Journal Logo


Twenty-Hour Growth Hormone Secretory Profiles after Aerobic and Resistance Exercise


Author Information
Medicine & Science in Sports & Exercise: October 2014 - Volume 46 - Issue 10 - p 1917-1927
doi: 10.1249/MSS.0000000000000315
  • Free


Exercise-induced elevations in circulating growth hormone (GH) have been well documented since the initial report of Hunter et al. (8) in Science. Exercise intensity and duration, fitness level, age, gender, nutritional status, and body composition all influence the GH response to acute exercise (10,14,19,21,34–36). As GH possesses pleiotropic properties encompassing both anabolic and lipolytic actions (5,23,28,33), the GH response to exercise is of basic, applied, and clinical scientific interest to better understand those specific muscle and adipose tissue adaptations resulting from altered physical activity patterns. Among the most desired outcomes from physical training, concomitant reductions in fat mass and increases in fat-free mass are thought to be at least partially mediated through GH and its somatogenic actions in exercising humans (16,19,21,37). It has been the suggestion of investigations that both the structure of the GH molecule and the amplitude its pulsatile release may be more physiologically important in driving responses at target tissues rather than sheer circulating concentrations (11,16,19,25).

Further mechanistic insight into the GH response to exercise can potentially aid in exercise prescription guidelines. For example, Weltman et al. (34) have demonstrated that postmenopausal women who regularly exercised above the lactate threshold for a year experienced significant increases in 24-h GH secretory release (i.e., total overall secretion rate and increased GH peak and mean amplitude), and this coincided with greater fat mass loss. More recently, the Weltman laboratory reported that both continuous and intermittent acute aerobic exercise totaling 30 min in duration were equally effective in increasing 24-h GH secretion (35). Further, although very few studies have examined exercise bouts >30 min, Wideman et al. (36) have demonstrated that when intensity is kept constant, the GH response to aerobic exercise when assessed during a 6-h period exhibited a duration dose–response relationship (120 > 60 > 30 min). This study is especially intriguing when one considers that exercise prescription guidelines from the 2005 Institute of Medicine of the National Academies have suggested more than 30 min of exercise a day, and perhaps as much as 60–90 min is necessary for optimal health and weight loss (18). Furthermore, a 2009 American College of Sports Medicine Position Stand also advocates for exercise of greater than 250 min·wk−1 for clinically significant weight loss (2). The clinical implication for an exercise-induced amplification in GH secretion serving as one mechanism contributing toward favorable body composition alterations after physical training is clear. It is also clear from the literature that in contrast to aerobic exercise, studies examining GH secretory dynamics after resistance exercise are lacking (30,37). We believe that further dose–response studies are required that examine GH secretion to both aerobic and resistance modes of exercise and may aid in the refinement of exercise prescription guidelines.

Because GH is released in a pulsatile, episodic manner throughout the day and this pattern of release is essential in optimizing its physiological cellular effects (i.e., pulsatile GH and increased pulse amplitude) results in greater increases in local insulin-like growth factor I and lipolysis than continuous exposure) (11,16,28), it is important to use frequent and prolonged blood sampling to accurately characterize the full influence that exercise may exhibit on GH-mediated outcomes. Using time series software, underlying hormonal secretion rates and pulsatile release characteristics can be accurately determined from serial measurements in the circulation (31,32). Although there are reports in the literature describing the impact of exercise on GH secretion using these analyses, these are usually limited by the inclusion of only one mode of exercise (i.e., aerobic) and sampling time courses of ≤12 h. To date, no studies have examined GH secretion recovery patterns for various bouts of aerobic and resistance exercise within the same experimental paradigm. We hypothesized that intensity-controlled aerobic and resistance exercise bouts of 2 h in duration would result in greater GH secretion during a 20-h recovery period than exercise bouts of 1 h in duration and that this dose–response relationship would be quantitatively similar within exercise modes. The primary aim of this investigation was to gain further mechanistic insight with regard to GH secretion patterns within the context of an exercise paradigm. Specifically, the purpose of this study was to quantify GH secretion parameters using deconvolution software during a 20-h recovery period after various acute exercise regimens (aerobic and resistance exercise bouts of two different durations: moderate (1 h of exercise) and long (2 h of exercise). This study was undertaken with the supposition that exercise-induced alterations in GH pulsatile secretion from the anterior pituitary gland represent an important modifiable physiological network that serves to modulate many of the somatogenic actions of altered physical activity patterns.


Experimental Design

A within-subjects, counterbalanced repeated-measures research design was implemented, in which subjects randomly completed five independent conditions: 1) control (no exercise; CON), 2) a moderate-duration aerobic exercise session (MA), 3) a long-duration aerobic exercise session (LA), 4) a moderate-duration resistance exercise session (MR), and 5) a long-duration resistance exercise session (LR) (refer to Fig. 1). Strictly controlled and quantified diet, sleep, and physical activity levels were superimposed on each of these conditions so that any potential differences in GH secretion could be solely attributed to the effects of exercise and not of other factors known to influence GH secretory dynamics.

Schematic of the experimental design for the study displaying the experimental conditions, information on the study conditions during each overnight trial, and the experimental questions that the design was intended to address.


