Kaplan-Meier survival curves (n = 2985) for tertiles of the ratio of oxidated low-density lipoprotein (oxLDL) to LDL cholesterol (LDL-C) (per mille) for onset of mobility limitation (A) (log rank P = .02) and severe mobility limitation (B) (log rank P = .01) events. For oxLDL/LDL-C, the first tertile (1) was less than 8.38‰; the second tertile (2), 8.38‰ to 11.48‰; and the third tertile (3), greater than 11.48‰.
Hazard ratios and 95% confidence intervals for the association of 4 different variables with the onset of mobility limitation (A) and severe mobility limitation (B) events. High levels of interleukin 6 (IL-6) or oxidated low-density lipoprotein (oxLDL) are defined as the highest IL-6 or oxLDL/LDL cholesterol (LDL-C) ratio tertiles. Normal levels of IL-6 or oxLDL are defined as the 2 lowest IL-6 or oxLDL/LDL-C ratio tertiles. All data are adjusted for age, sex, race, site, smoking, physical activity, coronary heart disease, diabetes, hypertension, osteoarthritis, peripheral artery disease, cerebrovascular disease, depression, pulmonary disease, body mass index, values of total cholesterol, creatinine, albumin, fasting glucose, and glycated hemoglobin, and use of nonsteroidal anti-inflammatory drugs, angiotensin-converting enzyme inhibitors, and statins.
Cesari M, Kritchevsky SB, Nicklas BJ, Penninx BWHJ, Holvoet P, Koh-Banerjee P, Cummings SR, Harris TB, Newman AB, Pahor M. Lipoprotein Peroxidation and Mobility LimitationResults From the Health, Aging, and Body Composition Study. Arch Intern Med. 2005;165(18):2148-2154. doi:10.1001/archinte.165.18.2148
Oxidative damage plays an important role in leading to major health-related events. The aim of this study was to assess the predictive value of a lipoprotein peroxidation marker, oxidized low-density lipoprotein (oxLDL) for incident mobility limitation (ML).
Data are from 2985 well-functioning elders enrolled in the Health ABC study (median follow-up, 4.1 years). All oxLDL levels were measured at the baseline assessment. The oxLDL/LDL cholesterol (LDL-C) ratio (log value) was used as a measure of lipoprotein peroxidation. Mobility limitation was defined by 2 consecutive semiannual reports of any difficulty either walking 1/4 mile or climbing up 10 steps without resting. Severe ML was defined by 2 consecutive reports of great difficulty or inability to do the same tasks. Cox proportional hazards models were performed to assess hazard ratios (HRs) and 95% confidence intervals (CIs).
The mean (SD) age of the sample was 74.2 (2.9) years. After adjustment for potential confounders (sociodemographic factors, smoking, physical activity, body mass index, clinical conditions, biological markers, and medications), the relationship between the oxLDL/LDL-C ratio and disability events was statistically significant (per log-unit difference in the oxLDL/LDL-C ratio) (for ML: HR, 1.22; 95% CI, 1.06-1.41; for severe ML: HR, 1.43; 95% CI, 1.15-1.79). Consistent results were found when interleukin 6 level was included as a covariate in the adjusted models (ML: HR, 1.13; 95% CI, 0.98-1.31; severe ML: HR, 1.31; 95% CI, 1.05-1.64). No significant sex, race, interleukin 6 level, or clinical conditions interaction was found with the oxLDL/LDL-C ratio and mobility disability.
Lipoprotein peroxidation predicts the onset of ML in older persons. The oxLDL predictive value for ML is partly explained by interleukin 6 levels.
