Br J Sports Med 43:521-525 doi:10.1136/bjsm.2007.041970
  • Original article

The role of lactate in the exercise-induced human growth hormone response: evidence from McArdle disease

  1. R J Godfrey1,
  2. G P Whyte2,
  3. J Buckley3,
  4. R Quinlivan4
  1. 1
    Centre for Sports Medicine and Human Performance, Brunel University, Uxbridge, UK
  2. 2
    School of Sport and Exercise Sciences, Liverpool John Moores University, Liverpool, UK
  3. 3
    Centre of Exercise and Nutrition Science, University of Chester, Cheshire, UK
  4. 4
    The Muscle Clinic, Robert Jones and Agnes Hunt Orthopaedic NHS Trust, Oswestry, UK
  1. Dr Richard Godfrey, Centre for Sports Medicine and Human Performance, School of Sport and Education, Brunel University, Uxbridge, UB8 3PH, United Kingdom; richard.godfrey{at}
  • Accepted 20 November 2007
  • Published Online First 9 January 2008


Purpose: Increased blood lactate concentration has been suggested as a primary stimulus for the exercise-induced growth hormone response (EIGR). Patients with McArdle disease are unable to produce lactate in response to exercise and thus offer a unique model to assess the role of lactate in the EIGR. Accordingly, McArdle’s patients were exercised to test the hypothesis that lactate is a major stimulus of the EIGR.

Methods: 11 patients with McArdle disease (3 male, 8 female; age: 35.5 (SD 13.9) years, height: 166 (8) cm, body mass: 75.2 (13.1) kg) were recruited for the study. The patients walked initially at 0.42 m/s, increasing by 0.14 m/s per 3 min stage. Exercise was terminated when participants completed 3 minutes at 1.80 m/s or when a Borg CR10 pain scale rating of “4” was reached. Stages were separated by 60 s for capillary blood sampling for analysis of hGH and blood lactate concentration.

Results: McArdle’s patients’ blood lactate levels remained at resting levels (0.3–1.2 mmol/l) as exercise intensity increased. Nine out of 11 participants failed to demonstrate an EIGR obtaining hGH values below the clinical definition of a response (>3 μg/l).

Conclusion: The absence of an EIGR in nine out of 11 participants suggests that lactate could play a major role in the EIGR in humans.

Recent evidence demonstrates an increasingly wide and varied role for lactate other than that of a simple metabolic intermediary. For example, Lassarre et al1 suggested that the level of circulating human growth hormone (hGH) during submaximal exercise is associated with anaerobic glycolysis in muscle. This suggestion implicated lactate, pyruvate and a small number of other metabolic intermediaries of the Embden–Meyerhoff pathway in the exercise-induced increase in circulating hGH. Of these, only lactate has demonstrated a seeming growth in ubiquity of function. Lactate has been observed to inhibit the post-tetanic reuptake of Ca2+ ions by sarcoplasmic reticulum,2 to stimulate insulin secretion3 and to stimulate testosterone secretion,4 and it has been suggested that it is involved in promoting wound repair.5 In addition, lactate has been shown to be an oxidisable fuel used by mitochondria during exercise.6 The ubiquitous role of lactate has led to the suggestion that it may also be involved in many more functions, including the exercise-induced growth hormone response (EIGR). In disease-free, healthy individuals it is well known that incremental exercise results in about a 10-fold increase in circulating lactate. In addition, individuals demonstrate a “lactate threshold” (LT). Below LT increasing exercise intensity results in a linear increase in circulating lactate, while above LT circulating lactate increases exponentially. Although circulating levels of hGH are reported to rise in disease-free, healthy individuals in response to low levels of exercise, a more pronounced (sevenfold) and sustained rise (60–90 min post-exercise) in hGH is observed with exercise above LT.7

The idea that lactate may be the stimulus for the EIGR was first suggested by Sutton in 19678 and was followed by the first publication presenting empirical support for the relationship between lactate and the EIGR in 1976.9 Subsequent studies have reported a high correlation between the pattern of response of hGH and lactate to exercise,10 11 which has led many to suggest that an increase in blood lactate concentration stimulates hGH secretion.12 However, a causal link between lactate concentration and the EIGR has not yet been demonstrated.