Eight healthy, recreationally trained (i.e., physical activity participation of three to five times per week, but not involved in competitive athletics) male volunteers (24 ± 5 yr, 182 ± 6 cm, 87 ± 9 kg, 21%BF ± 5%BF) volunteered to participate in the study (refer to Table 1 for description of physical and fitness characteristics of the subjects). The study was approved by the Human Use Review and Scientific Review Committees at the U.S. Army Research Institute of Environmental Medicine (Natick, MA) and by the Human Subjects Research Review Board of its parent organization, the U.S. Army Medical Research and Materiel Command (Fort Detrick, MD). Volunteers were briefed on the study purpose, methods, and procedures both orally and in writing, and written informed consent was obtained before participation. Each subject was medically screened by a physician before inclusion in the study. From the screening procedure, subjects were determined to be nonsmokers and free of any endocrine (i.e., glandular hypo/hyper secretion) and orthopedic (i.e., limited range of motion from preexisting injury) disorders. None of the subjects had taken supplements (i.e., nutritional and steroidal) during the preceding month (30 d) or had donated blood within the last 56 d.

Baseline subject physical and fitness characteristics.

Overnight Trials

For each of the conditions (one per month), all subjects slept overnight (two consecutive nights) in the metabolic/sleeping laboratory located in the Doriot Climatic Chamber facility at the Natick Soldier Systems Command (Natick, MA). All volunteers slept in the facility the night before each overnight serial blood draw to facilitate familiarization with the facility. Therefore, subjects had two overnight stays on each of the five conditions. One of these conditions served as a nonexercise control session. During each overnight trial, subjects were allowed to move freely, to watch television, to receive telephone calls, to read quietly, and so on. Bedroom lights were turned off at 2200 h, and the television was turned off at 2300 h. Subjects were awakened in the morning at 0600 h. All conditions were separated by at least 30 d.

Dietary Control

Subjects were instructed to maintain their normal diet during the 30-d period between each of the five sessions. All subjects completed 3-d dietary intake records (before each trial). Subjects were asked to replicate their first 3-d dietary intake before their subsequent (second through fifth) overnight visits. Dietary analyses (Nutritionist IV; First DataBank, San Bruno, CA) of these records verified that the caloric content and macro/microcomposition were similar for the 3 d before each overnight stay (20). Standardized meals were provided to all subjects the day before and the day of the trial. These meals were prepared by registered dietitians at the Massachusetts Institute of Technology (MIT) general clinical research center. These meals conformed to the following criteria: no caffeine, aspartame, or snacks; macronutrient distribution was 50% carbohydrate, 20% protein, and 30% fat; and sodium was controlled at 3 g. Caloric intake was calculated on the basis of the estimated 24-h caloric expenditure (basal metabolic needs plus physical activity) derived from resting metabolic rate measurements (20,22). The resting metabolic measures were obtained during the screening phase of the study. Meal times were breakfast at 0800 h, lunch at 1200 h, and dinner at 1900 h, and these times were consistent throughout the study. Lunch and dinner times were scheduled around the 1500- to 1600-h start time for the afternoon exercise workouts to ensure all subjects exercised in the postabsorptive state. Subjects consumed water ad libitum.

Preliminary Performance/Baseline Testing

One-repetition maximums (1-RMs) were tested for the following exercises (in this order): squat, bench press, leg press, and lat pulldown using both free weight (Sorinex, LA; back squat and bench press) and machine equipment (BodyMaster, Rayne, LA; leg press and lat pulldown). Our methodology for maximal strength testing has been reported elsewhere (22) with all 1-RM measurements performed on the same day. The mean 1-RM values for all tested exercises are presented in Table 1.

Maximal oxygen uptake (V˙O2peak) was determined using a discontinuous, progressive protocol on a cycle ergometer (Preference HRT-2000R). Subjects pedaled at a rate of 80 rpm against a load of 75 W for a period of 5–6 min. On the next stage, subjects pedaled at 80 rpm against a resistance of 150 W. On subsequent stages, resistance on the cycle ergometer was increased by 50 W while maintaining 80 rpm. Each stage lasted approximately 4–5 min, and the rest between each stage was 5–7 min. Midway into each stage, expired air was collected with an online metabolic measurement system consisting of a turbine (KL Engineering, Northbridge, CA) to measure expired gas volume (Applied Electrochemistry S-3A and Beckman LB-2) and gas analyzers to measure concentrations of oxygen and carbon dioxide, respectively, in the expired air. Heart rate was monitored via a three-lead ECG. The test was terminated upon volitional exhaustion or the attainment of a plateau in V˙O2 defined as less than 2.0 mL·kg−1·min−1 increase through two successive exercise loads. The mean V˙O2peak for the subjects is presented in Table 1.

The load corresponding to 70% V˙O2peak was estimated by a close examination of the physiological data (i.e., heart rate and energy cost) from each stage. The subjects were brought back on another day to confirm the selection of the load that would be used during the aerobic exercise sessions, and adjustments were made as needed.

Exercise Protocols

Acute heavy resistance exercise protocol.

