Coexistence of serotonin syndrome and neuroleptic malignant syndrome: does it exist?

Abstract

We report a 21-year-old man with bipolar disorder who was on a stable dose of escitalopram and risperidone. Tramadol and cough syrup (dextromethorphan) were added for his recent attack of upper respiratory tract infection. However, he developed various neurological symptoms. Haloperidol and ondansetron were added after hospitalisation. However, his condition deteriorated. A diagnosis of serotonin syndrome (SS) was made, and cyproheptadine was started. Cyproheptadine provided relief in most of the symptoms within 48 hours except for the presence of fever and rigidity. The addition of bromocriptine provided a complete resolution of the symptoms. We considered the presence of both SS and neuroleptic malignant syndrome (NMS) in this case. There are four similar cases in the literature. We discussed a diagnostic and therapeutic approach for patients who are on both serotonergic agents and neuroleptics and develop SS-like or NMS-like clinical features.

Background

Serotonin syndrome (SS) and neuroleptic malignant syndrome (NMS) are drug-induced clinical syndromes. The clinical presentations of SS and NMS overlap. Differentiation between the two is very important as treatment differs for both. SS is typically caused by serotonergic agents, while NMS is caused by the ingestion of dopamine antagonists (neuroleptics).¹ However, patients may receive both serotonergic and neuroleptic agents simultaneously. In such patients, SS or NMS or a combination of both may develop, and this can pose a clinical challenge for physicians.² However, the correct diagnosis is very important because both conditions can be fatal. Herein, we are reporting a patient who developed both SS and NMS simultaneously. We also discuss a practical approach for patients who are on both serotonergic agents and neuroleptics and develop SS-like or NMS-like clinical features.

Case presentation

A 21-year-old man, observed in pre-COVID-19 time, had a 4-year history of bipolar disorder. Since then, he had several episodes of depression and mania; the majority of relapses were related to the discontinuation of treatment. The patient received various antidepressants and antipsychotics over 3–4 years. He had been stable on a combined regimen of escitalopram 20 mg/day and risperidone 0.5 mg/day for 4 months. About 2 weeks ago, he was diagnosed with upper respiratory tract infections (URTI) because of sore throat, cough, fever and myalgia. The patient received a combination of azithromycin (500 mg/day), tramadol–paracetamol (37.5/325 mg thrice daily) and cough syrup (dextromethorphan) for URTI. A few days later, he developed agitation, sleep disturbances and headaches.

Given the possibility of relapse of mania with URTI, the dose of risperidone was increased to 1.0 mg/day. However, his condition deteriorated, and he became violent and incoherent. He was brought to the medical emergency unit, and on arrival, clinical features included agitation, raised temperature (38.1°C), tachycardia (126 beats/min) and neck rigidity. Considering the presence of neck rigidity, headache, fever and altered behaviour, meningoencephalitis of unclear aetiology was suspected. Antibiotics (injectable ceftriaxone and ampicillin) and injectable acyclovir were started empirically. In parallel, he was subjected to various laboratory investigations.

Investigations

Initial laboratory tests showed leucocytosis (white blood cell count: 14.4 ×10⁹/L) and hyperCKemia (creatine kinase (CK) levels: 840 IU/L). Laboratory tests for thyroid, liver and renal function were normal. Serological testing for HIV, Herpes simplex virus (HSV), Hepatitis B virus (HBV), tuberculosis and malaria parasite was negative. Cranial MRI, Cerebrospinal fluid (CSF) parameters and electroencephalogram were normal.

Treatment

Besides antibiotics and acyclovir, he received supportive care, including intravenous lorazepam (2 mg). However, there was no significant improvement with these measures. Injection of ondansetron was started to reduce nausea and prevent gastrointestinal stress. However, his symptoms worsened, and he developed a high-grade fever (40.0°C), diaphoresis, hypertension (154/94 mm Hg), incoordination and myoclonic jerks. Later, he became less communicative and drowsy. At this point, neurological consultation was sought. In addition to above mentioned physical symptoms and signs, we noted generalised stiffness, generalised hyperreflexia, inducible clonus at the ankles, mydriasis and hyperactive bowel sounds. The possibility of meningoencephalitis was less likely because of the normal CSF examination and normal cranial MRI. This patient met the criterion of both SS and NMS. However, considering the presence of hyperreflexia, inducible clonus, myoclonic jerks, mydriasis and hyperactive bowel sound, the possibility of SS was high.

