A systematic review 1 studied the effects of endurance training on serum HDL cholesterol concentrations. Medline search covered the years 1966 to 2005. The review included randomized controlled trials published in English, involving aerobic endurance training with a minimum duration of 8 weeks, each exercise episode lasting at least 15 minutes and with a control group that did not receive any advice for exercising (nor any other form of intervention). The study participants had to be at least 20 years of age without any particular health problems, medication or diet therapy. The analysis included 25 trials, with a total number of 1 400 participants (mean age range 23-75 years). The methodological quality of the trials was assessed to be rather modest. Only 12 trials reported dropout rates that varied between 4 and 39 percent.
The exercise period lasted on average 27 weeks, with an average of 3.7 exercise episodes per week. Each episode lasted on average 41 min, with an estimated intensity of 65% of the maximal aerobic capacity (VO2max) or 5.3 MET (metabolic equivalents), respectively. The estimated energy expenditure was 1000 kcal per week. The most common exercise forms comprised walking, jogging and exercise cycling.
The average BMI at the baseline was 26.5 (non-weighted mean of the different trials) and the measured or estimated VO2max 32.7 ml/(min × kg). The change in the body weight or VO2max after the exercise was not reported. At the baseline, the serum concentration (non-weighted mean of the different trials) of HDL cholesterol was 1.29 mmol/l, total serum cholesterol concentration 5.27 and triglyceride concentration 1.22 mmol/l. The change in HDL cholesterol during exercise was calculated as the net effect (the change in the control group was subtracted from the change in the exercise group), weighted by the sizes of the study populations (inverse of the variance).
Net change in HDL cholesterol was 0.065 mmol/l (95% CI 0.035-0.096 mmol/l), corresponding to a decrease of approximately 5% from the baseline. Funnel plot analysis revealed no publication bias. In further analyses, the weekly energy expenditure (EE) or weekly exercise length was categorized into 4 classes. In conclusion, weekly EE had to be at least 900 kcal or weekly total exercise length at least 120 minutes to produce a statistically significant increase in the HDL cholesterol concentration.
In addition, a weighted linear meta-regression analysis was performed to investigate which characteristics in the exercise programmes or in the study populations were associated with the change in HDL cholesterol. Univariate and multivariate analyses indicated that exercise duration was the most important predictive factor, and the net change in the HDL cholesterol concentration was not associated with the intensity or other characteristics of the exercise (which may, however, also reflect the difficulties in measuring these characteristics during exercise). In stepwise regression analysis, the increase in HDL cholesterol was greater in subjects who had a total cholesterol concentration of at least 5.7 mmol/l and BMI < 28 at the baseline.
Comment: The observed change in HDL cholesterol concentration was thus statistically significant and clinically rather modest. The realization of the study intervention was variable considering compliance and attrition rates. The effects of endurance training on physical fitness were measured with different methods (direct measurement of VO2max in maximal aerobic capacity test or estimation based on heart rate recordings or submaximal test). Most study subjects had only modest abnormalities in their lipid profiles. In the summary characteristics of the study populations it remained unclear what type of medications the subjects possibly used. At least in one study, all the subjects had severe hypertension. Data on total cholesterol and triglyceride concentrations were also collected but changes in these after training are not reported.
In another systematic review 2, the effects of aerobic exercise training on blood lipids were considered, with special emphasis on possible dose-response relationships. Medline search started from the year 1987. Exercise training studies employing mainly supervised group training with duration of at least 12 weeks were included in the review. The study subjects were sedentary adults in good general health. A total of 51 exercise studies were included in the analysis, including 28 randomized controlled trials (approximately 2 300 subjects). The total number of study subjects was approximately 4 700 (age between 18 and 80 years, 60% male), the most subjects being slightly overweight (BMI 25 to 30) Caucasians.
In only 9 studies the average total blood cholesterol concentration was increased (> 6.2 mmol/l) in the participating subjects at the start of the intervention. In most studies the training program followed the principles of endurance-enhancing exercise (intensity from moderate to high, frequency 3 to 5 times per week and duration of single episode at least 30 min). In 7 studies, different training intensities were compared but none of the studies compared programs with different energy expenditures. The average weekly energy expenditure of the programs was ca. 1 500 kcal in males and 1 200 kcal in females. VO2max improved on average by 16%. Change in body weight varied, partly depending on whether a diet change was included in the intervention. In the 61 study groups (with approximately 2 200 participants) where diet change was not included in the targets, body weight was decreased on average by 0.8 kg.
Baseline blood concentration of total cholesterol was on average 5.29, LDL cholesterol 3.53, HDL cholesterol 1.18 and triglycerides 1.41 mmol/l. The response to exercise seen in blood lipoproteins showed marked heterogeneity (i.e., different responses in different studies). The most common change in lipid profile (detected in 50% of the studies) was an increase in the blood HDL concentration. In 20 studies (with approximately 2 200 participants) where the diet was kept unchanged, blood HDL cholesterol increased on average by 4.6% from baseline (0.05 mmol/l). Respectively, the blood triglyceride concentration was decreases on average by 3.7%, LDL cholesterol by 5.0% and total cholesterol by 1.0%. The change in total cholesterol was not statistically significant. The change in HDL cholesterol was inversely associated with the lipid concentration at baseline. Age, sex or change in VO2max were not associated with the changes in lipid profiles. There were no dose-response relationships between the different energy expenditures during exercise and the changes in lipid profiles.
Comment: The review included both randomized and non-randomized exercise studies. A part of the study subjects were overweight or obese and a part also had a dietary intervention. The majority of participants had no major deviations in their lipid profiles. However, they may have had the metabolic syndrome. Training interventions in dyslipemic patients (without coronary heart diseases or diabetes) were very few. The evidence of the beneficial effects of low-intensity exercise as a part of everyday activities on the serum lipids is insufficient.
Primary/Secondary Keywords