Schizophrenia Research Forum - A Catalyst for Creative Thinking

Forum Discussion: Is Active Psychosis Neurotoxic?

View article

In our Forum discussion “journal club” series, the editors of Schizophrenia Bulletin or Schizophrenia Research provide access to the full text of a new article. A short introduction by a journal editor, below, gets us started, and then it's up to our readers to share their ideas and insights, questions and reactions to the selected paper. So read on…

View Comments By:
Robert McClure — Posted 10 November 2006
Karl-Ludvig Reichelt — Posted 22 November 2006
Mariano Scolari — Posted 27 November 2006

Background Text
By Gunvant Thaker, Maryland Psychiatric Research Center, and Associate Editor, Schizophrenia Bulletin

In the latest issue of Schizophrenia Bulletin, Tom McGlashan raised an important question: Is active psychosis neurotoxic? (See McGlashan, 2006.) This has been a controversial issue for some time in our field and has tremendous implications for how we focus our treatment in our patients and our research activities to identify novel treatment strategies. The hypothesis originated based on an interesting paper by Wyatt on the effects of antipsychotic medications on the natural course of schizophrenia (Wyatt, 1991). Subsequent early intervention and first episode studies further supported the hypothesis. However, most of this supporting evidence is derived from correlational analyses where causal effects are impossible to infer, or from uncontrolled and/or observational studies where random assignment to different treatments didn’t occur. Tom McGlashan further points out that several predictions based on the neurotoxic hypothesis are not validated by the existing data. Controlled clinical trials of schizophrenia that used random assignment and placebo arm can be informative in this regard. John Bola recently reviewed these studies and concluded that there was no evidence of long-term harm to individuals whose psychosis remained untreated during the double-blind trial (Bola, 2005; SRF related news story).

The onset of schizophrenia is defined by the onset of psychotic symptoms (“first break”), whereas impairments in other domains such as cognitive and neurophysiological function, as well as social and occupational deterioration and negative symptoms, emerge at different times. Whether treatment of one domain of symptoms/impairments affects the emergence or the course of the other domains remains unknown. This question is as much relevant for identifying treatments for the neurophysiological and cognitive deficits of schizophrenia that frequently precede psychosis as it is for the early and aggressive treatment of psychosis.

Comments on Online Discussion
Comment by:  Robert McClure (Disclosure)
Submitted 10 November 2006
Posted 10 November 2006

In 2001, in an Archives of General Psychiatry paper, Danny Mathalon and colleagues expressed concern that the neurodevelopmental hypothesis was threatening to overshadow the neurodegenerative hypothesis of schizophrenia (Mathalon et al., 2001). They made this statement while publishing their MRI results, demonstrating longitudinal increases in cerebrospinal fluid and decreases in regional brain volumes in schizophrenic subjects, measured with MRI. Their findings were consistent with other MRI studies showing progressive longitudinal brain volume change in schizophrenia. Although the paper was excellent overall, and very well-executed, I took issue with their interpretation of the findings (Weinberger and McClure, 2002). Specifically, we disagreed with the idea that quantitative measures of CSF space and regional brain volume with MRI could constitute evidence that brain tissue has degenerated. Danny Weinberger and I also stated that in order for neurodegeneration to be present, it should be observed at the tissue level. The question of what constitutes adequate evidence for the presence of neurodegeneration in the schizophrenic brain remains an open one. McGlashan addresses a closely related question—is active psychosis neurotoxic?—in his recent paper.

The current neurodegenerative hypothesis (see McClure and Lieberman, 2004) postulates that in the context of developmental maturation, abnormal genetic and/or epigenetic events lead to a limited neurodegenerative process that is associated with the expression of the symptoms and clinical course of schizophrenia. The suggested evidence for the existence of limited degenerative neuropathology (and its corollary, the presence of neurotoxicity) in schizophrenia includes (1) the resemblance of schizophrenia to other neurodegenerative illnesses; (2) the progressive clinical course; (3) the effects of treatment on the symptoms and clinical course; (4) the emergence and progression of cognitive dysfunction; (5) the association of more prevalent and severe brain pathomorphology with disease chronicity and severity found in neuroimaging studies of patients; (6) findings of the progression of brain pathomorphology in association with schizophrenia’s clinical course; (7) the presence of subtle abnormalities on postmortem studies of patients; and (8) the consistency of current pathophysiological theories of schizophrenia with a limited neurodegenerative process.

