The use of resistance exercise (RE) to improve health-related fitness has increased considerably over the past few years (Centers for Disease Control and Prevention, 2006). The American College of Sports Medicine has recommended RE as an effective means of improving muscular strength and endurance, fat-free mass, and bone mineral density (Pollock et al., 1998). Also, RE is recommended by the American Heart Association as complementary to aerobic exercise to reduce cardiovascular disease risk factors (Williams et al., 2007). Previous studies have shown that RE can reduce postprandial lipemia 16 h postexercise (Pafili et al. , 2009; Petitt et al., 2003; Zafeiridis et al., 2007), but the effect of RE on total cholesterol (TC) and cholesterol sub fractions concentration remains controversial. There is limited evidence concerning the acute effect of RE on TC, but most studies have failed to show an effect (Hill et al., 2005; Jurimae et al., 1990; Magkos et al., 2008; Wallace et al., 1991). However, when eccentric contractions are performed and muscle damage ensues, TC has been reported to decrease not immediately after, but in the following one to four days post-exercise (Nikolaidis et al., 2008; Paschalis et al., 2010; Shahbazpour et al., 2004; Smith et al., 1994). All these authors partly attributed this to the extraction of cholesterol from the blood in order to assist in the resynthesis of disrupted muscle cell membranes.

However, the exact mechanisms are not known and it remains speculative whether this effect is due to muscle damage or to the exercise energy expenditure. When examining the relevant literature, one problem in comparing the effects of different RE protocols on blood lipids is that the characteristics of the exercise are not adequately quantified and described and also vary significantly between studies (Durstine and Haskell, 1994). Even though there is evidence that the effect of RE on blood lipids depends on exercise intensity and volume (Lira et al., 2009; Tucker and Silvester, 1996), the few studies that examined the effect of eccentric RE on TC and cholesterol sub-fraction concentration (Nikolaidis et al., 2008; Paschalis et al., 2010; Shahbazpour et al., 2004; Smith et al., 1994). have not measured energy expenditure or other cardiorespiratory responses. Also two of the aforementioned studies have not controlled (Smith et al., 1994) or inadequately controlled (Shahbazpour et al., 2004) the participants’ diet the days preceding the blood sampling and therefore their results cannot be safely attributed to the effect of exercise or to a possible change in participants’ dietary intake.

Therefore, the purpose of this study was to examine the cholesterol responses to a heavy leg press exercise session emphasizing on the eccentric movement previously shown to reduce fasting TAG 40 hours postexercise (Pafili et al., 2009). Since the effects of exercise on TC may not be obvious earlier than 24 h following exercise (Durstine and Huskel, 1994; Nikolaidis et al., 2008; Paschalis et al., 2010) we present unpublished data regarding the effects of a heavy resistance exercise emphasizing on the eccentric movement protocol on total cholesterol and cholesterol sub-fraction concentration up to 48 h post exercise. Another purpose of the study was to quantify and present the cardiorespiratory responses of the exercise bout.

Nine male volunteers (age: 27.2 ± 1.1 yr, body mass: 76.2 ± 2.5 kg, height: 1.79 ± 0.02 m) took part in the study, which was approved by the institutional ethics committee. Participants gave written consent after being informed about the procedures, the risks involved, and their right to terminate participation at will. All procedures in this study were in accordance with the Helsinki declaration of 1975, as revised in 1996. Participants were nonsmokers, without known history of cardiovascular disease, and were not taking any medication or nutritional supplements known to affect lipid or carbohydrate metabolism. In addition, participants were selected only if they had refrained from regular RE for the past 12 months. None of the participants performed more than 2 h·wk^–1 of recreational physical activities, (e.g., walking, cycling, swimming) as reported in personal interviews.

Each participant took part it two conditions separated by at least one week in random order. In the control condition participants consumed a standardized meal at 8:00 AM and seven hours later (15:00 AM) a baseline blood sample was obtained for the measurement of cholesterol and cholesterol sub-fraction concentration. On a separate day, one week later, the resistance exercise protocol was performed. The exercise protocol was timed so that it would finish at 15:00 AM. This was done so that the 24 and 48 h post-exercise blood samples were obtained at the same time of the day as in the control condition (15:00 AM) and after the consumption of the standardized meal at 8:00 AM on both post-exercise days.

To avoid the accustomization of the participants with eccentric exercise and the ensuing muscle damage, i.e. the so called “repeated bout effect”, (Clarkson and Hubal, 2002), the load corresponding to 6 RM was calculated for each individual using the fat-free thigh cross sectional area (CSA). On the day of the resistance exercise trial, the lean thigh cross sectional area was estimated from mid thigh circumference and the corresponding anterior thigh skinfold thickness, using a previously validated regression equation developed by Housch et al., 1995:

CSA = (4.68 × mid thigh circumference (cm) – (2.09 anterior thigh skinfold (mm)) – 80.99.

