First psychotic episode: when the cause is organic

A 24-year-old young man is brought to the emergency department by his mother. About two weeks ago, he started sleeping little, becoming suspicious of his neighbors and hearing voices that comment on what he does. He has never had a psychiatric illness. On assessment, he is disoriented at times and, according to his mother, had a "fainting spell with shaking" the previous week.

The first psychotic episode is usually treated as the entry point of a primary psychiatric illness, such as schizophrenia. Most of the time that is indeed the case. But some patients have an identifiable organic cause, and it is precisely in the first episode that this cause has the best chance of being found and treated.

This text discusses how to think about the organic cause in the first psychotic episode: which red flags change the probability, what autoimmune encephalitis taught psychiatry, what role drugs play and what to expect from imaging and cerebrospinal fluid tests.

Key points

• Secondary psychosis has an identifiable organic cause, and the first episode is the best time to look for it.
• Abrupt onset, an atypical picture and neurological or clinical signs are the main red flags.
• Autoimmune encephalitis, especially anti-NMDA, can start as a purely psychiatric picture.
• Testing antibodies in every patient, without criteria, yields little.
• Substance-induced psychosis is not a benign diagnosis and can progress to a primary psychotic illness.

What secondary psychosis is

A German review on autoimmune encephalitis defines primary schizophreniform psychoses as pictures caused by complex interactions between genes and environment, and secondary forms as those based on a clearly identifiable organic cause (Endres et al., 2020). The cause may be a direct etiological factor or a relevant pathogenic factor.

The Freiburg group lists the main families of secondary causes: autoimmune diseases of the central nervous system, such as autoimmune encephalitis, systemic autoimmune diseases, such as lupus, infections, such as neurosyphilis and neuroborreliosis, endocrine diseases, such as hyperthyroidism and hyperparathyroidism, metabolic diseases, such as Wilson disease, and epilepsy (Endres et al., 2020). The list is long, and the reasoning is not about ordering everything, but about knowing what to look for in each patient.

Delirium is not primary psychosis

A basic distinction prevents many errors: hallucinations and delusions also appear in delirium, and delirium is, by definition, the expression of an organic disease. What sets it apart is the change in attention and level of consciousness, with a fluctuating course over the day.

In practice, this means that every patient with new psychosis should have attention and orientation carefully assessed. A patient who is easily distracted, loses the thread of the conversation and worsens at the end of the day is not just psychotic. He is showing a sign that the brain is suffering from some cause that needs to be found, whether infection, metabolic disorder, drug or inflammation.

Why this matters in the first episode

The schizophrenia label tends to stick. Once the psychiatric diagnosis enters the chart, new complaints are usually read through that lens, a phenomenon similar to premature closure and other reasoning shortcuts. The first episode is the moment when the physician still has the freedom to consider other hypotheses.

The weight of the diagnosis also shows up in treatment. A French review notes that an estimated at least 30 out of every 100 patients with schizophrenia do not respond to conventional antipsychotics (Ellul et al., 2017). For the authors, this reinforces the need to identify subgroups with specific biological mechanisms, including those of autoimmune origin.

The link with autoimmune diseases

The link between autoimmunity and psychosis goes beyond encephalitis. A Danish review summarizes registry studies in which the risk of psychosis after a diagnosis of an autoimmune disease was about 45 per 100 higher (Jeppesen and Benros, 2019). The authors point out that this association also draws attention to the medical conditions that accompany patients with psychosis.

A meta-analysis gathered 31 studies, with data from more than 25 million people, and found a positive association between non-neurological autoimmune diseases and psychosis, with an odds ratio of 1.26 (Cullen et al., 2019). The association varied by disease: it was positive for pernicious anemia, with an odds ratio of 1.91, and for Graves disease, with 1.33, and negative for rheumatoid arthritis, with 0.65.

The warning signs that change the probability

The central point of the reasoning is that an organic cause is not equally likely in every patient. Certain findings greatly raise this probability, and the investigation should rest on them, as the reasoning of pretest probability teaches.

