DISCUSSION

Since neuroleptic malignant syndrome was proposed as a clinical entity in the 1960s, various case descriptions and clinical studies have been reported from psychiatric and neurologic units. The pathophysiologic mechanisms of NMS still are unclear, but the dopaminergic system is thought to be pivotal in the etiology of NMS because all neuroleptics that can cause NMS act as antagonists at dopamine receptors in the central nervous system. Certain predisposing conditions such as dehydration, malnutrition, and exhaustion are risk factors for development of NMS. Psychomotor agitation, high or rapidly increasing neuroleptic doses, and frequent use of neuroleptic injections are other risk factors that tend to be interrelated (Kawanishi, 2003). Although minimizing risk factors could decrease NMS incidence to some extent, prediction and prevention of NMS still are difficult. On the other hand, a patient who has experienced NMS is at high risk for recurrence if neuroleptics are reintroduced. Thus, constitutional factors of NMS have been a focus of attention, and pharmacogenetic studies have suggested a genetic background for NMS occurrence (Kawanishi, 2003; Kishida et al., 2004).

Recent NMS cases reported from clinical settings apart from psychiatric and neurologic units underscore that NMS is possible wherever any neuroleptic drug is administered (Tenenbein, 1985; Zohar et al., 1992; Bakri et al., 1992; Tanaka et al., 1998; Russell et al., 2001; Onose et al., 2002). As diseases other than neuropsychiatric disorders could act as triggers if they significantly worsened the physical condition of the patient, individuals with advanced cancer could be considered at high risk. We recently reported development of NMS following neuroleptic treatment after bone marrow transplantation for acute myeloid leukemia; pretransplantation irradiation, chemotherapy, and graft-versus-host disease (GVHD) had resulted in both physical and mental stress (Onose et al., 2002). NMS in cancer patients has been reported only rarely despite frequent complications of cancer and its treatment by mental disorders managed with neuroleptic drugs. Tanaka et al. (1998) reported NMS occurrence complicating treatment of delirium that followed surgery for soft palate cancer, noting diagnostic difficulties arising from clinical oncologists' unfamiliarity with NMS and sometimes from the resemblance of NMS symptoms to those related to cancer itself, as we experienced in patients who underwent bone marrow transplantation (Onose et al., 2002).

We know of no previous reports of NMS occurrence in the intensive care unit on the day of surgery. In our patient, cerebral dysfunction might have triggered NMS after repeated haloperidol injections. Patients with organic brain diseases typically are naive to neuroleptics. Although organic brain diseases themselves have been suggested as risk factors of NMS, this issue has not been fully assessed by a large-scale study (Shalev & Munitz, 1986). NMS and its prodromal symptoms also become more difficult to diagnose when neurologic symptoms from organic brain disease could mask NMS symptoms. Prodromal symptoms in our case were tachycardia and diaphoresis observed beginning on day 7.

NMS probably is less rare overall than initially thought, with incomplete forms being particularly likely to be overlooked in nonpsychiatric settings. Although NMS is potentially lethal, recognition of prodromal NMS symptoms can permit action to decrease morbidity and mortality. In clinical oncologic practice, patients undergoing psychopharmacologic treatment require vigilance against NMS.