Plasma lactate is widely used as a biomarker in critical illness. The aims of the present study were to elucidate the usefulness of a three-compartment model for muscle lactate kinetics in humans and to characterize the response to an exogenous adrenaline challenge. Repeated blood samples from artery and femoral vein together with blood flow measurements and muscle biopsies were obtained from healthy male volunteers (n=8) at baseline and during an adrenaline infusion. Concentrations of lactate and enrichment of [13C]lactate were measured and kinetics calculated. Mitochondrial activity, glycogen concentration, oxygen uptake and CO2 release were assessed. The adrenaline challenge increased plasma lactate 4-fold as a result of a greater increase in the rate of appearance (Ra) than the increase in the rate of disappearance (Rd). Leg muscle net release of lactate increased 3.5-fold, whereas intramuscular production had a high variation but did not change. Mitochondrial state 3 respiration increased by 30%. Glycogen concentration, oxygen uptake and CO2 production remained unchanged. In conclusion a three-compartment model gives additional information to the two-compartment model but, due to its larger variation and invasive muscle biopsy, it is less likely to become a regularly used tool in clinical research. Hyperlactataemia in response to adrenergic stimuli was driven by an elevated lactate release from skeletal muscle most probably due to a redirection of a high intramuscular turnover rather than an increased production.

CLINICAL PERSPECTIVES

  • Although plasma lactate is an important and frequently used laboratory test, the metabolism of lactate is still not fully understood in the critically ill patient.

  • In the present study, we compared different methods for studying lactate metabolism using labelled lactate and investigated lactate metabolism under stress. Hopefully this, together with further clinical studies, will lead to better understanding of pathophysiology and optimization of treatment during resuscitation and intensive care.

  • Even though the three-compartment model gave additional information, it is associated with larger variation and the need for biopsies makes it more invasive which limits its usefulness in larger clinical studies. Therefore whole-body turnover and two-compartment models, using arteriovenous difference, are probably preferable in these cohorts.

INTRODUCTION

Lactate is a three-carbon compound formed from pyruvate by pyruvate dehydrogenase in the cytosol of cells. After being exported to the central circulation it can be taken up and used as an energy source in various tissues, such as heart and brain, or used for gluconeogenesis in the liver and kidneys. In situations of stress, its metabolic importance increases as it has been shown to be the main substrate used for gluconeogenesis [1]. Plasma lactate is frequently used as a biomarker in various clinical settings, and elevated plasma levels are associated with poor prognosis in a number of different conditions, e.g. sepsis, trauma and post cardiac arrest [2,3]. Of these conditions plasma lactate has been most studied in patients with sepsis, where it is often used for diagnosis and evaluation of resuscitation [46]. Although often used in clinical practice, the metabolism of lactate in critically ill patients is not well described. The previous misconception that elevated lactate is a strict anaerobic product caused by tissue hypoxia, is no longer accepted and adrenergic stimulation, through β2 receptor activation, have been shown to play a major role in animal models of sepsis and locally in patients with sepsis [79]. This might be to recruit lactate, a metabolic intermediate, to vital organs in situations of stress. It would therefore be of great interest to further characterize the effect of adrenaline on lactate metabolism in healthy human.

The use of isotopically labelled lactate introduces possibilities to study lactate kinetics at different levels, i.e. whole body as well as individual tissues, depending on sampling sites [10]. Labelled lactate has been used to study skeletal muscle and brain lactate kinetics in healthy volunteers during rest and exercise [11,12]. In these studies, sampling has been confined to extracellular compartments and therefore does not take intracellular recycling into consideration. Due to its metabolic importance, it could be useful to study lactate metabolism in individual organs.

In the present study, we elucidated whether a three-compartment model (arterial, venous and intramuscular) is useful to characterize muscle lactate kinetics in humans as suggested by Lehman [13] and practiced in anaesthetized dogs by Chinkes et al. [14]. Healthy volunteers were studied in the basal state and during an adrenaline challenge. The aims of the study were (1) to evaluate the three-compartment model for muscle lactate turnover in humans, and (2) to study the effects of adrenaline on human muscle lactate turnover in relation to substrate metabolism. For the second aim, measures of glucose metabolism and mitochondrial function were also included.

MATERIALS AND METHODS

Study subjects

The study was performed on non-obese healthy males (n=4 for the pilot study and n=8 for the main study) with a mean age of 33 years (range 24–49 years) and mean body mass index (BMI) of 23.6 (range 20.7–27.7) for the main study and a mean age of 24 years (range 21–28 years) and mean BMI 23.7 (range 20.1–25.8) in the pilot group. Exclusion criteria were diabetes or chronic muscle or heart disease, irregularities on physical examination, laboratory blood screening or ECG (electrocardiography). In one of the eight subjects, a second muscle biopsy could not be obtained; all calculations including lactate concentration and enrichment in muscle were therefore performed on the remaining seven subjects.

All participants gave consent after receiving both written and oral information. The study protocol, including amendments made after the pilot study, was approved by the regional Ethical Committee in Stockholm.

Pilot experiments

The main study was preceded by a pilot series with a 5-h continuous infusion of a gradually increasing dose of adrenaline in four healthy male volunteers. These volunteers were closely observed with haemodynamic monitoring and regular arterial blood gases (ABGs). The aim of this pilot was to establish a protocol where plasma lactate concentration would reach >3.5 mmol·l−1 in a safe way with minimal discomfort for the subjects. Results (not shown) were used to design a safe and reliable protocol for our main study.

Study protocol

Subjects arrived at our department in the morning, after an overnight fast and received a peripheral venous catheter, radial arterial catheter and a central venous catheter (one lumen, 4 Fr (French)) in the right femoral vein (Figure 1). All subjects were monitored with continuous blood pressure, oxygen saturation and three lead ECG. A trained physician was present at all times.

Schematic representation of study protocol

Figure 1
Schematic representation of study protocol

indicates blood sampling ◆ blood flow measurement and + skeletal muscle biopsy. *Blood sample for control of venous location. ‘A’ indicates arterial blood sample, and ‘V’ indicates venous blood sample.

Figure 1
Schematic representation of study protocol

indicates blood sampling ◆ blood flow measurement and + skeletal muscle biopsy. *Blood sample for control of venous location. ‘A’ indicates arterial blood sample, and ‘V’ indicates venous blood sample.

