Elsevier

Schizophrenia Research

Volume 193, March 2018, Pages 418-422
Schizophrenia Research

The association between cardio-respiratory fitness and cognition in schizophrenia

https://doi.org/10.1016/j.schres.2017.07.015Get rights and content

Abstract

Objective

Schizophrenia is associated with reduced cardio-respiratory fitness (CRF), and impaired cognition is a core feature of the disorder. Despite their particular significance to schizophrenia disparately, the relationship between these two variables has not yet been thoroughly assessed. In this study we aimed to investigate naturally occurring associations between CRF and all cognitive domains within this patient population.

Method

Eighty outpatients with schizophrenia spectrum disorders participated in the study. Neurocognition was assessed with the Wechsler Adult Intelligence Scale version 4 General Ability Index (WAIS GAI) and the MATRICS Consensus Cognitive Battery (MCCB). Oxygen uptake was measured directly by analyzing O2 and CO2 content in expired air during a maximum exercise session on a treadmill using a modified Balke protocol. Clinical symptom load was assessed with the Positive and Negative Syndrome Scale (PANSS). Hierarchical multiple regression analyses were conducted, controlling for sex and age, and negative psychotic symptom levels.

Results

CRF explained a significant 8.2% and 9.1% of the variance in general intellectual ability and state-sensitive cognitive functioning respectively, beyond the impact of negative psychotic symptom load.

Conclusion

The study indicates a direct relation between CRF and cognition in schizophrenia. Impaired cognition is a difficult-to-treat expression of the disorder, and identifying modifiable factors possibly mediating cognition, such as CRF, is of great clinical value.

Introduction

At group level, schizophrenia patients display profoundly low cardio-respiratory fitness (CRF; Scheewe et al., 2012, Heggelund et al., 2011, Strassnig et al., 2011). Low CRF increases the risk of cardiovascular disease (CVD; Laursen et al., 2014, Wildgust and Beary, 2010), which is the largest single cause of death in schizophrenia (Hennekens et al., 2005). Moreover, impaired cognition is a central clinical feature in schizophrenia (Dickinson et al., 2008). The naturally occurring associations between CRF and cognitive functions have not been thoroughly investigated in this particular patient group.

Several lines of research have indicated an association between CRF and cognition in the general population (Voss et al., 2011, Etnier et al., 2006). In a large-scale cohort study on young men aged 15 to 18 years by Åberg et al. (2009), small to moderate correlations were found between CRF and all measured cognitive functions, including full-scale IQ. In a review including all age groups and both cross-sectional, cohort, and intervention studies, Hötting and Röder (2013) concluded that physical exercise is positively associated with cognitive functioning, and that the effects are specific to both type of exercise and age group. Evidence is scarcer on the relationship between CRF and cognition in schizophrenia patients. Kimhy et al. (2014) investigated naturally occurring associations between CRF and the subset of cognitive functions included in the MATRICS Consensus Cognitive Battery (MCCB) in a sample of 32 schizophrenia patients. Positive correlations were reported for the specific domains of executive functioning, working memory, processing speed and social cognition, and CRF was found to explain a substantial 22% of the variance in the overall MCCB composite score. The authors conveyed the findings as preliminary and called for replication in a larger sample. Several exercise intervention studies have also shown positive effects from enhanced CRF on particular cognitive domains in schizophrenia. A recent meta-analysis by Firth et al. (2016) included 10 studies, whereof six reported positive changes in one or more cognitive domains. The meta-analysis was conducted on a total of 186 participants in CRF-enhancing exercise activities and 199 control subjects, and revealed that exercise interventions significantly improved overall cognitive performance. The authors concluded that the effect was robust, with an effect size of g = 0.33 and low statistical heterogeneity between studies. Regarding specific domains, significant effects were found for working memory, attention/vigilance, and social cognition.

