Effect of hemoglobin concentration on maximal uptake in canine gastrocnemius muscle in situ
0,
MICHAEL C. HOGAN, DONALD E. BEBOUT, AND PETER D. WAGNER Division of Physiology, Department of Medicine, University of California, San Diego, La Jolla, California 92093-0623 HOGAN,MICHAEL C.,DONALD E. BEBOUT,AND PETERD. WAGNER.Effect of hemoglobin concentration on maximal O2 uptake in canine gastrocnemius
muscle in situ. J. Appl. Physiol. to maximally working muscle was decreased by altering hemoglobin (Hb) concentration and arterial PO, (Pao,) to investigate whether the reductions in maximal 0, uptake (Vozmax) that occur with lowered [Hb] are in part related to changes in the effective muscle 0, diffusing capacity (Dmo,). Two sets of experiments were conducted. In the initial set (n = 8), three levels of Hb [5.8 t 0.3, 9.4 t 0.1, and 14.4 t 0.6 (SE) g/l00 ml] in the blood were used in random order to pump perfuse, at equal muscle blood flows and Pa,,, maximally working isolated dog gastrocnemius muscle. iTo, max declined with decreasing [Hb], but the relationship between TO, max and both the effluent venous PO, ( Pvo2) and the calculated mean capillary PO, (PC,,) was not linear through the origin and, therefore, not compatible with a single value of Dmo, (as calculated by Bohr integration using a model based on Fick’s law of diffusion). To clarify these results, a second set of experiments (n = 6) was conducted in which two levels of Hb (14.0 ~fi 0.6 and 6.9 t 0.6 g/l00 ml) were each combined with two levels of oxygenation (Pa,, 79 2 8 and 29 t ‘2 Torr) and applied in random sequence to again pump perfuse maximally working dog gastrocnemius muscle at constant blood flow. In these experiments, the relationship between VO2 max and both Pvo, and calculated Pi?,, for each [Hb] was consistent with a constant estimate of Dmoz as Pao, was reduced, but the calculated Dmo, for the lower [Hb] was 33% less than that at the higher [Hb] (P < 0.05). Whether the apparent reduction in Dmoz with lower [Hb] is due to increased erythrocyte spacing, reduced chemical off loading of 0, from Hb, or some other effect remains to be determined.
70(3): 1105-1112, 1991.-O, delivery
fatigue; skeletal muscle; gas exchange; sion limitation; hypoxia
exercise; lactate;
diffu-
IT IS WELL KNOWN that reduction in hemoglobin (Hb) concentration reduces maximal 0, uptake (Oo,,,,) in animals (8, 16) and humans (23). Clearly, one factor responsible is the reduced convective O2 delivery [flow X arterial O2 content (Ca,,)] to muscle, but it is also possible, as suggested by Stainsby et al. (21), that reductions in [Hb] have some additional effect on 0, diffusion from the erythrocyte to the mitochondria, thereby altering the muscle O2 diffusing capacity (Dmo,), which is an estimate of the diffusive conductance of the muscle for OZ. Cain (1) demonstrated a higher critical venous PO, (PvoZ) in resting whole animals when 0, delivery was reduced below critical levels (the point at which VO, begins to decline) by anemic hypoxia, than when O2 delivery was 0161-7567/91 $1.50 Copyright
reduced by hypoxic hypoxia. Cain (1) interpreted his results to suggest that the driving pressure for diffusion of O2 did not appear to have limited its transport to tissues and that VO, was more dependent on convective O2delivery. However, if the Dmo, is reduced when [Hb] is lowered, this could explain the difference Cain (1) found in critical PO, between anemic and hypoxic hypoxia. In fact, a recent report from Gutierrez et al. (10) suggests that the higher critical PvoZ found during anemic hypoxia than during hypoxic hypoxia in O,-limited resting muscle may be a result of an increase in the resistance to 0, diffusion (or a reduced DmoJ during anemic hypoxia. We have recently suggested (12, 13, 17) that 0, diffusion in the peripheral tissue is one important determinant of VO, max as 0, delivery is reduced and that O2delivery is not the unique determinant of VO, max (14). Gutierrez and colleagues (9-U) have also shown that 0, diffusion limitation may have an important role in determining the Vo2 of resting muscle when the 0, delivery is reduced below some critical level. These previous studies (9,12,17) demonstrated strong correlations between VO, and the driving pressure for 0, diffusion in the capillary, estimated by a calculation of the mean capillary PO, (PEo,), so the reduction in vo2 with each level of reduced O2 delivery was hypothesized to be in part due to the decreased capillary 0, driving pressure. In these studies (12, l7), the calculated Dmoz was unchanged with the varied levels of hypoxemia, possibly because of the imposed constancy of muscle blood flow during the different conditions. The purpose of the present experiments was to investigate whether the estimated DmoZ is in fact changed by [Hb] and whether this is an important factor determining V02 maXduring these conditions. METHODS Two sets of experiments were conducted; the first to investigate the changes in the relationships of \jozrnax to PvoZ and calculated PCoZthat result from reducing [Hb] under normoxic conditions and constant muscle blood flow and the second to more clearly interpret the alterations in these relationships by comparing normoxic and hypoxic conditions at each of two different Hb levels. The differences in the two sets are discussed in Experimental protocol. Adult mongrel dogs of either sex with a weight range of 15-24 kg were anesthetized with pentobarbital sodium (30 mg/kg); maintenance doses were given as required. The dogs were intubated with a cuffed
0 1991 the American
Physiological
Society
1105
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
1106
MUSCLE
MAXIMAL
0,
CONSUMPTION
endotracheal tube. Esophageal temperature was maintained near 37°C by heating pads. Heparin (1,500 U/kg) was given to the animals after the surgery described below. Ventilation was maintained with a Harvard 613 ventilator to keep blood gases and pH normal. Surgical preparation. The left gastrocnemius-flexor digitorum superficialis muscle complex (for convenience referred to as gastrocnemius) was isolated as described previously (15, 20). Briefly, a medial incision was made through the skin of the left hindlimb from midthigh to the ankle. The sartorius, gracilis, semitendinosus, and semimembranosus muscles, which overlie the gastrocnemius, were doubly ligated and cut between the ties. All vessels draining into the popliteal vein, except those from the gastrocnemius, were ligated to isolate the venous outflow from the gastrocnemius. The arterial circulation to the gastrocnemius was isolated by ligating all vessels from the femoral and popliteal arteries that did not enter the gastrocnemius. The left popliteal vein was cannulated, and the venous outflow was returned to the animal via a jugular catheter. A blood flowmeter (Micron Medical RC 2000) was inserted on this line to monitor muscle blood flow. The right femoral artery was catheterized for arterial blood sampling. This catheter was connected to the left femoral artery so that the isolated muscle was perfused by blood from this contralateral artery. Perfusion was accomplished either directly from the contralateral (systemic pressure- self-perfused) or via a Sigmamotor pump to control flow. A pressure transducer in this line at the head of the muscle constantly monitored perfusion pressure. A carotid artery was also catheterized to monitor systemic blood pressure. The left sciatic nerve, which innervates the gastrocnemius, was doubly ligated and cut between the ties. To prevent cooling and drying, all exposed tissues were covered with saline-soaked gauze and with a sheet of plastic wrap. After the muscle was surgically isolated, the Achilles tendon was attached to an isometric myograph (Statham 1360 transducer) to measure tension development. The hindlimb was fixed at the knee and ankle and attached to the myograph with struts to minimize movement. Weights were used at the end of each experiment to calibrate the tension myograph. Isometric muscle contractions (twitches) were elicited by stimulation of the sciatic nerve with square-wave impulses of 0.2-ms duration at 4-6 V. The stimulation protocol was the same as used previously (12). Briefly, the muscle contracted at 3 Hz for 3 min and then at 5 Hz until a IO-15% drop in developed tension occurred. If fatigue was not present at 5 Hz by the end of 2 min, the frequency was increased to 7 Hz until the drop in developed tension occurred (always within 2 min). Vo2 was measured at the time that the lo-15% decline in developed tension occurred. This contraction pattern., which is similar to the type of test used ,to measure VO, maxin humans, ensured attainment of Vozmaxfor any [Hb] or arterial PO, (Pa,J without an excessive fatigue development. It should be noted that our use of “Vo2 max”is relative to the use of isometric twitch contractions in this preparation. It may be that a different type of contrac-
AND
HB
CONCENTRATION
