|Home | About | Journals | Submit | Contact Us | Français|
Peak oxygen uptake (VO2peak) is a strong predictor of cardiovascular disease (CVD) and all-cause mortality, but is inadequately described in patients with schizophrenia. The aim of this study was to evaluate treadmill VO2peak, CVD risk factors and quality of life (QOL) in patients with schizophrenia (ICD-10, F20-29).
33 patients, 22 men (33.7 ± 10.4 years) and 11 women (35.9 ± 11.5 years), were included. Patients VO2peak were compared with normative VO2peak in healthy individuals from the Nord-Trøndelag Health Study (HUNT). Risk factors were compared above and below the VO2peak thresholds; 44.2 and 35.1 ml·kg-1·min-1 in men and women, respectively.
VO2peak was 37.1 ± 9.2 ml·kg-1·min-1 in men with schizophrenia; 74 ± 19% of normative healthy men (p < 0.001). VO2peak was 35.6 ± 10.7 ml·kg-1·min-1 in women with schizophrenia; 89 ± 25% of normative healthy women (n.s.). Based on odds ratio patients were 28.3 (95% CI = 1.6-505.6) times more likely to have one or more CVD risk factors if they were below the VO2peak thresholds. VO2peak correlated with the SF-36 physical functioning (r = 0.58), general health (r = 0.53), vitality (r = 0.47), social function (r = 0.41) and physical component score (r = 0.51).
Men with schizophrenia have lower VO2peak than the general population. Patients with the lowest VO2peak have higher odds of having one or more risk factors for cardiovascular disease. VO2peak should be regarded as least as important as the conventional risk factors for CVD and evaluation of VO2peak should be incorporated in clinical practice.
Patients suffering from schizophrenia have a mortality risk that is two to three times that of the general population and the leading cause of death is cardiovascular disease (CVD) [1,2]. Although, multifactor causes have been identified, reduced cardiorespiratory fitness has probably been overlooked as a risk factor for CVD in patients with schizophrenia .
Cardiorespiratory fitness, measured as peak oxygen uptake (VO2peak) is a strong predictor of CVD and all-cause mortality [4,5]. Improvements in VO2peak have indicated reduced risk of CVD, coronary heart disease and all cause mortality . VO2peak is often a stronger predictor of mortality than conventional risk factors for CVD . McAuley and Blair  recently pointed out reduced cardiorespiratory fitness as a greater health threat than obesity and suggested that more emphasis should be put on increasing VO2peak. This might be especially important considering that higher levels of VO2peak seems to attenuate or eliminate the increased health risk associated with obesity . Findings from the epidemiological Nord-Trøndelag Health Study (the HUNT Study) demonstrate that physical active people with a clustering of cardiovascular risk factors appears to have comparable risk of premature death as inactive individuals without risk factors . In the same cohort men with VO2peak below 44.2 ml·kg-1·min-1 were eight times more likely to have a cluster of CVD risk factors, compared to men above 50.5 ml·kg-1·min-1 .
Results from the Aerobics Center Longitudional Study further suggest that people with low VO2peak is characterized by depressive symptoms and low emotional well being . High levels of VO2peak are associated with high levels of quality of life (QOL) . Body mass index (BMI) are found inversely related to QOL in patients with schizophrenia  but the relation between VO2peak and perceived QOL are not evaluated.
Objective measures of VO2peak have rarely been presented in patients with schizophrenia. The classical study by Carlson et al.  were the first to describe oxygen uptake in patients with schizophrenia, but many of their patients did not reach values close to maximal oxygen uptake. Our research group revealed significant changes in VO2peak after eight weeks of high aerobic intensity training in patients with schizophrenia . Recently, Strassnig et al.  published measures of oxygen uptake in 117 patients with schizophrenia that were exceedingly low (4.4 metabolic equivalents ≈ 15.4 ml·kg-1·min-1). This VO2peak value are much lower than the VO2 required for walking in patients with schizophrenia , and at a level that may indicate a need for heart transplant in heart failure patients .
The primary aim of this study was to evaluate objectively measured VO2peak during walking or running in men and women with schizophrenia compared to VO2peak in healthy individuals from the Nord-Trøndelag Health Study (HUNT). We hypothesized that patients with schizophrenia had reduced VO2peak compared to normative healthy individuals. The secondary aim was to evaluate relationships between VO2peak, risk factors for cardiovascular disease, and quality of life.
We included 33 patients, 11 women and 22 men, with ICD-10 schizophrenia, schizotypal or delusional disorders (F20 to F29) in the study. Patients were in- and out-patients at a University hospital and had agreed to take part in exercise interventions studies. All patients were under antipsychotic medical treatment. 24 patients were smokers. Exclusion criteria were known coronary artery disease, known chronic obstructive pulmonary disease, and not being able to perform physical treadmill testing and exercise. Patients were examined by a physician at inclusion to the study and the exclusion criterions were confirmed by medical records.