As we have reported previously (20,21,25,30), the acute heavy resistance exercise protocols were designed to recruit and activate a large amount of muscle tissue. This was accomplished by performing multijoint exercises that required the use of large muscle groups in both the lower and the upper body (i.e., squat, leg press, bench press, and lat pulldown). These exercises were performed on the same apparatus as the 1-RM strength testing. The relative loads for each exercise alternated between 10- and 5-RM loads. The 10- and 5-RM loads were calculated as 70% and 85% of the 1-RM exercise, respectively. The exercises alternated between the squat and the bench press (five sets total for the 25-set protocol and 10 sets total for the 50-set protocol) then alternated between the leg press and the lat pulldown (5 sets total for the 25-set protocol and 50 sets total for the 50-set protocol). Three rotations were performed for the squat and bench press exercises, and two rotations were performed for the leg press and lat pulldown exercises. The exercise set was terminated upon the achievement of the number of repetitions at the particular RM load or muscle failure. If the desired number of repetitions was not achieved, the load was reduced before the commencement of the next set of that exercise. To provide the needed recovery, a 90-s rest period was given after each exercise as well as alternating muscle groups. Spotters were used to ensure the safety of the volunteer and also to offer motivation and encouragement throughout the workout. The moderate-duration resistance exercise condition was comprised of 25 total sets and lasted approximately 1 h, whereas the long-duration resistance exercise condition was comprised of 50 total sets and lasted approximately 2 h. As mentioned previously, all sessions ended at 1700 h.

Acute aerobic exercise sessions.

The acute aerobic exercise protocols were performed at 70% V˙O2peak on the same cycle ergometer that was used to determine the peak oxygen consumption. The moderate-duration aerobic exercise session consisted of three, 15-min bouts, and the long-duration aerobic exercise session consisted of six, 15-min bouts both performed at 70% V˙O2peak. Each bout was separated by a 5- to 7-min rest interval. Thus, the moderate-duration aerobic exercise bout lasted approximately 1 h (45 min of actual exercise), whereas the long-duration aerobic exercise bout lasted approximately 2 h (90 min of actual exercise). For all bouts, subjects exercised at approximately 70% of their previously determined V˙O2peak. Heart rate was monitored with a three-lead ECG (Hewlett Packard, Waltham, MA), and oxygen uptake was measured during the last 3 min of each bout to validate exercise intensity.

Blood Collection and Processing

Venous catheters were inserted at approximately 1700 h for each condition (immediately after exercise for the exercise conditions), and serial blood draws were collected every 20 min (q-20) for a 20-h period (60 total time points). The catheter lines were kept patent with a saline drip that was closely monitored by the nursing and phlebotomy staff. Blood (1.5 mL) was collected at each time point drawn for later analysis, and less than 1 mL of blood was used to clear the line for each time point. The total amount of blood drawn for GH analyses was approximately 150 mL for the 20 h period. After blood was drawn for each time point, it was allowed to clot at room temperature and centrifuged for 30 min at 1500g at 4°C. After centrifugation, serum was aliquoted into microfuge tubes. All serum samples were kept frozen at –80°C until analysis was performed.

Hormone Analyses

GH was measured using a chemiluminescent immunoassay on the Immulite (Diagnostic Products Corporation, Los Angeles, CA). Immunoassay sensitivity for GH was 0.01 ng·mL−1. Intra- and inter-assay variances were 6% and 5.5%, respectively, for GH. Assays were performed in duplicate, and each subject’s sample was run in one batch as to minimize interassay variance.

Cluster Analysis

The cluster analysis program was used to analyze the overall mean GH, 20-h integrated area under curve (AUC) via trapezoidal rule and characteristics of concentration “peaks” (32). The cluster algorithm searched for significant increases and decreases among data points in a series via pooled t-tests. We configured the algorithm using two points for the determination of a peak and one point to establish a nadir. A peak was defined as a series of concentrations that demonstrated an increase over time followed by a decrease, with the additional requirement that a nadir (i.e., a decrease followed by an increase) was present on each side of the peak. Cluster analysis was used to provide descriptive information regarding observed concentrations as a function of time, such as the number of peaks in 20 h, the mean interval between peaks, the mean peak width (i.e., duration), the mean peak height (i.e., amplitude), and the area under peaks.

Deconvolution Analysis

Multiple-parameter deconvolution analysis was used to estimate the secretion and elimination characteristics of the hormones based on the GH assay-measured concentrations. This method used a convolution integral, which is solved by nonlinear least squares parameter estimation (31). For this analysis, we used the same data file that was configured and used for the Cluster program. Half-life secretion burst number, interval between bursts, burst area, burst amplitude, amplitude of largest burst, basal secretion per minute, basal secretion (basal secretion × 1200 min), pulsatile secretion (burst area × secretion burst number), and total overnight secretion (basal secretion + pulsatile secretion) were determined for GH.

Body Composition

Percent body fat was measured at the beginning of the study using dual-energy x-ray absorptiometry (Lunar, Madison, WI). Scanning was done in 1-cm slices from head to toe using the 20-min scanning speed. Total body estimates of body fat percent were determined using manufacturer-described procedures and supplied algorithms (Total Body Analysis, version 3.6; Lunar).