tion pattern will result in higher Voz; however, our results still represent the highest Tjo, for these conditions. Before each contraction period, the resting muscle was passively stretched until a tension setting of 10 g/g muscle mass (muscle weight was estimated) was recorded. This ensured that the initial tension development was not affected by slippage in the system that might have occurred during the prior contraction period. This resting muscle length was slightly less than the length at which the contractile response was greatest. Before the first contraction period, the blood supply to the isolated muscle was switched from self-perfusion to pump-perfusion and enough time was allowed for conditions to stabilize at a blood flow similar to self-perfusion. Experimental protocol. The different levels of Hb were induced by two methods. The first method involved changing the [Hb] of the whole animal by adding plasma or packed erythrocytes from a donor dog to the blood of the experimental animal until the randomly selected [Hb] was achieved. This was done while the total blood volume was kept constant The second method involved use of a set ond pump to ‘dilute the bl.ood entering the muscle with plasma to the [Hb] desired. There was no difference in the experimental results between the two methods. These two methods of altering [Hb] were used to minimize any specific disadvantages inherent in either of the two methods. As mentioned, two separate sets of experiments were performed. In the first set (n = 8), three randomly ordered levels of [Hb] were used (-5,10, and 15 g/l00 ml). In four of the animals, [Hb] was altered by the first method, while in the other 4 animals, [Hb] was altered by the second method. The muscle was worked maximally as previously described under each of these three different Hb levels, with 15-20 min of rest between treatments. In th e second set of experiments, two different levels of Hb (-14 and 7 g/100 ml) were each combined with two different levels of PaoP for a total of four conditions in each of six animals studied in random sequence. In these experiments, [Hb] was altered in four of the animals by method 1 and in two of the animals by method 2. The Pao9 was altered by having the animal breathe room air ora gas mixture containing the fraction of 0, necessary to attain the desired Pa,, of -30 Torr. Again the muscle was worked maximally under the four different conditions, with 15-20 min of rest between each treatment. In the first contraction period in both sets of experiments, the blood flow was set to elicit a mean blood pressure to the muscle of -120 Torr. The same blood flow was used in all subsequent contraction periods so that a constant muscle blood flow was achieved, but muscle blood pressures were necessarily different (lower during hypoxemia). Measurements. Arterial blood samples were obtained from the arterial line entering the muscle, and venous samples were obtained from the left popliteal vein as close to the gastrocnemius as possible. Arterial and venous blood samples were drawn anaerobically during the last 20 s of each contraction and were kept on ice. As blood samples were drawn, venous blood flow measurements were made by timed blood collections into a gradu-
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
MUSCLE
MAXIMAL
0,
CONSUMPTION
1. Blood parameters corresponding to Hb levels in the first experimental series
TABLE
[Hb],
Pa,,, Torr Pace,, Torr PH, L a 1a, mM Muscle blood flow, ml 100 g-’ min-’ Cao2, ml/100 ml BPm, Torr Maximum developed tension, g/g l
l
g/100
ml
5.8t0.3*
9.4+0.1*
14.4+0.6*
77+3
79t3 371~2
74k4 36k2
39+2 7.41kO.02
7.37kO.02
7.38t0.02
4.3kO.4 122+11 7.4+0.5* 105+3
3.7kO.4 121+10 12.0+0.2* 123+12
2.8~10.6 120+13 17.9t0.8* 159s3*
12.8k0.9
13.OkO.9
13.3kO.9
Values are means + SE; n = 8. Paoz, arterial PO:!; Pat+, arterial pH,, arterial pH; [La],, arterial [lactate]; Cao2, arterial O2 content; BPm, muscle arterial blood pressure. * Significantly different (P < 0.05) from other conditions.
AND
1107
CONCENTRATION
HB
fusive pathway (in this case, phenomena related to Hb and the particulate nature of blood). The calculated Dmo2 does not take into account heterogeneity or 0, shunting between artery and vein, because no methods are currently available for the quantification of these phenomena. However, these calculated values are most useful as a comparison among conditions and for gas exchange analysis and have been predictive of specific experimental outcomes (14). Statistics. Two-way analysis of variance, at the 0.05 level of significance, was used. Duncan’s multiple-range test was used to determine where differences occurred. RESULTS
Pco~;
ated cylinder and corrected for the amount of venous blood sampled. Blood lactate concentrations were determined from the arterial and venous samples by means of a blood lactate analyzer (model Z3L, Yello w Springs Instruments). Blood PO,, Pco,, and pH were measured within 5-8 min with a blood gas analyzer (model 813, Instrumentation Laboratory) at 37”C, while [Hb], percent 0, and CO, saturation, and Ca,, were measured with an IL 282 COoximeter (model 282, Instrumentation Laboratory). These instruments were calibrated before each experiment and frequently throughout each experiment. Plasma bicarbonate concentration was calculated from the measured pH and PCO, values by use of the Henderson-Hasselbalch equation. The Fick principle was used to calculate muscle VO, and lactate release. The muscle was removed and weighed at the end of each experiment. A Bohr integration technique was used to calculate an estimate of both PC,, and Dm, for every contraction period. This technique has been discussed in detail elsewhere (12, 17) and is based on an analysis proposed by Wagner (22). Briefly, the drop in PO, along a capillary is calculated using Fick’s law of diffusion
. vo 2 = Dmoz(Pcoz -
PtioZ)
(1)
where PcoZ is the capillary PO, at the point along the capillary at which the calculation is being performed and Ptioz is the PO, at the point of utilization or mitochondrial cytochrome PO,. Pti,, in this calculation is set at 0 Torr at all points along %he capill ary, which 1s only slightly less than that measured in the cytoplasm of longitudinal sections of muscle fibers (7). In this process, a Dm ‘02l.s calculated that accounts for the drop in PO, from measured arterial to venous values with the assumption of a homogeneous muscle. Changes in the slope of the 0, dissociation curve due to changes in PCO, and pH are accounted for. It should be emphasized that the Dmo2 calculated from this analysis is a lumped-parameter estimate of a complex transport coefficient that may have several potentially variable components between the Hb molecule and the cytochromes (6). However, the very purpose of the calculation is to ask whether this variable is affected by identifiable components of the dif-
Mean weight of the exercised gastrocnemius muscles was 83 t 8 (SE) g for the first set of experiments (n = 8) and 78 t 6 g for the second set (n = 6). Table 1 presents the principal variables from the first set of experiments (3 levels of Hb) concerning the conditions of the inflowing blood to the muscle. There were no significant differences among the three Hb levels in the important blood acid-base parameters or gas partial pressures. By means of pump perfusion, muscle blood flow was held constant among the treatments, so there was no significant difference in this variable. The blood pressure at the head of the muscle required to maintain equal muscle blood flows was significantly greater at the highest [Hb], presumably because of the differences in blood viscosity. Table 2 presents the data concerning muscle 0, gas exchange, metabolism, and fatigue from the first set of experiments. The 0, delivery was significantly different among the three conditions, as was the.Vo, m8X.Figure 1 illustrates the relationship between Vo2maxand both Pvo2 and calculated Pc, . From Fig. 1 it can be seen that the fall in Vo2 max as [fib] declined was proportionally much greater than the fall in PcoZ and the line of best fit does not pass near the origin. This suggests that the fall in V02max that occurred with decreasing [Hb] was not simply a result of reduced driving pressure (as estimated by Pco2 at constant Dm,J, because a linear relationship 2. Summary of principal variables related to 0, transport and gas exchange corresponding to Hb levels in the first experimental series
TABLE
[Hb],
o2 delivery, ml 100 g-’ . min-’ vo 2 max, ml 100 g-’ min-’ Pvoz, Torr Pcoz, Torr DmoP, ml. 100 g- i min-’ Torr-’ o2 extraction, % La, pm01 100 g-’ min-’ l
l
l
l
l
l
g/100
ml
5.8iO.3*
9.4-tO.l”
14.4kO.6’
9.0+0.8* 6.1+0.7*
14.5k1.4” 9.1*1.0*
20.1*2.9* 12.2+1.0*
23+2* 35+2*
27?2* 40+2
30&2* 41+2
0.25-tO.03" 64+4 56+18
0.32-tO.O3* 59+4t 56klO
l
0.20t0.02* 68+4-j44+17
Values are means + SE; n = 8.0, delivery, Caoz X muscle blood flow; mean capillary PO,; DmOz, calcu9 venous Pas; PCoz, calculated lated muscle diffusing capacity; La, muscle lactate output (flow X venoarterial difference). * Significantly different (P < 0.05) from other conditions. t Significantly different (P < 0.05) from other condition with the same symbol.
