Both serotonin syndrome and NMS often present with muscle rigidity, hyperthermia, and altered mental status. Here's how to diagnose and treat them.
Published On: 8/16/2021
Duration: 24 minutes, 26 seconds
Related Article: "Serotonin Syndrome Versus NMS," The Carlat Hospital Psychiatry Report, January 2021 (free article)
Tormoehlen LM and Rusyniak DE, Handb Clin Neurol 2018;157:663–675.
Volpi-Abadie J et al, Ochsner J 2013;13(4):533–540.
Dr. Hendrick: Both serotonin syndrome and neuroleptic malignant syndrome, or NMS, often present with muscle rigidity, hyperthermia, and altered mental status. While the etiologies of these disorders are different, it may be difficult to differentiate between these two disorders due to their significant symptomatology overlap. The latest issue of The Carlat Hospital Psychiatry Report includes an interview with Dr. Laura Tormoehlen, who is a neurologist and an attending physician for the neurology service at Indiana University Health Methodist Hospital, as well as for the toxicology service at Methodist, University, Riley and Eskenazi Hospitals. Additionally, Dr. Tormoehlen completed a medical toxicology fellowship, and she runs the neurotoxicology clinic at Indiana University, and serves as the Vice Chair of Clinical Practice for the neurology department. In this podcast, I’ll discuss the highlights from our interview with Dr. Tormoehlen, and elaborate on how to diagnose and treat serotonin syndrome and NMS.
Welcome to The Carlat Psychiatry Podcast
This is a special episode from The Carlat Hospital Psychiatry Report.
I’m Dr. Victoria Hendrick, the Editor-in-Chief of The Carlat Hospital Psychiatry Report, and a clinical professor at the David Geffen School of Medicine at UCLA. I’m also the director of inpatient psychiatry at Olive View -- UCLA Medical Center.
We’ll begin this episode with a brief overview of serotonin syndrome and NMS.
Serotonin syndrome results from excessive serotonin activity in the central nervous system and can range from mild to severe, and it can even be potentially fatal. In our interview, Dr. Tormoehlen described many of the symptoms associated with the mild version of serotonin syndrome. And while some of these symptoms may seem benign in nature, they very well could signify that your patient has entered the beginning stages of serotonin syndrome.
Dr. Tormoehlen: So anytime you counsel your patients about “You might have diarrhea or tremor from the addition of this medication to your regimen” you’re telling them you might have mild serotonin syndrome when you add this. And so you know careful conversation around that is important because you certainly would want your patient to tell you that so that you either discontinue the other one or stop the new one or make an adjustment in your cross titration plan or that’s your patient’s body telling you that they’re at the beginning of serotonergic excess, and if you don’t make a therapeutic adjustment if they continue or increase their dose before their brain is ready then they might manifest with a more severe clinical syndrome that would necessitate presentation to the hospital.
Dr. Hendrick: Serotonin syndrome is predictable. If a patient takes enough serotonergic drugs, especially if they have different mechanisms, you will observe serotonin syndrome. Serotonin syndrome is more common than we realize because mild cases occur that we don’t think of as being serotonin syndrome.
On the other hand, NMS is much more idiosyncratic and difficult to predict. NMS is a rare but potentially life-threatening reaction to antipsychotic medications, due to blockage of dopamine receptors.
Dr. Hendrick: As I mentioned before, there’s a significant degree of clinical overlap between serotonin syndrome and NMS. For instance, if a patient’s vitals indicate that they’re hypertensive, hyperthermic, tachycardic, and/or tachypneic, they very well could have either one of these two disorders. And muscle rigidity and altered mental status is common in both conditions as well.
So what about lab findings? Well, unfortunately labs won’t provide you with any diagnostic clarity considering that both disorders can lead to elevated creatinine kinase, or creatinine phosphokinase, levels. The elevated levels of creatinine kinase results from muscle rigidity, as creatinine kinase is an enzyme that leaks out of damaged muscle tissue.
Once you see either an increased tone or an elevated temperature of unknown origin, then the next step is to ask the patient about their medication history, and perform a review of the patient’s medication history in their EHR, keeping in mind that many drugs that we might not think of as serotonergic can be serotonergic as a secondary mechanism.
There are several ways drugs can be proserotonergic (i.e. lead to increased serotonin activity). For example, they can increase serotonin release, inhibit serotonin metabolism, inhibit serotonin reuptake, or activate serotonin receptors. Fentanyl, linezolid and OTC drugs like dextromethorphan are all examples of proserotonergic medications that tend to be overlooked. You also want to get information on any supplements like St John’s wort.
Also, it’s important to recognize that the particular serotonergic potency of certain drugs should be considered when reviewing a patient’s medication history.
