As Robin Murray mentioned in his article discussed in a previous post, a 1976 paper by Eve Johnstone, Tim Crow et al showing cerebral ventricular enlargement in chronic schizophrenia "kickstarted the huge and ongoing endeavour of neuroimaging in psychosis". This finding was seen as validating the "view that schizophrenia was a neurodegenerative disorder".
As I've mentioned before (see previous post), it was this paper that also produced the remarkable change in the position of E. Fuller Torrey from being Szaszian to a biomedical advocate of forced psychiatric treatment. From his point of view, psychiatry stopped "groping in the dark" from this 1976 point in time (see my book extract). This meant he no longer doubted that schizophrenia is a brain disease.
However, the general acceptance over recent years that antipsychotic medication can reduce brain volume (eg. see previous post) has led to a rethink, exemplified by Robin Murray's article. The cerebral atrophy demonstrated by Johnstone et al (1976) was almost certainly due to antipsychotic medication, at least mainly, and was not an indication that schizophrenia is a brain disease. Johnstone et al were wanting to make out that schizophrenia led to a dementing illness, as they found a significant association of increased ventricular size with measures of cognitive impairment. But any cognitive impairment in schizophrenia is functional not organic, so the whole basis of the paper was flawed. And, its misinterpretation has misled a whole generation of psychiatric researchers.
I've always argued that any difference in ventricular volume may be non-specific rather than due to schizophrenia (see my article). As with any statistical association, a causal connection is not necessarily implied. The brain is a dynamic, not static, organ and ventricular enlargement can change over time affected by confounding variables, such as nutrition and hydration. In fact, ventricular enlargement can also be found in other psychiatric conditions, such as bipolar disorder.
Furthermore, like Robin Murray, maybe I have underestimated the effects of antipsychotic medication. As I said in my book review of The Bitterest Pills, Jo Moncrieff "makes a stronger case than even I was aware of for ventricular enlargement in schizophrenia being a drug-induced phenomenon". Robin, even though he accepts that "high-dose antipsychotics contribute ... to the subsequent 'progressive' changes", still believes there are "subtle brain changes present at onset of schizophrenia". But Jo shows (p. 158) that the interpretation of the results of studies of antipsychotic-naive patients are not clear cut, with inconsistency about the area of the brain identified. Some studies have not reported any difference in global brain volumes in patients prior to them receiving antipsychotic drugs, even though they may have been unwell for many years.
There does, at least, seem to be a need for a proper review of studies of brain scans of patients with psychosis or schizophrenia prior to them receiving antipsychotic drugs. If the Johnstone et al (1976) study is really the reason for the modern resurgence of biomedical psychiatry, its conclusion that cerebral ventricles are enlarged in schizophrenia may be totally based on an artefact.