two patients, and ketamine and nitrous oxide were avoided. In these anesthetic courses, as above, neither psychoneuronal modifications nor autonomic instabilities due to propofol were evident. The first case suffered from prolonged catatonic coma, convulsion, and autonomic instabilities that we treated with using continuous midazolam infusion. The cause of postoperative progress differences in our two cases is suggested that the first case had already more serious symptoms with autonomic nerve instability and central hypoventilation before surgery than the second one.

Although the anesthetic implications of caring for patients with anti-NMDAR encephalitis have not yet been defined, ketamine and nitrous oxide should be avoided because of the uncertainty about the pharmacodynamic response of a disabled NMDAR and benzodiazepines and opiates that do not interfere with the NMDA pathway are preferred in these patients. Propofol acting indirectly at the NMDAR is suggested to be safely used under prudent monitoring and the close titration of dosages judging from NMDAR-related pharmacological mechanism and the past case reports including ours.

We described two female patients with anti-NMDAR encephalitis who required surgical resection of ovarian teratoma under general anesthesia using propofol. In both of these anesthetic courses, neither psychoneuronal modification nor autonomic instability by propofol was evident. We imply that propofol is safely used in patients with anti-NMDAR encephalitis, although it has some pharmacological effects on NMDAR.