PVOz
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
1108
MUSCLE
MAXIMAL
0,
CONSUMPTION
AND
HB
CONCENTRATION
3. Blood parameters corresponding to conditions in the second experimental series
TABLE
Condition Norm norm
5
[Hb], g/l00 Pa,, , Torr Paco2, Torr
5
ml
[W,, mM
0
0
Low [Hb]norm Pao,
Low [Hb]low PaoP
2.9t0.4t
6.8+0.7-f 80&9-f 411~2 7.3720.01 4.7kO.5
6.9-tO.5-f 30+1 43+1 7.3EhO.02 5.2kO.8
111+7 8.4-tl.Ot 128&11*
lllst6 8.5+l.O”r 105+lOt
112k7 4.5+0.5* 97+7t
8.7-tO.8
8.9kO.8
8.6kO.8
14.2kO.6
78-+7-f
29t2 43k4 7.36kO.03
40&l
Muscle blood flow, ml 100 g -’ . mine’ Caoz, ml1100 ml BPm, Torr Maximum developed tension, g/g
Norm [Hb]low Paoz
13.9kO.7
7.37kO.01 3.3+0.3t
PH,
?
[Hb]Pa,?
l
5
10
15
20 PO2
25
30
35
40
45
(Tom)
FIG. 1. Relationship between measured maximal 0, uptake (Vo 2 ,,,) and both venous PO, (Pvos) and calculated mean capillary PO, (Pcos) for 3 levels of hemoglobin (Hb) used in the first series of experiments. VO, m8x fell with decreasing [ Hb].
through the origin would have been expected from Fick’s law (Eq. I), just as we have found previously with reduction in Paor, in both this preparation (12) and in humans (17). The estimated Dmol for the three conditions were thus significantly different (because the PCo2)swere not much different but VO 2maxwas different). Figure 2 illustrates the relationship between Vopmax and 0, delivery. 0, delivery is closely correlated with VO, maxas [Hb] was altered, as found by others (3, 15, 23). The time required for developed tension to fall to 10-E% below the maximum developed tension was significantly less in the reduced [Hb] conditions (4.3 t 0.2, 5.3 t 0.2, and 6.3 t 0.3 min; low [Hb] to high), but the lactate output was not significantly different (Table 2). Table 3 provides the corresponding information for the second series of experiments as was provided in Table 1. In this series, there were four conditions: two levels of Hb coupled with two levels of Paoz, so at each [Hb], there were two significantly different Pao2 levels. As in the first series, there were essentially no differences in acid-base parameters (a slight increase in arterial lactate levels was seen in the lower [Hb] conditions). Again, muscle blood flow was held constant. Caoz values were significantly different, except coincidentally the normal [Hb]-low PO, combination had the same 0, content as
11457 17.5+1.0* 145*12*
9.OkO.9
Values are means _+ SE; n = 6. See Table 1 footnote for definition of abbreviations. * Significantly different (P < 0.05) from other conditions. t Not significantly different from any other condition with the same symbol but significantly different from other conditions.
the low [Hb] -normal PO, combination. As before, the mean arterial blood pressure to the muscle was higher for the same flow in the higher [Hb] conditions. Table 4 provides the gas exchange, metabolic, and fatigue results for the second series. 0, delivery was significantly different among the treatments, except the normal [Hb]-low PO, and low [Hb]-normal PO,, which were the same. V02max followed the same pattern. Figure 3 illustrates the relationship between Vo2max and calculated PEo, for the four conditions (the relationship between VO, maxand Pvo, was similar). Regression analysis of all the data points relating ir0, m8xto PCo, for each of the two [Hb] conditions showed that the intercepts of the regression lines for each [Hb] were not significantly different from the origin. Therefore, it should be noted that in Fig. 3 the line drawn for each of the two [Hb] conditions is the line of best fit that passesthrough the origin. It is apparent that when the two different [Hb] condi4. Summary of principal variables related to 0, transport and gas exchange corresponding to conditions in the second experimental series TABLE
20
Condition
n
'i-.E
15
0, delivery, ml 100 g-’ min-’ V02,, ml*100 g -l. min-’ Pvo*, Torr Pz,, , Torr Dm, , ml 100 g-*0 min-’ Torr-’ (I2 extraction, % La, pm01 100 -1 min -1 g
Low [HbJnormPao,
Low[Hb]lowPa0,
19.9t1.5”
9.2+1.0t
9.4+1.2t
5.0t0.6”
12.3kl.O” 28t2* 39t2”
7.4+0.6t 12+2t 20+2p
7.1+0.7t 19*2* 34k2’
4.2t0.5’ 11+2t 18,+2t
0.33~0.03” 63t6*
0.40t0.03” 8225
0.24+0.03t 80t5
0.25*0.02-f 84t3
17t16
28t15
l
7
l
rJ, 8 -
Norm [Hb]- Norm [HbJnormPao, lowPa0,
10
i 5 OJ
l
l
*P 0
l
0
5
10
15 20 25 30 35 40 02 delivery (ml- 100 CJ- ’ emin- ’ )
FIG. 2. Relationship between measured the 3 levels of Hb used in the first series with decreasing [ Hb].
vozrnall and 0, delivery of experiments. vozrnax
45 for fell
l
19t16
41&15
Values are means + SE; n = 6. See Table 2 abbreviations. * Significantly different (P < tions. t Not significantly different from any same symbol but significantly different from
footnote for definition of 0.05) from other condiother condition with the other conditions.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
MUSCLE
' *;
MAXIMAL
0, CONSUMPTION
15
I CP
10
z
--i
5
F 0
5
10
15
20 PCO,
25
30
35
40
45
(Tow)
FIG. 3. Relationship between measured vozrnax and calculated ( Pco,) for the 2 Hb levels combined with the 2 arterial PO, (Pa,J levels in the second series of experiments. Line drawn is line of best fit for each [Hb] condition that passes through the origin, indicating reduced muscle O2 diffusing capacity (Dm,J for the lower [Hb] condition. VO 2 max decreased with lower [Hb] and lower Pa,,.
tions are considered separately, with 0, delivery reduced by hypoxemia for each [Hb], vo2 max falls in a manner proportional to the decline in PC, for each [Hb]. The calculated Dm,, values in the low [I!b] conditions (0.24 t 0.03 and 0.25 t 0.02 ml 100 g-l min-’ Torr-‘) were significantly less than with high [Hb] (0.33 t 0.03 and 0.40 t 0.03) and were not different between the two PO, conditions during the low [Hb] treatment. If a line is drawn to connect the Pco,‘s for the two high Pao2 points (different [Hb]), the same x-intercept is obtained as seen for the three PC,, points in Fig. 1. Figure 4 illustrates the relationship between V,Z m8Xand 0, delivery, indicating a strong relationship, as in the first series. As before, the fati . gue process occurred faster with the lower 0, deliverl
l
l
ies, and lactate output was not significantly
different.