Statistical Analyses

All results are reported as mean ± SE. A repeated-measures ANOVA with was used to determine the effects of condition (control, moderate volume resistance, high volume resistance, moderate volume aerobic, and high volume aerobic) on the subsequent hormonal responses using an exploratory P value <0.10. Subsequent post hoc analyses were performed, where appropriate, using a Tukey’s honestly significant difference test using a P value of P < 0.05. A power analysis using the Russ Lenth power calculator was conducted on the basis of the variables deemed most physiologically important (mean burst area, mean burst amplitude, peak burst amplitude, total pulsatile secretion, and total overall secretion) using a 50% change compared with the control condition and indicated a sample size of 6 to 8 was required to for a 0.80 power. All analyses were performed with Statistica software packages (StatSoft, Tulsa, OK).


Quality/integrity of exercise protocols.

The intent of the exercise experimental design was to keep intensity similar between the moderate- and the long-duration exercise protocols. Table 2 presents all of the physiological data for both moderate- and long-duration acute exercise for the resistance and aerobic exercise sessions. For both the moderate- and the long-duration resistance exercise bouts, similar repetition maximum (RM) sets as a percentage of their maximum were observed for the 10- and 5-RM sets for squat, 10- and 5-RM sets for leg press, 10- and 5-RM sets or lat pull downs, and for the 10-RM sets for bench press for the moderate- and long-duration exercise bouts, respectively. For the 5-RM sets for the bench press, the long-duration was significantly less than the moderate duration (70.5% ± 15.4% vs 82.2% ± 11.5%), respectively. For the aerobic exercise bouts, similar values were observed for V˙O2, V˙E, RER, HR, load, and RPE. Thus, these data demonstrate that the intensity for both aerobic and resistance exercise was similar during the moderate- and long-duration exercise bouts. The energy expenditure of the exercise protocols for the long-duration exercise session was approximately twice that of the moderate-duration exercise sessions for both aerobic (1328 ± 161 vs 667 ± 48 kcal) and resistance (793 ± 97 vs 403 ± 48 kcal) exercise modes.

Exercise intensity variables during the acute moderate- and long-duration resistance and aerobic exercise conditions.

Cluster analysis.

Cluster analysis parameters (mean GH [ng], total GH area [ng·mL−1·min−1], peak height [ng], peak number, peak width [min], and peak area [ng·mL−1·min−1]) for GH concentrations across the experimental conditions are presented in Table 3. Significant differences were observed within the aerobic exercise conditions. During both moderate- and long-duration exercise conditions, mean GH and total GH area were higher than the control conditions. In addition, mean GH (2.02 ± 031 vs 1.5 ± 0.13 ng) and total GH area (2303.1 ± 357.9 vs 1659 ± 158 ng·mL−1·min−1) were higher in the long-duration versus moderate-duration aerobic exercise condition, respectively. No significant differences were observed for the resistance exercise conditions.

Cluster analysis parameters for GH concentrations across the experimental conditions (n = 8).

Deconvolution analysis.

Deconvolution parameters (half-life [min], number of bursts, interval between bursts [min], mean area under bursts [ng·mL−1], mean amplitude of bursts [ng·mL−1], peak burst amplitude [ng·mL−1], basal secretion rate [ng·mL−1·min−1], total basal secretion [ng·mL−1·20 h−1], total pulsatile secretion [ng·mL−1·20 h−1], and total secretion [ng·mL−1·20 h−1]) are presented in Table 4. The long-duration aerobic exercise condition had higher GH peak burst amplitude (1.05 ± 0.09 vs 0.54 ± 0.08 ng ), GH basal secretion rate (0.0025 ± 0.0005 vs 0.0011 ± 0.0003 ng·mL−1·min−1 ), GH basal secretion (3.02 ± 0.59 vs 1.37 ± 0.36 ng·mL−1·min−1), GH total pulsatile secretion (61.77 ± 13.54 vs 32.89 ± 4.35 ng·mL−1·min−1), and GH total secretion (64.79 ± 13.83 vs 34.26 ± 4.59 ng·mL−1·min−1 ) when compared with the control condition. Further, the long-duration aerobic exercise condition had a significantly higher peak amplitude (1.05 ± 0.09 vs 0.75 ± 0.13 ng) and pulsatile (61.77 ± 13.54 vs 39.55 ± 5.98 ng·mL−1·min−1) and total secretion (64.79 ± 13.83 vs 41.50 ± 6.16 ng·mL−1·min−1; P = 0.07) than the moderate-duration aerobic exercise condition.

GH secretion parameters assessed by deconvolution analysis for the experimental conditions (n = 8).

Figure 2 illustrates a representative cluster and deconvolution profile for one representative subject measured q 20 for 20 h across a control condition (CON), moderate-duration (1-h) aerobic exercise condition (MA), long-duration (2-h) aerobic exercise condition (LA), moderate-duration (2-h) resistance exercise condition (MR), and long-duration (2-h) resistance exercise condition (LR). Note that the highest GH peak and the largest GH total area are observed for the LA exercise condition.