Then, the individual 6 RM load was calculated from CSA using the regression equation for untrained individuals developed by Dolezal et al., 2000:

Load = 1.231 × CSA + 173.664.

The predicted 6-RM weight was used for the first set and the participant was instructed to perform as many repetitions as possible. If the number of repetitions was different from six, then the load for the next set was adjusted according to the equation provided by Brzycki, 1993:

Care was taken so that the eccentric phase of the movement was emphasized.

The resistance exercise protocol was preceded by a standardized warm-up (5 min jogging on a treadmill at 8 Km·h^–1 and 5 min of stretching). Participants performed eight sets of six repetitions of leg presses that emphasized the eccentric movement the 6- RM load calculated from their CSA on a 45° inclined leg press machine. For each repetition, they were instructed to lower the weights in 4 s and to perform a forceful push when reversing from the eccentric to the concentric movement. This was done to maximize the eccentric load. A metronome was used to ensure that the 4 s phase was timed. A three minute break was used between sets. Two spotters ensured that the required repetitions were executed in all sets by assisting during the concentric phase when the participant was unable to lift the weight. The mechanical stops of the machine were adjusted so that knee angle (between femur and tibia) at the bottom of the downward movement was 60°. This exercise protocol has been used previously (Dolezal et al., 2000).

Throughout the exercise and for 3 min into the recovery, expired air was analyzed breath- by-breath using (Cosmed K4b² portable metabolic unit) and data were averaged every 5 s by the spirometer software. The spirometer was calibrated before each exercise protocol.

For the estimation of energy expenditure, aerobic plus anaerobic energy were summed, while protein contribution was ignored. Aerobic energy was calculated using an equivalent of 21.1 kJ per litre of oxygen consumed during exercise and 19.6 kJ per litre of oxygen consumed during recovery, while anaerobic energy was estimated by converting the increase in lactate concentration (from rest to the peak value) to oxygen equivalents assuming that one mmol·l^–1 of lactate is equal to 3 ml O>₂ kg^–1 body mass, oxygen values was then converted to kilojoules as 21.1 kJ per liter of O₂ (Scott, 2006).

Heart rate was continuously recorded using a Polar S610i heart rate monitor, and the recorded data were analysed with Polar Precision Performance SW Software (Version 4.00.024, Copyright Polar Electro Oy 2003, www.polar.fi).

Duplicate 20μL capillary blood samples were taken before exercise, in the middle of the rest period between sets 4 and 5, immediately after the last set and three minutes into recovery. Samples were deproteinized in 0.3 M pechloric acid and lactate was measured enzymatically at a later date (Sigma Diagnostics, St Louis, MO, USA).

Post-absorptive (7 hours after a standardized meal) blood samples were obtained in the control condition as well as after 24 and 48 hours after exercise. Samples (5 ml each) were taken from an antecubital vein and were placed in non-heparinized tubes to clot for 1 h. Serum was stored at –20 °C and was later analyzed for triglyceride, total cholesterol (TC), high density lipoprotein cholesterol (HDL-C) and creatine kinase (CK) using an automated analyzer (ACE^® Clinical Chemistry System, Alfa Wasserman, Inc, NJ, USA). Biochemical parameters were measured in duplicate. Intraassay coefficients of variation were as follows: lactate 3.1%, TAG, 1.9%; TC, 1.0%; HDL-C, 1.6% and CK, 2.0%. LDL was calculated according to the Friedewald formula (Friedewald et al., 1972).

To avoid the effect of previous diet and exercise on blood lipid values participants were asked to refrain from any form of physical exercise for 2 d before each condition, and to record their diet for 2 d prior to the first blood sample. This diet was replicated for 2 d before the other condition 1 wk later. In the exercise condition that required blood sampling for two consecutive days (24 and 48 h following exercise), food intake for the rest of the first day following exercise was prescribed individually so that the total daily energy and macronutrient intake for each participant was the same as in the previous day. By doing this, energy intake and diet composition were standardized for the day before each blood sample. Additionally, the last meal before each data collection day was administered according to body mass 7 hours before blood sampling. The cholesterol content of that meal was 0.42 mg/kg body mass.

Data were analysed using one-way ANOVA and differences between means were located using Tukey’s post hoc tests. Data was analysed using STATISTICA for Windows StatSoft, Inc. (Release 5.0, Tulsa, OK). Significance was accepted at the p < 0.05 level, and data are presented as mean ± SE, unless otherwise stated.