DISCUSSION
The importa .nt result of this study was the observati .on th at one of the principal factors that dete rmines the to tal diffusive flux of 0, from erythrocyte to mitochondria, DmOz, was reduced 33% with lowered [Hb]. At the same time, VO, maxand 0, delivery were also closely linearly correlated, as found by others (8, 16, 23). Although the model used to analyze the measured input and output variables and calculate PEoZ and Dm,, yields only indirect proof that, in fact, the Dm,, was reduced during reduced [Hb], this type of analysis can provide added insight and useful information concerning the dynamics of muscle 0, diffusion. 0, supp&limitation. of 30, max. The results of prior work (12, 13, 17) have suggested that 0, diffusion in the peripheral tissues is one important determinant of . vo 2m8Xwhen the supply of 0, to the mitochondria is inadequate. Our hypothesis has been that the perfusive conductance of 0, to the tissue, 0, delivery, interacts with the . diffusive conductance of 0, into the ceil to determine VO,. Thus, for a given 0, delivery, the muscle’s diffusing capacity and the capillary pressure head for 0, diffusion will determine the amount of 0, that can be extracted (see Eq. 1). In previous studies (12,17), the reductions in capillary driving pressure were postulated as the factor
AND
1109
HB CONCENTRATION
affecting the amount of 0, moved by diffusion, and there was little or no change in the calculated DmoP as PO, was reduced. Other recent work has also demonstrated that 0, diffusion limitation may be important in determining the fall in resting muscle VO, when 0, delivery is reduced below critical levels (9-U). It has been demonstrated (l4), as predicted by the diffusion limitation hypothesis, that different values of Vo2max can be obtained at the same 0, delivery when the latter is produced by different combinations of blood flow and Pa,,, showing that 0, delivery is not the unique determinant of VO, max. The question of whether 0, availability to the mitochondria is adequate at Vo2 maxunder normoxic normal [Hb] conditions remains quite controversial. As demonstrated by Honig and colleagues (2, 5, 7), the PO, within the myocyte is uniform, both radially and longitudinally, and very low at maximal work (similar to that conducted in this study). At some points along the longitudinal length of a myocyte frozen while working maximally (7), they showed that the PO, was less than the 0.5Torr value they had previously demonstrated as being necessary for adequate mitochondrial oxygenation (5) for this preparation, and this was during normal 0, delivery conditions. It is quite possible that there is not a unique resolution to this controversy and that, for different people (or muscles), 0, availability to the mitochondria may be sufficient or insufficient at normal VO, max, depending on such variables as muscle blood flow heterogeneity, amount of muscle mass being worked, capillary and muscle fiber morphometry, training level, and muscle enzymatic activity. As demonstrated by Rowe11 et al. (19), 0, supply may be limiting to muscle during whole-body exercise because of the constraints of the cardiac output; however, when a small muscle mass (such as our gastrocnemius preparation) works maximally with this constraint removed, the limiting factors may be different. Concerning VO, m8x during anemic hypoxia, Gregg et al. (8) recently showed reductions in V02mex in running rats and suggested that muscle oxidative capacity did not restrict ire, m8x. Altered [Hb] and 0, diffusion limitation. If the tissue PO, is uniformly low throughout the length of the myocyte (7) and limiting to mitochondrial respiration (especially when 0, delivery is compromised), then Fick’s law
-6
j
1’0
1’5 O2 delivery
2‘0
2’5
io
i5
(ml.100
g-’
.min -‘)
40
45
FIG. 4. Relationship between measured \jo2 mm and O2 delivery for the 4 levels of Hb and Pa,, in the second series of experiments. VO,,, decreased with lower [Hb] and lower Paoz.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
1110
MtJSCLE
MAXIMAL
0,
CONSUMPTION
of diffusion (Q. I), where Vo2max will be determined by the interaction of Dmo, and PCoZ (assuming Ptiog = 0), can be used to model tissue gas exchange. The results of the first set of experiments in this study demonstrate that the relationship found previously [ (12) V,Z M8Xfalling linearly through the origin with PEo,] was not found. According to the relationship shown in Fig. 1, as [Hb] progressively fell, V02 max would be zero even when the driving pressure for 0, diffusion (Pco,) was high (x-intercept = 28 Torr). It is obvious that the relationship between VO 2 maxand PcoZ was not as simple when 0, delivery was reduced with lowered [Hb] as with hypoxemia, as Cain (1) demonstrated in O,-compromised resting muscle. The results of these initial experiments could be interpreted in several ways. If 0, diffusion limitation were a component of the reduced Vozmax with lowered [Hb], then it must have been through changes in DmoZ and not simply changes in PCoz. Alternatively, it could be argued that 0, diffusion limitation was not an important determinant of VO, maxand that the relationship between . vo 2 m8Xand 0, delivery (Fig. 2) was an adequate explanation for the fall in VO,, as other investigators have also demonstrated with lowered [ Hb] (1,4, 15,22). It was the purpose of the second set of experiments to clarify these initial results. If decreasing [Hb] resulted in reductions in the Dmoz, as predicted from the theoretical modeling of Stainsby et al. (2l), then this should be revealed by imposing a hypoxic reduction in 0, delivery along with the anemia. With 0, delivery reduced by hypoxemia for a given [Hb], 0, diffusion limitation would be expected on the basis of our previous work (12, 17) to produce an essentially linear relationship, through the origin as predicted by Fick’s law (Eq. I), between the measured . vo 2 M8Xand the mean capillary driving pressure (calculated PC, ). This relationship was, in fact, demonstrated at each [fib] by the second series of experiments (Fig. 3). These results suggest that 0, diffusion remains an important determinant of Vo2max when 0, delivery is reduced by lowered [Hb]; however, the changes in VO, max during lowered [Hb] are influenced by decreases in Dmop. Once a new Dmq, is established for the given [Hb], then the strong relationship observed previously between V02 In8Xand PC,,, becomes evident as 0, delivery is subsequently lowered by hypoxemia. Anemic hypoxia vs. hypoxic hypoxia. As noted, anemic hypoxia appears quite different from hypoxic hypoxia. Serendipitously, the 0, delivery levels during the three [Hb] conditions of the first anemic experimental series matched three of the 0, delivery levels measured during the hypoxic hypoxia study conducted previously (12), and the relevant information is listed in Table 5. For a given 0, delivery, Vo2 max was not significantly different between anemic and hypoxic hypoxia; however, the estimate of mean capillary driving pressure, PC,, (or even PvoJ, indicated that during hypoxic hypoxia a lower driving pressure could produce the same VO, max. This, in turn, gives rise to the difference in estimated lower in anemic hypoxia. Dmoz 9 which was substantially These results are similar to those found by Cain (1) when resting animal VO, was reduced by lowering 0, supply through either anemia or hypoxemia. He demonstrated that irO, for the anemic and hypoxic conditions fell in a manner more predictable by 0, delivery than by an estimate of mean capillary driving pressure. Because of the weak
AND
HB CONCENTRATION
relationship between VO, and the mixed venous PO, below which VO, began to decrease, Cain (1) concluded that Vo2 was not limited by PO, gradients (diffusion limitation) during these conditions of reduced 0, delivery. However, we are suggesting that the PO, driving gradient remains a primary component determining VO, m8Xduring anemic hypoxia, with this relationship being affected by the changes in Dmo2. Why is Dmoz reduced with lowered [Hb]? In a recent model analysis of tissue oxygenation, Stainsby et al. (21) considered the origin of reduced Dm,, during anemic hypoxia “an enigma.” There may be several ways in which the lower [Hb] may decrease DmoZ. Dmoz is a complex parameter that encompasses a broad spectrum of variables. Some of the “[Hb] -dependent” components of Dmo, include the capillary hematocrit (and its distribution), erythrocyte spacing, 0, off-loading kinetics from Hb, and erythrocyte transit time, while some of the “[ Hb] -independent” components include muscle capillarity and the degree of capillary recruitment, diffusion distances and solubilities in the various compartments from Hb to mitochondria, heterogeneity, and myoglobinfacilitated 0, transport within the cell. A change in any one or more of these components could lower the effective Dmo2 during reduced [Hb]. With the equal muscle blood flows used in this study, a change in flow distribution and the diffusion distances involved is possible but seems unlikely; however, the inevitable increase in mean erythrocyte spacing with reduced [Hb] could also reduce Dmo , as proposed by Federspiel and Pope1 (3). Another possible explanation is that, simply due to the lower [Hb], the term “@Vc” (see Eq. 2) of Roughton and Forster (18) would be less, and this would reduce overall Dmoz accordingly. Whether any of these mechanisms or others not considered are important remains to be determined. Roughton and Forster (US), analyzing the diffusion pathway in the lung, divided the total resistance to 0, diffusion (l/Do,) into resistance imposed between the plasma and alveolus (1 /Dperice iiiav) and the resistance to 0, on-loading (l/@Vc), where B is the 0, on-loading reaction rate from Hb and Vc is the capillary blood volume. Thus l/D02
(2)
= 1 /Dpericapiiiary + I/OVC
A similar kind of analysis can be proposed in the muscle but formulated more generally as follows: the total resistance (1 /DmoJ is divided into a [Hb] -independent component (l/D) and a [Hb]-dependent component (11 K[Hb]). The assumption is made that the [Hbl-dependent component is directly proportional to the [Hb] and D is constant (at constant blood flow in a given muscle) by definition. Equation 2 can then be used to partition the diffusion resistance into [Hb]-dependent and [Hb]independent parts (3) 1/Dm,:, = l/D + l/K[Hb] In the second series of experiments, the calculated Dm,, at the normal [Hb] of 14.1 g/l00 ml was 0.36 ml. 100 g -l. min-l Torr-1 and 0.25 ml 100 g-’ mine1 Torr-’ at 48% of that [ Hb]. Equation 3 can be used for each of the two [Hb] conditions to simultaneously solve for the unknown values of D and K. When this was done, it was found that, at normal [ Hb], roughly 60% of the total resisl
l
l
l
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
MUSCLE
[Hb], g/100 ml Paoz, Torr Caoz, ml/100 ml Muscle blood flow, ml. 100 g-’ I)z delivery, ml 100 g-’ min-’ vo 2max, ml- 100 g-‘=min-’ Pvo,, Torr Pco2, Torr DO,, ml 100 g-’ min-’ Torr-’ l
l
l
min-’
l
l
l
MAXIMAL
0,
CONSUMPTION
5.8+0.3 77+3 7.4kO.5 122+11 9.OkO.8 6.1kO.7 23+2 35+2 0.20~0.02
13.3kO.6 29&l 7.9k0.6 118+15 9.3k1.6 7.2k1.3 14rtl 22&l 0.32kO.06
Values are means + SE; n = 8 during HH and n = 6 during previous study (12). Condition 3 represents control conditions
We gratefully acknowledge Dr. Steve Hempleman for helpful discussions concerning the analyses of some of these data. This research was supported by National Heart, Lung, and Blood Institute Grant HL-17731. Address for reprint requests: M. C. Hogan, Dept. of Medicine M-023A, University of California, San Diego, La Jolla, CA 92093-0623. 26 March
1990; accepted
in final
form
4 October
1990.