In an attempt to delineate the role of lactate in the EIGR, Luger et al13 infused sodium lactate to a plasma concentration similar to that observed during exercise at an intensity of 70–90% VO2max in sedentary volunteers at rest, which caused a rise in hGH. The reported increase in hGH was lower than that observed with exercise at 70–90% VO2max and led the authors to conclude that lactate may be one of the factors stimulating the EIGR, and that “other additional stimuli” may also be involved. The infusion of sodium lactate at rest, however, does not represent a true physiological model for the assessment of the role of lactate in the EIGR.

McArdle disease (type V glycogen storage disease) afflicts one in a million people and is caused by the absence of the enzyme myophosphorylase, responsible for the conversion of glycogen to glucose in skeletal muscle. Thus, in patients with McArdle disease there is no lactate production during exercise.14 McArdle disease, therefore, offers a true physiological model with which to assess the role of lactate in the EIGR. Hence, if lactate is the major stimulus for exercise-induced hGH secretion then one would not expect McArdle’s patients to exhibit an EIGR. The aim of the present study was to determine whether McArdle’s patients exhibit an EIGR, and hence establish whether or not lactate is a primary stimulus of the EIGR. The hypothesis being tested is that exercise-induced lactate production provides the major stimulus for the EIGR.



This work has been carried out in accordance with the Declaration of Helsinki (2000). Following ethical approval and written informed consent, 11 McArdle’s patients (three male, eight female; age 35.5 (SD 13.9) years, height 166 (8) cm, body mass 75.2 (13.1) kg, body mass index 27.2 (3.6) kg/m2), diagnosed by muscle biopsy and DNA analysis, volunteered to take part in the study. Eight of the patients were homozygous (for the R49X mutation) while three were heterozygous. Skeletal muscle biopsy samples were typical for McArdle’s patients, demonstrating an absence of myophosphorylase activity and a preponderance of sub-sarcolemmal glycogen.

In view of the risk of myoglobinuria and renal failure associated with strenuous exercise in this population, exclusion criteria for participation in the study were applied. Hence, only those participants who had previously achieved a distance of greater than 700 m in a 12 min shuttle walking test and who could sustain a walking velocity of greater than 0.83 m/s (3 km/h) without rhabdomyolysis (Buckley et al (unpublished data)) were included in the study. The 12 min shuttle walking test involves walking back and forth over a 20 m distance, aiming to cover the greatest possible distance in 12 min. Participants in this test use a self-determined pace and are able to stop for rest at their own discretion.

Experimental design

Participants in the present study performed a progressive incremental treadmill (Powerjog Ltd, Birmingham, UK) walking protocol. During the “warm-up” period the participants were asked to walk at 0.42 m/s (1.5 km/h) with regular rests (if needed) until “second wind” was achieved. This “second wind” phenomenon has been demonstrated in McArdle’s patients in previous research15 and can take 5–15 min to achieve, subsequently allowing patients to exercise for up to 2 h. In the present study, following the warm-up, the test protocol also began at 0.42 m/s (1.5 km/h), with a stage duration of 3 min. After each 3 min stage the participant stopped for 60 s, during which a capillary blood sample was taken simultaneously from each ear, one for determination of blood lactate concentration and the other for hGH concentration. The test continued, increasing by 0.14 m/s (0.5 km/h) every 3 min until either 3 min were completed at 1.80 m/s (6.5 km/h) or until a self-reported score of “4” was reached on the CR10 pain scale.16 At this point the test was terminated. The major symptom in McArdle’s patients is exercise-induced muscle pain. A perceived pain level of “4” on the CR10 scale was chosen because it has been used as a critical level in other clinical populations for exertion, breathlessness, claudication and angina pain.17 The whole exercise protocol used in the present study was conducted at 0% gradient and heart rate was measured throughout using a telemetric heart-rate system (Polar Electro Oy, Finland).

