Clin Physiol Funct Imaging (2016)
Single and combined associations of accelerometer-assessed physical activity and muscle-strengthening activities on plasma homocysteine in a national sample Samuel L. Buckner1, Jeremy P. Loenneke1 and Paul D. Loprinzi2 1 2
Department of Health, Exercise Science, and Recreation Management, Kevser Ermin Applied Physiology Laboratory, The University of Mississippi, and Department of Health, Exercise Science and Recreation Management, Center for Health Behavior Research, The University of Mississippi, University, MS, USA
Summary Correspondence Paul D. Loprinzi, Center for Health Behavior Research, Department of Health, Exercise Science, and Recreation Management, The University of Mississippi, 229 Turner Center, University, MS 38677, USA E-mail: [email protected]
Accepted for publication Received 18 September 2015; accepted 21 January 2016
Key words endothelial dysfunction; guidelines; health
Background: Aerobic exercise and resistance training (RT) are individually associated with positive health outcomes. It has been shown that combining both aerobic and RT exercise may produce greater outcomes than each type alone. Homocysteine has been shown to decrease with aerobic and RT exercise. However, no study has examined the combined effects of both types on homocysteine, which was the purpose of this study. Methods: Data from the 2003–2006 NHANES were used. Homocysteine values were obtained from a blood sample. Accelerometer-derived physical activity (PA) was measured using accelerometry, and MSA was measured via a questionnaire. Results: Meeting the recommendations for accelerometer-derived PA was associated with lower homocysteine (badjusted = 029; P = 004), as was meeting the recommendations for MSA (badjusted = 030; P = 0028). Compared to those meeting zero recommendations, there was no difference between those meeting 1 and those meeting both (P = 008). Further, higher homocysteine was linked with increased all-cause mortality (HR = 147; 95% CI: 129–169) and CVD-specific mortality (HR = 148; 95% CI: 120–181). Conclusion: These results suggest that PA and MSA are associated with lower homocysteine and that there may be a dose–response relationship when combining both forms of exercise, which is an important finding as higher homocysteine is predictive of increased all-cause and CVD-specific mortality.
Introduction Regular engagement in both aerobic exercise and musclestrengthening activities is individually associated with positive health outcomes. For example, physical inactivity is associated with negative health consequences due to coronary heart disease (CHD) (Powell et al., 1987), hypertension (Rossi et al., 2012), diabetes (Sigal et al., 2013), cancer (Clague & Bernstein, 2012) and arthritis (Suomi & Collier, 2003). Moreover, regular physical activity (PA) is associated with more favourable levels of biomarkers, such as decreases in C-reactive protein (CRP) (Loprinzi et al., 2013), decreases in homocysteine (Okura et al., 2006) and other markers of inflammation (Kasapis & Thompson, 2005) and endothelial dysfunction (Hambrecht et al., 2003). Similarly, increased muscular strength is associated with decreases in all-cause mortality (Katzmarzyk &
Craig, 2002; Newman et al., 2006), inflammatory markers such as CRP (Olson et al., 2007), total cholesterol (Fahlman et al., 2002), homocysteine (Vincent et al., 2003), systolic blood pressure (SBP) (Martel et al., 1999) and an increase in insulin sensitivity (Brooks et al., 2007). Recommendations from the United States Departments of Health and Human Services (USDHHS) suggest at least 150 min of moderate or 75 min of vigorous aerobic activity a week (or some combination of the two) for substantial health benefits and engagement in strength training at least two times a week to improve muscular strength (Physical Activity Guidelines Advisory Committee Report, 2009). Within the literature, a more complete understanding of the health benefits related to PA is developing (Bauman, 2004; Reiner et al., 2013); however, the combined effects of both aerobic exercise and musclestrengthening activities (MSA) have received little attention.