The total exercise duration was 25.4 ± 0.2 min. The average weight lifted per contraction was 2.80 ± 0.21 kg/kg body mass while the average angular velocity of the knee joint movement during the eccentric phase was 30°·s^–1 (as calculated from the total change in knee angle (from 180° to 60°) in the timed period of 4 s.) Total exercise energy expenditure (from the beginning to the end of exercise) was 136.72 ± 7.11 Kcal (7.86 Kcal·min^–1). Average energy expenditure during the 8 sets of exercise, the subsequent eight 3 min recovery periods and exercise + recovery is shown in Figure 1. The average lactate contribution to energy expenditure during exercise was 13.63 ± 1.70 Kcal.

The average oxygen consumption and the average metabolic energy equivalents (METs) during rest, the 8 sets of exercise, the subsequent 3 min recovery periods and exercise + recovery are shown in Figure 2. Since pure exercise duration per set was only about 30 s and the recovery period per set was 3 min, the arithmetic means of pure exercise + recovery oxygen consumption and MET's differ from the exercise + recovery values. The time course of oxygen uptake, CO₂ output and RER during the whole exercise session is shown in Figure 3. At this point it must be noted that the protocol of the study required the participants to perform as many repetitions as possible against a calculated 6RM load for the first exercise test. In fact, the number of repetitions performed in the first set was 13.6 ± 2.8 and this increased the duration of the first set to about 59.5 sec. Then the load for the next sets was readjusted to the actual individual 6RM. The increased number of repetitions performed in the first set, resulted in significantly higher oxygen uptake and CO₂ production during that set (Figure 3). Thereafter oxygen consumption remained unchanged throughout the test protocol (Figure 3). However, RER values decreased with the progression of the exercise and were significantly lower during the last four sets compared with the first four sets of the exercise session both during exercise (1.12 ± 0.06 vs. 1.00 ± 0.05 for first and last four sets respectively, p < 0.01), and recovery (1.48 ± 0.07 vs. 1.15 ± 0.05 for first and last four sets respectively, p < 0.01).

Average heart rate during the exercise session is shown in Figure 4. Rest average HR was 67 ± 2 bpm. Maximum heart rate during exercise was 168 ± 1 bpm. During exercise periods average HR was 152 ± 2 bpm, while during recovery periods average HR was 133 ± 1 bpm. The intensity of exercise as determined by % predicted HR max (as calculated from the formula: HR max= 220- age (yrs)), was 78.2 ± 2.1% HR max.

CK concentration was significantly higher on both days after exercise compared with the control value (197 ± 57 vs. 1090 ± 272, p < 0.03 and 1230 ± 440 U·L^–1 p < 0.01, for control, 24h and 48h following exercise, respectively).

TC, HDL cholesterol and calculated LDL cholesterol concentration did not change significantly over the 48 hours following exercise (Figure 5).

The main findings of this study were that this type of heavy strength exercise increased energy expenditure from aerobic sources to relatively high levels, considering the actual work done. Furthermore, total cholesterol and its sub-fractions HDL-C and LDL-C were not affected by this exercise protocol that caused considerable muscle damage.

The characteristics of this exercise program were typical of a maximal strength training session with heavy load (6 RM) and long rest periods (3 min) (Kraemer and Ratamess, 2004; Kraemer et al., 1987). The METs expended during exercise were equivalent to bicycling at <10mph (Ainsworth et al., 2000), and according to the oxygen uptake, exercise may be considered as “moderate intensity” according to ACSM's guidelines for aerobic exercise (Haskell et al., 2007). However, due to the nature of resistance exercise, HR responses were significantly higher than those of aerobic exercise of similar oxygen uptake (Braun et al., 2005). According to the energy calculations from VO₂ and blood lactate, this heavy RE session increased energy expenditure to relatively high levels, considering that only 50 repetitions were performed during the 8 exercise sets. It must be noted that the largest part of the aerobic energy expenditure was done during the recovery periods where oxygen uptake remained elevated.

Although a previous report from our laboratory showed a decrease in postprandial lipemia 16 hours after this exercise protocol (Pafili et al., 2009), TC concentration remained unchanged at both 24 and 48 h postexercise compared with rest. These results are in agreement with the results of previous studies of resistance exercise (Hill et al., 2005; Jurimae et al., 1990; Magkos et al., 2008; Wallace et al., 1991), and with results from studies of eccentric resistance exercise (Nikolaidis et al., 2008; Paschalis et al., 2010) as far as 24 h postexercise values are concerned. However, in the study of Nikolaidis et al., 2008 TC concentration was significantly reduced 48 h postexercise following the first exercise bout. This was attributed to a possible use of blood cholesterol for the repair of damaged muscle membranes, and this argument was reinforced by the fact that when the same exercise was performed again after 24-30 days there was no change in TC and this was accompanied with much less muscle damage, i.e. the “repeated bout effect” (Nikolaidis et al., 2008). In the other study by the same research group the more pronounced effect of the exercise on TC concentration in overweighed participants were also attributed to greater muscle damage (Paschalis et al., 2010).