Fifty years ago, Harman1 posited the “free radical theory” of aging, suggesting that endogenous oxygen radicals generated during normal cellular processes lead to an accumulation of oxidative damage.2 Free radicals, defined as “any species capable of independent existence that contains one or more unpaired electrons,”3 are able to disrupt the equilibrium of biological systems by damaging their major constituent molecules.4,5 The oxidative damage due to the excess of free radicals is thought to play an important role in leading to major health-related events such as mobility limitation, disability, and mortality. Current evidence also suggests the existence of a strong relationship between oxidative damage and inflammation as promoters of pathophysiologic changes occurring with aging.2,6,7
Oxidized low-density lipoproteins (oxLDLs) represent a commonly used marker of oxidative damage and have been shown to play a major role in atherosclerosis and cardiovascular disease.8,9 Furthermore, oxLDLs are not only implicated in the early development of atherosclerosis but also in late clinical manifestations by inducing inflammation and endothelial dysfunction.10
Inflammation represents a well-established risk factor for major health-related events in older persons, including physical disability.11 Mobility limitation is a common early stage of the disablement process. It predicts major physical disability12 and mortality,13 and it is associated with poor quality of life12,14 and with substantial social and health care costs.15
In the present study, we examine the hypothesis that oxidative damage, as measured by the level of lipoprotein peroxidation, predicts new mobility disability events in a large sample of community-dwelling and well-functioning older persons. We also explore whether the predictive value of lipoprotein peroxidation is independent of levels of interleukin 6 (IL-6), a marker of inflammation.
The present study uses data from the Health ABC study,16 a 10-year prospective cohort study investigating the impact of changes in body composition and health conditions on physiologic and functional status at increased age. Participants (n = 3075) aged 70 to 79 years were recruited from April 1997 through June 1998 from a list of Medicare beneficiaries residing in the areas surrounding Pittsburgh, Pa, and Memphis, Tenn. Eligibility criteria included (1) no self-reported difficulty walking 1/4 mile, climbing 10 steps, or performing basic activities of daily living, (2) no life-threatening illness, (3) no plans to leave the area for 3 years, and (4) no participation in a lifestyle intervention trial.
The present study is based on 2985 participants after 90 participants were excluded for missing data on oxLDL levels or outcome event variables. All participants provided written informed consent. The institutional review boards of the clinical sites approved the study protocol.
Levels of oxLDL were obtained from frozen stored plasma collected by venipuncture after an overnight fast at the baseline visit. Blood samples were obtained in the morning, and after processing the specimens were aliquoted into cryovials and frozen at −70°C. Plasma levels of oxLDL were measured using a monoclonal antibody(4E6)–based competition enzyme-linked immunosorbent assay. The high specificity of this assay has previously been described.8 The interassay coefficient of variation of oxLDL is 12%. Because the level of oxLDL is strongly correlated with that of its substrate (r = 0.519), the ratio (per mille [‰]) between oxLDLs and LDL cholesterol (LDL-C) levels (oxLDL/LDL-C ratio)17 has been used in the present analyses.
All of the Health ABC study participants had annual clinical visits and interim 6-month telephone contacts during which health status was assessed and information on interim hospitalizations and major outpatient procedures was collected. If participants were unable to come to the clinic, a home visit was performed, with a reduced set of key measures from the clinic visit, or a phone interview was conducted with the participant or a proxy. For the present study, we considered the following mobility outcomes:
Mobility limitation, defined by 2 consecutive semiannual reports of any difficulty either walking 1/4 mile or climbing up 10 steps without resting; and
Severe mobility limitation, defined by 2 consecutive reports of great difficulty or inability to walk 1/4 mile or to climb up 10 steps without resting.
Covariates included sociodemographic variables (age, sex, race, and study site), clinical conditions (presence of coronary heart disease, diabetes, hypertension, osteoarthritis, peripheral artery disease, cerebrovascular disease, depression, and pulmonary disease as defined by algorithms identified by the clinical investigators in the study18), and physical and biological parameters, including smoking status, body mass index (BMI, calculated as weight in kilograms divided by the square of height in meters), physical activity (calculated using the Harvard Alumni study19- 22 algorithm based on walking and exercise expenditure in kilocalories per week), and measurements for total cholesterol, creatinine, albumin, fasting glucose, and glycated hemoglobin (HbA1c). Total cholesterol, creatinine, and albumin values were determined by a colorimetric technique on a Vitros 950 analyzer (Johnson & Johnson, New Brunswick, NJ). Levels of LDL-C were calculated by the equation described by Friedewald et al.23 Fasting glucose levels were measured using an automated glucose oxidase reaction (YSI 2300 Glucose Analyzer; YSI, Yellow Springs, Ohio). Levels of HbA1c were measured by a fully automated analyzer (Variant; Bio-Rad Laboratories Inc, Hercules, Calif) using the principle of ion exchange high-performance liquid chromatography. Medications taken in the past 2 weeks were brought in, recorded, and coded according to the Iowa Drug Information System. Using this system, participants using nonsteroidal anti-inflammatory drugs, angiotensin-converting enzyme inhibitors, and statins were identified.
