Article Text

Download PDFPDF

Neuroleptic malignant syndrome in patients with cancer: a systematic review
  1. Izumi Sato1,2,
  2. Hideki Onishi3,
  3. Chiaki Kawanishi4,
  4. Shuhei Yamada5,
  5. Mayumi Ishida3 and
  6. Koji Kawakami1
  1. 1 Department of Pharmacoepidemiology, Graduate School of Medicine and Public Health, Kyoto University, Kyoto, Japan
  2. 2 The Keihanshin Consortium for Fostering the Next Generation of Global Leaders in Research (K-CONNEX), Kyoto, Japan
  3. 3 Department of Psycho-Oncology, Saitama Medical University International Medical Center, Hidaka, Saitama, Japan
  4. 4 Department of Neuropsychiatry, Sapporo Medical University Graduate School of Medicine, Sapporo, Hokkaido, Japan
  5. 5 Department of Quality and Patient Safety Management, Chiba Cancer Center, Chiba, Japan
  1. Correspondence to Dr Koji Kawakami, Department of Pharmacoepidemiology, Graduate School of Medicine and Public Health, Kyoto University Yoshida Konoecho, Kyoto 606-8501, Japan; kawakami.koji.4e{at}kyoto-u.ac.jp

Abstract

Background Antipsychotics potentially cause a low incidence of the side effect called neuroleptic malignant syndrome (NMS), which has a high mortality rate. However, few studies on NMS among patients with cancer exist.

Aims We aimed to examine the characteristics of antipsychotic-induced NMS among patients with cancer.

Methods We conducted a systematic review of published reports on NMS described during the treatment of patients with any type of cancer. Articles were identified by a comprehensive search of PubMed, Web of Science, the Cochrane Library and reference lists from relevant articles published until 25 December 2019. Original articles or case reports on humans published in English were included. This review summarises the symptoms, characteristics, treatment course and prognosis of patients with cancer with NMS.

Results Eleven patients with various cancer types from ten case reports published from 1988 to 2013 met the eligibility criteria. Mean age of the 11 patients was 52.5 (range, 32–83) years. NMS developed mostly during the postoperative period, and haloperidol and D2 receptor antagonists were determined as the common causative drugs. Ten patients survived following treatment that mostly involved discontinuing the causative drugs and administering dantrolene, if necessary.

Conclusion Although NMS intrinsically has a low incidence and high mortality, only few reports were available, with most patients surviving after early detection and appropriate treatment. Healthcare providers should consider NMS development while prescribing antipsychotics to ensure prompt recognition of the condition and rapid treatment for preventing unnecessary deaths.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Antipsychotics potentially cause neuroleptic malignant syndrome (NMS), well known to psychiatrists as a lethal adverse reaction requiring rapid treatment to save a life. The common symptoms are hyperthermia, muscle rigidity, altered consciousness, diaphoresis and tachycardia.1 The incidence of NMS is low, at 0.04%–0.2%,2 3 whereas its mortality has remained high at around 20% even several decades after NMS was discovered in the 1950s; it ranged from 5% to 9% in the 2000s.4–6 Thus, for management of NMS, a safety measure such as the ‘manual for handling disorders due to adverse drug reactions’ by the Ministry of Health, Labour and Welfare in Japan, was developed.7 Due to the difficulty in distinguishing NMS from other diseases with similar symptoms, NMS is usually detected by differential diagnosis.8 9

Antipsychotics are commonly prescribed for psychiatric symptoms, such as delirium and all-cause nausea and vomiting, among patients with cancer.10–15 D2 receptor antagonists (eg, prochlorperazine and metoclopramide) are commonly prescribed for all-cause nausea and vomiting. Additionally, antipsychotics, especially olanzapine, are sometimes recommended concomitantly with 5-hydroxytryptamine (5-HT3) receptor antagonists, one neurokinin-1 receptor (NK1R) antagonists and glucocorticoids for chemotherapy-induced nausea and vomiting.12–15 The prevalence rates of antipsychotic use were approximately 4% and 8% among patients with cancer in Japan and Denmark, respectively.16 17 The prevalence of delirium in patients with cancer ranged from 10% during the hospitalisation to 85% during the terminal phase.10 Risk factors of delirium include old age, cognitive impairment and hospitalisation.18