Cognitive impairment is deemed a core feature of schizophrenia, albeit its connections to the most prominent clinical features continue to be under investigation and debate (Bagney et al., 2015, Ventura et al., 2010, Reichenberg and Harvey, 2007). The impairment has been conveyed as global, although with differences in magnitude on the functions affected, and large individual variations in degree and profile of impairment (Dickinson et al., 2008, Reichenberg and Harvey, 2007, Weickert et al., 2000). It may be that these and other disorder-related circumstances elicit a different relationship between CRF and cognition in schizophrenia, compared to the general population. This may occur in at least two different ways, paved in opposite directions. Firstly, negative psychotic symptoms may compromise the participants' engagement and lead to subability performance in both the physical and neurocognitive testing situations. This may inflate the association between CRF and cognition, and thus create an artificial relationship between the two variables in a given sample. This risk may be countered through controlling for negative psychotic symptom load. Secondly, the direct expression of the disorder in impaired cognition may overshadow any associations to other factors such as CRF, and thus obscure a true relationship between the two variables. Furthermore, the association with CRF may differ between types of cognitive functions. The cognitive impairments in schizophrenia appear to be most pronounced in memory, attention, executive functions, and processing speed (Dickinson et al., 2008, Reichenberg and Harvey, 2007, Nuechterlein et al., 2004). Several other lines of evidence suggest that these functions are more responsive to changes in psychological and physiological state, as compared to the functions that comprise general intellectual ability (IQ). Heritability estimates are lower for the former (Knopik et al., 2013); the functions of verbal learning, memory and psychomotor speed appear more prone to be temporarily negatively influenced by other mental illnesses such as depression (Douglas and Porter, 2009); and in the described intervention studies the effect of improved CRF appeared limited to, or more pronounced in, the domains of working memory, attention/vigilance, processing speed, and visual and verbal learning (Firth et al., 2016). Thus, these functions may be more prone to the influence of factors such as CRF. For this reason, and for the sake of simplicity, we henceforth refer to memory, attention, executive functions, and processing speed as state-sensitive cognitive functions, as compared to the core IQ functions of verbal comprehension and perceptual reasoning.

Based on these theoretical implications, we expected that the naturally occurring positive association between CRF and IQ present in the general population would be occluded by the disorder-related impact on cognition in schizophrenia. However, we expected that the positive association between CRF and the state-sensitive cognitive functions would be retained. Specifically, we predicted that CRF as measured by VO2peak would explain a significant amount of the variance in the state-sensitive cognitive functions only, and that this influence would be present beyond the impact of negative psychotic symptom load.

Section snippets

Design

The study was conducted on baseline data from the randomized, controlled, observer-blinded clinical trial ‘Effects of Physical Activity in Psychosis’ (EPHAPS) (ClinicalTrials.gov, registration number NCT02205684; Engh et al., 2015).

Participants

Eighty participants aged 20–67 years were recruited from August 2014 through February 2017 from catchment area-based and publicly funded outpatient psychiatric clinics in Vestfold County, Norway. A subgroup of patients was referred from primary health care to the

Results

Basic descriptive statistics and standardized regression coefficients are displayed in Table 2. After entering age and gender into the equation, the target predictor variable of VO2peak explained 8.2% of the variance in WAIS GAI and 9.1% of the variance in MCCB NCS. Entering the PANSS measure into the analyses slightly reduced the unique contribution of VO2peak on cognition, but significant results were retained. PANSS negative factor scores independently explained 8.2% of the variance in the

Discussion

We predicted that CRF would explain a significant amount of the variance in the cognitive functions defined as state-sensitive, and that this influence would appear beyond the impact of negative psychotic symptom levels, in this sample of schizophrenia patients. These assumptions were confirmed. Furthermore, we predicted that CRF would not contribute significantly to the variance in IQ, due to the confounding effect of the disorder-related cognitive impairments in combination with these

Funding

This work was supported by Vestfold Hospital Trust, Division of Mental Illness and Drug Addiction and Vestfold Hospital Trust, Research and Development (grant number 2013/842), Norwegian Extra Foundation for Health and Rehabilitation through EXTRA funds (grant number 2013/2/0183), Norwegian Research Network in Severe Mental Illness (NORSMI; grant number 39383/2011–13) and Torgeir Lindvig's Trust (grant number 2013).

Contributors

All authors contributed to the manuscript in accordance with the Vancouver Protocol.

Conflict of interest

The authors declare that they have no competing interests.

