A case of modified electroconvulsive therapy for a patient with dementia with Lewy bodies and neuroleptic malignant syndrome

Dementia with Lewy bodies (DLB) is characterised by severe hypersensitivity to antipsychotic drugs, which sometimes causes neuroleptic malignant syndrome (NMS). NMS produces high fever, muscle rigidity, autonomic symptoms, and impaired consciousness and can be fatal if not appropriately treated. Dantrolene is often used to treat NMS, although modified electroconvulsive therapy (mECT) has been effective.1 mECT can be used to treat DLB symptoms, including psychiatric manifestations, parkinsonism, and dementia severity.2 However, there are no reports on mECT for DLB with NMS. In this report, we describe a patient with DLB who developed NMS and underwent mECT. Written consent was obtained from the patient's family members and sufficient care was taken to ensure patient and family anonymity. The patient was a retired 79-year-old man who lived with his wife and son. In 2023, he began experiencing visual hallucinations and irritability and visited his previous doctor on 22 August 2023. His medical history included diabetes mellitus and hypertension, and his family history was not available. Parkinsonism and rapid eye movement sleep behaviour disorder were not observed, although brain magnetic resonance imaging showed mild brain atrophy. His score for the revised Hasegawa Dementia Scale was 14 points, indicating cognitive impairment (30-point scale; the lower the score, the more severe the cognitive impairment), and DLB was suspected. After risperidone (1 mg/day) was prescribed for his psychiatric symptoms on 29 August 2023, the patient gradually became unable to move and was brought to his previous doctor on 25 September 2023. A fever in the 38°C range, bilateral muscle rigidity in the upper and lower limbs, generalised sweating, and impaired consciousness were observed, and blood tests showed an elevated creatinine kinase (CK) level of 17 648 U/L. When risperidone was discontinued and fluids and dantrolene 75 mg/day were administered because of suspected NMS, the CK level and fever decreased; however, his muscle rigidity and impaired consciousness did not improve. Lorazepam 10.5 mg/day was also administered and resulted in only mild muscle rigidity improvement. During hospitalisation, the patient also had decreased uptake on dopamine transporter single-photon emission computed tomography and metaiodobenzylguanidine myocardial scintigraphy; thus, he was diagnosed with DLB. The patient also developed pseudomembranous enterocolitis and cellulitis of the left upper limb, and his general condition deteriorated, including oedema of the extremities from malnutrition. Due to difficulty with food intake, a central venous catheter and gastric tube were placed. He was in a poor general condition and was transferred to our department on 13 November 2023, with a request to undergo mECT for NMS. At the time of transfer, the CK level was 27 U/L. However, muscle rigidity, a fever in the 38°C range, impaired consciousness, sweating, tachycardia, and hypertension were observed, and NMS was considered persistent. Since impaired consciousness was prominent, the psychiatric symptoms were not apparent and cognitive function was not assessed. Blood tests showed hypoalbuminaemia, and chest computed tomography revealed bilateral pleural effusions. A dermatologist was consulted regarding a counterpart for cellulitis, and because the patient's condition was improving, cefazolin was discontinued on the second day after admission. Vancomycin was discontinued on the sixth day due to pseudomembranous enterocolitis because the stool examination showed no signs of infection. Considering the risk of aspiration, central venous nutrition was continued using a central venous catheter for malnutrition. Although his general condition was poor and there were concerns about the risks associated with general anaesthesia, mECT was initiated by the anaesthesiologist on the fourth day after admission. mECT was started using a Thymatron System IV (Somatics, LLC, Lake Bluff, IL, USA) with right unilateral stimulation at a dose of 5% and a 0.25 ms pulse width. The first mECT produced an effective seizure, and there was no evidence of poor oxygenation or worsening of haemodynamics from the general anaesthesia. Therefore, mECT was continued, and the stimulation dose was increased to 30% for session 2. Flumazenil 0.8 mg was administered until session 15 because the patient was taking lorazepam. Anaesthetics were started with 100 mg of thiopental and adjusted according to the anaesthetic situation: 87.5, 100, and 125 mg in sessions 3, 7, and 9, respectively. Rocuronium 40 mg was administered as a muscle relaxant. The Richmond Agitation-Sedation Scale scores were −5 just before session 1, −3 just before session 2, and 0 after session 3 (a negative score indicates more impaired consciousness, and a positive score indicates more agitation). Therefore, the patient's impaired consciousness improved after session 3. Dantrolene was discontinued on the ninth day after admission, and lorazepam was tapered off on the 10th day after admission. Gradual