In both healthy controls and people with schizophrenia, physical activity influences white matter integrity. And, in patients, physical training improves PANNS scores. Improvement was seen particularly in positive symptoms, Hilleke Hulshoff Pol (University Medical Centre, Utrecht, the Netherlands) told the symposium on the neurobiological impact of exercise.

The TOPFIT study involved 33 people with schizophrenia and 48 normal controls – all of whom were physically inactive – randomised to an exercise programme or life as usual.

Tracking fibre tracts

The six-month endurance training programme consisted of an hour’s exercise twice weekly using a stationary bicycle and a choice of either a treadmill or rowing, plus anaerobic exercise with weights. To enter the analysis, subjects had to be at least 50% compliant with the programme. Patients and healthy controls were matched for standard demographic variables.

Fibre tracking data in healthy controls showed a slight decline in brain connectivity between baseline and six months. Among patients, who started from a lower baseline than controls, there was a significant increase in white matter fibres. This was especially evident in the corticospinal tract, which relates to motor function.

Both groups showed increased cardiovascular fitness. Neither experienced a change in BMI.

Motivating patients a key challenge

In response to a question from the audience, Dr Pol agreed that motivating people with schizophrenia to undertake such an exercise programme would prove a challenge: there were certainly patients who entered the study but did not comply and others who declined to take part.

The potential benefits of a shorter endurance training programme have been investigated by Berend Malchow (University of Munich, Germany) and colleagues in Munich and Göttingen. Twentyfive people with schizophrenia and 27 healthy controls were assigned six weeks of exercise training (three thirty-minute sessions per week) followed by six weeks of cognitive training.

An active control group of 26 people with schizophrenia were assigned to similar periods of table football followed by cognitive training. Patients had had several episodes of schizophrenia. Both they and the healthy controls were previously physically inactive.

Table football increased volume in the motor and anterior cingulate cortex

Whole brain morphometry

Healthy controls at baseline had a higher work output on the bicycle ergometer than patients. But, after training, patients exceeded the starting work capacity of the healthy controls. Table football produced no increase in exercise capacity.

The combined exercise and cognition programme led to improvement on the Global Assessment of Functioning scale (with better social adjustment and leisure and work activities) and improvement on some cognitive subdomains – which did not occur with table football.

Whole brain morphometry showed some subtle changes in grey matter, with increased volume in the temporal gyri compared to baseline in patients. Table soccer increased volume in the motor and anterior cingulate cortex.

These changes disappeared after three months with no further training, and only two patients continued to exercise of their own accord – despite many saying that they appreciated the benefit of greater fitness on everyday activities such as climbing the stairs.

There were no changes in hippocampal volume.

Delayed development, later degeneration

The North Finland Birth Cohort study, which has been following more than twelve thousand individuals born in 1966, continues to deliver thought-provoking information about risk factors for schizophrenia. On average, children who developed schizophrenia in adult life learned to stand or walk 2-3 weeks later than children who did not develop the disease during fifty years of follow-up, Matti Isohanni (University of Oulu, Finland) told the meeting.

Children who developed schizophrenia as adults learned to stand or walk 2-3 weeks later than those free of the disease

This suggests subtle pre-morbid developmental delay.

In people who subsequently suffer from schizophrenia, motor development at the age of one predicts motor development at the age of sixteen. This association is not found in those who remain free of the disease: late developers seem to have the opportunity to catch up. But, in children who later have schizophrenia, it is as if this is prevented by absence of plasticity in developmental pathways mediating motor function.

There are two further intriguing findings. In the schizophrenia subgroup, motor development at age one predicts cognitive deterioration in adulthood. It is as if abnormal neurodevelopment is in some way linked to later neurodegeneration.

Secondly, having good school grades in physical exercise at age 16 is associated with a 20% reduction in the risk of developing schizophrenia by middle age.