The patient had already stopped taking tramadol and dextromethorphan. In addition, we discontinued escitalopram, ondansetron and risperidone. Cyproheptadine was started (via nasogastric tube) with a loading dose of 12 mg, followed by 2 mg every 2 hours up to 24 hours. He was also started on lorazepam 1.0 mg intravenous 12 hourly.

Outcome and follow-up

After 12 hours of initiation of cyproheptadine, an improvement was noted with decreasing fever (38.6°C), decreasing muscle rigidity and resolution of clonus. After 24 hours of initiation of cyproheptadine, the patient was alert and oriented. The dosing schedule of cyproheptadine was changed to 8 mg three times a day after 24 hours. Hyperreflexia, myoclonus and hyperactive bowel sounds disappeared by 36 hours. Blood pressure also settled to a normal level. However, the patient still had a persistent low-grade fever (38.0°C−38.6°C), tachycardia (>110 beats/minute), intermittent diaphoresis and generalised rigidity. There was no improvement in these clinical parameters despite the continuation of cyproheptadine for 7–8 days. The repeat laboratory investigations were largely normal except for CK levels. The initial CK levels progressively increased to 1570 IU/L. Meanwhile, he accidentally received a single dose of risperidone (1.0 mg) that further increased fever (38.6°C–39.2°C), tachycardia (>118 beats/minute) and generalised rigidity. However, there was no hyperreflexia, inducible clonus, myoclonic jerks or mydriasis. It raised the possibility of associated NMS. Dantrolene was not available. So, bromocriptine 2.5 mg 6 hourly was added to cyproheptadine. The patient’s clinical symptoms and physical signs improved gradually. Fever and tachycardia subsided after the initiation of bromocriptine in 3 days. Rigidity improved markedly, and only mild rigidity remained by 7 days. He was discharged with advice to continue cyproheptadine (4 mg 8 hourly) and bromocriptine (2.5 mg 8 hourly) for another 2 weeks. After 2 weeks, his neurological deficits improved completely, and he returned to normal life.

After 3–4 months, amitriptyline 25 mg was started on the development of a depressive episode. He was monitored closely to see if he develops any drug-related side effects. There were no similar attacks in the next 12-month follow-up.

Discussion

Discussion

This case raises the diagnostic and therapeutic issues in patients who are on both serotonergic agents and dopamine antagonists and develop drug-induced clinical syndromes. Such patients may have SS or NMS or a combination of both. There is considerable overlap in clinical features between SS and NMS, and it is not easy to differentiate one disease from another.² However, a correct diagnosis is important as specific treatments for both disorders are different. There are no biological markers for both SS and NMS, and diagnosis is made by certain clinical criteria. Unfortunately, there is overlap even in the diagnostic criteria of these syndromes.^{3 4}

Table 1 compares the clinical characteristics of SS with NMS. Both syndromes may have a combination of cognitive impairment, rigidity, autonomic disturbances, fever and raised CK levels. It is said that high fever, severe rigidity and markedly raised CK levels favour NMS.¹ However, a few recent reviews suggest that these clinical features in severe SS may be similar to NMS.^{5 6} In a recent comparison of NMS to SS, the median CK levels were 15 059 IU/L and 10 888 IU/L, respectively.⁵ The median fever was 39.5°C for both SS and NMS groups. In a systemic review of fatal cases of SS, the mean temperature was 41.6°C.⁶ So, the severity of symptoms and signs (fever, rigidity and raised CK levels) may not be the true parameters for distinguishing NMS from SS.

Table 1.

A comparison between severe serotonin syndrome (SS) and neuroleptic malignant syndrome (NMS)

Parameters	Serotonin syndrome	Neuroleptic malignant syndrome
Similarities between SS and NMS
Cognitive impairment	Altered consciousness	Altered consciousness
Temperature	Hyperthermia	Hyperthermia
Autonomic disturbances	Autonomic instability	Autonomic instability
Rigidity	Present	Present
Laboratory abnormalities	Raised CK, leucocytosis	Raised CK, leucocytosis
Complication	Rhabdomyolysis, DIC, multiorgan failure	Rhabdomyolysis, DIC, multiorgan failure
Differences between SS and NMS
Type of medications	Serotonergic agents	Dopamine antagonists (neuroleptics)
Onset and evolution	Rapid onset and rapid progression	Insidious onset and slow progression
Reflexes/clonus	Hyperreflexia, clonus	Hyporeflexia
Myoclonus	Present	Absent
Shivering	Present	Absent
Pupil size	Mydriasis	Normal
Gastrointestinal features	Diarrhoea, nausea, vomiting, increased bowel sounds	No or minor gastrointestinal similarities
Treatment response	Rapid resolution of symptoms (<24 hours to a few days)	Slow resolution of symptoms (few days to 2 weeks)

CK, creatine kinase; DIC, Disseminated intravascular coagulation.