improvement in the oedema was observed, and rehabilitation was initiated on the 15th day after admission. In session 8, the stimulation dose was titrated to 70%, and some improvement in muscle rigidity was observed, including automatic shoulder joint and neck motion. Since the patient had persistent fever in the 37–38°C range, the stimulation dose was increased to 100%, although there was little improvement. Therefore, the mECT parameters were changed in session 12 to bilateral stimulation, a 40% dose, and 0.5 ms pulse width. The fever improved to approximately 37°C; however, little improvement was observed in muscle rigidity of the extremities. The stimulation dose was titrated to 100%, although muscle rigidity did not change. Pleural effusion was almost completely absent, the patient's general condition improved, and no psychiatric symptoms such as visual hallucinations and irritability were observed. However, the administration of cognitive function tests was not feasible; the patient was unable to engage in conversations, suggesting the presence of cognitive dysfunction. Since no further improvement was expected, mECT was terminated in session 18. The overall evaluation was not markedly different from that before the initiation of mECT (Fig. 1 and Table 1). The previous physician observed fever, muscle rigidity, impaired consciousness, and elevated CK levels after administering 1 mg/day of risperidone, and the patient met Levenson's diagnostic criteria for NMS. After transfer to our hospital, although there was no obvious elevation in the CK level, we considered NMS was still present because of the muscle rigidity, fever exceeding 38°C, impaired consciousness, and sweating. Antipsychotic drugs, including risperidone,3 have been reported to cause NMS in patients with DLB. In previous case reports, treatment included discontinuation of the causative agent, infusion, and administration of dantrolene, bromocriptine, and lorazepam for improved NMS within 2 weeks. Compared to previous case reports, the patient in this report was older, had a higher CK level, and did not improve with discontinuation of the causative agent, infusion, or dantrolene administration. Most NMS cases improve in 2–14 days if detected and treated early; however, if diagnosis and treatment are delayed, NMS can be prolonged, and stupor and parkinsonism may remain.4 Although the timing of NMS onset was unknown in this case, the patient may have been brought to the emergency room approximately 3 weeks after NMS onset, and the delayed response may have resulted in concurrent physical illness and lack of improvement in muscle rigidity. A systematic review of mECT for behavioural and psychological symptoms of dementia suggests mECT efficacy for depression, mania, screaming, and agitation.5 For DLB, mECT has been suggested to improve psychiatric symptoms, parkinsonism, and dementia severity.2 In a case report of mECT for Parkinson's disease with NMS and depression, improvement of NMS, depressive symptoms, and parkinsonism was observed.1 However, there are no reports on mECT for DLB with NMS. In this case, the patient underwent mECT, which improved impaired consciousness, fever, and sweating, although it did not improve muscle rigidity and cognitive dysfunction. At the time of transfer to our hospital, pseudomembranous enteritis and cellulitis improved; however, fever, impaired consciousness, and sweating were observed. These symptoms improved after mECT. Consciousness impairment could have been due to lorazepam oversedation but was improved by mECT before the dose reduction. Therefore, mECT was effective for treating some NMS symptoms. However, it cannot be ruled out that these symptoms may have improved via the natural process. The lack of improvement in muscle rigidity of the extremities and cognitive dysfunction may have been due to the late start of mECT because it was initiated 2 months after NMS onset. However, possible causes of cognitive dysfunction include adverse events of mECT, prolonged NMS, and progression of DLB. This case report has various limitations, including insufficient patient information and evaluations. The causal relationship between mECT and improvements in some NMS symptoms may be only speculative. In this case, DLB was complicated by NMS and treated with infusion and dantrolene; however, the effect was insufficient. mECT was performed approximately 2 months after NMS onset. mECT was effective for impaired consciousness, fever, and sweating, and the patient's general condition improved. However, muscle rigidity and cognitive impairment persisted. Therefore, the overall evaluation was unchanged from that before mECT initiation, although it appeared effective in treating some NMS symptoms. The efficacy of mECT for patients with DLB with NMS warrants further investigation. The authors declare no conflicts of interest for this article. Sufficient care was taken to ensure anonymity. Written consent was obtained from the family instead of ethics committee approval because this is a case report. Written consent was obtained from the patient's family members because of the patient's cognitive impairment. Research data are not shared.