Prior to walking on the treadmill participants were instructed on the use of the rating of perceived pain (CR10) scale,16 which included perceptual anchoring of the range descriptors. A rating of “0” equates to no pain at all and “10” is the most pain the participant can ever recall suffering. The scale has values in excess of “10”, which were described to the participant as being potential levels of pain beyond that which had previously been experienced. Patients were informed that overexertion is associated with muscle contractures, which in turn can lead to muscle breakdown (rhabdomyolysis), myoglobinuria and renal failure,18 and was to be avoided at all costs. Hence, exercise would be terminated either if a pain scale value of “4” was attained or if 3 minutes at 1.80 m/s (6.5 km/h) was completed, whichever came first. Because walking was used as the exercise modality, participants were requested to rate muscle pain specifically arising from the legs during the last minute of each stage.

Blood was collected into a glass capillary tube for measurement of whole blood lactate concentration and analysis was accomplished using a GM-7 enzymatic analyser (Analox Instruments Ltd, Hammersmith, London). Calibration was performed before each test using a 7 μl sample of 8 mmol/l lactate calibration solution which was checked against a quality control serum with a known lactate concentration range (2.3–2.7 mmol/l) (Analox Instruments Ltd, London, UK). The coefficient of variation for the measurement of lactate on the GM7 lactate analyser used in this study was 7%. Within-run precision is 2% at 2.5 mmol/l and accuracy for whole blood comparison, compared with a spectrophotometric method, reveals a correlation coefficient of r = 0.991 (Analox Instruments Ltd., Technical Data, London, UK).

Capillary samples for hGH were collected into 300 μl serum “Microvettes” (CB 300 LH, Sarstedt, Germany) placed on ice and spun down within 60 min in a microcentrifuge (Analox Instruments Ltd, London, UK) at 12 000 rpm for 10 min. Supernatant from capillary samples was then transferred to labelled Eppendorf tubes and frozen at –70°C for subsequent analysis of hGH concentration.

Serum hGH concentration was measured using an immunoradiometric assay (IRMA) kit (DSL-1900 ACTIVE, Diagnostic Systems Laboratories Inc., Texas, USA). This method of analysis utilises relatively small sample sizes of 50 μl, making it attractive for use with capillary samples, and has been previously validated.19

The radioactivity reflecting the amount of bound hGH was determined by a 1261 Multigamma on-line counter (Wallac Oy, Finland). A standard curve was produced using ten hGH standards, of known concentration, in the range 0.1–26.0 μg/l.

The theoretical sensitivity, or minimum detection limit, calculated by the interpolation of the mean plus two standard deviations of 10 replicates of the 0 μg/l hGH standards, is 0.01 μg/l. The intra-assay precision was determined from the mean of 12 replicates of three human serum samples. For 12 replicates, sample 1 was 2.91 (0.09) μg/l with a coefficient of variance (CV) of 3.1%, sample 2 was 5.58 (0.22) μg/l with a CV of 3.9%, and sample 3 was 25.84 (1.40) μg/l with a CV of 5.4%.

Statistical analyses

From analysis of blood hGH concentration it was apparent that only two subjects achieved values above the international consensus for the clinical definition of a response to a stimulus for hGH secretion (>3 μg/l).20 Accordingly, if subjects are divided into two groups, one of two and one of nine, inferential statistics are not appropriate, due either to the small numbers involved or to the aim of examining the role of lactate in the EIGR. Hence only descriptive statistics are applied.