Baseline blood samples were collected for determination of background isotopic enrichment. 1-13C-labelled sodium lactate ([13C]lactate; Cambridge Isotope Laboratory) was mixed with isotonic saline according to the subject's weight. At t=0 min, a primed continuous infusion was started (1 mg·kg−1+3.65 mg·h−1·kg−1). This was maintained for the entire experiment. At t=120 min, a continuous infusion of adrenaline was started at 0.02 μg·min−1·kg−1. The infusion was then increased by 0.02 μg·min−1·kg−1 every 20 min until a target dose of 0.08 μg·min−1·kg−1 was reached. This infusion rate was then ongoing for the remaining 2 h of the experiment (t=180–300 min).

Before the start (t=105–120 min) and end (t=285–300 min) of the adrenaline infusion, a series of four arterial and central venous blood samples and ABG and venous blood gas (VBG) were drawn at 5-min intervals. Simultaneously with the last samples of the series, a skeletal muscle biopsy from the vastus lateralis muscle of the right leg was obtained. Before and after the sampling series, blood flow of the right leg was determined using venous occlusion plethysmography, as previously described by our group [15]. Blood flow was measured ten times before and after the blood sampling periods, and a mean of the 20 measurements was used as blood flow before and during the adrenaline challenge. Plasma flow (PF) was calculated by correcting the blood flow for haematocrit content.

During the infusion of adrenaline, blood samples were drawn every 20 min. ABGs were immediately analysed to monitor electrolytes and the metabolic reaction. After the last blood flow measurement, all infusions were terminated and subjects remained under supervision for another 30 min until a final ABG was collected.

Blood samples were collected in EDTA tubes. The blood sample was centrifuged at 4°C for 10 min and plasma was then collected and stored at −80°C for later analysis.

Muscle biopsies were taken with a Bergström needle after application of local anaesthesia [Xylocaine (10 mg/ml), AstraZeneca]. Samples were cleaned from fat and divided into two portions. One part was snap-frozen in liquid nitrogen and stored at −80°C and the second part was put in a pre-chilled preservation buffer and immediately taken for analysis for mitochondrial respiration.

Sample analysis

Lactate concentration

Muscle samples were freeze-dried, and the remaining blood and connective tissue were removed under a microscope. Two milligrams of freeze-dried muscle was dissolved in 500 μl of 70% ethanol, vortex-mixed for 10 s at maximum speed three times and kept on ice for 20 min. Thereafter, the sample was centrifuged at 20000 g for 10 min. Supernatant was collected and evaporated by rotary evaporation (SpeedVac, ThermoSavant spd 11v, Thermo Scientific). Resulting solutions, as well as plasma samples, were analysed for lactate concentration through an enzymatic method using DiaSys Lactate FS kit and subsequent photo spectrometry analysis for NADH at 340 nm.

Isotopic enrichment

Two milligrams of freeze-dried and cleaned muscle was added to 500 μl of ethanol, homogenized using a mini bead beater (Biospec Products) and lactate-containing supernatant was collected after centrifugation (20000 g, 10 min) and evaporation in a SpeedVac.

Proteins in 100 μl of defrosted plasma was precipitated with 1 ml of ice-cold ethanol, centrifuged for 10 min at 20000 g and the remaining supernatant evaporated in a speedvac.

Enrichment of 13C-labelled lactate in the dissolved muscle solution and plasma were assessed using GC–MS (Inert XL MSD 5975C, Agilent Technologies). Volumes of 100 μl of pyridine and BSTFA (N,O-Bis(trimethylsilyl)trifluoroacetamide) (ratio 1:1) were used as derivatives during a 30-min incubation at room temperature. Peaks for 219 and 220 m/z (lactate and [13C]lactate) were compared to assess enrichment in each sample. Each muscle sample was analysed five times and the mean values were used for calculations.

Glucose concentration

Plasma glucose was analysed using a kit (God-Pod, Thermo Fisher Scientific) and analysed using photo spectrometry 510 nm (KONE lab 20 XTi, Thermo Scientific).

Glycogen concentration

Freeze-dried muscle was digested with 1 M KOH and then neutralized with 0.25 M HCl and 0.15 M sodium acetate/acetic acid buffer (pH 4.9). Glycogen was degraded with amyloglucosidase (Sigma–Aldrich). Samples were then analysed for glucose using the God-Pod kit as described above.

Catecholamine concentration

Plasma concentrations of adrenaline, noradrenaline and dopamine were determined using HPLC (Dionex) with electrochemical detection (Antech Solutions).

Mitochondrial analysis

Oxidative phosphorylation

Muscle sample was kept in a preservation buffer and mitochondria were isolated through homogenization followed by a series of dilutions and centrifugations as previously described [16]. Mitochondria were finally diluted in a buffer (225 mM mannitol, 75 mM sucrose, 10 mM Tris-base, 0.1 mM EDTA, 10 mM KCl and 10 mM, KH2PO4). The solution was placed in an Oxygraph chamber together with malate and MgCl2. Pyruvate (final concentration 3 mM) was used as energy substrate. ADP was added to induce oxidative phosphorylation. State 3 and 4 respiration was calculated as oxygen consumption per minute in the presence and absence of ADP respectively and correlated for the amount of citrate synthase (CS) in each sample. The respiratory control ration (RCR) was calculated as state 3 respiration divided by state 4 respiration. All assays were blinded for the examiner before calculation of respirational rates.

Mitochondrial CS assay

A portion of the mitochondria containing solution was frozen and stored for later analysis of CS. The samples were defrosted and diluted with a SET (Sucrose-EDTA-Triss) and phosphate buffer, and Triton X-100 was used for disruption of mitochondrial membranes. CS activity was then assessed, with photo spectrometry at 412 nm, for conversion from acetyl-CoA and oxaloacetic acid to citrate and CoA SH [17].

Blood gas analysis

A point-of-care blood gas analyser (ABL 800 flex, Radiometer Medical) was used for ABGs and VBGs in the experiment.

Calculations

Mean values for the four arterial and venous samples taken before and at the end of the adrenaline challenge were used to perform the lactate kinetic calculations, except for the non-steady-state whole-body calculations, where calculations were performed for kinetics between each data point and a mean from the three last values are presented. Examination of the individual data sets we concluded that [13C]lactate enrichment in plasma reached steady state (Figure 2), whereas concentration of lactate did not (Figure 3). We were therefore able to use steady-state calculations except for the whole-body rate of disappearance (Rd), where non-steady-state equations were used in accordance with [10].