Dr. Tormoehlen: A reasonable example of that would be triptans. Triptans are We sort of regard triptans as proserotonergic, and you know if you’ve tried to prescribe a triptan to a migraine patient who is also on amitriptyline as a prophylaxis, you might get an alert within your EMR will send you an alert. And triptans are only indeed proserotonergic, however in my experience they are pretty weakly proserotonergic, So you’ve got to take a whole lot of your triptan on top of other things, which I think just sort of lends itself to the conversation around serotonin syndrome.
Dr. Hendrick: So what should we do if a patient presents with the characteristic symptoms of NMS and serotonin syndrome, but we’re unable to acquire a thorough and accurate medication history?
Dr. Tormoehlen: I think it is important actually to consider these two diagnoses together and remove the possibility of a good history. So let’s assume we can’t get a good history, how might we be able to tell the difference? And the answer is there is no gold standard diagnostic test that will tell you and so the next things that I’m about to share are essentially a way that you can move towards a reasonable diagnostic certainty that will allow you to make management decisions, but if you can’t get there then you have to assume it could be both. But some ways that you might be able to tell the difference would be onset. So serotonin syndrome is much more likely to be acute in onset, so hours – on the order of hours, and NMS is likely to be subacute in onset – so days roughly speaking. And so if your patient or observers tell you they were fine 3 or 4 hours ago and now suddenly they are hyperthermic and rigid, it’s much more likely to be serotonin syndrome. That’s certainly not 100% diagnostic, but the time course can help you.
Dr. Hendrick: Yes, inquiring about the onset of symptoms can be informative. But what if the patient’s history is limited? Fortunately, although muscle rigidity is typical in both conditions, specific motor features can be used to differentiate between the two disorders.
Dr. Tormoehlen: So serotonin syndrome classically manifests with clonus: significant motor hyperactivity, sometimes even spontaneous – you can observe it happen from the door. And inducible clonus at the ankles is the thing that my residence and fellows observe me to add to the physical exam more than anything else I suppose. Interestingly, it tends to happen in the lower extremities more than the upper extremities so you have to check the legs – not just for tone, but also for clonus. Sometimes I think our bedside exam gets a little quick and we check for tone in the arms and maybe skip the legs, but if you do you might miss the increased tone and clonus in the lower extremities in serotonin syndrome. Conversely, NMS is rigidity less than clonus. You might observe a pretty substantial increase in tone even to the degree that you can’t reasonably check for inducible clonus at the ankles because the tone is so high.
Dr. Hendrick: To summarize, clonus specifically refers to involuntary, rhythmic muscle contractions, and it’s common in serotonin syndrome. You can induce clonus by briskly flexing the patient’s foot upward, you’ll see a rhythmic beating of the foot and ankle. Make sure you sustain that passive dorsiflexion of the foot. If the rhythmic beating continues beyond a couple of beats, that would be abnormal.
On the other hand, NMS is primarily about rigidity -- typically described as “lead pipe” -- rather than clonus. You might observe a substantial increase in tone even to the degree that you can’t check for inducible clonus at the ankles because the tone is so high.
As a side note, serotonin syndrome and many other disorders share similar features and symptoms. For instance, serotonin syndrome and sympathomimetic syndrome are clinically very similar, but sympathomimetic syndrome results from drugs that primarily cause increased activity of norepinephrine rather than serotonin. However, many drugs do both, so they have a significant clinical overlap. Anticholinergic syndrome also causes an agitated delirium with hyperthermia, hypertension, tachycardia, and mydriasis. But, it tends to be associated with dry mucous membranes, absent bowel sounds, and a milder degree of hyperthermia. Another would be GABA agonist withdrawal – so ethanol, of course, and benzo’s, barbiturates. Those patients can have profound hyperthermia, rigidity, seizures, and hypertension, tachycardia – the whole deal. GABA agonists like benzodiazepines will also treat those syndromes. There are many others, so nontoxic things would be thyrotoxicosis, sepsis of course, including CNS infection, then also stiff person syndrome, some other autoimmune conditions. There’s also tetanus and strychnine poisoning – those patients would probably have normal mentation.
Now that we’ve covered how you can differentiate between serotonin syndrome and NMS, let’s move on to how these two disorders are treated.