© 2016 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd
2 Physical Activity and Homocysteine, S. L. Buckner et al.
Specifically, the epidemiological investigation of meeting the guidelines for both forms of activity is lacking. Considering the individual benefits of both forms of exercise, it is plausible that the combination of both aerobic and MSA activities may lead to greater changes in favourable health outcomes. As such, in this study we examine the individual and combined associations of accelerometer-derived PA and MSA on homocysteine, a marker of endothelial dysfunction, which has recently been associated with increased risk of cardiovascular disease and mortality (Nygard et al., 1997; Anderson et al., 2000). Homocysteine is a sulphur-containing amino acid formed during the metabolism of methionine (Bolander-Gouaille, 2002), an essential amino acid which is involved in protein synthesis, the formation of polyamines and other metabolites. Prevalence of homocysteine in the blood was first observed in patients with homocystinuria; a condition involving an inborn error in metabolism, which leads to increased plasma homocysteine, and is associated with an increased incidence of cardiovascular disease early in life. Increases in this marker resulting from homocystinuria are well outside the normal physiological range (>100 lmol l 1) of 5–15 lmol l 1. Prompted by the findings in patients with homocystinuria, many subsequent studies have examined the associations between moderate increases in homocysteine and vascular disease (Nygard et al., 1997). For example, a meta-analysis of 27 studies found that a 5 umol increase in total homocysteine is associated with a 60% and 80% increase in coronary artery disease (CAD) for men and women, respectively (Boushey et al., 1995). Furthermore, Nygard et al. (1997) examined a strong graded dose–response relationship between total homocysteine and overall mortality in a prospective study of 587 adults, which is further validated by the findings of Anderson et al. (2000) who found that homocysteine is a significant independent predictor of mortality in patients with CAD. Homocysteine may be related to metabolic syndrome, and cardiovascular disease through its influence on endothelial function and platelet aggregation. Animal studies have indicated that high homocysteine levels inhibit endotheliumdependent anticoagulant reaction (Lentz & Sadler, 1991; Wald et al., 2002), induce pro-coagulant expression (Rodgers & Conn, 1990; Fryer et al., 1993) and inhibit endotheliumderived nitric oxide (Stamler et al., 1993), which acts as a vasodilator. Human studies also support an association between homocysteine and nitric oxide release (Rubanyi et al., 1986; Tawakol et al., 1997). Moreover, high homocysteine levels may also induce platelet aggregation and ultimately blood clotting, possibly through oxidative stress through the formation of reactive oxygen species (Edirisinghe, 2004; Wilson & Lentz, 2005). Thus, both experimental and epidemiological results provide compelling evidence that homocysteine is an important biomarker with regard to endothelial dysfunction and platelet aggregation. With regard to PA, Vincent et al. (2003) found that 6 weeks of resistance training decreased homocysteine by
approximately 5% in older (60–80 years) adults. Moreover, chronic aerobic exercise has been shown to reduce homocysteine levels in highly trained aerobic athletes, within the highest quartile of homocysteine levels (Konig et al., 2003; e Silva Ade & da Mota, 2014). However, results are inconsistent, as Okura et al. (2006) found slight increases in homocysteine with 20 weeks of aerobic exercise training in men and women ages 17–65. All studies had participants maintain their diet, as both folate and vitamin B12, are involved in the metabolism of homocysteine. In addition, Nygard et al. (1995) examined the associations between self-reported PA and homocysteine in a populationbased sample in the Hordaland Homocysteine Study (Norway), finding that individuals with the highest levels of PA had the lowest homocysteine levels; however, Vitamin B- status was not reported in this study. Considering the inconsistencies in the aforementioned studies, the purpose of this study was to examine the associations between physical activity and homocysteine, while investigating both the individual and combined associations of accelerometerdetermined PA and self-reported involvement in MSA. Previous epidemiological work has examined the individual association of PA with homocysteine (Nygard et al., 1995; Loprinzi & Cardinal, 2012), but to our knowledge, this is the first epidemiological study to examine the combined associations of accelerometer-determined PA and MSA on homocysteine levels. In addition, it has been acknowledged that inconsistent findings within the literature may be due to a lack of control for important covariates (Joubert & Manore, 2006). Therefore, we address this association using a nationally representative sample from the National Health and Nutrition Examination Survey (NHANES) of U.S. adults, while controlling for vitamin B12, folate and other relevant covariates. In addition to examining the association of PA and MSA with homocysteine, to further strengthen this study we use NHANES follow-up data to examine the extent to which lower homocysteine as a result of PA and MSA may influence all-cause and CVD-specific mortality.