Enrichment of 13C-labelled lactate in APE, n=8

Figure 2
Enrichment of 13C-labelled lactate in APE, n=8

Filled circles are arterial values, and empty circles venous values, as means ± S.D.

Figure 2
Enrichment of 13C-labelled lactate in APE, n=8

Filled circles are arterial values, and empty circles venous values, as means ± S.D.

Individual arterial plasma concentrations during the adrenaline infusion for all test subjects (n=8) in mmol·l−1

Figure 3
Individual arterial plasma concentrations during the adrenaline infusion for all test subjects (n=8) in mmol·l−1

Empty circles are individual values, and filled circles are mean values.

Figure 3
Individual arterial plasma concentrations during the adrenaline infusion for all test subjects (n=8) in mmol·l−1

Empty circles are individual values, and filled circles are mean values.

Whole-body calculations

Ra and Rd of lactate (μmol·min−1·kg of body weight−1) was calculated using the Steele equation for non-steady-state for each of the data points according to [10]:

 
formula
 
formula

In which i is the infusion rate of the tracer, pV is the distributional constant of lactate (0.1 l·kg−1), CA is the arterial concentration of lactate (μmol·ml−1), and EA is the isotopic enrichment in arterial plasma {[atom percentage excess (APE)]/100}.

Whole-body net balance (NB) is:

 
formula

For comparison, we also calculated Ra using steady-state equations:

 
formula

where Ei is the enrichments of the infused tracer (APE/100).

Leg calculations

Net uptake (NU) of lactate and glucose (μmol·min−1·100 ml of leg volume−1) over the leg was calculated:

 
formula

In which CA and CV are the arterial and venous concentrations respectively (μmol·ml−1) and PF (ml·min−1·100 ml of leg volume−1). A negative value would indicate that muscle is a net contributor to circulating lactate.

The RaL and RdL of lactate (μmol·min−1·100 ml of leg volume−1), over the leg, were calculated using a two-compartment model [14]:

 
formula
 
formula

In which EV is enrichment of the plasma lactate tracer in the vein ([APE]/100).

Leg skeletal muscle lactate utilization (LU) and lactate production (LP) (μmol·min−1·100 ml of leg volume−1) rates were calculated using a three-compartment model for lactate, all according to [14]:

 
formula
 
formula

In which EM is enrichment of the free lactate tracer in muscle ([APE]/100). Fluxes between the different compartments were calculated as:

 
formula
 
formula
 
formula
 
formula
 
formula

where Fin is the flux of lactate in to the leg, Fout is flux out from the leg, Fma is the lactate going into the muscle from artery, Fvm is the lactate going to the vein from the muscle, and Fva is the lactate shunted from artery to vein without being taken up by the muscle (Figure 4). All fluxes are expressed as μmol·min−1·100 ml of leg volume−1.

Schematic representation of the fluxes in a three-compartment model for lactate kinetics
Figure 4
Schematic representation of the fluxes in a three-compartment model for lactate kinetics

Fin is the lactate delivery to tissue, Fout is lactate flux leaving tissue, Fma is the lactate going into the muscle, Fvm is the lactate exported to the vein from the muscle, and Fva is the lactate passing through the tissue without being extracted from the plasma. Production represents the newly produced lactate in muscle, and the utilization represents the lactate being consumed within the muscle.

Figure 4
Schematic representation of the fluxes in a three-compartment model for lactate kinetics

Fin is the lactate delivery to tissue, Fout is lactate flux leaving tissue, Fma is the lactate going into the muscle, Fvm is the lactate exported to the vein from the muscle, and Fva is the lactate passing through the tissue without being extracted from the plasma. Production represents the newly produced lactate in muscle, and the utilization represents the lactate being consumed within the muscle.

Oxygen and CO2 consumption

The amount of oxygen in arterial (Art) and venous (Ven) samples (mmol·l−1) were calculated by:

 
formula
 
formula

Oxygen uptake (nmol·min−1·100 ml of leg volume−1) was then assessed as:

 
formula

The amount of CO2 (mmol−l) in arterial and venous blood and net uptake of CO2 over the leg (nmol·min−1·100 ml of leg volume−1) were calculated by:

 
formula
 
formula
 
formula

Statistics

Student's t test for paired samples was used for statistical comparison between groups. All data are expressed as means ± S.D., unless otherwise indicated.

RESULTS

Plasma levels of adrenaline increased from 0.41 (±0.19) to 10.6 (±2.9) nmol·l−1 during the experiment, whereas circulating levels of the other catecholamines remained unchanged. Heart rate increased approximately 50%, with an unchanged systolic and 10% decreased diastolic blood pressure (Table 1).

Table 1
Haemodynamics and results from analyses in plasma and muscle

Results from measurement in plasma samples and ABG and VBG. State 3 and 4 respiration are approximations from mitochondrial analyses. * indicates calculated value, whereas others are measured. P-values are presented for differences between baseline and at end of adrenaline infusion.

