Limb skeletal muscle adaptation
in athletes
after training at altitude MASAO BENTE August
MIZUNO, SEN, ERIK
MASAO, MYGIND,
MIZUNO, SCHIBYE, Krogh
CARSTEN BIRGER
Institute,
CARSTEN &EL, BENTE
University
THOMAS
SCHIBYE,
BIRGER
of Copenhagen,
BRO-RASMUSRASMUSSEN,
AND BENGT SALTIN. Limb skeletal muscle adaptation in athletes after training at altitude. J. Appl. Physiol. 68(2): 496-502,
1990.-Morphological and biochemicalcharacteristics of biopsies obtained from gastrocnemius(GAS) and triceps brachii muscle (TRI), as well as maximal O2 uptake (90~ max)and O2 deficit, were determinedin 10 well-trained cross-country skiers before and after a Z-wk stay (2,100 m above sea level) and training (2,700 m above sealevel) at altitude. On return to sea level, vo2 mBxwas the sameas the prealtitude value, whereas an increase in O2 deficit (29%) and in short-term running performance (17%) was observed (P c 0.05). GAS showed maintained capillary supply but a 10% decreasein mitochondrial enzyme activities (P c 0.05), whereas an increase in capillary supply (P < 0.05) but unchangedmitochondrial enzyme activities
were observed
in TRI.
Buffer
capacity
was
increasedby 6% in both GAS and TRI (P c 0.05). A positive correlation was found between the relative increasein buffer capacity of GAS and short-term
running
ERIK
JUEL, THOMAS BRO-RASMUSSEN, RASMUSSEN, AND BENGT SALTIN
time (P < 0.05). Thus
the presentstudy indicatesno effect of 2 wk of altitude training on W2 max but provides evidenceto suggestan improvement in short-term exercise performance, which may be the result of an increasein musclebuffer capacity.
DK-2100
Copenhagen
MYGIND,
0, Denmark
discrepancy in observations &y be attributed to the training status of subjects before altitude sojourn, the level of altitude, and the intensity as well as the duration of exercise performed at altitude, In a number of animal experiments an increased capillary network and an elevated oxidative potential of skeletal muscles, including myoglobin content, mitochondrial protein content, and oxidative enzyme activities, have been reported to accompany chronic exposure to hypoxia (for references see Ref. 28). The results Ied some authors to propose tissue hypoxia as the major stimulus for an enhancement of oxidative potentials and/ or an increase in capillary supply. In several other studies, no or less pronounced effects have been observed when this concept has been challenged. Thus, in human skeletal muscles, no change in the number of capillaries around the muscle fibers was found after long-term sojourn at high altitude (28) or after 40 days of progressive hypobaria (15). Mitochondrial enzyme activities have
been found to be unchanged (34) or even decreased (6, 15, 28). Data are also available describing decreased glycolytic enzyme activities in human skeletal muscles accompanied by an increase in short-term work capacity
maximal O2 uptake; O2 deficit; muscle morphology; skeletal after training performed in a low-pressure chamber (32). muscleenzyme activities; musclebuffer capacity; hypoxia The present study was therefore undertaken to evalu-
ate the effect of altitude O2 pressure in inspiratory air, such as that induced by exposure to hypoxic environments, is accom-
A REDUCED
panied by a decrease in arterial O2 tension. With acute
exposure to high altitude, it has been well documented that a decrease in maximal O2 uptake (Vo2 max) occurs (1, 10, 11, 13, 26, 31). Even at a lowered O2 pressure
corresponding to only 900 m above sea level, a reduction in VO~ maxcan be seen in well-trained individuals (33). After
training at altitude, however, it is still. unclear
whether an effect on VO, max and work performance capacity can be expected on return to sea level. Physical
training
for 2 wk on the limb
skeletal muscle morphology, enzyme activities, and mus-
cle buffer capacity as well as on VOW maxand 02 deficit. Because the training status and the physical condition of subjects before altitude training are important factors
for evaluating the effects of altitude training, elite crosscountry skiers were selected as subjects in this study. Moreover, before the training camp at altitude, 902 max
was regularly determined during a 5-mo training period at sea level. This step was taken to ensure that the subjects had reached a stable physical condition before
training
at altitude.