The identification of the offending agents is one of the clinical clues to differentiate SS from NMS. While SS is caused by serotonergic agents, NMS is caused by dopamine antagonists. This patient received both types of medications. He received at least four serotonergic drugs in recent times (escitalopram, tramadol, dextromethorphan and ondansetron). In parallel, he was administered a typical neuroleptic, haloperidol, after admission to the hospital. He was on risperidone for 4 months, and recently, the dose of risperidone was increased when he felt agitation. Risperidone is an atypical neuroleptic, which may cause NMS. Some cases of risperidone-induced SS have also been reported in the literature.⁷ However, a few authors believe that risperidone can be helpful in the resolution of SS.⁸ So, this patient had been on multiple drugs that can cause both SS and NMS. However, we considered the possibility of isolated SS in the initial evaluation because of the presence of hyperreflexia, inducible clonus, myoclonic jerks, mydriasis and hyperactive bowel sounds. These clinical features are typical of SS and are usually not found in patients with NMS (table 1). The disappearance of these symptoms after ingestion of cyproheptadine further reinforces the diagnosis of SS. SS usually resolves within a few days of discontinuation of the offending agent and initiation of therapies. Most of the symptoms in this case disappeared within a few days. Although fever and rigidity also improved, the improvement was not complete. Following the accidental ingestion of risperidone, increased fever and rigidity without the reappearance of hyperreflexia, inducible clonus, myoclonic jerks and mydriasis increased the possibility of NMS. A positive response by the administration of bromocriptine also strengthens the view that this patient had NMS. So, treatment with the 5-HT2A antagonist provided relief in some of the clinical symptoms, and the addition of bromocriptine completely resolved the symptom. Therefore, we believe that this patient had both SS and NMS. There was no better alternative diagnosis for this patient. In fact, even in the initial assessment, the patient met the Hunter criteria of SS as well as the DSM-V (Diagnostic and Statistical Manual of Mental Disorders-Fifth Edition) criteria of NMS.^{3 9} However, in the early stage, we treated him as a case of isolated SS. The incomplete response by cyproheptadine prompted us to think of NMS.

The differentiation of SS with NMS is difficult, and several authors have discussed this part. The coexistence of both syndromes at the same time makes it more challenging.^{1 2} To the best of our knowledge, there are at least four case reports in the literature where the authors believed that their patients had both SS and NMS at the same time. Mazhar et al reported a 64-year-old man with neuropathy and depression.¹⁰ The patient had recently received linezolid and metoclopramide for his diabetic foot infection. Thereafter, he developed febrile encephalopathy with various neurological symptoms and signs. Initially, the authors made a diagnosis of NMS and started dantrolene, carbidopa–levodopa and baclofen. However, it provided improvement in only rigidity and CK levels, and other symptoms persisted. Adding cyproheptadine provided improvement in fever, diaphoresis, altered behaviours, myoclonus and high blood pressure. In Nisijima et al case report, a 62-year-old man with a mood disorder received imipramine, lithium and metoclopramide.¹¹ He developed febrile encephalopathy with various autonomic and neuromuscular abnormalities. The initial diagnosis was NMS. However, the patient did not show any response to dantrolene, and cyproheptadine was added to it. He received a combination of cyproheptadine and dantrolene for about 10 days. The complete response was noted in about 20 days. The authors suggested that a delayed response was due to associated NMS. Dosi et al reported a 19-year-old patient who developed a combination of altered mental status, autonomic instability and neuromuscular hyperexcitability following ingestion of valproic acid, ondansetron, risperidone, olanzapine and lithium.¹² Cyproheptadine provided relief in most of the symptoms within 24 hours except for akinetic mutism and rigidity, which lasted for more than 2 weeks. The patient did not receive any specific therapy for NMS. The authors concluded that persistent rigidity and akinetic mutism were due to associated NMS. Sun et al reported a 12-year-old boy with a history of developmental delay and other psychiatric disorders.¹³ The patient developed a mixed presentation of SS and NMS following administration of various drugs. However, the patient did not receive any specific therapy for NMS. Kaufman et al reported a 23-year-old woman who developed various clinical symptoms following the ingestion of venlafaxine, topiramate, divalproex sodium, risperidone and carbamazepine. Initially, they treated the patient with a combination of cyproheptadine and dantrolene, considering a diagnosis of mixed SS and NMS. However, later, they considered it as a case of NMS.² The authors suggested an algorithm for cases that develop a toxidrome in the setting of polypharmacy.