BaselineAdrenalineΔP-value
Haemodynamic parameters (n=8) 
 Heart rate (beats/min) 58.2±8.3 87±14 29±13 <0.001 
 Systolic blood pressure (mmHg) 131.0±6.2 135.1±7.7 4.0±10 0.3 
 Diastolic blood pressure (mmHg) 65.5±3.9 58.3±4.5 −7.3±5.4 0.006 
 Blow flow in leg (ml·min−1·100 ml of leg volume−12.3±1.0 4.9±1.8 2.6±1.0 <0.001 
Plasma samples (n=8) 
 Glucoseart (mmol·l−15.6±0.34 11.5±2.2 6.0±2.0 <0.001 
 Glucoseven (mmol·l−15.47±0.53 11.3±1.5 5.8±1.4 <0.001 
 Adrenaline (nmol·l−10.41±0.19 10.6±2.9 10.2±3.0 <0.001 
 Noradrenaline (nmol·l−11.63±0.47 1.65±0.59 0.02±0.34 0.88 
 Dopamine (nmol·l−10.08±0.03 0.08±0.02 −0.01±0.04 0.59 
ABG (n=8) 
 pH 7.42±0.02 7.42±0.02 0.01±0.02 0.49 
 sO2 (%) 97.9±0.7 98.0±0.6 0.1±0.8 0.81 
 pO2 (kPa) 13.8±1.1 13.5±1.1 −0.2±1.2 0.62 
 pCO2 (kPa) 5.33±0.19 4.89±0.42 −0.43±0.26 0.02 
 Arterial base excess (mmol·l−1)* 0.8±1.4 −0.5±1.5 −1.3±2.3 0.14 
 HCO3 (mmol·l−1)* 25.2±1.2 24.0±1.2 −1.2±2.0 0.12 
 Ca2+ (mmol·l−11.19±0.02 1.12±0.01 −0.07±0.02 <0.001 
 Haemoglobin (g·l−1145.6±9.7 145.5±9.0 −0.1±2.4 0.90 
 K+ (mmol·l−13.95±0.09 3.26±0.18 −0.70±0.15 <0.001 
 Na+ (mmol·l−1137.6±1.0 138.2±1.0 0.64±0.89 0.08 
 Cl (mmol·l−1107.8±1.9 105.2±2.0 −2.6±1.9 0.005 
 Haematocrit* 0.45±0.03 0.45±0.03 −0.0±0.01 0.94 
 Anion gap (mmol·l−1)* 5.0±2.6 9.7±3.3 4.7±3.1 0.003 
VBG (n=8) 
 sO2 (%) 71.9±8.6 88.5±6.6 16.6±7.2 <0.001 
 pO2 (kPa) 5.23±0.75 7.56±1.3 2.33±1.09 <0.001 
 pCO2 (kPa) 6.11±0.21 5.25±0.41 −0.86±0.35 <0.001 
NU over leg (n=8) 
 NU [Glu] (μmol·min−1·100 ml of leg volume−1)* 0.2±1.3 0.5±4.2 0.3±3.4 0.78 
 NU [O2] (nmol·min−1·100 ml of leg volume-1)* 530±270 500±480 37±226 0.64 
 NU [CO2] (nmol·min−1·100 ml of leg volume-1)* −3.3±2.9 −3.3±3.4 −0.0±5.4 0.99 
Intramuscular measurements (n=7) 
 Glycogen (μg·g of dry weight−1292±163 283±153 −10±40 0.54 
 State 3 respiration (nmol of O2·min−1·units of CS−146±13 61±17 15±12 0.01 
 State 4 respiration (nmol of O2·min−1·units of CS−14.8±1.3 5.5±1.6 0.7±1.7 0.32 
 RCR* 9.7±1.9 11.5±3.7 1.9±4.5 0.32 
BaselineAdrenalineΔP-value
Haemodynamic parameters (n=8) 
 Heart rate (beats/min) 58.2±8.3 87±14 29±13 <0.001 
 Systolic blood pressure (mmHg) 131.0±6.2 135.1±7.7 4.0±10 0.3 
 Diastolic blood pressure (mmHg) 65.5±3.9 58.3±4.5 −7.3±5.4 0.006 
 Blow flow in leg (ml·min−1·100 ml of leg volume−12.3±1.0 4.9±1.8 2.6±1.0 <0.001 
Plasma samples (n=8) 
 Glucoseart (mmol·l−15.6±0.34 11.5±2.2 6.0±2.0 <0.001 
 Glucoseven (mmol·l−15.47±0.53 11.3±1.5 5.8±1.4 <0.001 
 Adrenaline (nmol·l−10.41±0.19 10.6±2.9 10.2±3.0 <0.001 
 Noradrenaline (nmol·l−11.63±0.47 1.65±0.59 0.02±0.34 0.88 
 Dopamine (nmol·l−10.08±0.03 0.08±0.02 −0.01±0.04 0.59 
ABG (n=8) 
 pH 7.42±0.02 7.42±0.02 0.01±0.02 0.49 
 sO2 (%) 97.9±0.7 98.0±0.6 0.1±0.8 0.81 
 pO2 (kPa) 13.8±1.1 13.5±1.1 −0.2±1.2 0.62 
 pCO2 (kPa) 5.33±0.19 4.89±0.42 −0.43±0.26 0.02 
 Arterial base excess (mmol·l−1)* 0.8±1.4 −0.5±1.5 −1.3±2.3 0.14 
 HCO3 (mmol·l−1)* 25.2±1.2 24.0±1.2 −1.2±2.0 0.12 
 Ca2+ (mmol·l−11.19±0.02 1.12±0.01 −0.07±0.02 <0.001 
 Haemoglobin (g·l−1145.6±9.7 145.5±9.0 −0.1±2.4 0.90 
 K+ (mmol·l−13.95±0.09 3.26±0.18 −0.70±0.15 <0.001 
 Na+ (mmol·l−1137.6±1.0 138.2±1.0 0.64±0.89 0.08 
 Cl (mmol·l−1107.8±1.9 105.2±2.0 −2.6±1.9 0.005 
 Haematocrit* 0.45±0.03 0.45±0.03 −0.0±0.01 0.94 
 Anion gap (mmol·l−1)* 5.0±2.6 9.7±3.3 4.7±3.1 0.003 
VBG (n=8) 
 sO2 (%) 71.9±8.6 88.5±6.6 16.6±7.2 <0.001 
 pO2 (kPa) 5.23±0.75 7.56±1.3 2.33±1.09 <0.001 
 pCO2 (kPa) 6.11±0.21 5.25±0.41 −0.86±0.35 <0.001 
NU over leg (n=8) 
 NU [Glu] (μmol·min−1·100 ml of leg volume−1)* 0.2±1.3 0.5±4.2 0.3±3.4 0.78 
 NU [O2] (nmol·min−1·100 ml of leg volume-1)* 530±270 500±480 37±226 0.64 
 NU [CO2] (nmol·min−1·100 ml of leg volume-1)* −3.3±2.9 −3.3±3.4 −0.0±5.4 0.99 
Intramuscular measurements (n=7) 
 Glycogen (μg·g of dry weight−1292±163 283±153 −10±40 0.54 
 State 3 respiration (nmol of O2·min−1·units of CS−146±13 61±17 15±12 0.01 
 State 4 respiration (nmol of O2·min−1·units of CS−14.8±1.3 5.5±1.6 0.7±1.7 0.32 
 RCR* 9.7±1.9 11.5±3.7 1.9±4.5 0.32 

Plasma lactate concentration increased approximately 4-fold in a response to the adrenaline challenge (Figure 3); however, with a large inter-individual variation. No changes in lactate concentration in muscle were observed (Table 2). Plasma enrichment reached a steady state (Figure 2) before as well as during adrenaline infusion. This was confirmed by the fact that the Ra was identical when derived from steady-state and non-steady-state equations respectively (Table 2). The plasma concentration of lactate correlated statistically with Ra (R=0.78, P=0.02), but not with the plasma concentration of adrenaline (R=0.62, P=0.1).