activities performed during altitude sojourn, including
climbing, and skiing, have been rean improvement in work performance and/or an increase in VO 2 m8Xon return to sea level (5, 17). In contrast, unchanged VOW max has been observed
SUBJECTS
after mountain expeditions (6, 8). Several investigations in which endurance-trained athletes were studied have also demonstrated an improved running performance and an increased i'o 2 maxafter altitude training (10, ll), although others have observed no change (1,13,26). The
Ten well-trained male cross-country skiers, including six members of the Danish national team, were studied. The subjects were informed about possible risks involved in this study before they gave their oral consent to participate. The median age was 22 (range 18-33) yr,
hiking, mountain ported to facilitate
496
0161-7567/90
$1.50 Copyright
AND
METHODS
Subjects
0 1990 the American
Physiological
Society
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 16, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
SKELETAL
MUSCLE
ADAPTATION
height 184 (173-190) cm, weight 73 (67-82) kg, and . vo 2 max72 (66-77) ml. kg-’ min. The subjects stayed at 2,100 m above sea level and trained at 2,700 m above sea level for 2 wk after a 5-mo controlled training period at sea level. The type of training performed at sea level and during the altitude sojourn included roller skiing or cross-country skiing and running. At sea level, some strength training, cycling, swimming, canoeing, and rowing were performed as well. During training and maximal exercise at sea level and at altitude, heart rate was registered to ensure that the time as well as the intensity of training were equivalent. l
Methods * vo 2 maX.Treadmill
running was used for the determination of V02,max according to the protocol described by Saltin and Astrand (27), in which a constant speed and step increase in the elevation were employed. Measurements were performed in May, August, and September during the training period at sea level as well as immediately before (prealtitude) and 2 days after the stay at altitude (postaltitude). Expiratory gas was collected into Douglas bags and the air volume was measured in a Tissot spirometer. O2 and CO2 concentrations were determined with a paramagnetic O2 analyzer (Servomex) and an infrared CO, analyzer (Beckman LB-II), respectively. These analyzers were regularly calibrated during the day by known concentrations of gas determined by the Scholander technique. 02 deficit was calculated by using the nomogram described by Margaria et al. (20). A constant electrocardiogram was recorded during treadmill running. Short- term performance. Running time to exhaustion during the VO 2 maxmeasurement (range 240-380 s) was used as an index for short-term running performance in the August, September, and pre- and postaltitude tests. Muscle biopsies. Muscle biopsies were obtained from gastrocnemius (GAS) and triceps brachii muscle (TRI) by the needle technique (4). The samples were taken 4 days before and within 36 h after the altitude sojourn. The muscle tissue was immediately dissected into two parts. One piece was rolled on a filter paper for a couple of seconds to remove the blood. Thereafter, it was frozen directly in liquid N2 and used for biochemical analyses. The time elapsing before freezing was 15-20 s. The remaining piece was mounted in an embedding medium (OCT Compound, Miels Tissue Tek) for histochemical analyses and frozen in isopentane at freezing point cooled with liquid N2. The samples were stored at -80°C until analyses were performed. In two subjects, the small size of muscle biopsy did not allow all analyses to be performed. Thus the number of observations is eight for some variables. Histochemical analyses. Serial transverse sections, 10 pm thick, were cut at -20°C by using a Cryo-cut Cryostat Microtome (American Optical) and incubated for myofibrillar ATPase reaction at pH 9.4 (23). Applying preincubations at pH 4.3, 4.6, and 10.3 (7), the fibers were classified under microscopy as slow twitch type I (ST), fast twitch type IIa (FTa), type IIb (FTb), and type IIc (FTC). The relative occurrence of fiber types was deter-
WITH
ALTITUDE
497
TRAINING