REFERENCES 1. CAIN, S. Oxygen delivery and uptake in dogs during anemic and hypoxic hypoxia. J. Appl. Physiol. 42: 228-234, 1977. 2. CONNEIT, R. J., T. E. J. GAYESKI, AND C. R. HONIG. An upper bound on the minimum PO, for 0, consumption in red muscle. Adv. Exp. Med. Biol. 191: 291-300, 1985.
1111
HB CONCENTRATION
9.420.1 79+3 12.OkO.2 121+10 14.5k1.4 9.Wl.O 27+2 40+2 0.25-tO.02
AH. AH, anemic for the 2 studies.
tance was associated with the [ Hb] -independent component and thus 40% with the [Hb]-dependent component. At the lower [Hb], there was an approximate reversal of these relative resistances, with the [ Hb] -dependent component now becoming more important. Intuitively, if Dmoz was determined wholly by [Hb] -dependent factors, then the percent change in Dmoz would have been equal to the percent change in [Hb]. The 50% reduction in [Hb] did not result in a 50% drop in DmoP, which fell by only 33%, indicating that [Hb] is not the only factor affecting DmOz. Fatigue. The time before fatigue set in and the lactate output followed the same pattern with lowered [Hb] as found previously in hypoxemia (12). Fatigue development occurred significantly sooner with the lower 0, deliveries, but lactate output was not different. In fact, a comparison of the anemic hypoxia (4.3 t 0.2, 5.3 t 0.2, and 6.3 t 0.2 min; lowest 0, delivery to highest) and the hypoxic hypoxia fatigue times (4.7 t 0.2, 5.3 t 0.2, and 6.2 + 0.5; lowest OS delivery to highest) demonstrates lit&difference in fatigue times for the two methods of 0, delivery. This suggests that the development of fatigue was dependent only on 0, delivery and not on the manner in which it was altered. Conclusion. This study suggests that reducing the [Hb] as a means of lowering the 0, delivery to working muscle decreases the effective Dmo2. These results suggest that . vo 2max during anemia is reduced not only by the expected effect from the decreased convective 0, delivery to the muscle but also by these reductions in Dmq,. Whether this reduction in Drn,? relates to changes in oxyhemoglobin off-loading kinetics, altered erythrocyte spacing, or other effects of altering [Hb] remains to be determined.
Received
AND
conditions
13.5k0.6 38&l 11.6t0.6 122-tl5 14.lrt2.0 10.2-tl.2 20+1 29&l 0.31kO.05 of this study;
HH,
14.4t0.6 74+4 17.9t0.8 120+13 2O.lk2.9 12.2U.O 3Ok2 4122 0.32kO.03 hypoxic
hypoxia
13.5+0.8 79+4 16.9kO.8 124+16 21.Ok2.9 14.Ok1.5 28k2 44+1 0.30~0.05 conditions
of the
3. FEDERSPIEL, W. J., AND A. S. POPEL. A theoretical analysis of the effect of the particulate nature of blood on oxygen release in capillaries. Microvasc. Rex 32: 164-189, 1986. 4. GAEHTGENS, P., F. KREUTZ, AND K. H. AI~BRECHT. Optimal hematocrit for canine skeletal muscle during rhythmic isotonic exercise. Eur. J. Appl. Ph.ysiol. Occup. Physiol. 41: 27-39, 1979. 5. GAYESKI, T. E. J., R. J. CONNETT, AND C. R. HONE. Minimum intracellular PO, for maximum cytochrome turnover in red muscle in situ. Am. J. Yhysiol. 252 (Heart Circ. Physiol. 21): H906-H915, 1987. 6. GAYESKI, T. E. J., W. J. FEDERSPIEL, AND C. R. HONIG. A graphical analysis of the influence of red cell transit time, carrier-free layer thickness, and intracellular PO, on blood-tissue 0, transport. Adv. Exp. Med. Viol. 222: 25-35, 1988. 7. GAYESKI, T. E. J., AND C. R. HONIG. Intracellular PO, in long axis of individual fibers in working dog gracilis muscle. Am. J. Physiol. 254 (Heart Circ. Physiol. 23): H1179-H1186, 1988. 8. GREGG, S. G., R. S. MAZZEO, T. F. BUDINGER, AND G. A. BROOKS. Acute anemia increases lactate production and decreases clearance during exercise. J. Appl. Physiol. 67: 756-764, 1989. G., AND ,J. M. ANDRY. Increased hemoglobin 0, affin9. GUTIERREZ, ity does not improve 0, consumption in hypoxemia. J. Appl. Physiol. 66: 837-843, 1989. 10. GUTIERREZ, G., C. MARINI, A. L. ACERO, AND N. LUND. Skeletal muscle PO, during hypoxemia and isovolemic anemia. J. Appl. Physiol. 68: 2047-2053, 1990. 11. GUTIERREZ, G., R. J. POHIL, AND R. STRONG. Effect, of flow on 0, consumption during progressive hypoxemia. J. Appl. Physiol. 65: 601-607, 1988. 12. HOGAN, M. C., D. E. BEBOUT, P. D. WAGNER, AND J. R. WEST. Maximal 0, uptake of in situ dog muscle during acute hypoxemia with constant perfusion. J. Appl. Physiol. 69: 570-576, 1990. 13. HOGAN, M. C., ,J. ROCA, P. D. WAGNER, AND J. B. WEST. Limitation of maximal 0, uptake and performance by acute hypoxia in dog muscle in situ. J. Appl. Physiol. 65: 815-821, 1988. 14. HOGAN, M. C., J. ROCA, J. B. WEST, AND P. D. WAGNER. Dissociation of maximal 0, uptake from 0, delivery in canine gastrocnemius in situ. J. Appl. Physiol. 66: 12 19-1226, 1989. 15. HOGAN, M. C., AND H. G. WELCH. Effect of altered arterial 0, tensions on muscle metabolism in dog skeletal muscle during fatiguing work. Am. J. Physiol. 251 (Cell Physiol. 20): C216-C222, 1986. 16. HORSTMAN, D. H., M. GLESER, D. WOLFE, T: TRYON, AND J. DELEHUNT. Effects of hemoglobin reduction on VO, max and related hemodynamics in exercising dogs. J. Appl. Physiol. 37: 97-102, 1974. 17. ROCA, J., M. C. HOGAN, D. STORY, D. E. BEBOUT, P. HAAB, R. GONZALEZ, 0. UENO, AND P. D. WAGNER. Tissue 0, diffusion limitation of maximal 0, uptake in man. J. Appl. Physiol. 67: 291-299, 1989. 18. ROUGHTON, F. J. W., AND R. E. FORSTER. Relative importance of diffusion and chemical reaction rates in determining rate of exchange of gases in the human lung, with special reference to true diffusing capacity of pulmonary membrane and volume of blood in the lung capillaries. J. Appl. Physiol. 11: 290-302, 1957. 19. ROWELL, L. B., B. SALTIN, B. KIENS, AND N. J. CHRISTENSEN. IS peak quadriceps flow in humans even higher during exercise with hypoxemia? Am. J. Physiol. 251 (Heart Circ. Physiol. 20): H 1038H1044, 1986. 20. STAINSBY, W. N., AND A. B. OTIS. Blood flow, blood oxygen ten-
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
1112
MUSCLE
MAXIMAL
0,
CONSUMPTION
sion, oxygen uptake, and oxygen transport in skeletal muscle. Am. J. Physiol. 206: 858-866, 1964. 21. STAINSBY, W. N., B. SYNDER, AND H. G. WELCH. A pictographic essay on blood and tissue oxygen transport. Med. Sci. Sports Exercise 20: 213-221,198s. 22. WAGNER, P. D. An integrated view of the determinants of maxi-
AND
HB
CONCENTRATION
mum oxygen uptake. In: Oxygen Transfer from Atmosphere to Tissues, edited by N. C. Gonzalez and M. R. Fedde. New York: Plenum, 1988. 23. WOODSEN, R. D., R. E. WILLS, AND C. LENFANT. Effect of acute and established anemia on 0, transport at rest, submaximal and maximal work. J. Appl. Physiol. 44: 36-43, 1978.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
0,
MICHAEL C. HOGAN, DONALD E. BEBOUT, AND PETER D. WAGNER Division of Physiology, Department of Medicine, University of California, San Diego, La Jolla, California 92093-0623 HOGAN,MICHAEL C.,DONALD E. BEBOUT,AND PETERD. WAGNER.Effect of hemoglobin concentration on maximal O2 uptake in canine gastrocnemius
muscle in situ. J. Appl. Physiol. to maximally working muscle was decreased by altering hemoglobin (Hb) concentration and arterial PO, (Pao,) to investigate whether the reductions in maximal 0, uptake (Vozmax) that occur with lowered [Hb] are in part related to changes in the effective muscle 0, diffusing capacity (Dmo,). Two sets of experiments were conducted. In the initial set (n = 8), three levels of Hb [5.8 t 0.3, 9.4 t 0.1, and 14.4 t 0.6 (SE) g/l00 ml] in the blood were used in random order to pump perfuse, at equal muscle blood flows and Pa,,, maximally working isolated dog gastrocnemius muscle. iTo, max declined with decreasing [Hb], but the relationship between TO, max and both the effluent venous PO, ( Pvo2) and the calculated mean capillary PO, (PC,,) was not linear through the origin and, therefore, not compatible with a single value of Dmo, (as calculated by Bohr integration using a model based on Fick’s law of diffusion). To clarify these results, a second set of experiments (n = 6) was conducted in which two levels of Hb (14.0 ~fi 0.6 and 6.9 t 0.6 g/l00 ml) were each combined with two levels of oxygenation (Pa,, 79 2 8 and 29 t ‘2 Torr) and applied in random sequence to again pump perfuse maximally working dog gastrocnemius muscle at constant blood flow. In these experiments, the relationship between VO2 max and both Pvo, and calculated Pi?,, for each [Hb] was consistent with a constant estimate of Dmoz as Pao, was reduced, but the calculated Dmo, for the lower [Hb] was 33% less than that at the higher [Hb] (P < 0.05). Whether the apparent reduction in Dmoz with lower [Hb] is due to increased erythrocyte spacing, reduced chemical off loading of 0, from Hb, or some other effect remains to be determined.