Table 2
Concentrations and kinetics of lactate before and during adrenaline infusion

Calculations of Ra, Rd and NB in whole body. Two- and three-compartment calculations of lactate production, utilization and fluxes between leg muscle and plasma before and after adrenaline infusion. P-values indicate differences between baseline and after adrenaline infusion. Ra for whole body is calculated with non-steady-state (NSS) and steady-state (SS) equations. n=8 unless otherwise indicated.

BaselineAdrenalineΔP-value
Whole-body kinetics (μmol·min−1·kg−1
Ra (NSS) 10.3±1.7 32.6±9.9 22.3±9.3 <0.001 
Ra (SS) 10.8±1.9 32.8±9.5 22.0±8.7 <0.001 
Rd 10.8±1.6 28.3±8.9 19.1±9.2 <0.001 
 NB −0.0±0.6 4.5±4.0 4.5±3.7 0.01 
Leg kinetics (two–compartment; μmol·min−1·100 ml of leg volume−1
RaL 0.43±0.12 1.24±0.55 0.81±0.45 0.001 
RdL 0.47±0.14 0.88±0.50 0.42±0.46 0.039 
 NU [lac] 0.03±0.09 −0.37±0.44 −0.40±0.47 0.048 
Leg kinetics (three-compartment) n=7 (μmol·min−1·100 ml of leg volume−1
 Production 5.7±3.2 3.9±3.3 −1.8±3.7 0.25 
 Utilization 5.7±3.2 3.5±3.5 −2.2±3.7 0.17 
Fin 1.19±0.53 9.6±4.1 8.3±4.0 <0.001 
Fout 1.18±0.52 9.9±4.2 8.7±4.1 <0.001 
Fma 0.48±0.15 1.32±0.79 0.85±0.73 0.016 
Fvm 0.47±0.14 1.69±0.96 1.22±0.86 0.005 
Fva 0.71±0.39 8.2±3.5 7.5±3.5 <0.001 
Concentrations 
 Lactateart (mmol·l−10.97±0.17 3.9±2.2 2.9±2.0 0.005 
 Lactateven (mmol·l−10.95±0.18 4.0±2.2 3.0±2.0 0.004 
 Lactatemusc (nmol·mg dry weight−1; n=7) 8.1±5.0 9.1±3.1 0.9±4.6 0.61 
Enrichments (APE of [13C]lactate, %) 
 Arterial 5.11±0.77 1.78±0.54 −3.33±0.84 <0.001 
 Venous 3.13±0.48 1.53±0.43 −1.60±0.42 <0.001 
 Muscle (n=7) 0.52±0.24 0.47±0.09 −0.05±0.29 0.66 
BaselineAdrenalineΔP-value
Whole-body kinetics (μmol·min−1·kg−1
Ra (NSS) 10.3±1.7 32.6±9.9 22.3±9.3 <0.001 
Ra (SS) 10.8±1.9 32.8±9.5 22.0±8.7 <0.001 
Rd 10.8±1.6 28.3±8.9 19.1±9.2 <0.001 
 NB −0.0±0.6 4.5±4.0 4.5±3.7 0.01 
Leg kinetics (two–compartment; μmol·min−1·100 ml of leg volume−1
RaL 0.43±0.12 1.24±0.55 0.81±0.45 0.001 
RdL 0.47±0.14 0.88±0.50 0.42±0.46 0.039 
 NU [lac] 0.03±0.09 −0.37±0.44 −0.40±0.47 0.048 
Leg kinetics (three-compartment) n=7 (μmol·min−1·100 ml of leg volume−1
 Production 5.7±3.2 3.9±3.3 −1.8±3.7 0.25 
 Utilization 5.7±3.2 3.5±3.5 −2.2±3.7 0.17 
Fin 1.19±0.53 9.6±4.1 8.3±4.0 <0.001 
Fout 1.18±0.52 9.9±4.2 8.7±4.1 <0.001 
Fma 0.48±0.15 1.32±0.79 0.85±0.73 0.016 
Fvm 0.47±0.14 1.69±0.96 1.22±0.86 0.005 
Fva 0.71±0.39 8.2±3.5 7.5±3.5 <0.001 
Concentrations 
 Lactateart (mmol·l−10.97±0.17 3.9±2.2 2.9±2.0 0.005 
 Lactateven (mmol·l−10.95±0.18 4.0±2.2 3.0±2.0 0.004 
 Lactatemusc (nmol·mg dry weight−1; n=7) 8.1±5.0 9.1±3.1 0.9±4.6 0.61 
Enrichments (APE of [13C]lactate, %) 
 Arterial 5.11±0.77 1.78±0.54 −3.33±0.84 <0.001 
 Venous 3.13±0.48 1.53±0.43 −1.60±0.42 <0.001 
 Muscle (n=7) 0.52±0.24 0.47±0.09 −0.05±0.29 0.66 

In the basal state, Ra and Rd of lactate were similar in leg and whole body. Measured as whole-body turnover, both the Ra and the Rd roughly triple with a positive NB of approximately 4.5 μmol·min−1·kg of body weight−1 in response to the adrenaline challenge. In the leg, however, RaL increased 3-fold whereas RdL barely doubled when estimated with a two-compartment model. Thus rendering in an NU of 0.40 (±0.47) μmol·min−1·100 ml of leg volume−1 (Table 2), the leg thereby acts as a net contributor to the plasma lactate pool.

The three-compartment estimation rendered no detectable change in intramuscular lactate production and utilization after the adrenaline infusion. Compared with basal state, adrenaline increased muscle uptake of lactate (Fma from 0.5 to 1.3 μmol·min−1·100 ml of leg volume−1) as well as release of lactate from muscle to plasma (Fvm from 0.5 to 1.3 μmol·min−1·100 ml of leg volume−1), thus rendering skeletal muscle a net contributor of approximately 0.4 μmol·min−1·100 ml of leg volume−1. These changes were considerably smaller than the increase in lactate shunted past the muscle during adrenaline infusion as Fva increased nearly 12-fold from 0.7 to 8.2 μmol·min−1·100 ml of leg volume−1 (Figure 5). This was attributed to increase in blood flow in the leg as well as higher plasma concentrations of lactate (Tables 1 and 2)

Lactate fluxes before (numbers on top) and during (bold numbers) infusion of adrenaline

Figure 5
Lactate fluxes before (numbers on top) and during (bold numbers) infusion of adrenaline

All fluxes are expressed as μmol·min−1·100 ml of leg volume−1 ± S.D. *P<0.05 for difference between time points.