Increased IL-6 level was considered an additional potential explanatory covariate in the adjusted models to evaluate whether inflammation was able to explain the association between oxLDL levels and the onset of (severe) mobility limitation events. It has been demonstrated that increased level of IL-6, a multifunctional cytokine and a marker of the geriatric syndrome of frailty,24 can be used to predict incident physical disability in older persons.11,25,26 Levels of IL-6 were measured in duplicate by an enzyme-linked immunosorbent assay (R&D Systems, Minneapolis, Minn). The detectable limit for IL-6 (HS600 Quantikine kit; R&D Systems) was 0.10 pg/mL. Blind duplicate analyses (n = 150) for IL-6 level showed interassay coefficients of variation of 10.3%. Circulating IL-6 levels obtained from 1 time point have been shown to be reproducible and representative over extended time periods.27
Given the nonnormal distribution of oxLDL, the oxLDL/LDL-C ratio, and IL-6, these biomarker levels were log-transformed to ensure equality of variances and to make the errors approximately normally distributed. Pearson correlation analyses were performed between oxLDL (log value), the oxLDL/LDL-C ratio (log value), and IL-6 (log value). Kaplan-Meier survival analyses were performed to evaluate the onset of (severe) mobility limitation events according to the oxLDL/LDL-C ratio tertiles. Unadjusted and adjusted Cox proportional hazard models were performed to evaluate the predictive value of the oxLDL/LDL-C ratio for incident mobility disability events. Sex, race, and clinical condition interactions between the oxLDL/LDL-C ratio and mobility disability were evaluated adding the interaction term to the fully-adjusted models. High levels of IL-6 and high oxLDL/LDL-C ratios were defined as the highest IL-6 and oxLDL/LDL-C ratio tertiles.
The main characteristics of the sample population (n = 2985) are reported in Table 1. The mean (SD) age of the sample population was 74.2 (2.9) years. Prevalences of women and white patients were 51.6% and 58.1%, respectively. During the follow-up (median, 4.1 years), 1231 (41.2%) and 527 (17.7%) participants reported onset of mobility limitation and severe mobility limitation events, respectively.
Pearson correlation analyses showed statistically significant correlations of oxLDL with IL-6 (all log values; r = 0.072; P = .001), oxLDL/LDL-C ratio with IL-6 (all log values; r = 0.140; P<.001), and oxLDL with oxLDL/LDL-C ratio (all log values; r = 0.767; P<.001). Because the level of oxLDL was strongly correlated with that of its substrate (r = 0.547; P<.001), we used for the present analysis the oxLDL/LDL-C ratio (log value).
Kaplan-Meier survival curves for the onset of (severe) mobility limitation events across tertiles of the oxLDL/LDL-C ratio are shown in Figure 1. Participants with higher levels of lipoprotein peroxidation presented a significantly higher risk of developing mobility limitation events than those in the lowest tertiles (P<.05 for all).
Results from Cox proportional hazard models testing the predictive value of the oxLDL/LDL-C ratio for the onset of mobility disability events are reported in Table 2. The oxLDL/LDL-C ratio was significantly and strongly associated with an increased risk of mobility limitation (P<.01 for all). Minor differences were reported between unadjusted models and those taking into account all the potential confounders. No significance was reported for LDL-C in the prediction of mobility limitation events. Oxidated LDL levels were able to predict (severe) mobility limitation events but to a lesser extent than the oxLDL/LDL-C ratio.