70(3): 1105-1112, 1991.-O, delivery
fatigue; skeletal muscle; gas exchange; sion limitation; hypoxia
exercise; lactate;
diffu-
IT IS WELL KNOWN that reduction in hemoglobin (Hb) concentration reduces maximal 0, uptake (Oo,,,,) in animals (8, 16) and humans (23). Clearly, one factor responsible is the reduced convective O2 delivery [flow X arterial O2 content (Ca,,)] to muscle, but it is also possible, as suggested by Stainsby et al. (21), that reductions in [Hb] have some additional effect on 0, diffusion from the erythrocyte to the mitochondria, thereby altering the muscle O2 diffusing capacity (Dmo,), which is an estimate of the diffusive conductance of the muscle for OZ. Cain (1) demonstrated a higher critical venous PO, (PvoZ) in resting whole animals when 0, delivery was reduced below critical levels (the point at which VO, begins to decline) by anemic hypoxia, than when O2 delivery was 0161-7567/91 $1.50 Copyright
reduced by hypoxic hypoxia. Cain (1) interpreted his results to suggest that the driving pressure for diffusion of O2 did not appear to have limited its transport to tissues and that VO, was more dependent on convective O2delivery. However, if the Dmo, is reduced when [Hb] is lowered, this could explain the difference Cain (1) found in critical PO, between anemic and hypoxic hypoxia. In fact, a recent report from Gutierrez et al. (10) suggests that the higher critical PvoZ found during anemic hypoxia than during hypoxic hypoxia in O,-limited resting muscle may be a result of an increase in the resistance to 0, diffusion (or a reduced DmoJ during anemic hypoxia. We have recently suggested (12, 13, 17) that 0, diffusion in the peripheral tissue is one important determinant of VO, max as 0, delivery is reduced and that O2delivery is not the unique determinant of VO, max (14). Gutierrez and colleagues (9-U) have also shown that 0, diffusion limitation may have an important role in determining the Vo2 of resting muscle when the 0, delivery is reduced below some critical level. These previous studies (9,12,17) demonstrated strong correlations between VO, and the driving pressure for 0, diffusion in the capillary, estimated by a calculation of the mean capillary PO, (PEo,), so the reduction in vo2 with each level of reduced O2 delivery was hypothesized to be in part due to the decreased capillary 0, driving pressure. In these studies (12, l7), the calculated Dmoz was unchanged with the varied levels of hypoxemia, possibly because of the imposed constancy of muscle blood flow during the different conditions. The purpose of the present experiments was to investigate whether the estimated DmoZ is in fact changed by [Hb] and whether this is an important factor determining V02 maXduring these conditions. METHODS Two sets of experiments were conducted; the first to investigate the changes in the relationships of \jozrnax to PvoZ and calculated PCoZthat result from reducing [Hb] under normoxic conditions and constant muscle blood flow and the second to more clearly interpret the alterations in these relationships by comparing normoxic and hypoxic conditions at each of two different Hb levels. The differences in the two sets are discussed in Experimental protocol. Adult mongrel dogs of either sex with a weight range of 15-24 kg were anesthetized with pentobarbital sodium (30 mg/kg); maintenance doses were given as required. The dogs were intubated with a cuffed
0 1991 the American
Physiological
Society
1105
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
1106
MUSCLE
MAXIMAL
0,
CONSUMPTION
endotracheal tube. Esophageal temperature was maintained near 37°C by heating pads. Heparin (1,500 U/kg) was given to the animals after the surgery described below. Ventilation was maintained with a Harvard 613 ventilator to keep blood gases and pH normal. Surgical preparation. The left gastrocnemius-flexor digitorum superficialis muscle complex (for convenience referred to as gastrocnemius) was isolated as described previously (15, 20). Briefly, a medial incision was made through the skin of the left hindlimb from midthigh to the ankle. The sartorius, gracilis, semitendinosus, and semimembranosus muscles, which overlie the gastrocnemius, were doubly ligated and cut between the ties. All vessels draining into the popliteal vein, except those from the gastrocnemius, were ligated to isolate the venous outflow from the gastrocnemius. The arterial circulation to the gastrocnemius was isolated by ligating all vessels from the femoral and popliteal arteries that did not enter the gastrocnemius. The left popliteal vein was cannulated, and the venous outflow was returned to the animal via a jugular catheter. A blood flowmeter (Micron Medical RC 2000) was inserted on this line to monitor muscle blood flow. The right femoral artery was catheterized for arterial blood sampling. This catheter was connected to the left femoral artery so that the isolated muscle was perfused by blood from this contralateral artery. Perfusion was accomplished either directly from the contralateral (systemic pressure- self-perfused) or via a Sigmamotor pump to control flow. A pressure transducer in this line at the head of the muscle constantly monitored perfusion pressure. A carotid artery was also catheterized to monitor systemic blood pressure. The left sciatic nerve, which innervates the gastrocnemius, was doubly ligated and cut between the ties. To prevent cooling and drying, all exposed tissues were covered with saline-soaked gauze and with a sheet of plastic wrap. After the muscle was surgically isolated, the Achilles tendon was attached to an isometric myograph (Statham 1360 transducer) to measure tension development. The hindlimb was fixed at the knee and ankle and attached to the myograph with struts to minimize movement. Weights were used at the end of each experiment to calibrate the tension myograph. Isometric muscle contractions (twitches) were elicited by stimulation of the sciatic nerve with square-wave impulses of 0.2-ms duration at 4-6 V. The stimulation protocol was the same as used previously (12). Briefly, the muscle contracted at 3 Hz for 3 min and then at 5 Hz until a IO-15% drop in developed tension occurred. If fatigue was not present at 5 Hz by the end of 2 min, the frequency was increased to 7 Hz until the drop in developed tension occurred (always within 2 min). Vo2 was measured at the time that the lo-15% decline in developed tension occurred. This contraction pattern., which is similar to the type of test used ,to measure VO, maxin humans, ensured attainment of Vozmaxfor any [Hb] or arterial PO, (Pa,J without an excessive fatigue development. It should be noted that our use of “Vo2 max”is relative to the use of isometric twitch contractions in this preparation. It may be that a different type of contrac-
AND
HB
CONCENTRATION