An individualized protocol was applied to measure VO2peak and peak heart rate (HRpeak), using the Cortex Metamax II portable metabolic test system (Cortex Biophysik GmbH, Leipzig, Germany) and the Polar S610i heart rate monitor (Polar Electro, Finland), respectively. The protocol has previously been described in patients with schizophrenia as well as in healthy individuals [14,16].
The patients were carefully familiarized with the test procedures and the treadmill when entering the laboratory. Warm-up was ten minute walking or running on the treadmill at an intensity corresponding to 60-70% HRpeak. The test started from warm-up speed (with minimum 5% inclination) after which the speed or the inclination was increased every minute (0.5-1 km·h-1 and 1-2%, respectively) to a level that brought the patient to exhaustion. The highest oxygen uptake and heart rate (HR) recorded during the last minute of the test were determined as VO2peak and HRpeak, respectively. VO2peak where also presented as ml·kg-0.75·min-1 to normalise for the differences in bodyweight between the patients .
We compared the patients VO2peak with age and sex specific strata from the Nord-Trøndelag Health Study (the HUNT Study) . The HUNT study is an epidemiological study of the general population in the neighbouring county to the university hospital. The HUNT Fitness study tested VO2peak in 4 631 healthy individuals (20 to 90 years) using mixing chamber gas-analyzer ergospirometry (Cortex MetaMax II, Cortex, Leipzig, Germany) and an individualised protocol that has close resemblance to the protocol used in the present study. 14.1% of the participants reported to be inactive, defined as no activity or exercising less than once per week. For each patient with schizophrenia, we estimated a normative VO2peak, namely the mean value defined in the HUNT Fitness study strata for the corresponding sex and age. We titled the VO2peak estimated from sex and age strata independent of physical activity level, as HUNT general. The VO2peak from age and sex strata for healthy inactive men and women were titled HUNT inactive. The percent of HUNT general and HUNT inactive VO2peak was calculated as: (achieved VO2peak ÷ age predicted VO2peak) · 100.
In the HUNT Fitness study men and women below 44.2 ml·kg-1·min-1 and 35.1 ml·kg-1·min-1, respectively, were associated with higher cardiovascular risk factor profile . The same VO2peak values were used as threshold values when evaluating conventional CVD risk factors.
Morning fasting blood levels were taken. Serum glucose was analysed using Reflotron Plus system (Roche Diagnostics, Mannheim, Germany). HDL (high-density-lipoprotein) cholesterol, total cholesterol and triglyceride concentrations in serum were measured using a Modular P chemistry analyzer (Roche Diagnostics, Mannheim, Germany). LDL cholesterol was calculated using the Friedewald equation . BP (blood pressure) was measured using a Maxi-Stabil 3 (Welch Allyn, Jungingen, Germany). Patients were sitting and had rested for at least 5 minutes. Risk factors were classified as follows: hypertension, diastolic pressure ≥ 90 mmHg and/or systolic pressure ≥ 140 mmHg; elevated blood glucose, > 6.0 mmol·L-1; elevated total cholesterol, > 6.1 mmol·L-1 in patients < 30 years old, > 6.9 mmol·L-1 in patients 30-49 years old and > 7.8 mmol·L-1 in patients ≥ 50 years old; elevated LDL-cholesterol, 4.3 > mmol·L-1 in patients < 30 years old, 4.7 > mmol·L-1 in patients 30-49 years old and > 5.3 mmol·L-1 in patients ≥ 50 years old; reduced HDL-cholesterol, < 1.0 mmol·L-1; elevated triglyceride, > 2.6 mmol·L-1; obesity, BMI ≥ 30.0 kg·m-1 [19,20].
The short form (SF-36) was used to assess the physical health and mental health aspects of health related quality of life . SF-36 consists of eight sub scores and can also be divided into a physical component score (PCS) and mental component score (MCS). 0 reflect the poorest health whereas 100 reflect the best health.
The Positive and Negative Syndrome Scale (PANSS) was used to evaluate the severity of symptoms of schizophrenia . PANSS constitutes three scales measuring positive (productive symptoms), negative symptoms (deficit features) and general severity of illness. A total of 30 items are evaluated on a likert scale ranging from 1 (absent) to 7 (extreme) and added up to a total score as well as the three sub scores. In this study we used the positive and negative sub scores (7 items each) as well as the total score (30 items).