First-generation antipsychotics (FGAs) are the most common causative drugs of NMS19 20 among the other causative medications, such as second-generation antipsychotics (SGAs), D2 receptor antagonists, benzodiazepine, serotonin-selective receptor inhibitors, lithium and other dopamine antagonists).20–22 In addition, higher or cumulative doses of antipsychotics, rapid dose increases of antipsychotics, multiple antipsychotics and discontinuation of antiparkinson or antipsychotic drugs are assumed to be risk factors of NMS.16 23 24

The clinical signs of NMS developed in most patients include hyperthermia, changes in the mental status, extrapyramidal symptoms and autonomic dysfunction. Elevation of creatine kinase and white cell count and myoglobinuria are often observed as laboratory findings.9 The diagnosis of NMS is made based on clinical and laboratory findings.2 19 25 26 NMS should be managed based on a hierarchy of symptom severities.9 The essential treatment consists of (1) discontinuation of the causative drug, (2) whole-body control (eg, monitoring of renal function, consciousness level and cardiorespiratory function; body cooling and management of the electrolyte balance) and (3) drug treatment. Commonly, dantrolene is useful for moderate rigidity. In some cases, dopaminergic agents (eg, bromocriptine) and benzodiazepines (eg, lorazepam) are useful for reverse Parkinsonism and for controlling agitation, respectively.

While many studies on NMS have been reported, there have been few studies on NMS in cancer patients to the best of our knowledge. NMS is not well known to healthcare providers, although they commonly prescribe antipsychotics. It is assumed that many patients with cancer are at risk of NMS because they are treated with antipsychotics, which are necessary for delirium, all-cause nausea and vomiting, and any other reason.16 27 Thus, NMS may be overlooked and/or insufficient provision of appropriate treatment may be made for NMS.

We planned to examine the characteristics of NMS caused by FGAs and SGAs among patients with cancer to better understand the condition and to provide appropriate treatment for NMS in a real-world setting using medical databases. We conducted a systematic review of NMS cases among patients with cancer to understand NMS characteristics within this population. We summarised the symptoms, characteristics, course of treatment and prognosis for NMS in patients with cancer.

Materials and methods

Eligibility criteria

We considered an article to be eligible for this review if it described NMS among patients with any cancer during cancer treatment. Original articles or case reports regarding humans including qualitative and quantitative studies published in English were included. There might be reporting bias regarding the case reports owing to high mortality in NMS. Due to this, only cases with a good prognosis might have been reported. However, we aimed to identify previous studies on NMS, including the conditions in which they were published; therefore, we included case reports. We excluded reviews of articles, articles published only as abstracts, and articles in which the eligible patients were under 18 years of age.

Search strategy

Articles were identified by a comprehensive search of PubMed, Web of Science, the Cochrane Library and reference lists from relevant articles from database inception to 25 December 2019. In the search strategy, we used two kinds of terms with or without Medical Subject Heading (MeSH) and linked the two terms using the “AND” operator: (1) ‘neoplasms,’ ‘cancer,’ (2) ‘neuroleptic malignant syndrome’.

Process

Titles and abstracts were independently screened by two researchers (IS and HO). If an article could not be judged for eligibility based on the title and abstract, the full text was screened. We searched relevant original articles and case reports from the reference lists of articles identified through a database search. We judged the eligibility for inclusion of articles based on their titles, and only if such an assessment was not possible were the abstract and full-text screened. Articles were selected by IS and HO. First, IS identified the records through a database search and did the screening. Thereafter, IS and HO separately searched relevant articles for screening. Additionally, IS and HO separately screened the articles and compared each result. In case of discrepancies, the eligibility of articles was decided after a discussion between IS and HO.

Data extraction

IS extracted available data from each of the identified articles. The following data were included or extracted: (1) study design, (2) country, (3) year of NMS onset, (4) patients’ characteristics (age, sex, cancer type, comorbidity and pharmacotherapy for comorbidity), (5) symptom being treated, 6) pharmacotherapy (present/absent), (7) referral to a psychiatric professional, (8) definite diagnosis of NMS, (9) diagnostic criteria for NMS, (10) time to onset of NMS from initiation of pharmacotherapy, (11) treatment course of NMS (pharmacotherapy (present/absent)), (12) prognosis, (13) time to recovery or death, (14) suspected symptom of NMS, (15) vital signs and (16) laboratory test data. To avoid misclassification bias, our other researchers (HO, CK, SY and MI) checked the data and added variables on clinical aspects, as necessary, after selection of articles by IS and HO. All data presented were taken from the selected articles.

Only descriptive statistics were used to calculate the mean and median ages of the identified patients. This study was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.28 The flow diagram of patient selection was prepared based on the PRISMA 2009 flow diagram.