Analyses were then reperformed adding levels of IL-6 to the models to evaluate whether the association between the oxLDL/LDL-C ratio and mobility disability events was independent of this proinflammatory cytokine. Results are reported in Table 3. Unadjusted models showed statistically significant associations for both oxLDL/LDL-C ratio and IL-6 with mobility outcomes, even if a reduction in the associations strength was reported. Borderline significance was found when testing the predictive value of the oxLDL/LDL-C ratio for mobility limitation in a partially adjusted model (hazard ratio [HR], 1.13; 95% confidence interval [CI], 0.98-1.29; P = .09), and in the fully adjusted model (HR, 1.13; 95% CI, 0.98-1.31; P = .10). Analyses exploring severe mobility limitation showed significant results for the oxLDL/LDL-C ratio even after adjustment for all the potential confounders (HR, 1.31; 95% CI, 1.05-1.64; P = .02). Levels of IL-6 were strongly associated with both outcomes (P<.001 for all), independently of potential confounders and lipoprotein peroxidation. No significant sex, race, or IL-6 interaction was found between oxLDL/LDL-C ratio and mobility disability events.
Figure 2 shows results from fully adjusted Cox proportional hazard models exploring the relationship between mobility events and a 4-level categorical variable defined by the number of markers (oxLDL/LDL-C ratio and IL-6) in their highest tertiles (none [reference group], oxLDL/LDL-C ratio only, IL-6 only, or oxLDL/LDL-C ratio and IL-6). Participants with high levels of IL-6 alone or in combination with a high oxLDL/LDL-C ratio presented an increased risk of developing mobility limitation (HR, 1.29; 95% CI, 1.11-1.50 and HR, 1.47; 95% CI, 1.24-1.74, respectively) compared with participants with normal levels of both markers. Similar findings were reported for severe mobility limitation. In fact, participants with high levels of IL-6 alone or in combination with a high oxLDL/LDL-C ratio had a higher risk of developing severe mobility disability than the reference group (HR, 1.47; 95% CI, 1.16-1.85 and HR, 1.69; 95% CI, 1.31-2.19, respectively). No significant differences were found between participants with only a high oxLDL/LDL-C ratio compared with the reference group for the prediction of (severe) mobility limitation (P>.10).
To evaluate whether findings were driven by the presence of atherosclerotic diseases, we tested the interaction terms between oxLDL/LDL-C ratio and atherosclerotic diseases (coronary heart disease and cerebrovascular disease) for the onset of mobility disability events. No significant results were reported (P>.10 for all), suggesting that the predictive value of oxLDL/LDL-C ratio for mobility disability is independent of atherosclerotic diseases presence. Restricted analyses were also performed after exclusion of 289 participants who developed myocardial infarction and/or cerebrovascular events during the follow-up. Results from adjusted (model 2 adjustment) Cox proportional hazard models confirmed that oxLDL/LDL-C ratio was predictive for mobility limitation (HR, 1.21; 95% CI, 1.04-1.42; P = .02) and severe mobility limitation (HR, 1.47; 95% CI, 1.15-1.88; P = .002) events, independently of the presence of atherosclerotic process.
Other clinical conditions potentially mediating the relationship between oxLDL/LDL-C ratio and mobility limitation outcomes were also evaluated. Analyses were reperformed in restricted samples after exclusion of participants with diabetes, hypertension, osteoarthritis, peripheral artery disease, depression, and/or pulmonary disease. Consistent results with previous findings were reported, confirming the independent association between oxLDL/LDL-C ratio and incidence of mobility limitation.
The present study is the first to our knowledge to explore the predictive value of a marker of oxidative damage (oxLDL) for the onset of mobility limitation. We have shown that lipoprotein peroxidation is associated with an increased risk of developing mobility limitation in older persons. Moreover, we have explored whether this association was independent of IL-6 levels, a major and well-established marker of inflammation. Our findings show that IL-6 is a stronger predictor of mobility disability than oxLDL. However, the best prediction of incident mobility limitation events was obtained when both markers were considered together. Moreover, we reported a significant association between lipoprotein peroxidation and incident severe mobility limitation independent of IL-6 levels.