We used the independent samples T-test to compare differences between men and women, between patients below and above the VO2peak thresholds as well as between measured VO2peak and HUNT general and HUNT inactive VO2peak. We used the Pearson chi-square test to detect whether there was a significant association between patients above/below the VO2peak threshold and prevalence of risk factors. We calculated the odds ratio for having one or more risk factors in the patients below threshold. The analysis was adjusted for age and sex. In multiadjusted analysis we also adjusted for the potential cofounding effect of smoking.
We used Pearson r to analyse correlations between VO2peak (ml·kg-0.75·min-1) and each domain of the SF-36. The significance level (α) was set at p <0.05 (2-tailed). Data are described as mean and standard deviation (SD), unless otherwise noted. SPSS statistical package, version 18.0 (SPSS Inc.), was applied to analyse results.
The study was approved by the regional committees for medical and health research ethics, middle Norway and conducted according to the Helsinki declaration. Written informed consent was obtained from all the included patients after the procedures were fully explained.
Age was 33.7 ± 10.4 years and 35.9 ± 11.5 years in men and women, respectively. The total PANSS, total positive PANSS and total negative PANSS score was 65 ± 17, 15 ± 6 and 17 ± 8 in men, and 68 ± 23, 16 ± 6 and 18 ± 8 in women, respectively.
The VO2peak for the men and women with schizophrenia are presented in Table Table1.1. Individual VO2peak values are plotted against age as well as normative VO2peak strata from the HUNT Fitness study in Figure Figure1.1. VO2peak in the men with schizophrenia was 84 ± 21% of age predicted HUNT inactive (p < 0.001) and 74 ± 19% of HUNT general (p < 0.001). The VO2peak in the women with schizophrenia was not different from HUNT inactive (101 ± 28%) and HUNT general (89 ± 25%; n.s.). Age predicted VO2peak was 44.5 ± 2.9 in HUNT inactive men, 50.3 ± 4.1 ml·kg-1·min-1 in HUNT general men, 35.2 ± 1.8 in HUNT inactive women and 40.0 ± 3.2 ml·kg-1·min-1 in HUNT general women.
Risk factor assessment was lost in one male patient. Risk factors were present in 24 of 32 patients and of these five were above and 19 were below the thresholds. Among the eight patients without risk factors, six were above and two were below the thresholds (χ2 = 7.6, df = 1, p = 0.006). Based on the odds ratio adjusted for age and sex patients were 24.2 (95% CI = 1.5-505.6) times more likely to have one or more risk factors if they were below the VO2peak threshold. When we also adjusted for smoking the odds ratio was 28.3 (95% CI = 1.6-505.6). Among the patients below the VO2peak thresholds 10 patients had hypertension, 11 elevated glucose, 12 reduced HDL-cholesterol, 11 elevated triglyceride and 14 had obesity. Above the thresholds 2 patients had hypertension, 2 elevated glucose and 1 was obese. There were 8 smokers above the thresholds and 16 below. Differences in mean levels are presented in Table Table22.
The present results highlight reduced VO2peak as a major risk factor for CVD in patients suffering from schizophrenia. The VO2peak was 37.1 ± 9.2 and 35.6 ± 10.7 ml·kg-1·min-1 in men and women, respectively. These values are considerable higher than previous assumptions [3,13]. Strassnig et al.  reported VO2 values of 18.7 ± 6.8 and 13.4 ± 4.6 ml·kg-1·min-1in the men and women, respectively (mean age of 45.1 ± 10.1 years). These low VO2peak values is to some degree explained by the high body weight (mean BMI of 36.7 ± 7.5 m·kg2). However, there are some indications of an underrating of these patients' VO2peak. First, the patients only reached a low peak heart rate (142 ± 21 beats·min-1). Secondly, both Carlsson et al.  and Strassnig et al.  applied a cycle ergometer test which is known to depend more on the patients motivation than a treadmill test. Patients with schizophrenia terminate cycle tests already at submaximal work loads, in contrast to health subjects . Thirdly, subjects tested on a cycle ergometer achieve 7-16% lower VO2max compared with a maximal treadmill test, even when HRpeak is not significantly different [24,25].
In contrast to Strassnig et al. , the present results demonstrate that the mean VO2peak in the women was similar to the men with schizophrenia, even though the age was similar (36 years in women versus 34 years in men). Women normally have about 10 ml·kg-1·min-1 lower VO2peak compared to men at the same age . The mean body weight was 97.2 and 74.5 kg in men and women, respectively, which partially explain the difference in VO2peak.
