Skip to content


Annals of General Psychiatry

Open Access

Progressive brain volume changes in the first year of illness predict five-year outcome of schizophrenia

  • W Cahn1,
  • NEM van Haren1,
  • H Pol1,
  • HG Schnack1,
  • E Caspers1,
  • DAJ Laponder1 and
  • RS Kahn1
Annals of General Hospital Psychiatry20032(Suppl 1):S2

Received: 1 November 2003

Published: 23 December 2003


SchizophreniaGray MatterNegative SymptomGray Matter VolumeTotal Brain


In first-episode schizophrenia progressive brain volume changes have been found after the first year of the illness. This study examined associations between early progressive brain volume changes and five-year outcome in schizophrenia.


Thirty-one first-episode patients with schizophrenia were included in this longitudinal study. MRI brain scans were obtained at inclusion (T0) and after 1-year (T1). Intracranial, total brain, cerebral gray and white matter, cerebellar, lateral and third ventricle volumes were measured. After a follow-up period of 5.3 (SD = 0.8) years global outcome was measured with the Camberwell Assessment of Need (CAN) and clinical outcome was measured with the Positive and Negative Syndrome Scale (PANSS). To examine associations between early progressive brain volume changes (T1 minus T0) and five-year outcome Pearson product-moment correlations were performed with intracranial volume and age as covariates.


Total brain volume decreases over the first year correlated significantly with negative symptoms (r = -0.38, df = 27, p = 0.04) at five year follow-up. Gray matter volume decreases correlated significantly with positive symptoms (r = -0.40, df = 27, p = 0.03) and negative symptoms (r = -0.54, df = 27, p = 0.002). Lateral ventricle volume increases correlated significantly with the total score of the CAN (r = 0.54, df = 27, p = 0.003).


These findings suggest that medium-term symptomatic and global outcome is predicted respectively by early gray matter loss and lateral ventricular enlargement. It furthermore underscores the importance examining dynamic rather than static changes in brain structures in relation to predicting outcome of schizophrenia.

Authors’ Affiliations

University of Utrecht, Utrecht, The Netherlands


© The Author(s) 2003