A diagnostic approach for neurotoxic syndrome in patients receiving both serotonergic and neuroleptic agents

The diagnosis of SS and NMS is challenging. After a MEDLINE search, Birbeck and Kaplan found at least 17 cases of SS that were reported as NMS during 1987–1991.¹⁴ None of the patients were correctly diagnosed by treating physicians on the initial evaluation of 24 patients with SS in an intensive care unit (ICU) setting in a study by Prakash et al.¹⁵ In Pedavally et al case series of 33 patients with SS in an ICU, SS was suspected in only four patients on initial evaluation.¹⁶ A diagnostic confusion will be very high if patients receive both serotonergic and neuroleptic agents simultaneously. NMS is typically found in patients with psychiatric disorders. Although SS is commonly found in psychiatric patients, several new clinical settings have emerged where the patient may have SS.¹⁷ Several drugs with serotonergic properties are not marketed as serotonergic agents. All these complicate the diagnostic issue.

The demonstration of the temporal relationship between drug ingestion and the development of clinical features is vital for the diagnosis of both SS and NMS. SS typically develops within 24 hours of the administration of serotonergic agents.¹⁸ Similarly, the time limit for the development of NMS is 72 hours after exposure to dopamine antagonists.³ Therefore, it is important to know what kind of medicine the patient had taken recently. However, we should be careful that both SS and NMS may occur even with stable doses. In a recent review of 56 cases of fatal SS, seven (13%) patients had been on a stable dosage, and they had indolent onset.⁶

All clinical features and laboratory abnormalities described in NMS are also found in patients with SS or at least in patients with severe SS. There are no clinical features favouring NMS over SS. A typical case of SS may also fulfil the criteria of NMS. However, there are at least two clinical features that are particularly found in SS compared with NMS.¹⁹ The presence of neuromuscular abnormalities (hyperreflexia, clonus, myoclonus and shivering) and gastrointestinal symptoms (nausea, vomiting, diarrhoea and increased bowel sounds) favours the diagnosis of SS. Clonus is the central feature in the Hunter criteria of SS, and it is not found in patients with NMS.⁹ About 90% of the body’s serotonin is stored in the enterochromaffin cells of the gastrointestinal tract, and it controls several gastrointestinal functions. Therefore, gastrointestinal features are common in SS, whereas these are not found in patients with NMS.¹⁹ Diarrhoea is one of the symptoms in Sternbach’s criteria of SS.⁴ If you do not focus on clonus and gastrointestinal symptoms, most of the severe cases of SS are diagnosed as NMS. After a MEDLINE search, Kontaxakis et al reviewed olanzapine-induced NMS and observed that 65% of such patients also met Sternbach’s criteria.²⁰ So, if a patient meets both criteria, it doesn’t always indicate the coexistence of both diseases.

We are suggesting an algorithm for patients who are receiving both serotonergic and neuroleptic agents and develop a neurotoxic syndrome (figure 1). As gastrointestinal symptoms and clonus are slightly specific for SS, its presence favours the diagnosis of SS. However, such patients still may have associated NMS if they had received multiple types of drugs. The evolution in SS is very fast. In a systematic review of 56 cases of fatal SS, 50% of patients died within 24 hours of the onset of symptoms.⁶ Considering the seriousness of SS, besides supportive measures, specific therapy (5-HT2A antagonists) for SS must be started immediately. Cyproheptadine, chlorpromazine and atypical antipsychotics (olanzapine and risperidone) are frequently used as 5-HT2A antagonists.¹⁸ Cyproheptadine should be preferred. Chlorpromazine, olanzapine and risperidone should be avoided in such cases as they may aggravate associated NMS. In fact, in our case, accidental ingestion of risperidone aggravated the hidden (or undiagnosed) associated NMS. It is very difficult to diagnose associated NMS in patients with SS because most of the symptoms can be explained by SS itself. Therefore, treatment with atypical antipsychotics should be avoided in such circumstances.