Representative (data from subject 7) cluster-analyzed growth hormone (GH) concentration peak profiles (left panels) and their associated deconvolved secretion events (right panel) for control, moderate-duration (1-h) aerobic exercise, long-duration (2-h) aerobic exercise, moderate-duration (1-h) resistance exercise, and long-duration (2-h) resistance exercise conditions. Cluster and deconvolution analyses were performed on data measured every 20 min for 20-h from 1700 day 1 until 1300 day 2 (note that exercise was performed on day 1 during 1500–1700). Hence, these data represent post-exercise recovery profiles. Note that the individual data from representative subject 7 parallels the group mean data in that the long-duration aerobic exercise condition visually presents greater GH secretion.


Although the physiological importance of the pulsatile secretion of GH has been firmly established in both rodent and human models by comparing continuous versus intermittent patterns of exogenous delivery of GH and by demonstrating differential growth (i.e., liver insulin-like growth factor I production) and metabolic (i.e., lipolysis in adipose tissue) actions in GH-sensitive peripheral tissues (11,16,28), more information is required in regard to exercise. In particular, increased peak and mean pulse amplitudes appear to be the characteristics than elicit greater target tissue effects. The emergent question of importance to the health and welfare of the public is to what extent physical activity can maintain optimal GH secretion profiles throughout the life span. The plasticity of the endocrine system, specifically GH secretion from the anterior pituitary gland, is being increasingly recognized by an accumulating body of literature demonstrating that exercise of sufficient intensity and duration can amplify GH secretion during a prolonged postexercise recovery period (12,34,35,37). The observation in the current study that the most prominent amplification of GH secretion was observed after 2 h of aerobic exercise at an intensity of ∼70% V˙O2peak would appear to provide empirical endocrinological data in support of the 2005 Institute of Medicine physical activity guidelines, suggesting that more than 30 min (perhaps as much as 60–90 min) of vigorous activity a day are required for optimal health, fitness, and weight maintenance (18). Also, the 2009 American College of Sports Medicine Position Stand suggests advocating more than 250 min·wk−1 for clinically significant weight loss (2), if GH were to mediate the positive associative changes after exercise of such intensity and duration. We further believe that the physiological significance of GH secretory amplification during recovery from the long-duration aerobic exercise resides in the metabolic role that GH possesses (4,6,12,24,29) to provide adequate fuel for postexercise restorative processes rather than the conventional anabolic role that is most typically associated with postexercise increases in GH.

A clear dose–response relationship was only observed during recovery for mean 20-h GH concentrations in the aerobic exercise mode (CON < MA < LA; P ≤ 0.05) and for GH total area (CON < MA < LA). More importantly, however, the long-duration aerobic exercise mode resulted in a significant amplification of GH secretion as evidenced by increased GH burst peak amplitude, basal GH secretion rate, total GH basal secretion, total pulsatile secretion, and total GH secretion over the control (i.e., no exercise) condition. In addition, the long-duration aerobic exercise bout also resulted in an increase for GH burst peak amplitude, total GH pulsatile secretion, and total GH secretion during the moderate-duration aerobic exercise session.

Although there are many reports of acute (<6 h) GH recovery responses to aerobic exercise in the literature (35,37), only one other report is available that examined the effect of acute aerobic exercise on GH secretion for an entire day. Kanaley et al. (12) measured GH every 10 min for 24 h in young men after both sequential aerobic exercise and after delayed aerobic exercise and reported increases during daytime, but not nocturnal, GH secretion. This increase was mechanistically attributed to an increased GH pulse amplitude and mass of GH secreted per pulse. Although the data from the current study and that of Kanaley et al. (12) are in general agreement and similarly conclude that aerobic exercise lasting a total duration of ∼90 min results in greater GH secretion for an extended period (>6 h postexercise), several methodical differences between the two studies deserve mention. First, we made a deliberate decision not to measure GH release during the exercise bouts with the rationale that the well-known exercise-induced GH bolus of release might bias our results toward a GH secretion amplification, which might mainly be attributable to GH release during exercise rather than GH secretory release during the postexercise recovery period (the study question herein). Second, Kanaley et al. (12) reported data on total GH secretion, whereas the report herein also reports on the underlying components of total secretion (basal secretion + pulsatile secretion). Third, although subjects in both studies actually exercised for the same total length of time (90 min), at the same intensity (70% max V˙O2), and with identical modes (cycle ergometry), the exercise exposures were administered in dissimilar manners (15-min exercise bouts conducted for a 2-h period in the study herein, whereas Kanaley et al. (12) conducted 30-min exercise bouts for a period of either 3 or 8 h). Nonetheless, we believe both studies are congruent in demonstrating that although the acute exercise-induced increase in GH is known to restore toward baseline values within a short period (2 h), extended blood sampling combined using deconvolution secretory analysis offer a more robust assessment of underlying glandular secretion events and actually indicate that aerobic exercise of sufficient intensity and duration can significantly amplify GH secretion. Our data extend upon the previous report by demonstrating that both basal and pulsatile secretions are increased and that these increases are perceptible for a 20-h postexercise recovery period.