Dr. Tormoehlen: So stopping the drugs, all of them. Anything that’s proserotonergic or antidopaminergic or you just don’t know. These two things are not alone in the differential diagnosis of each other. So sympathomimetic syndrome, antimuscarinic, anticolinergic syndrome would also live on that differential. And so we want to stop all of those drugs too. First-line treatment is benzodiazepines. Second-line treatment is benzodiazepines. Third-line treatment is benzodiazepines. So essentially what you’re trying to do is remove the offending agents, and maybe this is a little bit too simple, but “chill the brain out.” So there is too much of something going on and a good way to make the brain just be a little quieter regardless of which neurotransmitter system is overreacting is GABA agonist therapy. And if your patient is hyperthermic then you’ve got to get them cooled. So benzo’s will help with that – not only in the central nervous system but also can help decrease the tone. A lot of the hyperthermia is overgeneration of heat from muscle activity. So if you can help relax the muscles then you might decrease the temperature. External cooling is really important. If your patient is really, really hot then sometimes prophylactically intubating, mechanically ventilating, and pharmacologically paralyzing is necessary, although that doesn’t happen particularly commonly. Most of the time patients who are going to need that need to be intubated anyway because their mental status is so depressed or they’re so agitated that they are not adequately protecting their airway. So those are the treatments that have a common theme between serotonin syndrome and NMS. The good news is that treatment strategy will also treat anticholinergic syndrome and it will also treat serotonin syndrome. So you don’t have to have magically figured it out as long as you aggressively support your patient and use GABA agonists. So after the patient is intubated then propofol is a reasonable substitute. Phenobarb is actually a reasonable substitute.
Dr. Hendrick: Given that it’s difficult to differentiate between serotonin syndrome and NMS, we are quite fortunate that the first-line treatment for both disorders is benzodiazepines. This enables us to initiate treatment when we are uncertain or unable to definitively deduce what specific disorder our patient is presenting with.
However, what if we do know what disorder they have? If we are certain of what condition we are treating, we can make use of antidotes that target the specific neurotransmitters causing our patient’s condition.
Dr. Tormoehlen: For antidotal therapy, if you are sure of what you’re treating then there are antidotes that can be sort of aimed directly at the receptors or the neurotransmitters causing the problem. For serotonin syndrome that is cyproheptadine which is an enterally available – you can only give it p.o. – and it’s primarily an antihistamine. It is a bit anticholinergic so if you think there’s a chance that your patient has anticholinergic syndrome then it’s probably not the best idea.
So I will use GABA agonists, aggressive supportive care, external cooling plus or minus cyproheptadine in the management of serotonin syndrome.
In the management of NMS, same: GABA agonist therapy, aggressive supportive care, external cooling, obviously discontinuation of the offending drugs. And then if I’m convinced that it’s NMS I would probably first choose bromocriptine as an antidote. Dopamine agonist also interally administered. The reason for that would be that in the same way that NMS is subacute in onset, it tends to be subacute in recovery especially for patients who have received Depo injection they are gonna be stuck with it for a while and so bromocriptine can be effective in managing their symptoms until the effect of the antipsychotic wears off.
And then the second agent that could be added if necessary and only if the rigidity is particularly severe would be dantrolene. So dantrolene remember is direct in the myosite muscle relaxant, so it does not help with any of the CNS symptoms; it only treats the rigidity. So if the temperature is so high and the rigidity is so severe then dantrolene can be a useful adjunct for that. But most of the time we get away with using GABA agonists plus or minus bromocriptine without the dantrolene which is nice because of the need for monitoring of hepatic tests and some of the rest of that.
Dr. Hendrick: Patients will usually improve quickly with aggressive management in either syndrome, but NMS takes longer to completely recover, so patients need a management plan on the order of weeks. If serotonin syndrome is severe then it may take several days to recover in the ICU, but once patients are medically stable and ready for discharge, they typically do not require any outpatient management; once it’s done it’s done.
We have now covered the main topics concerning the differentiation between serotonin syndrome and NMS, how these conditions are treated, and the prognosis with treatment for each disorder. But there is one more critical component we need to touch on.
We discussed how the quantity and potency of certain drugs may induce these disorders, and how the first treatment step is to discontinue the culprits that may be responsible for causing these disorders. Yet, what are we supposed to do after our patients have recovered from one of these conditions? I mean, there is a reason why these patients were started on these particular medications in the first place. So our next step is to re-introduce the most important pharmacotherapies needed to treat our patients' psychiatric disorders.
Dr. Tormoehlen: It’s much easier in serotonin syndrome, much, much easier typically because those patients there had an overdose or a drug interaction. The drug interaction patients tend to be much less severe; the really severe serotonin syndrome that I’ve seen is generally in the overdose patient. So typically we wait 24-48 hours after complete resolution of their syndrome and then they can go back to previously prescribed medicines as long as there are a reasonable number of proserotonergics. So sometimes patients have a polysubstance overdose and those substances include amitriptyline, citalopram, trazodone, tramadol – all of which are proserotonergic and potentially in the different things. So a patient who comes in with a polysubstance ingestion and was prehospital stay on 4 serotonergics, we should not resume 4 serotonergics after this. But you can pick the most important one and start that one pretty soon after resolution of the syndrome.