Nine out of 11 participants failed to demonstrate an EIGR as defined by the international consensus for the clinical definition of a response.20 Two individuals demonstrated peak serum hGH values above 3 μg/l of 4.65 and 8.40 μg/l respectively. Data on all participants are reported in table 1.

Table 1 Data resulting from exercise in all participants

Figure 1 illustrates mean hGH and lactate values using percentage of peak treadmill speed on the x-axis to facilitate normalisation of results across the range of exercise intensities used.

Figure 1

Mean serum growth hormone and whole blood lactate concentrations at 20, 40, 60, 80 and 100% of peak treadmill speed.

Resting blood lactate concentrations were “normal” (0.27–1.18 mmol/l) in all patients but none demonstrated a rise in blood lactate in response to exercise. Peak heart rate values were 146 (27) beats/min (ie 80% (12%) of age-predicted maximal heart rate (220−age)) (table 1).


The aim of this study was to determine whether or not lactate is a primary stimulus for the exercise-induced hGH response (EIGR). McArdle’s patients were used as a research model since they do not produce lactate in response to exercise and, as a physiological model, require only minimal invasive intervention. Nine out of 11 participants failed to achieve the clinical definition of a response (circulating values greater than 3 μg/l20). It is the contention of the present authors that these findings provide evidence that lactate is a major stimulus of the EIGR.

Peak hGH concentrations in all participants were lower than those observed in normal individuals and athletes.21 22 The values for rowers21 22 are peak values in individuals without any pathological restriction and so these individuals were able to exercise to maximal heart rate. However, if adjusted to represent values associated with 80% of peak HR (as was achieved in the McArdle’s patients) values for rowing (pooled data from two studies21 22), mean values are 6.7 (4.5) μg/l. Thus the mean value seen in the McArdle’s patients in the current study is less than 25% of the concentration that would be expected from disease-free, healthy individuals in response to exercise.

The fact that two patients achieved values in excess of 3 μg/l (patients 10 and 11, table 1) in response to incremental exercise may be a reflection of the range of mutations possible in the coding region or splice sites of the PYGM gene.23 In McArdle disease the R49X mutation is the most common form found in Europe and North America24 and it is specifically this mutation that was screened for in volunteers in this present study. While this adequately confirms the diagnosis of McArdle disease, it does not rule out other additional mutations in any of the patients. Little is known of the phenotype (physical manifestation) of McArdle’s generally and even less about the consequences of specific mutations or their interactive effects. Indeed Vorgerd et al25 have suggested that even where clinical homogeneity is demonstrated in McArdle’s patients molecular heterogeneity can exist. This leads to speculation that very variable responses to the same stimuli are extremely likely in this population. Hence a relatively consistent finding (such as an hGH value below 3 μg/l in nine out of 11 patients) could be considered compelling evidence.

Some might suggest that the exercise intensity was insufficient to elicit an EIGR. However, in the present study a mean peak heart rate of 80% of predicted maximum was attained. A minimum exercise intensity for significant hGH secretion (ie circulating values sustained above baseline for 60–90 min post-exercise) has been determined to require an intensity equivalent to or greater than lactate threshold (LT).7 In a healthy, unconditioned population LT occurs between 50–60% of maximal oxygen consumption.26 The American College of Sports Medicine, in providing guidelines on exercise to improve and maintain cardiopulmonary “fitness”, suggests that a percentage of 55–85% VO2max is equivalent to 60–90% of maximum heart rate.27 Maximal heart rate data, in the light of these recommendations, suggests that McArdle’s patients in the present study exercised to 60% VO2max, or above. Together with their poor level of conditioning, this demonstrates that participants exercised to a level above a relative intensity that would, in disease-free healthy individuals, be associated with lactate threshold. Accordingly, it is likely that all participants in the present study worked at an intensity that was sufficient to elicit an EIGR had they been able to do so.