In contrast to the 2-h (90 min of exercise) aerobic exercise bout, no significant alterations in GH secretory dynamics were observed after either 1 or 2 h of resistance exercise. Since the seminal work of Kraemer et al. (14,15) reporting the influence of acute resistance program variables on subsequent growth factor responses, the implication has been that the GH response to resistance exercise was linked with anabolism via protein accretion/muscle hypertrophy. In stark contrast to aerobic exercise, very little has been reported with regard to GH secretion after resistance exercise (7,21,30,36,37). Our laboratory has published the only other GH pulsatility/secretion results after resistance exercise (25,30). Important differences between our current and previous reports include the following: differing sampling frequencies (every 20 vs every 10 min), differing sampling durations (20 vs 12 h), and differences in subject characteristics (the subjects in the current study were less fit than our previously published study as indicated by body composition, strength, and aerobic fitness measures).

An important aspect of GH physiology that was not considered in this investigation is the degree to which GH exhibits a great deal of molecular heterogeneity (7,8,15,21,25). This study measured the main immunoreactive GH isoform (i.e., 22 kDa), and GH is known to circulate in more than 100 different isoforms. Hymer et al. (9), Kraemer et al. (14), and Pierce et al. (25) have all published reports demonstrating that acute resistance exercise can alter the relative proportions of circulating GH isoforms (i.e., greater increase in disulfide-linked GH isoforms). Further, McCall et al. (17) propose that afferent input from skeletal muscle proprioceptors can modulate bioassayable GH from the anterior pituitary. Thus, it is possible that muscle contractile activity related to either muscle fiber composition or muscle type can elicit the muscle-specific regulation of certain biologically active GH molecules that would not be identified by deconvolution secretory analyses, which rely on the use of 22-kDa specific antibodies used in the current study.

The finding that 2 h (90 min of exercise) of aerobic exercise bout, but not resistance exercise, resulted in an amplification of GH secretion may at first seem somewhat surprising, but one explanation for this result may be the differences in energy expenditure between the exercise modes. The greater energy expenditure from the aerobic exercise bout would obviously necessitate a greater disruption in metabolic homeostasis and place a larger demand for mobilizing postexercise energy stores via lipolysis (and possibly other routes of gluconeogenesis). There is strong evidence from the literature linking the exercise-induced GH release with lipid mobilization/fat metabolism (4,6,10,27,34–37). Elevated glycerol concentrations, indicative of lipolysis, have been observed in both abdominal and femoral adipose tissue interstitial fluid after intravenous GH administration (to mimic a physiological pulse) versus a NaCl placebo solution (6). In GH-deficient patients, Kanaley et al. (12,13) have demonstrated that acute GH exposure during exercise stimulates free fatty acid release and turnover. They concluded that exercise-induced GH secretion plays a significant role in the regulation of fatty acid metabolism. Gibney et al. (3) used a double-blind, placebo-controlled design using subjects who had received long-term GH replacement and observed that GH withdrawal for 3 months resulted in reduced release of glycerol and nonesterified fatty acids during intense exercise. They noted that these changes were accompanied by decreased lean mass and increased total body and trunk fat. Using healthy, moderately trained volunteers, Wee et al. (33) reported that both the GH response to exercise (20 min at 70% V˙O2 max) and the GH infusion during rest resulted in an increased lipolysis as measured by glycerol rate of appearance (Ra). Hence, although other hormones (i.e., catecholamines, cortisol) are also intimately involved with postexercise mobilization of fuel sources acting in contrainsular mechanisms, we speculate (due to an absence of fat metabolism measures in the current study) that an amplification of GH secretion after the long-duration aerobic exercise could be closely linked to an increase in fuel mobilization to provide the energy to replenish depleted carbohydrate stores and postexercise muscle protein. Further evidence that pulsatile GH secretion is required for lipolysis is supported by the findings of Cersosimo et al. (1) who, using a pancreatic clamp protocol, demonstrated the an increase in [3H]palmitate after pulsatile (four consecutive busts) GH infusion but not continuous GH infusion. In addition, Gravholt et al. (6) reported an increase in interstitial glycerol after a physiological pulse of GH. The finding that longer duration aerobic exercise resulted in greater GH secretion than moderate-duration aerobic exercise has obvious clinical implications for certain populations (e.g., obesity, metabolic syndrome) in terms of prescribing exercise sessions that favor GH-stimulated fat mobilization (9,13,27,35).

A major strength of this study was that strictly controlled diet, sleep, and activity regimens were superimposed upon each of the repeated-measures, randomized experimental conditions. This is the first study in the literature to present GH secretion profiles for both aerobic and resistance modes of exercise. We elected to match the aerobic and resistance moderate-duration (∼1 h) and long-duration (∼2 h) exercise trials by time and as mentioned previously. There were significant differences in energy expenditure between the aerobic and the resistance exercise bouts. From an experimental design perspective, it is important to underscore the fact that within each exercise mode, our desire to equate relative exercise intensity was also achieved (refer to Table 2). Thus, within each exercise mode, our results are a direct consequence of the differences in exercise duration exclusively (60 vs 120 min). As mentioned previously, it is still possible to consider that aerobic versus resistance modes of exercise may elicit different influences on GH physiology. However, our data would suggest that these differences might be better investigated in terms of examining GH molecular heterogeneity rather than GH secretory dynamics (3,15,21,26,30–32).