What the Freiburg review points out

The review on autoimmune encephalitis describes as typical signs the acute onset of paranoid and hallucinatory symptoms, a polymorphic and atypical presentation, psychotic episodes in someone who has already had encephalitis, and associated neurological and clinical symptoms such as catatonia, seizures, dyskinesia and autonomic instability (Endres et al., 2020). The relatively rapid development of the picture and the atypical symptoms should lead to a broad investigation.

Clues that come from the history

Not every warning sign appears on physical examination. Some clues come from talking with the family: an onset outside the usual age range, the absence of gradual changes in the preceding months, fever or headache in the days before the picture, or a rapid worsening that leaves the family unable to recognize the person.

The type of symptom also helps. Predominant visual hallucinations, confusion and memory changes draw more attention to an organic cause than isolated auditory hallucinations in a well-oriented patient. None of these clues confirms the cause, but each one raises the probability and reinforces the decision to investigate further.

Possible, probable and definite autoimmune psychosis

The same review reproduces criteria for autoimmune psychosis at three levels (Endres et al., 2020). The possible form requires a psychotic episode of abrupt onset, in less than 3 months, with at least one finding such as tumor, catatonia or dyskinesia, a severe adverse reaction to antipsychotic suggestive of neuroleptic malignant syndrome, disproportionate cognitive deficit, lowered consciousness, new seizures or significant autonomic dysfunction.

The probable form additionally requires pleocytosis in the CSF above 5 cells per mm³ or bilateral changes restricted to the medial temporal lobes on MRI, or else two findings among an encephalopathic electroencephalogram, oligoclonal bands or an increased IgG index in the CSF, and antineuronal antibody in the serum. The definite form requires antineuronal IgG antibodies in the CSF, after other causes have been excluded (Endres et al., 2020).

Level	What is required	Examples of findings
Possible	Psychosis of abrupt onset, in less than 3 months, with at least one warning finding	Tumor, catatonia, dyskinesia, new seizure, lowered consciousness, autonomic dysfunction
Probable	Possible criteria with a CSF or imaging change, or two auxiliary findings	CSF pleocytosis, bilateral medial temporal change, encephalopathic electroencephalogram, antibody in the serum
Definite	Probable criteria with antibody in the CSF	Antineuronal IgG antibodies in the CSF, after excluding other causes

Table source: Endres et al., 2020.

Without a warning sign, the yield drops

Red flags work in both directions. When they are absent, the probability of finding an autoimmune cause is low. A Swedish and British study evaluated 71 patients with a first psychotic episode without traditional neurological signs and found a weak link to antibodies in the serum of 3 of them, with no antibodies in the CSF (Theorell et al., 2021). The authors concluded that these findings do not support screening for neuronal surface antibodies in unselected psychotic patients.

Autoimmune encephalitis that begins in psychiatry

Anti-NMDA encephalitis is the model that changed the way secondary psychosis is understood. It can mimic, at least at the start, variants of primary psychosis (Endres et al., 2020).

Psychiatric symptoms are the rule

A Chinese study gathered 108 patients with a definitive diagnosis of anti-NMDA encephalitis (Wang et al., 2019). Of these, 103 had one or more psychiatric symptoms, and 62 began with psychiatric symptoms, such as mood change, aggression, mania, delirium and hallucinations. The authors concluded that psychiatric symptoms are the most common clinical feature of the disease.

The same study describes that, after the psychiatric phase, neurological symptoms usually appear, such as memory problems, seizures, reduced consciousness, dyskinesia, autonomic instability and hypoventilation (Wang et al., 2019). The patient in the case, with disorientation and a probable seizure event, already shows two red flags.

A normal MRI does not rule it out

A common mistake is to take a normal MRI as proof that there is no organic disease. An American and German review states that, in anti-NMDA encephalitis, the MRI shows no structural changes in about 89 out of every 100 patients at initial presentation and in 79 out of every 100 at follow-up (Najjar et al., 2018). The authors propose the term autoimmune psychosis precisely to remind us that neurological investigation, with electroencephalogram and CSF, must not be forgotten.