With moderate exercise intensities McArdle’s patients can experience pain, and pain itself can elevate heart rate. However, pain was unlikely to have been a mediator of a differential heart rate response during exercise in this study because i) exercise was stopped if level “4” was reached on the CR10 pain scale16 and ii) all participants were subject to the same relative stimulus.

McArdle’s patients demonstrated resting blood lactate concentrations which are similar to those observed in the normal population. The reason for this resting level may be attributed to lactate being produced by red blood cells and the liver. Red blood cells use only glycolysis (95% of red blood cell glucose use) and the hexose monophosphate shuttle (5% of red blood cell glucose use) to provide ATP for normal function.28 Despite normal resting blood lactate concentrations, the absence of myophosphorylase in the McArdle’s patients resulted in a plateau or decrease in blood lactate during exercise of increasing intensity (fig 1). Thus, although McArdle’s patients may demonstrate normal resting blood lactate concentrations they do not produce any lactate in response to muscular work.

In McArdle’s patients exercise is limited both by pathology and poor conditioning. Hence, across all of the McArdle’s participants in the present study a group mean HR, as a percentage of maximal heart rate, of 80% was achieved during walking. Most disease-free, healthy individuals would have to run to achieve such a percentage and this difference in locomotion would affect muscle metabolism and, arguably, the resultant hormonal milieu would render any direct comparison using a disease-free control group of questionable value.

Present findings provide a strong link between lactate and the EIGR, and are consistent with the postulate of Brooks (2002) that lactate could be considered a pseudo-hormone.29 However, in the present study, a rise above 3 μg/l in hGH in response to incremental exercise in two patients also suggests a minor role for “additional stimuli”. A number of other candidates have been suggested as having a role in the EIGR. These include afferent stimulation, reduced pH, increased nitric oxide and elevated catecholamine levels and these have been reviewed elsewhere.30 Current findings cannot exclude other metabolic intermediaries of the glycolytic pathway but growing evidence, some cited above, suggests that lactate is the most likely candidate in this case.

In the light of current findings future research should concentrate on the exact role played by each of the potential additional stimuli to lactate’s major role. Finally, further research is needed to fully elucidate lactate’s signalling role in mammalian physiological systems generally and with respect to the EIGR specifically.


The aim of this study was to determine whether lactate is the major stimulus for the EIGR. If one assumed that lactate was the major stimulus then it could be expected that McArdle’s patients would not secrete hGH in response to incremental exercise. In the present study, lactate’s major role in the EIGR is strongly implied. Nine out of 11 participants did not achieve the clinical definition for a response to a stimulus for hGH.

What is already known on this topic

Growth hormone secretion occurs throughout life and is important, even in adulthood, to maintain normal physiology and good health. Exercise elicits a growth hormone response. The higher the intensity of exercise, above that equivalent to the lactate threshold, the greater the growth hormone response. The exact mechanism for this exercise-induced growth hormone secretion is largely unknown although a number of candidates have been suggested, each as major stimuli. These include catecholamines, nitric oxide, afferent stimulation, reduced pH and lactate.

What this study adds

This study adds further evidence to the growing ubiquity of roles for lactate as an intercellular and intracellular metabolic signal, including recent suggestions naming it a “pseudo-hormone”. Most important, however, is that it provides compelling evidence for lactate as a major stimulus for the exercise-induced growth hormone response (EIGR). This study is particularly novel, even perhaps unique, in its use of McArdle’s patients, as these provide a better physiological model than has previously been used in examining the role of lactate in almost any exercise setting. Hence, stronger evidence for lactate as a stimulus for the EIGR is provided than has been previously possible.


The authors would like to thank the British Olympic Medical Trust for supporting this research and colleagues (Lee Romer, Tony Blazevich, Liz Gough, José González-Alonso and Kelly Ashford) at Brunel University for valuable comments on the manuscript.


  • Funding: The British Olympic Medical Trust, British Olympic Association, 1 Wandsworth Plain, London SW18 1EH.

  • Competing interests: None.