McGlashan uses his knowledge of the schizophrenia literature to provide a seasoned perspective on how clinical observations can instruct the scientific community about psychosis and neurotoxicity. Weinberger and I addressed some of the same points in our paper. With respect to (2), the progression of clinical aspects of the syndrome is certainly evident, but, as McGlashan notes, progression does not generally continue throughout the course of the illness, as would be expected if psychosis induced (or were caused by) neurotoxicity. With respect to (3), progression of clinical aspects of the syndrome are reduced with antipsychotic treatment. McGlashan emphasizes that antipsychotic treatment may allow patients to simply reengage with the world by reducing symptoms and modifying brain circuitry through reversible/plastic processes, rather than halting an irreversible/neurotoxic process. With respect to (4) the emergence and progression of cognitive dysfunction, Rund (1998) has noted that longitudinal studies of cognitive function do not suggest that progression continues, at least beyond the first 20 or so years of illness. McGlashan notes that alternative hypotheses must be tested before we can conclude that the progression observed results from neurotoxicity. With respect to (5), the association of more prevalent and severe brain pathomorphology with disease chronicity and severity found in neuroimaging studies of patients and (6), findings of the progression of brain pathomorphology in association with schizophrenia’s clinical course, Weinberger and I noted that the results of longitudinal MRI volume measurements vary from study to study; no two studies show the same pattern of changes across all of these measures; each study appears to have a unique combination of results; correlations between changes in symptoms and in measurements reported in some studies vary considerably from one study to another; and the magnitude of the progression of volume changes are too large to constitute the changes observed with MRI later in the disease. With respect to (7), neither McGlashan nor Weinberger and I found the postmortem change in schizophrenia consistent with a neurotoxic or neurodegenerative process.

McGlashan notes that the concept of psychosis being neurotoxic is not implausible, as suggested by (1) and (8). Jarskog and Lieberman (2006) have recently noted that though the underlying mechanisms of gray matter loss in schizophrenia remain uncertain, evidence of improved outcomes with antipsychotic treatment suggests that neuroprotection—the maintenance of the functional integrity of the brain in response to neurobiological stress—in schizophrenia is also possible. Opinions are likely to continue to differ, hopefully prompting constructive debate among investigators. Concepts of neurodevelopment, neurodegeneration, neurotoxicity, and neuroprotection all have heuristic value; may increase understanding of the neurobiology of schizophrenia; can stimulate research directed towards finding its etiology; or hamper communication within the scientific community if used in the context of rhetorical excess. The ultimate value of these concepts will rest on their ability to stimulate scientific inquiry that leads to discovery of the etiology of schizophrenia.


Mathalon D, Sullivan EV, Lim KO, Pfefferbaum A,. Progressive brain volume changes and the clinical course of schizophrenia in men. Archives General Psychiatry. 2001;58:148-157. Abstract

Weinberger DR, McClure RK. Neurotoxicity, neuroplasticity and magnetic resonance imaging morphometry. What is happening in the schizophrenic brain? Arch Gen Psychiatry 2002:553-558. Abstract

Jarskog LF, Lieberman JA. Neuroprotection in Schizophrenia., J Clin Psychiatry. 2006 Sep;67(9) Abstract

McClure RK and Lieberman JA. Neurodevelopmental and neurodegenerative hypothesise of schizophrenia: a review and critique. (2004) Current Opinion in Psychiatry: 16 (Supplement 2) S15-S28.

Rund B. A review of longitudinal studies of cognitive functions in schizophrenia patients. Schizophrenia Bulletin. 1998;24:425-435. Abstract

View all comments by Robert McClureComment by:  Karl-Ludvig Reichelt (Disclosure)
Submitted 17 November 2006
Posted 22 November 2006

Most of the evidence collected shows a considerable genetic disposition for schizophrenia (Onstad et al., 1991; Kendler and Diehl, 1993). This means there must be chemical changes secondary to some protein deviations. (If not, it is not genetic at all.) It is therefore not the psychosis which is toxic, but the psychosis is due to toxic/damaging substances affecting the brain. The increase in opioid receptor binding compounds of peptide nature (opioids) can explain this (Lindstrom et al., 1986; Reichelt et al., 1996), not the least because opioids induce Fos antigen in many of the key nuclei of the brain (Sun et al., 1999). Opioids have been shown to be trophic substances in brain development (Zagon and McLaughlin, 1987; Hauser et al., 1989), and in schizophrenia we believe that these are exorphins (Cade et al., 2000). In postpartum psychosis the exorphin is probably casomorphine (Lindstrom et al., 1984).