When only one test shows the disease

A report from the Freiburg group describes a 22-year-old patient who developed an acute polymorphic psychotic episode, with heard thoughts and paranoid ideas (Endres et al., 2019). Anti-NMDA antibodies appeared in the serum but not in the CSF, and FDG-PET showed the brain involvement that supported the diagnosis.

The patient had a rapid and convincing improvement after treatment, with normalization of neuropsychological tests and a drop in antibody titers (Endres et al., 2019). For the authors, the case shows the importance of a multimodal investigation, with more than one test, in similar situations.

Drugs: the most remembered cause and the most underestimated

In emergency department practice, the question about substance use comes before any imaging test. Substance-induced psychosis is frequent, and the patient with cannabis in the emergency department is a daily example of this presentation.

Drug-induced does not mean transient

The most common mistake is to consider substance-induced psychosis a benign diagnosis that resolves when the drug leaves the body. A South Korean national study followed 2244 people with a first episode of substance-induced psychosis (Piao et al., 2026). At a median follow-up of 9.3 years, conversion to a first episode of psychosis, such as schizophrenia or bipolar disorder, reached 23.3 out of every 100 at 5 years and 28.8 out of every 100 at 10 years.

All-cause mortality was much higher than in the general population, with a standardized mortality ratio of 33.3 (Piao et al., 2026). Younger age, female sex, disability, longer initial hospitalization and use of other stimulants were predictors of conversion.

What this changes in the reasoning

Identifying the drug does not end the investigation. A positive toxicology test explains part of the picture, but does not rule out an associated organic cause or an early primary psychotic disorder. Follow-up must continue after discharge, with attention to the first years, as the authors themselves recommend (Piao et al., 2026). A non-judgmental approach, discussed in stigma and language, helps the patient report what they actually used.

Medications are also on the list

The question about substances is not limited to illicit drugs. Corticosteroids, some antiparkinsonian drugs, stimulants and other commonly used medications can trigger psychotic symptoms, as can withdrawal from alcohol and sedatives. The medication list, including those bought without a prescription, is part of the history-taking for the first episode.

This care also applies to the patient already using an antipsychotic. A severe adverse reaction, with rigidity, fever and autonomic instability, is one of the findings that enter the criteria for possible autoimmune psychosis (Endres and colleagues, 2020) and calls for immediate reassessment of the diagnosis.

What the tests show and what they do not show

The workup of the first psychotic episode ranges from simple tests to complex ones, and each has a different yield depending on the patient.

The broad Freiburg protocol

The Freiburg group applied a broad protocol to 76 consecutive patients with schizophrenia spectrum disorders, with blood tests, metabolic and immunological markers, cerebrospinal fluid, MRI and electroencephalogram (Endres and colleagues, 2020). Vitamin and trace element deficiency appeared in 92 out of every 100, general cerebrospinal fluid changes in 54 out of every 100, especially signs of blood-brain barrier dysfunction, and chronic inflammatory lesions on MRI in 4 out of every 100.

The combination of electroencephalogram, MRI and cerebrospinal fluid showed some abnormality in 76 out of every 100 patients, and there was suspected autoimmune psychosis in three of them (Endres and colleagues, 2020). Cerebrospinal fluid was collected in 48 of the 76 patients. The authors advocate broad investigation, but acknowledge that larger studies are needed to validate the approach.

The basic workup that must not be missing

Before the more complex tests, there is a basic workup that makes sense in almost every first episode. The Freiburg protocol includes blood tests and metabolic and immunological markers (Endres and colleagues, 2020), and in practice this usually involves complete blood count, glucose, electrolytes, renal, hepatic and thyroid function, serologies for infections such as syphilis and HIV and toxicology screening.

These tests are cheap, fast and able to reveal treatable causes. The reasoning changes after them: if the basic workup is normal and there are no red flags, the probability of an organic cause falls. If there are red flags, a normal result does not end the search and opens the way to the electroencephalogram, the MRI and the cerebrospinal fluid.