Figure 5
Lactate fluxes before (numbers on top) and during (bold numbers) infusion of adrenaline

All fluxes are expressed as μmol·min−1·100 ml of leg volume−1 ± S.D. *P<0.05 for difference between time points.

Muscle mitochondrial state 3 respiration increased by 33% (±25%, P=0.01; Figure 6). State 4 respiration, RCR, oxygen and glycogen consumption, as well as CO2 production, all remained unchanged during the adrenaline challenge (Table 1).

Mitochondrial state 3 respiration for individual subjects before and during adrenaline infusion

Figure 6
Mitochondrial state 3 respiration for individual subjects before and during adrenaline infusion

*P<0.05 for differences between time points.

Figure 6
Mitochondrial state 3 respiration for individual subjects before and during adrenaline infusion

*P<0.05 for differences between time points.

DISCUSSION

In the present study the use of isotopically labelled lactate enabled us to estimate whole-body lactate kinetics (Ra, Rd and NB) as well as leg-muscle lactate kinetics, through the use of arteriovenous blood sampling combined with muscle biopsies. The three-compartment model showed a high intramuscular lactate turnover in relation to the release rate. However, variations were large using the three-compartment model. Results show that adrenaline infusion increases whole-body lactate turnover with a larger increase in Ra than in Rd, resulting in increased plasma lactate levels. Both the release and the uptake of lactate by leg muscle were increased with the adrenaline infusion, but, due to a larger increase in the release, leg muscle became a net contributor to the increased lactate appearance into the circulation. This increased release was not accompanied by a measurable change in intramuscular muscle lactate production, suggesting that, during the adrenaline infusion, a larger part of the intramuscular lactate turnover is released into the circulation.

Adrenaline infusion (0.08 μg·min−1·kg−1) increased plasma lactate to approximately 4 mM. Whole-body Ra and Rd increased with a positive NB at the end of the experiment, thus not reaching steady-state plasma concentration. This demonstrated that both plasma recruitment of lactate and uptake in tissues increased under continuous stimulation of adrenaline and that the increase in plasma concentration was an effect of a larger increase in Ra than in Rd. When Meyer et al. [1] studied the whole-body kinetics of lactate in healthy volunteers during a continuous infusion of adrenaline, Ra increased from 12 to 42 μmol·min−1·kg−1, in accordance with our results.

In the two-compartment model, plasma was sampled from artery and femoral vein thus reflecting metabolism in the leg. Similarly to others [12,14,18], we assumed this to be an estimation of metabolism in skeletal muscle. RaL roughly tripled (as Ra in whole body), whereas RdL barely doubled, rendering skeletal muscle a net contributor of lactate under the influence of adrenaline. This was expected and confirmed previous studies with local stimulation of adrenergic receptors in skeletal muscle in human subjects [19,20].

The use of a three-compartment model to characterize lactate metabolism in skeletal muscle of human subjects has not been reported before. The advantage of this technique is that it enables estimation of the internal turnover of lactate within the muscle as well as lactate fluxes between the arterial, venous and intramuscular compartments. This estimates how much of the produced lactate is actually released into the plasma compartment. In our study, the intramuscular production was higher than the amount leaving the muscle, 10% (±4%) in the basal state compared with 53% during adrenaline infusion (±28%, P=0.005; Figure 5). The finding of Chinkes et al. [14] that a 15% fraction of produced lactate is exported from muscle in the anaesthetized dog supports these results.

In our experimental situation, the increase in plasma lactate correlated with an increased net export from skeletal muscle. This could be interpreted to reflect increased muscular lactate production if assessed solely from whole-body or two-compartment models. However, in our three-compartment calculations, with additional muscle biopsies, we were not able to detect an actual increase in lactate production, even though we see an increase in transmembranous export (Figure 5). These results suggest that the adrenaline increases the release of muscle lactate (Fvm) from 0.5 to 1.7 μmol·min−1·100 ml of leg volume−1 without increasing the production of lactate in skeletal muscle. The only explanation for this is that that a larger part of the production is redirected to plasma rather than an internal degradation. Since the intramuscular recycling of lactate is high in skeletal muscle, this would allow for an immediate increased release of lactate into the circulation without a need to activate a net production of extra lactate. The consequence of this should be that less of the lactate is used for recycling. However, the large variation in the calculations of the intramuscular fluxes does not allow us to measure these differences with enough accuracy for any definite conclusions. In the present study, we do not see any increase in glucose uptake or decrease in glycogen stores in muscle supplying the extra precursors needed for the increased lactate release, again supporting the hypothesis that less of the lactate is recycled. The unchanged glycogen stores in muscle and glucose uptake confirms previous data, where a 2-h continuous infusion of adrenaline does not change muscular glycogen [21] and adrenaline-induced insulin resistance is repeatedly demonstrated during hyperinsulinaemic euglycaemic clamps [21,22].

In addition, we measured mitochondrial respiration and muscle oxygen extraction and CO2 production to link the increased lactate to a potential anaerobic-induced production. Adrenaline caused an increase in skeletal muscle mitochondria state 3 respiration. This has previously not been examined in human muscle, but both Breton et al. [23] and Wang et al. [24] report similar changes in rat liver mitochondria in response to adrenaline challenges.

This increase in respiratory rate was not accompanied by measurable increases in oxygen consumption or CO2 production over leg muscle, indicating an unaltered in vivo mitochondrial activity. This is probably due to the lack of energy depletion in the resting muscle as the up-regulation of mitochondrial activity is seen in the presence of ADP and abundance of energy substrates during ex vivo analysis. Although metabolically stressed through pharmacological stimulation, the subjects in this trial were bed-rested and their ADP/ATP ratio was probably too low for the up-regulation of potential state 3 respiration to actually result in an increased activity. Still, in a situation of adrenaline stimulation and increased energy demands, our ex vivo results suggest a potentially more rapid in vivo response if needed. The lack of an inhibition of mitochondrial function and the unchanged oxygen uptake over leg muscle suggest that the increased lactate production due to the adrenaline is not due to an increased anaerobic metabolism.