mined from a median 312 (range 203-731) fibers in each biopsy. The fibers identified as FTb or FTC amounted to 09.1% and O-1.7%, respectively. Capillaries were visualized by using the amylaseperiodic acid-Schiff method (2). Photomicrographs with a magnification of 150 were taken to determine the number of capillaries per fiber and the number found around each fiber type as well as to measure fiber crosssectional area and fiber circumference (22). The variables related to each fiber type were determined at least in 20 fibers of each type. To evaluate the reliability of the determined fiber occurrence, fiber area, and number of capillaries found around each fiber type, a comparison between two parts of each biopsy was performed (Table 1) . Biochemical analyses. The muscle samples for quantitative determinations of enzyme activities and buffer capacity were freeze-dried in a thermoelectric freezedryer (Hetosicc CD 52) for 48 h at -50°C and a pressure of 0.01 Torr. Thereafter, each sample was dissected under a microscope to remove blood, fat, and connective tissue in a room where the temperature (20°C) and humidity (~30%) were controlled. There was no difference in water content between samples obtained before and after altitude training (median 75.6, range 73.3-79.5%). On the samples obtained pre- and postaltitude, the following analytic procedures were performed to determine enzyme activities and buffer capacity. Enzyme actiuities. Fiber fragments with a median weight of 2.7 (1.4-4.3) mg were homogenized in an icechilled 0.3 M phosphate buffer adjusted to pH 7.7 and containing 0.5 mg/ml of bovine serum albumin. The volume of the buffer was equal to 400 times that of the fiber weight in milligrams. Total lactate dehydrogenase, citrate synthase, and 3-hydroxyacyl-CoA dehydrogenase activities were determined by using fluorometric methods of NAD-NADP coupled reactions as described by EssenGustavsson and Henriksson (12). A radioactive assay (14) was employed to determine total phosphorylase 1. Reliability of methods used for morphological characteristics, enzyme activities, and buffer capacity in gastrocnemius and triceps brachii muscle TABLE
Analysis
n
Muscle morphology Relative fiber occurrence ST FTa Mean fiber area ST FTa No. of capillaries around ST FTa Muscle enzyme activities Phosphorylase Lactate dehydrogenase Citrate synthase 3-Hydroxyacyl-CoA dehydrogenase Muscle buffer capacity n, No. of observations. nations.
Two parts
of biopsy
Coefficient Variation,
9 9
3.7
9 9
2.9 5.0
9 9
2.6
20 20 20
4.7 2.0 2.3 2.0 1.7
20 14
were
of %
6.1
3.4
used for determi-
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 16, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
498
activity.
SKELETAL
Enzyme activities
expressed
as micromoles
l
MUSCLE
ADAPTATION
WITH
ALTITUDE Maximal
were measured at 25°C and rnirPo g-’ dry weight.
TRAINING
oxygen
uptake
6t
Buffer capacity. Fiber fragments with a median weight of 1.3 (0.4-2.5) mg were homogenized in a salt solution containing
145 mM
KCl,
10 mM
NaCl,
and 5 mM
iodoacetic acid adjusted to pH 7.0. The volume of the solution was equal to 200 times that of the fiber weight in milligrams.
Glycolysis
was inhibited
by the addition
of iodoacetic acid (19). The loo-p1 homogenized dilution was adjusted to pH 7.0 with 0.01 N NaOH or HCl and titrated
to pH 6.0 with
0.01
Oxygen
N HCl. The sample was
mixed well by an electric blender at each step of titration. The pH electrode was made of tri-n-dodecylamine in a phthalate-polyvinyl
deficit I
6
chloride membrane placed in a glass
capillary having an inner diameter of -1 mm. Muscle buffer capacity was calculated as the number of moles of H+ required to change the pH from 7.0 to 6.0 per g of
dry weight tissues (pm01 . g dry wt-’ pH-l). l
Statistics Running
Values are expressed as medians with a range. Comparisons between related observations and between independent observations were performed by the Wilcoxon matched-pairs signed-rank test and the Mann- Whitney U test, respectively. The interdependence of observations was expressed by the Spearman rank correlation coefficient. In these analyses, the statistical tests described by Siegel (30) were employed. A level of 5% was considered significant for two-tailed tests.
time
100 set 300
Aug
May
RESULTS
FIG.
running
Preal titude .