Cerebrospinal fluid on a large scale

The same group analyzed the cerebrospinal fluid of 992 patients, 456 with schizophrenia-like presentations and 536 with affective presentations (Endres and colleagues, 2020). Cell count was increased in 4 out of every 100, and 8 out of every 100 had signs of neuroinflammation. Antibodies against neuronal surface antigens appeared in the serum of 1 out of every 100 patients and in the cerebrospinal fluid of 0.3 out of every 100.

Among the patients with antibodies, 92 out of every 100 had at least one sign of brain involvement on other tests (Endres and colleagues, 2020). The practical conclusion is that classic antibodies are rare, but when they appear, they are rarely alone.

Structural imaging in young patients

A Portuguese study reviewed the CT and MRI scans of 32 patients aged 18 to 48 admitted with a first psychotic episode (Coentre and colleagues, 2016). None had an imaging change responsible for the symptoms, and 37 out of every 100 had incidental findings unrelated to the psychosis, such as arachnoid cyst and falx calcification, with no need for referral.

The authors conclude that structural imaging is especially useful in patients with neurological symptoms, an atypical presentation and older age (Coentre and colleagues, 2016). The result speaks to the risk of overdiagnosis: tests without a clear indication produce findings that do not change management.

The open debate

There is a real tension between the studies. The Freiburg group advocates broad investigation and finds frequent changes, many of them nonspecific (Endres et al., 2020). The Swedish and British group, in patients without neurological signs, found no cases that justified routine antibody testing (Theorell et al., 2021). The most prudent synthesis is that red flags define who needs extended investigation, and that is where diagnostic uncertainty must be acknowledged and recorded.

Building the reasoning step by step

The reasoning in the face of a first psychotic episode can be summed up in a sequence.

1. Take the history with the family: time of onset, course and symptoms beyond psychosis.
2. Ask about substances, medications and known medical conditions.
3. Perform a complete physical and neurological examination, with attention to consciousness, cognition and movements.
4. Look for red flags: abrupt onset, polymorphic presentation, seizures, catatonia, dyskinesia, reduced consciousness, autonomic dysfunction or severe reaction to antipsychotics.
5. Without red flags, proceed with basic investigation and reassess if the picture changes.
6. With red flags, extend to electroencephalogram, MRI and CSF, with antibody testing in serum and CSF.
7. Record the hypothesis and what remains to be ruled out in the medical record.

Back to the patient in the case

The 24-year-old man has psychosis with onset over two weeks, fluctuating disorientation and a probable seizure event. By the criteria described in the Freiburg review, he already meets the definition of possible autoimmune psychosis: abrupt onset, in less than 3 months, with a new seizure and altered consciousness (Endres et al., 2020).

In this scenario, extended investigation is justified: electroencephalogram, MRI, CSF and antibody testing. A normal MRI does not end the reasoning (Najjar et al., 2018), and neither does a positive toxicology screen. The diagnosis of schizophrenia may even be the final outcome, but only after treatable causes have been ruled out.

Talking with the family about the investigation

The family usually arrives with a direct question: is it schizophrenia? Explaining that a first episode requires ruling out treatable causes helps build trust and reduces anxiety about tests such as lumbar puncture. It is worth stating clearly what each test looks for and what happens if the result comes back normal.

This conversation also protects the patient from an early label. When the team records that the hypothesis of organic disease was considered and investigated, the next professional knows what has already been done and what remains. If new signs appear after discharge, the family knows they need to return.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by searching for the ICD-10 code.

The limits of using the game are described on the NEXO economy page. If you research the teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information in curation and in creators, and institutions can learn about the proposal on the partnerships page. Training clinical decision-making and searching for the right ICD-10 code complete this practice.

Frequently asked questions

Does every first psychotic episode require a lumbar puncture?

No. CSF gains value when there are red flags, such as abrupt onset, seizures, catatonia, reduced consciousness or autonomic dysfunction (Endres et al., 2020). In patients without these signs, antibody testing had a very low yield (Theorell et al., 2021).

Does a normal MRI rule out autoimmune encephalitis?

