Cortisol link with psychosis according to a study

antidepresseur-news-nov2008-BTOWNSVILLE (AUSTRALIA) – JCU Associate Professor Zoltan Sarnyai said it was the first meta-analysis study to compare the level of cortisol in a waking patient’s body with the stage of schizophrenia they are suffering. Dr Sarnyai said it means doctors may be able to eventually identify those who will develop full-blown psychosis from amongst those who present with early stages of the disease. “Only some 20 to 30 per cent of individuals who are at high-risk of developing psychosis due to their clinical presentation or family history actually do so. Identifying those people early is where the cortisol measurement comes in.
“Biomarkers are very few and far between in psychiatry, so even though a huge amount of work is still needed, this could become a valuable technique,” said Dr Sarnyai.

Researchers at the Psychiatric Neuroscience Laboratory at the Australian Institute of Tropical Health and Medicine (AITHM) at JCU, conducted a meta-analysis of 11 studies.

The resulting paper, published in Neuroscience & Biobehavioral Reviews, shows that patients have different levels of the stress hormone after awakening (Cortisol Awakening Response, CAR) relative to healthy controls.

Co-author of the study, JCU’s Dr Maximus Berger, said scientists had suspected cortisol had a role in psychotic disorders for a long time, but until now, some results had been contradictory.

“We were able to show that patients with psychosis fail to produce cortisol after they wake up in the morning. We found this even in patients with recent onset of the illness,” said Dr Berger.

The paper identified some evidence to suggest that high-risk individuals who later develop psychosis already have changes in cortisol before they develop the illness.

Dr Sarnyai said low CAR levels are also an indicator of risk for other chronic diseases and have been linked to systemic inflammation and changes in the gut flora — which meant there was the potential for early diagnosis and treatment of these conditions too.


Creapharma made an interview with Dr Zoltan Sarnyai:

Creapharma – In general, medicine orientates itself always more in the direction of (bio)markers especially in genetics to identify risks of diseases (i.e. breast cancer) but in psychiatry that seems much more seldom as you mentioned in the press release, why do you think that is like that?
ZS – You are right. There are very few, if any, biomarkers in psychiatry. This is not because psychiatric disorders are any less “biologically-based” then breast cancer. In breast cancer we can say, there is cancer, and then we can also say exactly what type of cancer there is and how it will respond to treatment, on the basis of genetic or histological markers. In psychiatry, we can say with high degree of certainty that there is a psychiatric disease present but we cannot say, with absolute certainty what type of psychiatric disease there is. In breast cancer we know the disease mechanisms, the pathways and molecules involved and we can specifically look for them. So we have good biomarkers. In psychosis/schizophrenia we do not have a detailed understanding of the mechanism of the disease, we do not know the pathways and molecules involved so we cannot really develop biomarkers easily that are specific to psychosis/schizophrenia (do not recognise other psychiatric disorders) and sensitive enough to pick up the difference between healthy individuals and the ones suffering from psychosis.

In the blood (serum) or saliva, what do you (or the physician) measure exactly, is it the cortisol or a metabolism of this hormone?
We used in our meta-analysis studies that measured cortisol from the saliva. This is easy to obtain, it does not hurt and can be repeated as frequently as you want. Then an immune-reaction based essay can be used to detect cortisol with high specificity. Cortisol in saliva represents the blood cortisol levels very well. In fact, most of the cortisol in the blood are in bound form, a situation when cortisol is attached to a large protein (cortisol binding globulin), which makes cortisol unable to penetrate the cell membrane, bind to its receptor and exert its effect. Cortisol in the saliva, on the other hand, is free, un-bound cortisol, so in a sense salivary cortisol level represents the cortisol that is biologically active in the body.

The important point is that we analysed studies in which cortisol was measured from the saliva right at the moment when the patient woke up and then again 30 min later. Normally, the second cortisol value (30 min after awakening) is higher than the first one. This is called the cortisol awakening response (CAR). This can be found normally in all of us. It is a typical hormonal response just like the increase in oxytocin level in women during breast feeding. It should always happen. If it does not happen, or it does not happen strongly enough, meaning that the increase of cortisol after 30 min of awakening is too small, then we may have a problem. That is what we identified in our meta-analysis. Individuals with psychosis had lower CAR, in other words their cortisol level barely increased after waking up. Other studies suggest that this morning rise of cortisol, the CAR, is necessary to prepare the body and the brain to deal with the pressures and stresses of the day to come.

According to your results, would you recommend that everybody should in the future measure its cortisol in the blood or only people with clinical presentation (of the symptoms) or family history of psychosis?
I would be hesitant to recommend any such things just at the moment. A lot more research will be needed, primarily using longitudinal studies, which is to follow the same group of individuals over the years and see how their CAR correlates with the development and the progression of psychosis. In this meta-analysis we demonstrated that lower CAR was present in individuals with different stages of psychosis. This does not yet mean that individuals with other psychiatric conditions cannot have abnormal CAR. We do not yet know that. So we cannot claim just right now that blunted CAR is diagnostically specific for psychosis. The longitudinal studies mentioned above could help to understand if individuals who are high risk to develop psychosis and have abnormal CAR are more likely to transition to psychosis. Again, this is something we do not know yet.

Do you think this lack of cortisol after awakening is the cause or the consequence of schizophrenia? Other said, can it be that the cause of schizophrenia might be a hormonal trouble at least by some patients?
This is a very good question! On the basis of our study we cannot tell. Cortisol is both the marker of stress in the body and a very powerful molecule that influence brain functions profoundly. Our study suggests that the blunted CAR is not due to the long-term antipsychotic treatment and to the chronic nature of the disease as it is present at the first psychotic episode. It may not be the results of the psychotic episode, which is no doubt very stressful, because it is still present in treated patients who are not psychotic any more. There is a hypothesis of schizophrenia which proposes that chronic stress, especially early in life, has an influence on brain development and neurotransmission and it may interact with the genetic factors involved to bring out the disease in the late teens, early twenties. It is not completely unlikely that with blunted CAR we see a scar on the hormonal system in response to previous events in life, or even before birth.

Some recent theories suggest that schizophrenia and other psychiatrist diseases like depression could have an inflammation origin, studies noticed that by about 40% of patients suffering from schizophrenia or other psychiatric diseases the C reactive protein or cytokine levels were in higher concentration compared with healthy individuals. Do you think these observations are or can be connected with the low Cortisol Awakening Response shown in your study?
You are absolutely right! The inflammatory component can be quite important. Indeed they can very well be linked. Again, we do not know if chronic, low grade inflammation that has been seen in schizophrenia influences the cortisol activity or the altered cortisol activity impacts on the inflammatory processes. Genetic polymorphisms that drive the expression of abnormal amount of C-reactive protein has been associated with lower CAR in a recent study, which makes the link between inflammation and CAR even stronger ( A recent study looking at the link between CAR and inflammatory/immune markers in first-onset patients that later responded or not responded to treatment has found Non-Responders had markedly lower CAR and higher IL-6 and IFN-γ levels when compared with Responders. After 12 weeks, Non-Responders show persistent lower CAR and higher IL-6 and IFN-γ when compared with Responders. This study recommends that cortisol and inflammatory biomarkers at the onset of psychosis should be considered as possible predictors of treatment response, as well as potential targets for the development of novel therapeutic agents. (

June 5, 2016. Source: Press release of the study. Interview carried out by Xavier Gruffat (Pharmacist from ETH Zurich, Switzerland) between June 5th and June 8th 2016.

Informations sur la rédaction de cet article et la date de la dernière modification: 10.01.2022