Acknowledgments

The authors would like to give thanks to the patients participating in the study. We owe many thanks to Gry Bang-Kittilsen for contributing to the recruitment of patients and clinical testing, and to Ole-Jakob Bredrup, Jan Freddy Hovland, Bjørn Einar Oscarsen, Helge Bjune, Merete Rønningen Bergstad, Ellen Gurine Faervik, Camilla Lahn-Johannessen, and Siri Øyhus for contributing to the recruitment of patients and conducting the VO2peak tests.

References (36)

  • J. Ventura et al.

    Disorganization and reality distortion in schizophrenia: a meta-analysis of the relationship between positive symptoms and neurocognitive deficits

    Schizophr. Res.

    (2010)
  • M.W. Voss et al.

    Aerobic fitness is associated with greater efficiency of the network underlying cognitive control in preadolescent children

    Neuroscience

    (2011)
  • R.S. Wallwork et al.

    Searching for a consensus five-factor model of the Positive and Negative Syndrome Scale for schizophrenia

    Schizophr. Res.

    (2012)
  • M.A. Åberg et al.

    Cardiovascular fitness is associated with cognition in young adulthood

    Proc. Natl. Acad. Sci. U. S. A.

    (2009)
  • B. Balke et al.

    An experimental study of physical fitness of air force personnel

    US Armed Forces Med. J.

    (1959)
  • C. Bouchard et al.

    Individual differences in response to regular physical activity

    Med. Sci. Sports Exerc.

    (2001)
  • C. Bouchard et al.

    Familial aggregation of VO2max response to exercise training: results from the HERITAGE family study

    J. Appl. Physiol.

    (1999)
  • K.M. Douglas et al.

    Longitudinal assessment of neuropsychological function in major depression

    Aust. N C J. Psychiatry.

    (2009)
  • Cited by (11)

    • Meta-analysis of transcriptional regulatory networks for lipid metabolism in neural cells from schizophrenia patients based on an open-source intelligence approach

      2022, Neuroscience Research
      Citation Excerpt :

      The other two remaining loci showed no genetic association with smoking. Moreover, studies have shown that impaired cardio-respiratory fitness is associated with cognitive dysfunction (Filik et al., 2006; Holmen et al., 2018) and, in fact, patients with schizophrenia have a higher prevalence of angina and respiratory symptoms, and often have impaired lung function. It is also believed that there are genetic factors that increase the risk of both becoming a smoker and developing schizophrenia (Hartz et al., 2018).

    • Physical performance is more strongly associated with cognition in schizophrenia than psychiatric symptoms

      2019, European Psychiatry
      Citation Excerpt :

      Furthermore, there is a need to consider the association between cognitive function and disorder-related clinical symptoms in patients with schizophrenia. Both depression and psychotic symptoms appear to negatively affect memory and psychomotor speed [18,34]. In addition, health indicators such as total cholesterol levels and body mass index (BMI) may impact neurocognition in patients with schizophrenia [35,36].

    • Cardio-respiratory fitness is associated with a verbal factor across cognitive domains in schizophrenia

      2019, Schizophrenia Research
      Citation Excerpt :

      The cross-sectional design of the study leaves any propositions about causal pathways inferential. Several possible third-variable influences on the relationship between CRF and cognition in schizophrenia have been controlled for in a previous study (Holmen et al., 2018), but the possibility of other third variables contributing to the association, such as genetic factors, cannot be ruled out. Furthermore, in the present study we controlled for age and sex in all analyses, to account for general sex- and age-differences in CRF as measured by VO2peak.

    • Physical activity pattern and cardiorespiratory fitness in individuals with schizophrenia compared with a population-based sample

      2018, Schizophrenia Research
      Citation Excerpt :

      Blair (2009) noted that low CRF was a stronger predictor of mortality than any other risk factors (i.e., smoking, fat accumulation) (Blair, 2009). Furthermore, in people with schizophrenia, a low CRF level has been associated with negative symptoms (Vancampfort et al., 2015), cognitive impairment (Holmen et al., 2017), increased BMI (Vancampfort et al., 2015), and impaired daily functioning (Kimhy et al., 2014). The very low CRF levels reported in this study would therefore most likely not only increase the risk of developing lifestyle diseases, but also affect both disease-specific symptoms and the ability to move around without strain and discomfort.

    View all citing articles on Scopus
    View full text