It does not. In anti-NMDA encephalitis, the MRI is normal at initial presentation in most patients (Najjar et al., 2018). The electroencephalogram and CSF can show what imaging does not.

Does anti-NMDA encephalitis always begin with neurological symptoms?

No. In the Chinese study, 62 of the 108 patients started with psychiatric symptoms (Wang et al., 2019). Neurological symptoms usually appear later, which is why the first appointment is often psychiatric.

If the toxicology screen is positive, does the investigation end?

No. Substance-induced psychosis can progress to a first episode of primary psychosis in a significant share of patients over the years (Piao et al., 2026). Drug use also does not rule out an associated organic cause.

Is it worth ordering every test for every patient?

The evidence is not unanimous. A broad protocol found many abnormalities, a good portion of them nonspecific (Endres et al., 2020), and imaging in young patients without neurological signs rarely showed the cause (Coentre et al., 2016). Red flags help decide who needs extended investigation.

Take-home messages

• The first psychotic episode is the best moment to look for an organic cause.
• Abrupt onset, an atypical picture and neurological or clinical signs greatly increase the likelihood of a secondary cause.
• Anti-NMDA encephalitis often begins with psychiatric symptoms.
• A normal MRI does not rule out autoimmune encephalitis.
• Without red flags, testing everyone for antibodies yields little.
• Substance-induced psychosis requires follow-up, because it can progress to primary psychotic illness.

References

1. Coentre R, Silva-Dos-Santos A, Talina MC. Retrospective study on structural neuroimaging in first-episode psychosis. PeerJ. 2016. PMID 27257547.
2. Cullen AE, Holmes S, Pollak TA, et al. Associations Between Non-neurological Autoimmune Disorders and Psychosis: A Meta-analysis. Biol Psychiatry. 2019. PMID 30122288.
3. Ellul P, Groc L, Tamouza R, et al. The Clinical Challenge of Autoimmune Psychosis: Learning from Anti-NMDA Receptor Autoantibodies. Front Psychiatry. 2017. PMID 28469581.
4. Endres D, Leypoldt F, Bechter K, et al. Autoimmune encephalitis as a differential diagnosis of schizophreniform psychosis: clinical symptomatology, pathophysiology, diagnostic approach, and therapeutic considerations. Eur Arch Psychiatry Clin Neurosci. 2020. PMID 32166503.
5. Endres D, Matysik M, Feige B, et al. Diagnosing Organic Causes of Schizophrenia Spectrum Disorders: Findings from a One-Year Cohort of the Freiburg Diagnostic Protocol in Psychosis (FDPP). Diagnostics (Basel). 2020. PMID 32937787.
6. Endres D, Meixensberger S, Dersch R, et al. Cerebrospinal fluid, antineuronal autoantibody, EEG, and MRI findings from 992 patients with schizophreniform and affective psychosis. Transl Psychiatry. 2020. PMID 32782247.
7. Endres D, Rauer S, Kern W, et al. Psychiatric Presentation of Anti-NMDA Receptor Encephalitis. Front Neurol. 2019. PMID 31749755.
8. Jeppesen R, Benros ME. Autoimmune Diseases and Psychotic Disorders. Front Psychiatry. 2019. PMID 30949074.
9. Najjar S, Steiner J, Najjar A, et al. A clinical approach to new-onset psychosis associated with immune dysregulation: the concept of autoimmune psychosis. J Neuroinflammation. 2018. PMID 29433523.
10. Piao YH, Le TH, Li L, et al. Predictors of conversion to psychosis and mortality in first-episode substance-induced psychosis: a nationwide register-based study in South Korea. Schizophrenia (Heidelb). 2026. PMID 42091914.
11. Theorell J, Ramberger M, Harrison R, et al. Screening for pathogenic neuronal autoantibodies in serum and CSF of patients with first-episode psychosis. Transl Psychiatry. 2021. PMID 34741015.
12. Wang W, Zhang L, Chi XS, et al. Psychiatric Symptoms of Patients With Anti-NMDA Receptor Encephalitis. Front Neurol. 2019. PMID 32038450.