High-density lipoprotein cholesterol and objective measures of lower extremity performance in older nondisabled persons: the InChianti study

Abstract

Objectives: To evaluate the independent association between high-density lipoprotein cholesterol (HDL-C) levels and objective measures of lower extremity performance.

Design: Cross-sectional cohort study.

Setting: Community-based.

Participants: Eight hundred thirty-six nondisabled women and men aged 65 and older enrolled in the Invecchiare in Chianti study.

Measurements: Lower extremity performance was assessed using 4-m walking speed at fast pace, 400-m walking speed, and knee extension torque. Fasting HDL-C levels were determined using commercial enzymatic tests.

Results: The mean age of participants was 73.7 (65-92), and 55.6% were women. After adjusting for potential confounders (sociodemographic factors, smoking, physical activity, body composition, and clinical conditions including cardiovascular and cerebrovascular disease, inflammatory markers, and serum testosterone) HDL-C levels were significantly associated with knee extension torque in men and women and with 4-m and 400-m walking speed in men. Men in the highest tertile of the HDL-C distribution (>55 mg/dL) had, on average, a three times greater probability of belonging to the best tertile of all indexes of lower extremity performance, including 4-m fast walking speed (odds ratio (OR)=2.57, 95%=confidence interval (CI)=1.07-6.17), 400-m walking speed (OR=3.74, 95% CI=1.20-11.7), and knee extension torque (OR=3.63, 95%=CI 1.41-9.33). Path analysis suggested a direct relationship between HDL-C and knee extension torque.

Conclusion: In older nondisabled persons, HDL-C levels are highly correlated with knee extension torque and walking speed. Further research should focus on the biological mechanism of this association.

Figure 1

Box-plots showing age- and sex-adjusted performance values in 4-m fast walking speed, 400-m walking speed, and knee extension torque according to sex-specific high-density lipoprotein cholesterol (HDL-C) tertiles. The central box extends from the 25th to the 75th percentile; the line within the box represents the median value; lines projecting outside the box on either side are the “adjacent values”; points outside the lines are the “outliers.” P from age- and sex-adjusted analysis of covariance.

Figure 2

Adjusted odds ratios (95% confidence intervals) comparing the likelihood of belonging to the highest (black) and intermediate (gray) tertile with that of belonging to the lowest tertile for 4-m fast walking speed, 400-m walking speed, and knee extension strength according to high-density lipoprotein cholesterol (HDL-C) levels and sex. Models are adjusted for variables included in Model 4 as described in Table 2.

Figure 3

Structural equation model examining the relationship between high-density lipoprotein cholesterol (HDL-C) levels and knee extension torque. Each variable has a series of arrows that reflect the proposed direction of the relationship between the variables. For continuous variables, path coefficient represents the change that occurs in the variable at the head of the arrow for each unit change in the variable at the tail of the arrow. For binary variables, path coefficient represents the log(OR) for the variable at the head of an arrow for each unit change in the variable at the tail of the arrow. Solid lines are for statistically significant coefficients (P <.1). To assess the goodness-of-fit model, the covariance matrix that resulted from the model was compared with the covariance matrix of the actual data. Because the chi-square (.947, degrees of freedom = 1) was not significant (P = .33), this model was accepted as reasonably fitting the data. CVD = cardiovascular disease.