The most prominent changes in brain morphology are apparently seen in the first years after the diagnosis has been made (Madsen et al., 1999). See especially Haug (1982), who followed his patients over 12 years and found progressive changes using ventriculography and doing the analysis himself to make sure it was the same observer (Haug, 1982). There is no reason why food-derived exorphins should not affect the fetus and child and induce prodromal states extensively discussed elsewhere (Reichelt et al., 1996). Describing a symptom (psychosis) as toxic does not make sense to me.


Onstad S, Skre I, Torgersen S, Kringlen E. Twin concordance for DSM-III-R schizophrenia. Acta Psychiatr Scand. 1991 May;83(5):395-401. Abstract

Kendler KS, Diehl SR. The genetics of schizophrenia: a current, genetic-epidemiologic perspective. Schizophr Bull. 1993;19(2):261-85. Review. Abstract

Lindstrom LH, Besev G, Gunne LM, Terenius L. CSF levels of receptor-active endorphins in schizophrenic patients: correlations with symptomatology and monoamine metabolites. Psychiatry Res. 1986 Oct;19(2):93-100. Abstract

Reichelt KL, Seim AR, Reichelt WH. Could schizophrenia be reasonably explained by Dohan's hypothesis on genetic interaction with a dietary peptide overload? Prog Neuropsychopharmacol Biol Psychiatry. 1996 Oct;20(7):1083-114. Review. Abstract

Sun Z et al. (1999) Autism 3: 67-83.

Zagon IS, McLaughlin PJ. Endogenous opioid systems regulate cell proliferation in the developing rat brain. Brain Res. 1987 May 26;412(1):68-72. Abstract

Hauser KF, McLaughlin PJ, Zagon IS. Endogenous opioid systems and the regulation of dendritic growth and spine formation. J Comp Neurol. 1989 Mar 1;281(1):13-22. Abstract

Cade RJ et al. (2000) Nutr neurosci. 2: 57-72

Lindstrom LH, Nyberg F, Terenius L, Bauer K, Besev G, Gunne LM, Lyrenas S, Willdeck-Lund G, Lindberg B. CSF and plasma beta-casomorphin-like opioid peptides in postpartum psychosis. Am J Psychiatry. 1984 Sep;141(9):1059-66. Abstract

Madsen AL, Karle A, Rubin P, Cortsen M, Andersen HS, Hemmingsen R. Progressive atrophy of the frontal lobes in first-episode schizophrenia: interaction with clinical course and neuroleptic treatment. Acta Psychiatr Scand. 1999 Nov;100(5):367-74. Abstract

Haug JO. Pneumoencephalographic evidence of brain atrophy in acute and chronic schizophrenic patients. Acta Psychiatr Scand. 1982 Nov;66(5):374-83. Abstract

View all comments by Karl-Ludvig ReicheltComment by:  Mariano Scolari
Submitted 24 November 2006
Posted 27 November 2006

If active psychosis is neurotoxic, it is very important to find out the molecular mechanisms that promote it, because they would provide new therapeutic strategies for the treatment of schizophrenia. The question that arises is: What distinguishes a neurotoxic disease from a neurodegenerative one? Is it the processes implicated? The number of neurons lost with time? Maybe one could consider neurotoxicity as an acute process and neurodegeneration as a chronic event.

In light of the data reviewed in this interesting paper, it is probable that a change of terminology will be necessary. Consider the following fragments in the text: “Neuropathology, like longitudinal course, does not support the hypothesis that untreated psychosis is neurotoxic” and “…the theory assumes that reduced connectivity precedes symptom formation and is generative of characteristic symptoms.” These statements establish the necessity of distinguishing neurotoxicity and the loss of connectivity or synaptic contacts. Although both processes imply structural changes, in the case of neurotoxicity, these changes have a wider extension. For these reasons, and considering the lines cited above, it is possible that active psychosis would be a “synaptodegenerative” process and not a neurotoxic event.

View all comments by Mariano Scolari