Our study is interesting in that it is the first one examining lactate metabolism using a three-compartment model in human. It also contributes to a deeper understanding of lactate metabolism under influence of adrenaline. It is the first study examining the effect of adrenaline on oxidative phosphorylation in human skeletal muscle mitochondria. However, there are some limitations of our study. There was a large variation in blood flow between subjects, which resulted in a large variation of fluxes. Especially in combination with the low enrichment intramuscularly (Table 2) and limited number of samples, this resulted in a rather large variation in the three-compartment model. This happened despite our efforts to get a reliable intramuscular measurement by cleaning the muscle samples and measuring in 5-fold. To get around this, more muscle samples could be obtained, but this will increase the invasiveness of the method. An alternative is to include more patients, but also this is limited by the necessity to take a muscle biopsy. Tracer enrichment steady state was obtained, whereas plasma lactate concentrations did not reach a definite plateau; this, however, was handled by using non-steady-state equations for whole-body kinetics.

In summary, we examined lactate metabolism using three methods: whole-body turnover, a two-compartment and a three-compartment model. Our results show that adrenaline increases lactate levels in plasma due to an increased appearance/release of lactate into the plasma pool. Utilization of a two-pool model showed that skeletal muscle increases its release of lactate and thereby significantly contributes to the adrenaline induced increase in lactate levels. The three-pool model suggests that this increase is most probably not the result of an increased production of lactate, but a redirection of a larger part of the production to a release into plasma. However, the latter conclusion is an indirect conclusion because the variation in the three-pool model makes small changes in the intramuscular fluxes not detectable.

For future physiological or clinical studies, the most appropriate technique to answer the question can be used. The whole-body method is, due to its lesser invasiveness, possible to apply to larger groups of patients, but will not reveal any source/tissue that contributes to changes Ra or clearance from the plasma pool. The two-pool model allows the study of tissues or tissue beds and can elucidate which of these contribute with a net release of utilization of lactate and contributing to possible plasma changes. The three-pool model gives detailed information on whether the actual production or tissue utilization rates change but, due to its invasiveness and large variation in especially the intracellular fluxes, is less useful to detect small changes in small groups of subjects.

AUTHOR CONTRIBUTION

Jonathan Grip co-designed the study, carried out the clinical study, analysed data, performed kinetic calculations, wrote the manuscript and co-wrote ethical application. Towe Jakobsson validated and performed laboratory analysis and mitochondrial analysis. Christina Hebert validated and performed analysis for glycogen and lactate. Maria Klaude validated and performed laboratory analysis for lactate enrichment. Gustaf Sandström performed blood flow calculations, assisted in sample collections and laboratory analysis. Jan Wernerman performed medical supervision, regular consulting input, co-designed the study, performed procedures, co-wrote the manuscript and co-wrote ethical application. Olav Rooyackers designed the study, analysed the data, performed kinetic calculations, performed blood flow measurements and calculations.

FUNDING

This work was supported by the European Society of Clinical Nutrition and Metabolism through a personal fellowship to JG; and the Stockholm County Council [grant number 20100033]. Neither European Society of Clinical Nutrition and Metabolism nor Stockholm County Council had any influence on study design, study performance or interpretation of the data.

Abbreviations

     
  • ABG

    arterial blood gas

  •  
  • APE

    atom percentage excess

  •  
  • BMI

    body mass index

  •  
  • CS

    citrate synthase

  •  
  • ECG

    electrocardiography

  •  
  • LP

    lactate production

  •  
  • LU

    lactate utilization

  •  
  • NB

    net balance

  •  
  • NU

    net uptake

  •  
  • PF

    plasma flow

  •  
  • Ra

    rate of appearance

  •  
  • RCR

    respiratory control ration

  •  
  • Rd

    rate of disappearance

  •  
  • VBG

    venous blood gas

References

References
1
Meyer
 
C.
Stumvoll
 
M.
Welle
 
S.
Woerle
 
H.J.
Haymond
 
M.
Gerich
 
J.
 
Relative importance of liver, kidney, and substrates in epinephrine-induced increased gluconeogenesis in humans
Am. J. Physiol. Endocrinol. Metab.
2003
, vol. 
285
 (pg. 
E819
-
E826
)
[PubMed]
2
Jansen
 
T.C.
van Bommel
 
J.
Mulder
 
P.G.
Lima
 
A.P.
van der Hoven
 
B.
Rommes
 
J.H.
Snellen
 
F.T.
Bakker
 
J.
 
Prognostic value of blood lactate levels: does the clinical diagnosis at admission matter?
J. Trauma
2009
, vol. 
66
 (pg. 
377
-
385
)
[PubMed]
3
Wacharasint
 
P.
Nakada
 
T.A.
Boyd
 
J.H.
Russell
 
J.A.
Walley
 
K.R.
 
Normal-range blood lactate concentration in septic shock is prognostic and predictive
Shock
2012
, vol. 
38
 (pg. 
4
-
10
)
[PubMed]
4
Nguyen
 
H.B.
Rivers
 
E.P.
Knoblich
 
B.P.
Jacobsen
 
G.
Muzzin
 
A.
Ressler
 
J.A.
Tomlanovich
 
M.C.
 
Early lactate clearance is associated with improved outcome in severe sepsis and septic shock
Crit. Care Med.
2004
, vol. 
32
 (pg. 
1637
-
1642
)
[PubMed]
5
Trzeciak
 
S.
Dellinger
 
R.P.
Chansky
 
M.E.
Arnold
 
R.C.
Schorr
 
C.
Milcarek
 
B.
Hollenberg
 
S.M.
Parrillo
 
J.E.
 
Serum lactate as a predictor of mortality in patients with infection
Intensive Care Med.
2007
, vol. 
33
 (pg. 
970
-
977
)
[PubMed]
6
Jones
 
A.E.
Shapiro
 
N.I.
Trzeciak
 
S.
Arnold
 
R.C.
Claremont
 
H.A.
Kline
 
J.A.
 
Lactate clearance vs central venous oxygen saturation as goals of early sepsis therapy: a randomized clinical trial
JAMA
2010
, vol. 
303
 (pg. 
739
-
746
)
[PubMed]
7
Levy
 
B.
Gibot
 
S.
Franck
 
P.
Cravoisy
 
A.
Bollaert
 
P.E.
 