The comparison with normalised VO2peak from the HUNT Fitness study, confirm our hypothesis that VO2peak is reduced in men with schizophrenia. The VO2peak in the women with schizophrenia was almost identical (101%) to inactive healthy HUNT women. Even lower VO2peak in men with schizophrenia compared to normative inactive men might suggest that more than just inactivity contribute the reduced VO2peak. The VO2peak in the men with schizophrenia is similar to normative healthy men aged 60-69 years . In other words, the VO2peak in the men with schizophrenia is comparable to healthy men that are about 30 years older. Patients with schizophrenia actually have 15-25 years shorter life expectancy than the general population [26,27]. It is noteworthy that the VO2peak presented in the HUNT Fitness study is somewhat higher than previous described populations with regard to objectively measured VO2peak [28-31].
People with reduced VO2peak are consistently being associated with increased risk of cardiovascular and all-cause mortality. Kodama et al.  found that 3.5 ml·kg-1·min-1 (1 MET) increases were associated with 13% and 15% reductions in all-cause mortality and CVD/coronary heart disease, respectively. Aspenes et al.  found that 5 ml·kg-1·min-1 lower VO2peak correspond to 56% higher odds of having a cluster of cardiovascular risk factors.
The comparison of patients with schizophrenia below and above the VO2peak thresholds suggested by Aspenes et al.  confirm that patients below these thresholds have higher prevalence of risk factors compared with patients above the thresholds. Based on the odds ratio patients were 28.3 times more likely to have one or more risk factors if they were below the VO2peak thresholds. When comparing mean levels above and below thresholds, all risk factors, except glucose, was better in the patients above the thresholds. These findings suggest a strong connection between the patients VO2peak and the conventional risk factors for CVD, as confirmed in other populations [10,32].
Our data are not quite consistent with findings from US suggesting that especially women with schizophrenia are at high risk of developing metabolic syndrome . This is most likely caused by the women's fitness level in the present study, as VO2peak have been described as a strong independent predictor of metabolic syndrome .
These results emphasize that evaluation of VO2peak should be incorporated into routine clinical practice for risk prediction. The prognostic value of VO2peak is beyond that predicted from other conventional risk factors [6,34]. Even in individuals with present risk factors, the higher levels of VO2peak seem to confer a significant protective effect . Reduced VO2peak is a modifiable risk factor, and eight weeks aerobic high intensity interval training has provided significant improvements of VO2peak both in healthy populations  and in patients with schizophrenia . Furthermore, to reduce the risk of CVD, the interventions are probably more dependent on improving VO2peak than increasing physical activity level alone [35,36].
Our findings of lower SF-36 social function, role emotion and mental component score among women than among men might reflect a sex difference in the general population. Lower scores for women than for men have been identified in normative adults . The gender-specific correlations between items of SF-36 and VO2peak suggest major gender differences in self-perception. Only the correlation with between SF-36 physical functioning and VO2peak was significant in men, whereas six correlations with the SF-36 were significant in women. In all subjects together the VO2peak correlated with the patient's perception of physical function, general health, vitality, social function, and physical component score. With some exceptions, these findings are consistent with correlations between SF-36 variables and BMI in patients with schizophrenia . In line with Strassnig et al.  we found a significant correlation with the physical component score but not the mental component score, suggesting that reduced VO2peak mainly is perceived as a physical health problem, not mental. Contrary, both the mental and physical health components of QOL are found related to estimated VO2peak in healthy men . An interesting note is, however, that the patients with lower VO2peak seemed to experience lower vitality and social functioning. Sedentary people are associated with greater risk of low vitality . QOL are found to improve in a dose dependent manner in sedentary women when increasing physical activity level .
There are some limitations of the study. First, the sample size is low. Secondly, the patients were included in the study based on request to take part in exercise intervention studies. However, all eligible patients at the department were asked to participate in these studies. Thirdly, severe ill patients with schizophrenia, with poor insight to their illness, might have difficulties to evaluate their perception of QOL.
Men with schizophrenia have lower VO2peak than men in the general population. Patients with a VO2peak below 44.2 ml·kg-1·min-1 (men) and 35.1 ml·kg-1·min-1 (women) have higher odds of having one or more risk factors for cardiovascular disease. Low VO2peak compromise patients' perceived physical health. VO2peak should be regarded as least as important as the conventional risk factors for CVD and evaluation of VO2peak should be incorporated in clinical practice. Finally, these finding represent an urging need for developing effective physical training interventions for patients with schizophrenia.
The authors declare that they have no competing interests.
GM, JH, JH and JH designed the study. JH and GEN recruited patients, performed VO2peak testing and other data acquisition. GM and JH undertook the statistical analysis and JH wrote the first draft of the paper. All authors have contributed to and have approved the final manuscript.
The pre-publication history for this paper can be accessed here:
Thanks to the patients that volunteered to take part.