The hypothesis of muscle membrane repair using blood cholesterol is supported by recent studies that examined the gene expression profiling in human skeletal muscle during recovery from eccentric exercise. These studies show that following eccentric exercise a number of genes involved in lipid metabolism are activated, some of them potentially promoting cholesterol synthesis in order to fix the disrupted membranes (Mahoney et al., 2008). However, muscle damage in the present study as well as in the other relevant studies was quantified with indirect indices, and thus the results may not be directly comparable. The indirect index of plasma CK concentration was lower in the present study compared with the study of Paschalis et al., 2010, but greater compared with the study of Nikolaidis et al., 2007. Therefore, a safe conclusion cannot be drawn regarding the possible effect of the degree of muscle damage on the serum concentration of TC and its sub-fractions. Furthermore, the possible effects of energy expenditure on TC concentration can not be compared with previous studies, since none of them measured energy expenditure.

Another factor that may have affected the responses of the participants in the present study is the longer rest interval employed (3 min) compared with the other studies that have used shorter rest intervals: 2 min Smith et al., 1994, 1 min Shahbazpour et al., 2004, 2 min Nikolaidis et al., 2008 and Paschalis et al., 2010. Rest-interval length could affect the lipoprotein response to exercise since it has been shown that RE with shorter rest intervals affects more the post exercise leukocyte levels (Mayhew et al., 2005), and the hormonal responses to exercise (Bottaro et al., 2009; Buresh et al., 2009), compared to the same exercise with longer rest intervals. The increase in leukocyte levels after exercise is considered a part of the inflammatory response to injured muscle cells (Nieman, 1997) and there is evidence to suggest that it is intermediated by exercise-dependent secretion of stress hormones (Suzuki et al., 1999). Both inflammation and increased hormone secretion following exercise have been shown to affect lipoprotein levels (Frey et al., 1983; Khovidhunkit et al., 2004), so a greater increase in both these parameters could also mean a greater effect on blood lipoproteins.

It must also be noted that insulin resistance observed after eccentric exercise (Asp et al., 1996; Kirwan et al., 1992; Pafili et al., 2009) could increase blood cholesterol, since it has been associated with increased cholesterol synthesis and decreased cholesterol absorption in normoglycemic men (Pihlajamäki et al., 2004). This may counterbalance the possible reduction of TC due to the repair of damaged muscle membranes.

Another interesting possible explanation for the differences between our study and the studies of Nikolaidis et al., 2008 and Paschalis et al., 2010 is the interaction between resistance exercise and gender of the participants. Both those studies used female participants whereas in the present study the participants were males. Recent reviews (Tambalis et al., 2009) and meta-analyses of previous studies (Kelley and Kelley, 2009) suggest that the only studies that have shown an effect of RE training on blood TC concentration had female and elderly participants. Since there are studies that have shown different responses between men and women as far as the acute effects of eccentric RE are concerned (Rinard et al., 2000; Sewright et al., 2008), and even though hormonal responses have not been measured in response to eccentric resistance exercise in women, evidence from aerobic exercise studies suggest that men and women have different hormonal responses to exercise (Davis et al., 2000), so gender may be one important factor for changes in blood TC with heavy RE.

Previous studies have shown an increase of HDL concentration 24 h following RE only when exercise was of high volume and of longer duration than the present study (e.g. 90 min, Wallace et al., 1991). This fact, combined with evidence from aerobic studies showing that exercise energy expenditure needs to be at least 350 Kcal in order to have an effect on HDL concentration (Thompson et al., 2001) suggests that the energy expenditure of our protocol was not sufficient to affect HDL concentration.

In conclusion, a small volume of heavy resistance exercise that caused muscle damage had no effect on TC and cholesterol subfraction concentration 24 and 48 hours after exercise, even though it has previously been shown to reduce postprandial triglyceride concentration 16 hours following exercise. This is in contrast with the few studies that reported a decrease in TC in two up to five days after muscle damaging exercise. The lack of an effect of exercise on TC and HDL-C in the present study may be due to a lower degree of muscle damage, lower energy expenditure or a possible effect of gender on the TC responses. Our results point out the necessity of future studies to control for exercise characteristics and take into account factors such as participants’ gender, diet and hormonal responses that could affect the cholesterol response to exercise, in order to draw a sound conclusion regarding the effects of resistance exercise on blood lipids.