Results

We included 11 patients from nine case reports published from 1988 to 2013 (figure 1). We could not find any original articles. Three of the nine studies were from the UK, USA and Saudi Arabia and were reported from 1988 to 1998; the other cases had been reported in Japan from 1997 to 2013 (table 1).29–38 Mean age of the 11 patients was 52.5 years; their ages ranged from 32 to 83 years (median: 50 years); five patients were female, six male. Only one patient had mental disease (adjustment disorder) as a comorbidity. Comorbidities were not mentioned for most patients. Three cases had been referred to a psychiatric professional; such referral-related information was not available in the other case reports.

Table 1

Published cases of NMS and patients’ characteristics

Figure 1

Flow diagram for patient selection. NMS, neuroleptic malignant syndrome.

Six patients developed NMS during the course of postoperative treatment following the prescription of psychotropic medication for complications such as agitation, delirium, nausea and vomiting (online supplementary table 1). Three patients developed NMS during palliative care, when treatment was prescribed for delirium, pain, nausea, vomiting, anxiety, irritation and insomnia. The other two patients developed NMS after receiving high-dose chemotherapy and undergoing autologous stem cell transplantation and treatment was prescribed for anxiety, agitation, insomnia, nausea and vomiting. The last patient developed NMS when haloperidol was prescribed for symptoms of delirium after bone marrow transplant (BMT). Of the 11 patients, nine received FGAs (haloperidol, eight patients; droperidol, one patient; chlorpromazine, one patient) and four received D2 receptor antagonists (prochlorperazine and metoclopramide).

Supplemental material

After commencing treatment with a drug suspected of causing NMS, the patients’ vital signs were commonly found to show hyperthermia, tachycardia, tachypnea and change in blood pressure (online supplementary table 2). Additionally, patients had symptoms such as muscle rigidity, diaphoresis, tremor, disorientation, agitation, incontinence, consciousness disturbance, dysarthria and hallucinations. Definite diagnosis of NMS was made in six patients according to the following previously proposed criteria: Levenson’s criteria19; Diagnostic and Statistical Manual of Mental Disorders-IV (DSM-IV) criteria37; Caroff’s criteria2 and Pope’s criteria.38 The mode of the number of days required for NMS occurrence was 2 days. The longest time was 14 days among patients following postoperative treatment. In two cases, there was no information concerning the time of NMS onset.

Supplemental material

NMS was mostly addressed by discontinuing the medication suspected to cause NMS (table 2). Half of the patients also received dantrolene or diazepam. Two patients were started on supportive care. Ten patients recovered, and one died. The time to recovery or death and definition of recovery varied in each case. The time ranged from the next day to the 14th day after starting NMS treatment.

Table 2

NMS: treatment, prognosis and time to recovery or death

Discussion

This was a systematic review of NMS in cancer patients. Only 11 patients met the eligibility criteria for reported cases from 1988 to 2013. Age and types of cancer varied, and no gender difference was observed among the patients. NMS mostly developed during the postoperative period and was found in patients receiving palliative care, high-dose chemotherapy or undergoing autologous stem cell transplantation or BMT. Typical NMS symptoms were observed among the patients; diagnoses using any of NMS criteria were made only for six patients. The causative drugs were FGAs, especially haloperidol and D2 receptor antagonists, and they were mostly discontinued when NMS treatment was started. Dantrolene was also administered to five patients. Ten patients recovered within 2–14 days.

Although the incidence of NMS is low intrinsically, there might be under-reporting for NMS. It was estimated that there might be at least a few hundred cases in patients with cancer treated with FGA or SGA during the last several decades according to the annual incidence of cancer (approximately 370 000 patients in 1988–860 000 patients in 2013) in Japan,39 4%–8% of antipsychotics use among patients with cancer,16 17 and 0.04%–0.2% of estimated incidence of NMS.2 3

Some possible reasons for under-reporting are (1) limited knowledge about NMS among healthcare providers, (2) the decreasing incidence of NMS and (3) difficulty in making a diagnosis of NMS. First, healthcare providers should be careful while prescribing haloperidol for delirium. The incidence of delirium among patients with cancer ranges from 10% to 85%.10 11 40 Haloperidol was the causative drug in most NMS cases in our review. High incidence of NMS caused by FGAs.11 Moreover, healthcare providers should know that D2 receptor antagonists, commonly used for all-cause nausea and vomiting, carry a risk of inducing NMS. The abrupt reduction of dopamine activity in the central nerve system may cause NMS.41 In our review, D2 receptor antagonists were prescribed to four patients for nausea and vomiting. Furthermore, the reduction may be enhanced by the concomitant use of D2 receptor antagonist and FGA/SGA, which have to block the effect of D2 dopamine receptors in the brain.42 Although we found only one case of simultaneous use of both medications, such use would be common in the real-world setting. Additionally, olanzapine is commonly used with 5-HT3 receptor antagonists, NK1R antagonists and/or glucocorticoids for chemotherapy-induced nausea and vomiting.12 13 15 43