(Aug),
VO2 maX.In the prealtitude experiments, a median of 5.0 (range 4.5-5.3) l/min in VOW m&xwas obtained. This value did not differ from those determined during training at sea level, except for the value in September, which was 6% higher (P < 0.05; Fig. 1). The median for body weight varied
in athletes
after training at altitude MASAO BENTE August
MIZUNO, SEN, ERIK
MASAO, MYGIND,
MIZUNO, SCHIBYE, Krogh
CARSTEN BIRGER
Institute,
CARSTEN &EL, BENTE
University
THOMAS
SCHIBYE,
BIRGER
of Copenhagen,
BRO-RASMUSRASMUSSEN,
AND BENGT SALTIN. Limb skeletal muscle adaptation in athletes after training at altitude. J. Appl. Physiol. 68(2): 496-502,
1990.-Morphological and biochemicalcharacteristics of biopsies obtained from gastrocnemius(GAS) and triceps brachii muscle (TRI), as well as maximal O2 uptake (90~ max)and O2 deficit, were determinedin 10 well-trained cross-country skiers before and after a Z-wk stay (2,100 m above sea level) and training (2,700 m above sealevel) at altitude. On return to sea level, vo2 mBxwas the sameas the prealtitude value, whereas an increase in O2 deficit (29%) and in short-term running performance (17%) was observed (P c 0.05). GAS showed maintained capillary supply but a 10% decreasein mitochondrial enzyme activities (P c 0.05), whereas an increase in capillary supply (P < 0.05) but unchangedmitochondrial enzyme activities
were observed
in TRI.
Buffer
capacity
was
increasedby 6% in both GAS and TRI (P c 0.05). A positive correlation was found between the relative increasein buffer capacity of GAS and short-term
running
ERIK
JUEL, THOMAS BRO-RASMUSSEN, RASMUSSEN, AND BENGT SALTIN
time (P < 0.05). Thus
the presentstudy indicatesno effect of 2 wk of altitude training on W2 max but provides evidenceto suggestan improvement in short-term exercise performance, which may be the result of an increasein musclebuffer capacity.
DK-2100
Copenhagen
MYGIND,
0, Denmark
discrepancy in observations &y be attributed to the training status of subjects before altitude sojourn, the level of altitude, and the intensity as well as the duration of exercise performed at altitude, In a number of animal experiments an increased capillary network and an elevated oxidative potential of skeletal muscles, including myoglobin content, mitochondrial protein content, and oxidative enzyme activities, have been reported to accompany chronic exposure to hypoxia (for references see Ref. 28). The results Ied some authors to propose tissue hypoxia as the major stimulus for an enhancement of oxidative potentials and/ or an increase in capillary supply. In several other studies, no or less pronounced effects have been observed when this concept has been challenged. Thus, in human skeletal muscles, no change in the number of capillaries around the muscle fibers was found after long-term sojourn at high altitude (28) or after 40 days of progressive hypobaria (15). Mitochondrial enzyme activities have
been found to be unchanged (34) or even decreased (6, 15, 28). Data are also available describing decreased glycolytic enzyme activities in human skeletal muscles accompanied by an increase in short-term work capacity
maximal O2 uptake; O2 deficit; muscle morphology; skeletal after training performed in a low-pressure chamber (32). muscleenzyme activities; musclebuffer capacity; hypoxia The present study was therefore undertaken to evalu-
ate the effect of altitude O2 pressure in inspiratory air, such as that induced by exposure to hypoxic environments, is accom-
A REDUCED
panied by a decrease in arterial O2 tension. With acute
exposure to high altitude, it has been well documented that a decrease in maximal O2 uptake (Vo2 max) occurs (1, 10, 11, 13, 26, 31). Even at a lowered O2 pressure
corresponding to only 900 m above sea level, a reduction in VO~ maxcan be seen in well-trained individuals (33). After
training at altitude, however, it is still. unclear
whether an effect on VO, max and work performance capacity can be expected on return to sea level. Physical
training
for 2 wk on the limb
skeletal muscle morphology, enzyme activities, and mus-
cle buffer capacity as well as on VOW maxand 02 deficit. Because the training status and the physical condition of subjects before altitude training are important factors
for evaluating the effects of altitude training, elite crosscountry skiers were selected as subjects in this study. Moreover, before the training camp at altitude, 902 max
was regularly determined during a 5-mo training period at sea level. This step was taken to ensure that the subjects had reached a stable physical condition before
training
at altitude.