Methods Design and participants Data were obtained from the 2003–2006 NHANES, which was approved by the National Center for Health Statistics ethics committee. For the mortality analyses, follow-up data through 31 December 2011 were used. Mortality data were ascertained via linkage with the National Death Index. The NHANES is an ongoing survey conducted by the Center for Disease Control and Prevention, which uses a multistage, complex clustered probability design to select a representative sample of non-institutionalized United States civilians. A total of 2431 consented adults (20–85 years) provided data on the study variables.
© 2016 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd
Physical Activity and Homocysteine, S. L. Buckner et al. 3
Homocysteine Total plasma homocysteine was measured using the ‘Abbot Homocysteine (HCY) assay’, a fully automated fluorescence polarization immunoassay (FPIA) from Abbott Diagnostics (Abbott Diagnostic, IL, USA). Plasma total homocysteine concentrations were calculated using the Abbott Axsym, which utilizes a machine stored calibration curve. It has been demonstrated that this method is similar to other commonly utilized methods (i.e. HPLC-FD, HPLC-ED, GC/MS) (Ueland et al., 1993). Assessment of physical activity Objectively measured PA was assessed via accelerometery, with details provided elsewhere (Troiano et al., 2008; Loprinzi et al., 2014). Briefly, participants who were not prevented by impairments of walking or wearing an accelerometer were issued an ActiGraph 7164, which was worn for a period of up to 7 days. Individuals that met the USDHHS guideline of 150 min week 1 of accelerometer-assessed moderate-to-vigorous physical activity (MVPA), with MVPA defined as activity counts/min ≥ 2020, were classified as active. Only those with 4+ days of 10+ h day 1 of monitoring data were included in the analyses to ensure habitual movement patterns were captured. Non-wear was defined by a period of a minimum of 60 consecutive minutes of zero activity counts, with the allowance of 1–2 min of activity counts between 0 and 100. Measurement of muscle-strengthening activities During a household interview, participants were asked the following questions related to engagement in MSA: (i) ‘Over the past 30 days, did you do any physical activities specifically designed to strengthen your muscles such as lifting weights, push-ups or sit-ups?’ (response option: yes or no), and (ii) among those answering yes to this first question, they were asked, ‘Over the past 30 days, how many times did you do these activities designed to strengthen your muscles such as lifting weights, push-ups, or sit-ups?’ (range = 0–60). MSA were collected using subjective measures due to the complex nature of quantifying strength-type exercise in an epidemiological context. For example, there is no objective measure of MSA other than recording sets, repetitions completed, speed of repetitions and the relative resistance used for each individual. Meeting the recommendations for MSA was considered engagement in such activities at least twice a week. The NHANES MSA items have provided evidence of construct validity (Loprinzi et al., 2015). Covariates Covariates included in the analytic models were as follows: age (years), gender, race, C-reactive protein (mg dl 1), cotinine (ng ml 1), plasma folate (lmol l 1), plasma vitamin B12 (pg ml 1), metabolic syndrome (Grundy et al., 2005),
coronary artery disease and measured waist circumference (cm). Covariates were included based on known associations with homocysteine (Joubert & Manore, 2006). Analysis All statistical analyses were computed in Stata (v. 12, StataCorp LP, College Station, TX, USA) and accounted for the complex NHANES survey design. Multivariable linear regression was used to individually examine the associations between accelerometerderived PA and MSA with homocysteine (outcome variable). In addition, multivariable linear regression was used to examine the associations between the combined participation in accelerometer-derived PA and MSA with homocysteine. Cox proportional hazards models were used to examine the association between homocysteine and all-cause and CVD-specific mortality. Schoenfeld’s residuals were used to verify the proportional hazards assumption, as a key assumption in Cox models is that the effect of any predictor variable on mortality is constant over time. Thus, a nonzero slope is an indication of a violation of the proportional hazard assumption. Statistical significance was set at a nominal alpha of 005.
Results Table 1 displays the weighed characteristics of the analysed sample. Meeting the recommendations for accelerometerderived PA was associated with lower homocysteine (badjusted = 029; P = 004), as was meeting the recommendations for MSA (badjusted = 030; P = 0028). In addition, meeting 1 or 2 of the guidelines was associated with lower plasma homocysteine (badjusted = 033; P = 0007; badjusted = 057; P