tion pattern will result in higher Voz; however, our results still represent the highest Tjo, for these conditions. Before each contraction period, the resting muscle was passively stretched until a tension setting of 10 g/g muscle mass (muscle weight was estimated) was recorded. This ensured that the initial tension development was not affected by slippage in the system that might have occurred during the prior contraction period. This resting muscle length was slightly less than the length at which the contractile response was greatest. Before the first contraction period, the blood supply to the isolated muscle was switched from self-perfusion to pump-perfusion and enough time was allowed for conditions to stabilize at a blood flow similar to self-perfusion. Experimental protocol. The different levels of Hb were induced by two methods. The first method involved changing the [Hb] of the whole animal by adding plasma or packed erythrocytes from a donor dog to the blood of the experimental animal until the randomly selected [Hb] was achieved. This was done while the total blood volume was kept constant The second method involved use of a set ond pump to ‘dilute the bl.ood entering the muscle with plasma to the [Hb] desired. There was no difference in the experimental results between the two methods. These two methods of altering [Hb] were used to minimize any specific disadvantages inherent in either of the two methods. As mentioned, two separate sets of experiments were performed. In the first set (n = 8), three randomly ordered levels of [Hb] were used (-5,10, and 15 g/l00 ml). In four of the animals, [Hb] was altered by the first method, while in the other 4 animals, [Hb] was altered by the second method. The muscle was worked maximally as previously described under each of these three different Hb levels, with 15-20 min of rest between treatments. In th e second set of experiments, two different levels of Hb (-14 and 7 g/100 ml) were each combined with two different levels of PaoP for a total of four conditions in each of six animals studied in random sequence. In these experiments, [Hb] was altered in four of the animals by method 1 and in two of the animals by method 2. The Pao9 was altered by having the animal breathe room air ora gas mixture containing the fraction of 0, necessary to attain the desired Pa,, of -30 Torr. Again the muscle was worked maximally under the four different conditions, with 15-20 min of rest between each treatment. In the first contraction period in both sets of experiments, the blood flow was set to elicit a mean blood pressure to the muscle of -120 Torr. The same blood flow was used in all subsequent contraction periods so that a constant muscle blood flow was achieved, but muscle blood pressures were necessarily different (lower during hypoxemia). Measurements. Arterial blood samples were obtained from the arterial line entering the muscle, and venous samples were obtained from the left popliteal vein as close to the gastrocnemius as possible. Arterial and venous blood samples were drawn anaerobically during the last 20 s of each contraction and were kept on ice. As blood samples were drawn, venous blood flow measurements were made by timed blood collections into a gradu-
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
MUSCLE
MAXIMAL
0,
CONSUMPTION
1. Blood parameters corresponding to Hb levels in the first experimental series
TABLE
[Hb],
Pa,,, Torr Pace,, Torr PH, L a 1a, mM Muscle blood flow, ml 100 g-’ min-’ Cao2, ml/100 ml BPm, Torr Maximum developed tension, g/g l
l
g/100
ml
5.8t0.3*
9.4+0.1*
14.4+0.6*
77+3
79t3 371~2
74k4 36k2
39+2 7.41kO.02
7.37kO.02
7.38t0.02
4.3kO.4 122+11 7.4+0.5* 105+3
3.7kO.4 121+10 12.0+0.2* 123+12
2.8~10.6 120+13 17.9t0.8* 159s3*
12.8k0.9
13.OkO.9
13.3kO.9
Values are means + SE; n = 8. Paoz, arterial PO:!; Pat+, arterial pH,, arterial pH; [La],, arterial [lactate]; Cao2, arterial O2 content; BPm, muscle arterial blood pressure. * Significantly different (P < 0.05) from other conditions.
AND
1107
CONCENTRATION
HB
fusive pathway (in this case, phenomena related to Hb and the particulate nature of blood). The calculated Dmo2 does not take into account heterogeneity or 0, shunting between artery and vein, because no methods are currently available for the quantification of these phenomena. However, these calculated values are most useful as a comparison among conditions and for gas exchange analysis and have been predictive of specific experimental outcomes (14). Statistics. Two-way analysis of variance, at the 0.05 level of significance, was used. Duncan’s multiple-range test was used to determine where differences occurred. RESULTS
Pco~;
ated cylinder and corrected for the amount of venous blood sampled. Blood lactate concentrations were determined from the arterial and venous samples by means of a blood lactate analyzer (model Z3L, Yello w Springs Instruments). Blood PO,, Pco,, and pH were measured within 5-8 min with a blood gas analyzer (model 813, Instrumentation Laboratory) at 37”C, while [Hb], percent 0, and CO, saturation, and Ca,, were measured with an IL 282 COoximeter (model 282, Instrumentation Laboratory). These instruments were calibrated before each experiment and frequently throughout each experiment. Plasma bicarbonate concentration was calculated from the measured pH and PCO, values by use of the Henderson-Hasselbalch equation. The Fick principle was used to calculate muscle VO, and lactate release. The muscle was removed and weighed at the end of each experiment. A Bohr integration technique was used to calculate an estimate of both PC,, and Dm, for every contraction period. This technique has been discussed in detail elsewhere (12, 17) and is based on an analysis proposed by Wagner (22). Briefly, the drop in PO, along a capillary is calculated using Fick’s law of diffusion
. vo 2 = Dmoz(Pcoz -
PtioZ)
(1)
where PcoZ is the capillary PO, at the point along the capillary at which the calculation is being performed and Ptioz is the PO, at the point of utilization or mitochondrial cytochrome PO,. Pti,, in this calculation is set at 0 Torr at all points along %he capill ary, which 1s only slightly less than that measured in the cytoplasm of longitudinal sections of muscle fibers (7). In this process, a Dm ‘02l.s calculated that accounts for the drop in PO, from measured arterial to venous values with the assumption of a homogeneous muscle. Changes in the slope of the 0, dissociation curve due to changes in PCO, and pH are accounted for. It should be emphasized that the Dmo2 calculated from this analysis is a lumped-parameter estimate of a complex transport coefficient that may have several potentially variable components between the Hb molecule and the cytochromes (6). However, the very purpose of the calculation is to ask whether this variable is affected by identifiable components of the dif-
Mean weight of the exercised gastrocnemius muscles was 83 t 8 (SE) g for the first set of experiments (n = 8) and 78 t 6 g for the second set (n = 6). Table 1 presents the principal variables from the first set of experiments (3 levels of Hb) concerning the conditions of the inflowing blood to the muscle. There were no significant differences among the three Hb levels in the important blood acid-base parameters or gas partial pressures. By means of pump perfusion, muscle blood flow was held constant among the treatments, so there was no significant difference in this variable. The blood pressure at the head of the muscle required to maintain equal muscle blood flows was significantly greater at the highest [Hb], presumably because of the differences in blood viscosity. Table 2 presents the data concerning muscle 0, gas exchange, metabolism, and fatigue from the first set of experiments. The 0, delivery was significantly different among the three conditions, as was the.Vo, m8X.Figure 1 illustrates the relationship between Vo2maxand both Pvo2 and calculated Pc, . From Fig. 1 it can be seen that the fall in Vo2 max as [fib] declined was proportionally much greater than the fall in PcoZ and the line of best fit does not pass near the origin. This suggests that the fall in V02max that occurred with decreasing [Hb] was not simply a result of reduced driving pressure (as estimated by Pco2 at constant Dm,J, because a linear relationship 2. Summary of principal variables related to 0, transport and gas exchange corresponding to Hb levels in the first experimental series
TABLE
[Hb],
o2 delivery, ml 100 g-’ . min-’ vo 2 max, ml 100 g-’ min-’ Pvoz, Torr Pcoz, Torr DmoP, ml. 100 g- i min-’ Torr-’ o2 extraction, % La, pm01 100 g-’ min-’ l
l
l
l
l
l
g/100
ml
5.8iO.3*
9.4-tO.l”
14.4kO.6’
9.0+0.8* 6.1+0.7*
14.5k1.4” 9.1*1.0*
20.1*2.9* 12.2+1.0*
23+2* 35+2*
27?2* 40+2
30&2* 41+2
0.25-tO.03" 64+4 56+18
0.32-tO.O3* 59+4t 56klO
l
0.20t0.02* 68+4-j44+17
Values are means + SE; n = 8.0, delivery, Caoz X muscle blood flow; mean capillary PO,; DmOz, calcu9 venous Pas; PCoz, calculated lated muscle diffusing capacity; La, muscle lactate output (flow X venoarterial difference). * Significantly different (P < 0.05) from other conditions. t Significantly different (P < 0.05) from other condition with the same symbol.