Relation between muscle Na+K+ ATPase activity and raised lactate concentrations in septic shock: a prospective study
Lancet
2005
, vol. 
365
 (pg. 
871
-
875
)
[PubMed]
8
Levy
 
B.
Perez
 
P.
Gibot
 
S.
Gerard
 
A.
 
Increased muscle-to-serum lactate gradient predicts progression towards septic shock in septic patients
Intensive Care Med.
2010
, vol. 
36
 (pg. 
1703
-
1709
)
[PubMed]
9
McCarter
 
F.D.
Nierman
 
S.R.
James
 
J.H.
Wang
 
L.
King
 
J.K.
Friend
 
L.A.
Fischer
 
J.E.
 
Role of skeletal muscle Na+-K +ATPase activity in increased lactate production in sub-acute sepsis
Life Sci.
2002
, vol. 
70
 (pg. 
1875
-
1888
)
[PubMed]
10
Wolfe
 
R.R.
Chinkes
 
D.L.
 
Isotope Tracers in Metabolic Research: Principles and Practice of Kinetic Analysis
1992
Wiley-Liss
11
van Hall
 
G.
Stromstad
 
M.
Rasmussen
 
P.
Jans
 
O.
Zaar
 
M.
Gam
 
C.
Quistorff
 
B.
Secher
 
N.H.
Nielsen
 
H.B.
 
Blood lactate is an important energy source for the human brain
J. Cereb. Blood Flow Metab.
2009
, vol. 
29
 (pg. 
1121
-
1129
)
[PubMed]
12
van Hall
 
G.
Jensen-Urstad
 
M.
Rosdahl
 
H.
Holmberg
 
H.C.
Saltin
 
B.
Calbet
 
J.A. L.
 
Leg and arm lactate and substrate kinetics during exercise
Am. J. Physiol. Endocrinol. Metab.
2003
, vol. 
284
 (pg. 
E193
-
E205
)
[PubMed]
13
Lehman
 
S.L.
 
Measurement of lactate production by tracer techniques
Med. Sci. Sports Exerc.
1991
, vol. 
23
 (pg. 
935
-
938
)
[PubMed]
14
Chinkes
 
D.L.
Zhang
 
X.J.
Romijn
 
J.A.
Sakurai
 
Y.
Wolfe
 
R.R.
 
Measurement of pyruvate and lactate kinetics across the hindlimb and gut of anesthetized dogs
Am. J. Physiol.
1994
, vol. 
267
 (pg. 
E174
-
E182
)
[PubMed]
15
Vesali
 
R.
Klaude
 
M.
Rooyackers
 
O.
Tjäder
 
I.
Barle
 
H.
Wernerman
 
J.
 
Longitudinal pattern of glutamine/glutamate balance across the leg in long-stay intensive care unit patients
Clin. Nutr.
2002
, vol. 
21
 (pg. 
505
-
514
)
[PubMed]
16
Tonkonogi
 
M.
Fernstrom
 
M.
Walsh
 
B.
Ji
 
L.L.
Rooyackers
 
O.
Hammarqvist
 
F.
Wernerman
 
J.
Sahlin
 
K.
 
Reduced oxidative power but unchanged antioxidative capacity in skeletal muscle from aged humans
Pflugers Arch.
2003
, vol. 
446
 (pg. 
261
-
269
)
[PubMed]
17
Fredriksson
 
K.
Hammarqvist
 
F.
Strigard
 
K.
Hultenby
 
K.
Ljungqvist
 
O.
Wernerman
 
J.
Rooyackers
 
O.
 
Derangements in mitochondrial metabolism in intercostal and leg muscle of critically ill patients with sepsis-induced multiple organ failure
Am. J. Physiol. Endocrinol. Metab.
2006
, vol. 
291
 (pg. 
E1044
-
E1050
)
[PubMed]
18
Vesali
 
R.
Cibicek
 
N.
Jakobsson
 
T.
Klaude
 
M.
Wernerman
 
J.
Rooyackers
 
O.
 
Protein metabolism in leg muscle following an endotoxin injection in healthy volunteers
Clin. Sci.
2010
, vol. 
118
 (pg. 
421
-
427
)
19
Gjedsted
 
J.
Buhl
 
M.
Nielsen
 
S.
Schmitz
 
O.
Vestergaard
 
E.T.
Tonnesen
 
E.
Moller
 
N.
 
Effects of adrenaline on lactate, glucose, lipid and protein metabolism in the placebo controlled bilaterally perfused human leg
Acta Physiol.
2011
, vol. 
202
 (pg. 
641
-
648
)
20
Qvisth
 
V.
Hagström-Toft
 
E.
Enoksson
 
S.
Bolinder
 
J.
 
Catecholamine regulation of local lactate production in vivo in skeletal muscle and adipose tissue: role of β-adrenoreceptor subtypes
J. Clin. Endocrinol. Metab.
2008
, vol. 
93
 (pg. 
240
-
246
)
[PubMed]
21
Laurent
 
D.
Petersen
 
K.F.
Russell
 
R.R.
Cline
 
G.W.
Shulman
 
G.I.
 
Effect of epinephrine on muscle glycogenolysis and insulin-stimulated muscle glycogen synthesis in humans
Am. J. Physiol.
1998
, vol. 
274
 (pg. 
E130
-
E138
)
[PubMed]
22
Jensen
 
J.
Ruge
 
T.
Lai
 
Y.C.
Svensson
 
M.K.
Eriksson
 
J.W.
 
Effects of adrenaline on whole-body glucose metabolism and insulin-mediated regulation of glycogen synthase and PKB phosphorylation in human skeletal muscle
Metabolism
2011
, vol. 
60
 (pg. 
215
-
226
)
[PubMed]
23
Breton
 
L.
Clot
 
J.P.
Bouriannes
 
J.
Baudry
 
M.
 
Adrenaline activates oxidative phosphorylation of rat liver mitochondria through alpha 1-receptors
C. R. Seances Soc. Biol. Fil.
1987
, vol. 
181
 (pg. 
242
-
248
)
[PubMed]
24
Wang
 
X.M.
Yang
 
L.
Chen
 
K.M.
 
Catecholamines: important factors in the increase of oxidative phosphorylation coupling in rat-liver mitochondria during the early phase of burn injury
Burns
1993
, vol. 
19
 (pg. 
110
-
112
)
[PubMed]