activities performed during altitude sojourn, including
climbing, and skiing, have been rean improvement in work performance and/or an increase in VO 2 m8Xon return to sea level (5, 17). In contrast, unchanged VOW max has been observed
SUBJECTS
after mountain expeditions (6, 8). Several investigations in which endurance-trained athletes were studied have also demonstrated an improved running performance and an increased i'o 2 maxafter altitude training (10, ll), although others have observed no change (1,13,26). The
Ten well-trained male cross-country skiers, including six members of the Danish national team, were studied. The subjects were informed about possible risks involved in this study before they gave their oral consent to participate. The median age was 22 (range 18-33) yr,
hiking, mountain ported to facilitate
496
0161-7567/90
$1.50 Copyright
AND
METHODS
Subjects
0 1990 the American
Physiological
Society
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 16, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
SKELETAL
MUSCLE
ADAPTATION
height 184 (173-190) cm, weight 73 (67-82) kg, and . vo 2 max72 (66-77) ml. kg-’ min. The subjects stayed at 2,100 m above sea level and trained at 2,700 m above sea level for 2 wk after a 5-mo controlled training period at sea level. The type of training performed at sea level and during the altitude sojourn included roller skiing or cross-country skiing and running. At sea level, some strength training, cycling, swimming, canoeing, and rowing were performed as well. During training and maximal exercise at sea level and at altitude, heart rate was registered to ensure that the time as well as the intensity of training were equivalent. l
Methods * vo 2 maX.Treadmill
running was used for the determination of V02,max according to the protocol described by Saltin and Astrand (27), in which a constant speed and step increase in the elevation were employed. Measurements were performed in May, August, and September during the training period at sea level as well as immediately before (prealtitude) and 2 days after the stay at altitude (postaltitude). Expiratory gas was collected into Douglas bags and the air volume was measured in a Tissot spirometer. O2 and CO2 concentrations were determined with a paramagnetic O2 analyzer (Servomex) and an infrared CO, analyzer (Beckman LB-II), respectively. These analyzers were regularly calibrated during the day by known concentrations of gas determined by the Scholander technique. 02 deficit was calculated by using the nomogram described by Margaria et al. (20). A constant electrocardiogram was recorded during treadmill running. Short- term performance. Running time to exhaustion during the VO 2 maxmeasurement (range 240-380 s) was used as an index for short-term running performance in the August, September, and pre- and postaltitude tests. Muscle biopsies. Muscle biopsies were obtained from gastrocnemius (GAS) and triceps brachii muscle (TRI) by the needle technique (4). The samples were taken 4 days before and within 36 h after the altitude sojourn. The muscle tissue was immediately dissected into two parts. One piece was rolled on a filter paper for a couple of seconds to remove the blood. Thereafter, it was frozen directly in liquid N2 and used for biochemical analyses. The time elapsing before freezing was 15-20 s. The remaining piece was mounted in an embedding medium (OCT Compound, Miels Tissue Tek) for histochemical analyses and frozen in isopentane at freezing point cooled with liquid N2. The samples were stored at -80°C until analyses were performed. In two subjects, the small size of muscle biopsy did not allow all analyses to be performed. Thus the number of observations is eight for some variables. Histochemical analyses. Serial transverse sections, 10 pm thick, were cut at -20°C by using a Cryo-cut Cryostat Microtome (American Optical) and incubated for myofibrillar ATPase reaction at pH 9.4 (23). Applying preincubations at pH 4.3, 4.6, and 10.3 (7), the fibers were classified under microscopy as slow twitch type I (ST), fast twitch type IIa (FTa), type IIb (FTb), and type IIc (FTC). The relative occurrence of fiber types was deter-
WITH
ALTITUDE
497
TRAINING