Currently, SGA use is increasing and that might have reduced NMS incidence due to its potential safety profile.44–46 Additionally, the risk of incidence of NMS among patients taking SGA is lower than that among patients treated with FGA.47 The decreasing the incidence of NMS and the limited number of reports on NMS studies has resulted in healthcare providers being less careful in noticing NMS. Prescription of SGAs to patients with cancer should also be monitored.

NMS is not easily detected, although various diagnostic criteria exist because other diseases have similar symptoms. In our review, the NMS criteria used were only reported for six patients; use of multiple NMS criteria were reported for some patients. They probably prioritised rapid treatment to save lives. Because of the prompt action that is often required for this fatal disease, each criterion is often used for patients irrespective of whether the patient meets the criteria in a clinical setting or not. There are various diagnostic criteria: ‘hyperthermia’, ‘rigidity’, ‘autonomic imbalance’ and ‘abnormal laboratory finding’.8 9 47 According to these criteria, it seems that most patients included in our review received the standard treatment for NMS. However, it is not clear whether unreported patients or patients who were supposed to have been diagnosed with NMS received appropriate treatment or not. Dissemination of an appropriate NMS management approach is necessary for practitioners in the field of oncology.

Our review had several limitations. The most important limitation is the lack of studies to establish valid evidence. As mentioned above, there is a reporting bias as we could only select case reports and thus, the information only included subjective data. Only patients with good prognoses might have been selected for these case reports. In this situation, it was impossible to evaluate a patient as having poor prognosis. Additionally, we might have missed some studies or cases because of the following reasons. First, we only included papers written in English. Second, we did not consult any information specialist/subject librarian when we made the search strategy. Third, we only searched three databases. Lastly, we did not include the exact words without MeSH, such as “malignan*” or “tumor/”tumour.” Although these words are linked with the MeSH of “Neoplasms,” which is the same as “cancer” and “neoplasms,” which we used as search terms. The other limitation was missing data. We could not judge whether the data was really missing, or non-existent. For example, comorbidities were not mentioned for eight patients. It was not clear whether the patients did not have any comorbidities or whether the patients’ comorbidities were not reported. However, due to the small number of case reports, we could not evaluate the epidemiology as well as risk factors of NMS among patients with cancer. Further reports and studies are required to evaluate these factors.

Conclusion

No clinical studies and only a few case reports were included in the systematic review of NMS in patients with cancer. Although age and cancer type varied among cases, NMS mostly developed during the postoperative period; the most common causative drug was haloperidol. Furthermore, most patients survived after early detection and appropriate treatment. Healthcare providers should consider NMS development while prescribing antipsychotics to ensure prompt recognition of the condition and rapid treatment for preventing unnecessary deaths.

References

Footnotes

  • Contributors Titles and abstracts were independently screened by two researchers (IS and HO). If an article could not be judged for eligibility based on the title and abstract, the full-text was screened. Similarly, reference lists from relevant original articles and case reports were screened. Articles were selected by IS and HO. To avoid misclassification bias, our other researchers (HO, CK, SY and MI) checked the data and added variables on clinical aspects, as necessary, after selection of articles by IS and HO. KK supervised all processes of this study.

  • Funding This work was supported by both the Keihanshin Consortium for Fostering the Next Generation of Global Leaders in Research (K-CONNEX), established by Human Resource Development Program for Science and Technology, MEXT, and Grants-in-Aid for Young Scientists (KAKENHI Grant Number 18K15416) from the Japan Society for the Promotion of Science.

  • Competing interests CK reports other from Sumitomo Dainippon Pharma, other from Stella Pharma Corporation, other from CMIC, other from Suntory Beverage & Food, other from Kaken Pharmaceutical, other from Astellas Pharma, other from Mitsubishi Tanabe Pharma Co, other from AbbVie, other from Santen Pharmaceutical, other from Daiichi Sankyo, other from Takeda Pharmaceutical, other from Boehringer Ingelheim Japan, other from School Health Record Centre, other from Real World Data, outside the submitted work.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.