Author + information
- Christopher J.K. Hammett, MB, ChB1,
- Helen C. Oxenham, MD,
- J. Chris Baldi, PhD,
- Robert N. Doughty, MD, MRCP, FRACP,
- Rohan Ameratunga, MB, ChB, PhD,
- John K. French, MB, PhD, FACC and
- Harvey D. White, DSc, FACC
- Ralph A.H. Stewart, MD* ()
To the Editor:
Nonspecific markers of inflammation such as C-reactive protein (CRP) are now recognized as major risk factors for cardiovascular events. General population studies have reported an inverse association between serum CRP levels and self-reported physical activity or physical fitness (1,2). These studies suggested that regular physical exercise might lower CRP levels by an anti-inflammatory action. However, an alternative explanation is that exercise lowers CRP levels by reducing total or abdominal fat. We performed a randomized controlled trial to assess the effects of six months' regular exercise training on serum CRP levels and body fat measured by dual-energy X-ray absorptiometry (DEXA) in a healthy elderly population.
Healthy elderly subjects age 60 to 85 years were recruited from the general population by newspaper advertisements. The exclusion criteria included inability to undertake an exercise-training program, current illness known to be associated with a systemic inflammatory response, a history of hypertension or diabetes, any current cardiovascular medication (including aspirin), or any evidence of cardiovascular disease on physical examination, electrocardiography, treadmill exercise testing, or echocardiography. The study protocol was approved by the University of Auckland ethics committee, and all participants gave written informed consent.
Subjects were assigned randomly to either an exercise group or a control group. The participants in the control group were asked to maintain their usual level of physical activity. The exercise group underwent six months' exercise training, consisting of three supervised sessions and one unsupervised session each week. Exercise intensity was monitored by heart rate response. The training intensity was increased gradually so that by the fourth month, participants were training for 45 min at a heart rate of 80% of their current estimated maximum oxygen consumption (VO2max).
At the start and end of the study, the patients' VO2max was measured during graded treadmill exercise, body mass index (BMI) was calculated (weight/height2[kg/m2]), and total body fat and trunk fat composition were determined by DEXA scanning using a Lunar DPX-IQ scanner (GE Medical Systems, Waukesha, Wisconsin). Venous blood samples were collected for measurement of CRP levels at least 24 h after the last exercise session. In participants with any illness or injury associated with an acute phase response, blood sampling was delayed for two weeks. The samples were stored at −80°C before being analyzed as a single batch using a high-sensitivity assay (Dade Behring, Inc., Deerfield, Illinois).
Because the distribution of CRP levels was skewed, natural-log-transformed CRP was used for statistical analysis. The Pearson correlation coefficient was used to assess the association between natural-log-transformed CRP and other baseline variables. Changes in CRP levels and other variables were compared between groups using a Mann-Whitney Utest. All analyses were performed using SAS Release 8.0 Software (SAS Institute Inc., Cary, North Carolina).
Two participants were lost to follow-up, leaving a final cohort for analysis of 30 exercise subjects and 31 control subjects. Analyses were performed on an intention-to-treat basis and included all subjects with follow-up data. A sensitivity analysis (excluding four dropouts from the exercise program) did not alter the study findings.
The baseline characteristics of the exercise and control groups were similar (Table 1).Greater physical fitness (VO2max) was associated with a lower BMI (r = −0.27, p = 0.03), less body fat (r = −0.61, p < 0.001), and less trunk fat (r = −0.55, p < 0.001). C-reactive protein levels correlated with measures of adiposity, including the BMI (r = 0.40), body fat percentage (r = 0.46), and trunk fat percentage (r = 0.48, p ≤ 0.001 for all). In contrast, there was only a weak trend toward lower CRP levels with greater fitness (r = −0.22, p = 0.09).
At six months, there was an 18% improvement in cardiorespiratory fitness in the exercise group and no change in the control group (Table 2).There were no significant changes in body weight, body fat percentage, lipid profile, or serum glucose levels in either group. Considerable individual variability in CRP levels existed between baseline and six months but, on average, serum CRP levels did not change in either group. In addition, no association between changes in fitness and changes in CRP levels during the six-month trial was found. The results were similar for men and women.
In this study, six months' exercise training in healthy elderly participants did not lower serum CRP levels despite a significant improvement in their cardiorespiratory fitness. This finding contrasts with those of several large epidemiologic studies that reported an inverse association between serum CRP levels and regular physical activity (1,3,4) and/or physical fitness (2,5). There are several possible explanations for this difference. The association between exercise and inflammatory markers observed in epidemiologic studies may not be causal. Poor health of any severity may cause inflammation, as well as decreasing regular exercise and physical fitness. Alternatively, the cumulative effects of exercise may influence serum CRP levels over the course of many years, although not by shorter-term changes in exercise levels. This possibility is consistent with the hypothesis that physical exercise acts predominantly by reducing body fat. It is supported by the strong inverse association between physical fitness and body fat percentage observed in this study and the results of clinical trials that have reported reductions in CRP levels after weight loss (6). Although observational studies that reported an association between physical activity or fitness and serum CRP levels adjusted for BMI or waist girth (1,2), these measures of body fat are relatively imprecise, and residual confounding is possible. In the current study, the more precise DEXA scanning method was used to measure body fat and trunk fat. These DEXA measures correlated more strongly than BMI with both CRP levels and physical fitness. A limitation of the current study was variability in serum CRP levels. This variability reduced the statistical power of the study and, therefore, a small effect of improved physical fitness on CRP levels, independent of changes in body fat, cannot be excluded. Few previous studies have reported the effects of exercise training on CRP levels. Mattusch et al. (7) found a significant reduction in CRP levels after nine months of marathon training in 12 athletes. Smith et al. (8) reported a trend toward lower CRP levels in 43 volunteers after six months of exercise training. These studies did not include a control group and did not measure body fat. In conclusion, our findings suggest that the association between greater physical fitness and lower serum CRP levels is explained, at least in part, by long-term regular exercise reducing body fat.
The authors thank Roy The, BSc, for CRP analysis, Teena West, MSc, for statistical input, Charlene Nell for secretarial assistance, and Anna Breckon, ELS, for editorial assistance.
↵1 Please note: this work was supported by Project Grant No. 990 from the National Heart Foundation of New Zealand (Auckland, New Zealand) and by the generous loan of a CRP autoanalyser from Dade Behring Diagnostics Limited (Auckland, New Zealand). Dr Hammett was the recipient of a Cardiac Society of Australia and New Zealand/Merck Sharp and Dohme Fellowship.
- American College of Cardiology Foundation
- Wannamethee S.G.,
- Lowe G.D.,
- Whincup P.H.,
- et al.
- Church T.S.,
- Barlow C.E.,
- Earnest C.P.,
- et al.
- Geffken D.F.,
- Cushman M.,
- Burke G.L.,
- et al.
- LaMonte M.J.,
- Durstine J.L.,
- Yanowitz F.G.,
- et al.
- Tchernof A.,
- Nolan A.,
- Sites C.K.,
- et al.