In conclusion, using a repeated-measures, randomized experimental design in which 20 h of GH secretion was assessed in young healthy men after five experimental trials—control (no exercise) condition, a moderate-duration (1-h) aerobic exercise bout consisting of 45 min of exercise, a long-duration (2-h) aerobic exercise bout consisting of 90 min of exercise, a moderate-duration (1-h) resistance exercise bout, and a long-duration (2-h) exercise bout, with diet, sleep, and activity regimens strictly controlled—only the long-duration aerobic exercise bout resulted in a significant amplification of GH secretion (i.e., amplitude of largest burst, basal secretion rate, total basal secretion, total pulsatile secretion, and total overall secretion). That a similar GH secretion amplification was not observed for resistance exercise of similar duration may be somewhat surprising, given that both GH and resistance exercise are associated with anabolic outcomes thought to be closely linked. We believe that the underlying cause may be due to the greater energy expenditure achieved during aerobic exercise and a greater demand for postexercise fat mobilization. The physiological significance of GH pulsatile secretion and its important metabolic role with regard to fuel provision are further reinforced by the findings of Surya et al. (28), who reported that only pulsatile, not continuous, delivery of GH stimulates lipolysis in humans. The manner of GH presentation to peripheral tissues is an important parameter of GH action, and our results are in concurrence with Wideman et al. (36) suggesting that exercise duration and associated energy expenditure are important determinants for GH secretion in postexercise recovery.

Funding for this project was provided by the Military Operational Research Program from the U.S. Army Medical Research and Materiel Command under Task Area S: Physiological Mechanisms of Musculoskeletal Injuries. The authors gratefully acknowledge the motivated research volunteers who participated in the study.

There is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.

The views, opinions, and/or findings contained in this publication are those of the authors and should not be construed as an official Department of the Army position, policy, or decision unless so designated by official documentation. The investigators have adhered to the policies for protection of human subjects as prescribed in Army Regulation 70-25, and the research was conducted in adherence with the provisions of 32 CFR Part 219.