Figure 1.

An algorithm for treatment of suspected serotonin syndrome or neuroleptic malignant syndrome after polypharmacy. CK, creatine kinase.

Kaufman suggested a combined treatment with cyproheptadine and dantrolene for undifferentiated cases of SS/NMS at an early stage.² They suggested adding bromocriptine if there is no significant improvement in 48 hours. However, adding both bromocriptine and dantrolene may worsen SS very rapidly, especially if given in the early stage.^{6 18 21} In Mazhar et al and Nisijima et al case reports, dantrolene was the first to be given.^{10 11} But dantrolene did not have much effect on them. However, adding cyproheptadine led to a complete resolution of symptoms. In Dosi et al and Sun et al reports, the patients received only cyproheptadine, and no specific therapy for NMS was given.^{12 13}

So, based on these case reports and considering the possibility of aggravation of SS by dantrolene/bromocriptine, we suggest starting with only cyproheptadine. Bromocriptine can be added later, depending on the response by cyproheptadine. If there is no response to cyproheptadine in the first 24–48 hours, SS is less likely, and bromocriptine or dantrolene can be started. If there is a response to cyproheptadine, it indicates the presence of SS. Therefore, adding bromocriptine or dantrolene at this juncture may aggravate the symptoms complex of SS. Cyproheptadine should be continued until the symptoms are completely resolved or further improvement stops. If there is no further improvement and some symptoms persist, there is a possibility of an associated NMS. Bromocriptine should be started in this scenario. However, it would be better to continue cyproheptadine as long as patients are getting bromocriptine or dantrolene. Continuation of cyproheptadine with bromocriptine or dantrolene will reduce the possibility of reappearance or exacerbations of SS. Several drugs used in ICU settings have serotonergic properties.¹⁸ Therefore, physicians should be very cautious before starting any drug in such patients. Benzodiazepines can be used safely to control agitation in both the syndrome. It may also reduce rigidity and rigidity-associated fever.¹⁸

The absence of clonus and gastrointestinal symptoms does not favour the diagnosis of SS, and the possibility of isolated NMS is more likely. Such patients should be treated like NMS. If there is no response to bromocriptine or dantrolene, a possibility of SS should be considered. However, absent clonus must be interpreted with caution as severe rigidity of SS may mask clonus and hyperreflexia.⁶ Peripheral neuropathy also masks hyperreflexia and clonus. Therefore, if a patient with peripheral neuropathy develops SS, there may not be clonus and hyperreflexia.²² So, physicians should be careful in interpreting ‘absent’ clonus in patients having severe rigidity. So, the presence of clonus is in favour of SS, but its absence does not dismiss SS. Unfortunately, clonus examination is not a regular phenomenon in ICU and other clinical settings.²³ It further creates diagnostic confusion.

This case illustrates the complexities in diagnosing the drug-induced clinical syndromes in patients who are on both serotonergic agents and dopamine antagonists. As both SS and NMS are life-threatening conditions, a sound diagnostic and therapeutic approach is required.

Limitation

A diagnostic issue is an inherent property of both NMS and SS. There are no biochemical markers to confirm these syndromes, and diagnosis is made by clinical criteria. The clinical criteria of both syndromes include the exclusion of a number of other diseases. For various reasons, all differential diagnoses of these serious clinical syndromes cannot be excluded in a limited time frame. So, most of the time, the diagnosis of SS and NMS remains provisional. We believe that our case had both SS and NMS and there are no better alternative diagnoses. However, we cannot rule out other possibilities in this patient.

Patient’s perspective.

I am so thankful for timely intervention by neurologists.

Learning points.

• The clinical features of serotonin syndrome and neuroleptic malignant syndrome overlap.

• Serotonin syndrome and neuroleptic malignant syndrome may coexist.

• Differentiation between the two is very important as treatment differs for both.

• Treatment of one syndrome may aggravate the symptoms of another one.

Ethics statements

Patient consent for publication

Obtained.