mined from a median 312 (range 203-731) fibers in each biopsy. The fibers identified as FTb or FTC amounted to 09.1% and O-1.7%, respectively. Capillaries were visualized by using the amylaseperiodic acid-Schiff method (2). Photomicrographs with a magnification of 150 were taken to determine the number of capillaries per fiber and the number found around each fiber type as well as to measure fiber crosssectional area and fiber circumference (22). The variables related to each fiber type were determined at least in 20 fibers of each type. To evaluate the reliability of the determined fiber occurrence, fiber area, and number of capillaries found around each fiber type, a comparison between two parts of each biopsy was performed (Table 1) . Biochemical analyses. The muscle samples for quantitative determinations of enzyme activities and buffer capacity were freeze-dried in a thermoelectric freezedryer (Hetosicc CD 52) for 48 h at -50°C and a pressure of 0.01 Torr. Thereafter, each sample was dissected under a microscope to remove blood, fat, and connective tissue in a room where the temperature (20°C) and humidity (~30%) were controlled. There was no difference in water content between samples obtained before and after altitude training (median 75.6, range 73.3-79.5%). On the samples obtained pre- and postaltitude, the following analytic procedures were performed to determine enzyme activities and buffer capacity. Enzyme actiuities. Fiber fragments with a median weight of 2.7 (1.4-4.3) mg were homogenized in an icechilled 0.3 M phosphate buffer adjusted to pH 7.7 and containing 0.5 mg/ml of bovine serum albumin. The volume of the buffer was equal to 400 times that of the fiber weight in milligrams. Total lactate dehydrogenase, citrate synthase, and 3-hydroxyacyl-CoA dehydrogenase activities were determined by using fluorometric methods of NAD-NADP coupled reactions as described by EssenGustavsson and Henriksson (12). A radioactive assay (14) was employed to determine total phosphorylase 1. Reliability of methods used for morphological characteristics, enzyme activities, and buffer capacity in gastrocnemius and triceps brachii muscle TABLE
Analysis
n
Muscle morphology Relative fiber occurrence ST FTa Mean fiber area ST FTa No. of capillaries around ST FTa Muscle enzyme activities Phosphorylase Lactate dehydrogenase Citrate synthase 3-Hydroxyacyl-CoA dehydrogenase Muscle buffer capacity n, No. of observations. nations.
Two parts
of biopsy
Coefficient Variation,
9 9
3.7
9 9
2.9 5.0
9 9
2.6
20 20 20
4.7 2.0 2.3 2.0 1.7
20 14
were
of %
6.1
3.4
used for determi-
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 16, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
498
activity.
SKELETAL
Enzyme activities
expressed
as micromoles
l
MUSCLE
ADAPTATION
WITH
ALTITUDE Maximal
were measured at 25°C and rnirPo g-’ dry weight.
TRAINING
oxygen
uptake
6t
Buffer capacity. Fiber fragments with a median weight of 1.3 (0.4-2.5) mg were homogenized in a salt solution containing
145 mM
KCl,
10 mM
NaCl,
and 5 mM
iodoacetic acid adjusted to pH 7.0. The volume of the solution was equal to 200 times that of the fiber weight in milligrams.
Glycolysis
was inhibited
by the addition
of iodoacetic acid (19). The loo-p1 homogenized dilution was adjusted to pH 7.0 with 0.01 N NaOH or HCl and titrated
to pH 6.0 with
0.01
Oxygen
N HCl. The sample was
mixed well by an electric blender at each step of titration. The pH electrode was made of tri-n-dodecylamine in a phthalate-polyvinyl
deficit I
6
chloride membrane placed in a glass
capillary having an inner diameter of -1 mm. Muscle buffer capacity was calculated as the number of moles of H+ required to change the pH from 7.0 to 6.0 per g of
dry weight tissues (pm01 . g dry wt-’ pH-l). l
Statistics Running
Values are expressed as medians with a range. Comparisons between related observations and between independent observations were performed by the Wilcoxon matched-pairs signed-rank test and the Mann- Whitney U test, respectively. The interdependence of observations was expressed by the Spearman rank correlation coefficient. In these analyses, the statistical tests described by Siegel (30) were employed. A level of 5% was considered significant for two-tailed tests.
time
100 set 300
Aug
May
RESULTS
FIG.
running
Preal titude .
(Aug),
VO2 maX.In the prealtitude experiments, a median of 5.0 (range 4.5-5.3) l/min in VOW m&xwas obtained. This value did not differ from those determined during training at sea level, except for the value in September, which was 6% higher (P < 0.05; Fig. 1). The median for body weight varied