1. Cersosimo E, Danou F, Persson M, Miles JM. Effects of pulsatile delivery of basal growth hormone on lipolysis in humans. Am J Physiol. 1996; 271: E123–6.
2. Donnelly JE, Blair SN, Jakicic JM, Manore MM, Rankin JW, Smith BK. American College of Sports Medicine Position Stand. Appropriate physical activity intervention strategies for weight loss and prevention of weight regain for adults. Med Sci Sports Exerc. 2009; 41 (2): 459–71.
3. Gibney J, Healy ML, Sonksen PH. The growth hormone/insulin-like growth factor-I axis in exercise and sport. Endocr Rev. 2007; 28: 603–24.
4. Gibney J, Healy ML, Stolinski M, et al. Effect of growth hormone (GH) on glycerol and free fatty acid metabolism during exhaustive exercise in GH-deficient adults. J Clin Endocrinol Metab. 2003; 88: 1792–7.
5. Giustina A, Veldhuis JD. Pathophysiology of the neuroregulation of growth hormone secretion in experimental animals and the human. Endocr Rev. 1998; 19: 717–97.
6. Gravholt CH, Schmitz O, Simonsen L, Bulow J, Christiansen JS, Moller N. Effects of a physiological GH pulse on interstitial glycerol in abdominal and femoral adipose tissue. Am J Physiol. 1999; 277: E848–54.
7. Hartman ML, Veldhuis JD, Thorner MO. Normal control of growth hormone secretion. Horm Res. 1993; 40: 37–47.
8. Hunter WM, Fonseka CC, Passmore R. Growth hormone: important role in muscular exercise in adults. Science. 1965; 150: 1051–3.
9. Hymer WC, Kraemer WJ, Nindl BC, et al. Characteristics of circulating growth hormone in women after acute heavy resistance exercise. Am J Physiol Endocrinol Metab. 2001; 281: E878–87.
10. Irving BA, Weltman JY, Patrie JT, et al. Effects of exercise training intensity on nocturnal growth hormone secretion in obese adults with the metabolic syndrome. J Clin Endocrinol Metab. 2009; 94: 1979–86.
11. Isgaard J, Carlsson L, Isaksson OG, Jansson JO. Pulsatile intravenous growth hormone (GH) infusion to hypophysectomized rats increases insulin-like growth factor I messenger ribonucleic acid in skeletal tissues more effectively than continuous GH infusion. Endocrinology. 1988; 123: 2605–10.
12. Kanaley JA, Dall R, Moller N, et al. Acute exposure to GH during exercise stimulates the turnover of free fatty acids in GH-deficient men. J Appl Physiol. 2004; 96: 747–53.
13. Kanaley JA, Weltman JY, Veldhuis JD, Rogol AD, Hartman ML, Weltman A. Human growth hormone response to repeated bouts of aerobic exercise. J Appl Physiol. 1997; 83: 1756–61.
14. Kraemer WJ, Marchitelli L, Gordon SE, et al. Hormonal and growth factor responses to heavy resistance exercise protocols. J Appl Physiol. 1990; 69: 1442–50.
15. Kraemer WJ, Nindl BC, Marx JO, et al. Chronic resistance training in women potentiates growth hormone in vivo bioactivity: characterization of molecular mass variants. Am J Physiol Endocrinol Metab. 2006; 291: E1177–87.
16. Maiter D, Underwood LE, Maes M, Davenport ML, Ketelslegers JM. Different effects of intermittent and continuous growth hormone (GH) administration on serum somatomedin-C/insulin-like growth factor I and liver GH receptors in hypophysectomized rats. Endocrinology. 1988; 123: 1053–9.
17. McCall GE, Gosselink KL, Bigbee AJ, Roy RR, Grindeland RE, Edgerton VR. Muscle afferent-pituitary axis: a novel pathway for modulating the secretion of a pituitary growth factor. Exerc Sport Sci Rev. 2001; 29 (4): 164–9.
18. National Research Council. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids (Macronutrients). Washington, D.C.: The National Academies Press; 2005.
19. Nindl BC. Exercise modulation of growth hormone isoforms: current knowledge and future directions for the exercise endocrinologist. Br J Sports Med. 2007; 41: 346–8; discussion 48.
20. Nindl BC, Alemany JA, Tuckow AP, Kellogg MD, Sharp MA, Patton JF. Effects of exercise mode and duration on 24-h IGF-I system recovery responses. Med Sci Sports Exerc. 2009; 41 (6): 1261–70.
21. Nindl BC, Hymer WC, Deaver DR, Kraemer WJ. Growth hormone pulsatility profile characteristics following acute heavy resistance exercise. J Appl Physiol. 2001; 91: 163–72.
22. Nindl BC, Kraemer WJ, Arciero PJ, Samatallee N, Leone CD, Mayo MF, Hafeman DL. Leptin concentrations experience a delayed reduction after resistance exercise in men. Med Sci Sports Exerc. 2002; 34 (4): 608–13.
23. Nindl BC, Kraemer WJ, Marx JO, Tuckow AP, Hymer WC. Growth hormone molecular heterogeneity and exercise. Exerc Sport Sci Rev. 2003; 31 (4): 161–6.
24. Ormsbee MJ, Thyfault JP, Johnson EA, Kraus RM, Choi MD, Hickner RC. Fat metabolism and acute resistance exercise in trained men. J Appl Physiol. 2007; 102: 1767–72.
25. Pierce JR, Tuckow AP, Alemany JA, et al. Effects of acute and chronic exercise on disulfide-linked growth hormone variants. Med Sci Sports Exerc. 2009; 41 (3): 581–7.
26. Pritzlaff CJ, Wideman L, Weltman JY, et al. Impact of acute exercise intensity on pulsatile growth hormone release in men. J Appl Physiol. 1999; 87: 498–504.
27. Stokes KA, Tyler C, Gilbert KL. The growth hormone response to repeated bouts of sprint exercise with and without suppression of lipolysis in men. J Appl Physiol. 2008; 104: 724–8.
28. Surya S, Horowitz JF, Goldenberg N, et al. The pattern of growth hormone delivery to peripheral tissues determines insulin-like growth factor-1 and lipolytic responses in obese subjects. J Clin Endocrinol Metab. 2009; 94: 2828–34.
29. Trepp R, Fluck M, Stettler C, et al. Effect of GH on human skeletal muscle lipid metabolism in GH deficiency. Am J Physiol Endocrinol Metab. 2008; 294: E1127–34.
30. Tuckow AP, Rarick KR, Kraemer WJ, Marx JO, Hymer WC, Nindl BC. Nocturnal growth hormone secretory dynamics are altered after resistance exercise: deconvolution analysis of 12-hour immunofunctional and immunoreactive isoforms. Am J Physiol Regul Integr Comp Physiol. 2006; 291: R1749–55.
31. Veldhuis JD, Carlson ML, Johnson ML. The pituitary gland secretes in bursts: appraising the nature of glandular secretory impulses by simultaneous multiple-parameter deconvolution of plasma hormone concentrations. Proc Natl Acad Sci U S A. 1987; 84: 7686–90.
32. Veldhuis JD, Johnson ML. Cluster analysis: a simple, versatile, and robust algorithm for endocrine pulse detection. Am J Physiol. 1986; 250: E486–93.
33. Wee J, Charlton C, Simpson H, et al. GH secretion in acute exercise may result in post-exercise lipolysis. Growth Horm IGF Res. 2005; 15: 397–404.
34. Weltman A, Weltman JY, Schurrer R, Evans WS, Veldhuis JD, Rogol AD. Endurance training amplifies the pulsatile release of growth hormone: effects of training intensity. J Appl Physiol. 1992; 72: 2188–96.
35. Weltman A, Weltman JY, Watson Winfield DD, et al. Effects of continuous versus intermittent exercise, obesity, and gender on growth hormone secretion. J Clin Endocrinol Metab. 2008; 93: 4711–20.
36. Wideman L, Consitt L, Patrie J, et al. The impact of sex and exercise duration on growth hormone secretion. J Appl Physiol. 2006; 101: 1641–7.
37. Wideman L, Weltman JY, Hartman ML, Veldhuis JD, Weltman A. Growth hormone release during acute and chronic aerobic and resistance exercise: recent findings. Sports Med. 2002; 32: 987–1004.


© 2014 American College of Sports Medicine