NMS is much harder, because as I mentioned before, it’s an idiosyncratic reaction, very hard to predict who’s gonna get it, and then once they’ve had it we tend to sort of classically have this trepidation about resuming, which I think is an appropriate trepidation but you know better than I do patients who require antipsychotic therapy require antipsychotic therapy, so you can’t necessarily say, “Well, you’ve had NMS once before so you can never have an antipsychotic again.” I try really hard not to hamstring my psychiatry colleagues by saying that. So you know I think you would want to wait until their symptoms are completely resolved. You would want to pick a second-generation antipsychotic less likely to cause NMS. You probably would want to choose something - not the drug that just caused the NMS, low dose, slow titration, very careful observation, and then probably no depo’s for that patient.
And then the last thing would be there are some “risk factors” for NMS – really just sort of retrospective observations of which patients developed it. Depo administration was certainly on the list. Younger age was certainly on the list – can’t really change that. But dehydration is also on the list and so careful counseling of your patient: “You know you’ve had this side effect before, but we think you will need this kind of medication. We’re gonna need to watch really closely and here are some things that we want you to take care to avoid.”
The risk factors for serotonin syndrome are how many serotonergic meds you’re on and then overdosing on them. Like I said, serotonin syndrome is fairly predictable. You take any person and give them enough proserotonergic drugs they’ll get it. So not the same for NMS, it is much more difficult to predict, but for serotonin syndrome it’s how many different drugs that are proserotonergic and how many different mechanisms. So if you take a serotonin reuptake inhibitor and a monoamine oxidase inhibitor and serotonin receptor agonist you’re probably gonna get more profound serotonin syndrome than 3 drugs that operate in the same mechanism.
Dr. Hendrick: Overall, both conditions present similarly, and there are an abundance of disorders that also overlap with serotonin syndrome. Nevertheless, we can still differentiate between these disorders by conducting a thorough search of our patient’s medication history, inquiring about the onset of their symptoms, and by interpreting the motor features associated with their muscle rigidity.
Remember, the archetypal motor feature of serotonin syndrome is clonus, while “lead-pipe”-like muscle rigidity is associated with NMS. Even if we are unable to conclude which disorder is present, we can initiate treatment with benzodiazepines and reduce our patient’s hyperthermic symptoms.
If we are aware of what disorder is present, then we can utilize cyproheptadine for serotonin syndrome, which is dosed at 12 mg followed by 2 mg every 2 hours until improvement and then 8 mg every 6 hours for maintenance dosing.
For the management of NMS, bromocriptine – a dopamine agonist – is the potential antidote (2.5-5 mg orally or by NGT every 8 hours ). The second agent we might add, but only if the rigidity is severe, would be dantrolene (3-5 mg/kg IV divided TID, or orally at 100-400 mg/day QID). Dantrolene is a muscle relaxant, so it only treats the rigidity; it does not help with any of the CNS symptoms. If dantrolene is needed, you will need to monitor hepatic function as dantrolene can cause liver toxicity. In addition, bromocriptine does have some proserotonergic activity, so should be avoided if serotonin syndrome remains possible.
In serotonin syndrome, we can restart medications after 24-48 hours of complete resolution of their syndrome, and then patients can go back to previously prescribed medicines as long as you limit the number of proserotonergics and restart only the most important medications. If possible, start with one agent at low dose. And, in NMS, we only want to restart antipsychotics if the indication to resume is strong. As long as you start at a low dose and titrate up slowly, you can restart an antipsychotic. You want to wait until the symptoms are completely resolved, at least two weeks, and then pick a second-generation antipsychotic because they appear less likely to cause NMS. Also, you want to choose a different drug than the one that caused the NMS, and don’t use depot injections.
These disorders may seem complicated to treat, but the aforementioned tips and guidelines will help you navigate the immensely overlapping symptomologies of these disorders, and allow you to provide your patients with the best clinical care you can give.
The print interview is available for subscribers to read in The Carlat Hospital Psychiatry Report. Hopefully people check it out. Subscribers get print issues in the mail and email notifications when new issues are available on the website. Subscriptions also come with full access to all the articles on the website and CME credits.
And everything from Carlat Publishing is independently researched and produced. There’s no funding from the pharmaceutical industry.
That’s right, the newsletters and books we produce depend entirely on reader support. There are no ads and our authors don’t receive industry funding. That helps us bring you unbiased information you can trust.
Go to www.thecarlatreport.com to sign up. You can get a full subscription to any of our three newsletters for $30 off using the coupon code LISTENER.
There are a lot of doctors and nurse practitioners listening, and quite a few students and residents too. Many of you are interested in getting some kind of CME credit for listening.
So we’re going to do it! We’re working on putting that program together now and as soon as it’s available you’ll be able to take a post-test for listening and get credits. Everyone who took the survey is going to get early notification when it’s ready because we have their email addresses.
If you’d like to get that early notification you can still take the survey at www.thecarlatreport.com/podcastsurvey.
As always, the links you need are in the episode description. Thanks for listening and have a great day!
Got feedback? Take the podcast survey.