How can oxLDL level predict mobility limitation? It might be argued that the predictive value of oxLDL levels for the onset of mobility limitation might be solely owing to the atherogenic action and consequent cardiovascular status associated with this lipoprotein peroxidation marker. Oxidative damage in general, and oxLDL in particular, have been associated with atherosclerosis, which in turn has shown a graded relationship with the likelihood of maintaining intact mobility function.28 However, no significant results were found when we tested the interactions between oxLDL/LDL-C ratio and atherosclerotic diseases (coronary heart disease, cerebrovascular disease) for the incidence of mobility disability. Moreover, restricted analyses performed in participants with no incident myocardial infarction and/or cerebrovascular events occurring during the follow-up led to results consistent with those obtained from the overall sample, suggesting that the association between oxLDL level and mobility limitation is independent of atherosclerotic processes. We also evaluated other clinical conditions as potential mediators of the studied relationship. However, the independent association between lipoprotein peroxidation and mobility limitation was confirmed even after exclusion of participants with diabetes, hypertension, osteoarthritis, peripheral artery disease, depression, and/or pulmonary disease.
On the basis of current evidence we may hypothesize that inflammation and oxidative damage, strongly related to each other,7,29,30 may represent promoters of the disabling process. Confirming previous reports, we showed in the present study the strong predictive value of IL-6 levels for the onset of mobility disability. The relationship of inflammation with the onset of mobility disability has been widely demonstrated.11,26,31 Inflammatory markers have been shown to negatively effect skeletal muscle mass and quality, namely by accelerating changes that are typical of the aging process.11,32- 34 In fact, a direct influence of cytokine levels on muscle mass has been demonstrated in humans33,35 as well as in animal models.36- 38
A synergistic relationship between inflammation and oxidative damage can also be found at the beginning of several pathophysiologic changes and clinical conditions potentially mediating the onset of physical function loss. For example, obesity, a major risk factor for disability, is associated with oxidative damage as well as with inflammation.39 Similarly, several clinical conditions (such as cardiovascular disease, diabetes, cancer, pulmonary disease, and neurodegenerative disease) have been associated with increased levels of oxidative damage and inflammation. The burden of comorbidity is a well-established etiologic risk factor for the frailty syndrome, which has disability as major outcome.40
For the present analyses, we used plasma levels of oxLDL, a cardiovascular risk factor commonly adopted to measure oxidative damage. However, other markers may represent better markers of oxidative damage than oxLDL (eg, isoprostanes).41,42
If further studies will confirm oxidative damage as a causative agent of the disabling process, new potential targets for future interventions aimed at preventing the onset of physical disability can be considered. Antioxidant supplementation might represent a potential approach to simultaneously reduce oxidative damage levels and increase physical performance. Unfortunately, although some findings have shown improvements,43- 45 other studies do not support a beneficial effect of increased antioxidant intakes on physical performance.46- 48 Nevertheless, it has been suggested that an adequate antioxidant intake is needed to maintain healthy muscular activity.49 Physical exercise may represent another intervention improving physical performance by reducing oxidative damage. In fact, even if physical exercise is associated with an abnormal production of free radicals in the short term, elderly individuals who are physically active benefit from exercise-induced adaptations in the cellular antioxidant defense system.50 Wang and colleagues51 demonstrated in animal models that regular physical exercise is able to decrease levels of lipid peroxides. Finally, weight loss, which has been shown to improve physical function in older persons,52 is also associated with reductions of oxidative damage and inflammation levels.39
In conclusion, our study demonstrates that lipoprotein peroxidation is predictive for the onset of mobility disability in well-functioning nondisabled older persons. This association is strongly associated with levels of inflammation. Further studies are needed to confirm our findings, especially using other markers of oxidative damage, and to provide a basis to verify whether intervention aimed at reducing or preventing oxidative damage could be addressed for the prevention of disability in older persons.
Correspondence: Matteo Cesari, MD, PhD, Department of Aging and Geriatric Research, College of Medicine, Institute on Aging, University of Florida, 1329 SW 16th St, Room 5273, Gainesville, FL 32608 (firstname.lastname@example.org).
Accepted for Publication: May 28, 2005.
Financial Disclosure: None.
Funding/Support: This work was supported through the National Institute on Aging (contract Nos. N01-AG-6-2106, N01-AG-6-2101, and N01-AG-6-2103), the Intrauniversitaire Attractiepolen Programma of the Belgian Federal Government (P05/02), and the Funds voor Wetenschappelijk Onderzoek-Vlaanderen (G.0263.01). The data analyses and the work of Dr Cesari were supported by the Wake Forest University Claude D. Pepper Older Americans Independence Center (National Institute on Aging grant P30-AG-021332-02).