PVOz
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
1108
MUSCLE
MAXIMAL
0,
CONSUMPTION
AND
HB
CONCENTRATION
3. Blood parameters corresponding to conditions in the second experimental series
TABLE
Condition Norm norm
5
[Hb], g/l00 Pa,, , Torr Paco2, Torr
5
ml
[W,, mM
0
0
Low [Hb]norm Pao,
Low [Hb]low PaoP
2.9t0.4t
6.8+0.7-f 80&9-f 411~2 7.3720.01 4.7kO.5
6.9-tO.5-f 30+1 43+1 7.3EhO.02 5.2kO.8
111+7 8.4-tl.Ot 128&11*
lllst6 8.5+l.O”r 105+lOt
112k7 4.5+0.5* 97+7t
8.7-tO.8
8.9kO.8
8.6kO.8
14.2kO.6
78-+7-f
29t2 43k4 7.36kO.03
40&l
Muscle blood flow, ml 100 g -’ . mine’ Caoz, ml1100 ml BPm, Torr Maximum developed tension, g/g
Norm [Hb]low Paoz
13.9kO.7
7.37kO.01 3.3+0.3t
PH,
?
[Hb]Pa,?
l
5
10
15
20 PO2
25
30
35
40
45
(Tom)
FIG. 1. Relationship between measured maximal 0, uptake (Vo 2 ,,,) and both venous PO, (Pvos) and calculated mean capillary PO, (Pcos) for 3 levels of hemoglobin (Hb) used in the first series of experiments. VO, m8x fell with decreasing [ Hb].
through the origin would have been expected from Fick’s law (Eq. I), just as we have found previously with reduction in Paor, in both this preparation (12) and in humans (17). The estimated Dmol for the three conditions were thus significantly different (because the PCo2)swere not much different but VO 2maxwas different). Figure 2 illustrates the relationship between Vopmax and 0, delivery. 0, delivery is closely correlated with VO, maxas [Hb] was altered, as found by others (3, 15, 23). The time required for developed tension to fall to 10-E% below the maximum developed tension was significantly less in the reduced [Hb] conditions (4.3 t 0.2, 5.3 t 0.2, and 6.3 t 0.3 min; low [Hb] to high), but the lactate output was not significantly different (Table 2). Table 3 provides the corresponding information for the second series of experiments as was provided in Table 1. In this series, there were four conditions: two levels of Hb coupled with two levels of Paoz, so at each [Hb], there were two significantly different Pao2 levels. As in the first series, there were essentially no differences in acid-base parameters (a slight increase in arterial lactate levels was seen in the lower [Hb] conditions). Again, muscle blood flow was held constant. Caoz values were significantly different, except coincidentally the normal [Hb]-low PO, combination had the same 0, content as
11457 17.5+1.0* 145*12*
9.OkO.9
Values are means _+ SE; n = 6. See Table 1 footnote for definition of abbreviations. * Significantly different (P < 0.05) from other conditions. t Not significantly different from any other condition with the same symbol but significantly different from other conditions.
the low [Hb] -normal PO, combination. As before, the mean arterial blood pressure to the muscle was higher for the same flow in the higher [Hb] conditions. Table 4 provides the gas exchange, metabolic, and fatigue results for the second series. 0, delivery was significantly different among the treatments, except the normal [Hb]-low PO, and low [Hb]-normal PO,, which were the same. V02max followed the same pattern. Figure 3 illustrates the relationship between Vo2max and calculated PEo, for the four conditions (the relationship between VO, maxand Pvo, was similar). Regression analysis of all the data points relating ir0, m8xto PCo, for each of the two [Hb] conditions showed that the intercepts of the regression lines for each [Hb] were not significantly different from the origin. Therefore, it should be noted that in Fig. 3 the line drawn for each of the two [Hb] conditions is the line of best fit that passesthrough the origin. It is apparent that when the two different [Hb] condi4. Summary of principal variables related to 0, transport and gas exchange corresponding to conditions in the second experimental series TABLE
20
Condition
n
'i-.E
15
0, delivery, ml 100 g-’ min-’ V02,, ml*100 g -l. min-’ Pvo*, Torr Pz,, , Torr Dm, , ml 100 g-*0 min-’ Torr-’ (I2 extraction, % La, pm01 100 -1 min -1 g
Low [HbJnormPao,
Low[Hb]lowPa0,
19.9t1.5”
9.2+1.0t
9.4+1.2t
5.0t0.6”
12.3kl.O” 28t2* 39t2”
7.4+0.6t 12+2t 20+2p
7.1+0.7t 19*2* 34k2’
4.2t0.5’ 11+2t 18,+2t
0.33~0.03” 63t6*
0.40t0.03” 8225
0.24+0.03t 80t5
0.25*0.02-f 84t3
17t16
28t15
l
7
l
rJ, 8 -
Norm [Hb]- Norm [HbJnormPao, lowPa0,
10
i 5 OJ
l
l
*P 0
l
0
5
10
15 20 25 30 35 40 02 delivery (ml- 100 CJ- ’ emin- ’ )
FIG. 2. Relationship between measured the 3 levels of Hb used in the first series with decreasing [ Hb].
vozrnall and 0, delivery of experiments. vozrnax
45 for fell
l
19t16
41&15
Values are means + SE; n = 6. See Table 2 abbreviations. * Significantly different (P < tions. t Not significantly different from any same symbol but significantly different from
footnote for definition of 0.05) from other condiother condition with the other conditions.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
MUSCLE
' *;
MAXIMAL
0, CONSUMPTION
15
I CP
10
z
--i
5
F 0
5
10
15
20 PCO,
25
30
35
40
45
(Tow)
FIG. 3. Relationship between measured vozrnax and calculated ( Pco,) for the 2 Hb levels combined with the 2 arterial PO, (Pa,J levels in the second series of experiments. Line drawn is line of best fit for each [Hb] condition that passes through the origin, indicating reduced muscle O2 diffusing capacity (Dm,J for the lower [Hb] condition. VO 2 max decreased with lower [Hb] and lower Pa,,.
tions are considered separately, with 0, delivery reduced by hypoxemia for each [Hb], vo2 max falls in a manner proportional to the decline in PC, for each [Hb]. The calculated Dm,, values in the low [I!b] conditions (0.24 t 0.03 and 0.25 t 0.02 ml 100 g-l min-’ Torr-‘) were significantly less than with high [Hb] (0.33 t 0.03 and 0.40 t 0.03) and were not different between the two PO, conditions during the low [Hb] treatment. If a line is drawn to connect the Pco,‘s for the two high Pao2 points (different [Hb]), the same x-intercept is obtained as seen for the three PC,, points in Fig. 1. Figure 4 illustrates the relationship between V,Z m8Xand 0, delivery, indicating a strong relationship, as in the first series. As before, the fati . gue process occurred faster with the lower 0, deliverl
l
l
ies, and lactate output was not significantly
different.
DISCUSSION
The importa .nt result of this study was the observati .on th at one of the principal factors that dete rmines the to tal diffusive flux of 0, from erythrocyte to mitochondria, DmOz, was reduced 33% with lowered [Hb]. At the same time, VO, maxand 0, delivery were also closely linearly correlated, as found by others (8, 16, 23). Although the model used to analyze the measured input and output variables and calculate PEoZ and Dm,, yields only indirect proof that, in fact, the Dm,, was reduced during reduced [Hb], this type of analysis can provide added insight and useful information concerning the dynamics of muscle 0, diffusion. 0, supp&limitation. of 30, max. The results of prior work (12, 13, 17) have suggested that 0, diffusion in the peripheral tissues is one important determinant of . vo 2m8Xwhen the supply of 0, to the mitochondria is inadequate. Our hypothesis has been that the perfusive conductance of 0, to the tissue, 0, delivery, interacts with the . diffusive conductance of 0, into the ceil to determine VO,. Thus, for a given 0, delivery, the muscle’s diffusing capacity and the capillary pressure head for 0, diffusion will determine the amount of 0, that can be extracted (see Eq. 1). In previous studies (12,17), the reductions in capillary driving pressure were postulated as the factor
AND
1109
HB CONCENTRATION
affecting the amount of 0, moved by diffusion, and there was little or no change in the calculated DmoP as PO, was reduced. Other recent work has also demonstrated that 0, diffusion limitation may be important in determining the fall in resting muscle VO, when 0, delivery is reduced below critical levels (9-U). It has been demonstrated (l4), as predicted by the diffusion limitation hypothesis, that different values of Vo2max can be obtained at the same 0, delivery when the latter is produced by different combinations of blood flow and Pa,,, showing that 0, delivery is not the unique determinant of VO, max. The question of whether 0, availability to the mitochondria is adequate at Vo2 maxunder normoxic normal [Hb] conditions remains quite controversial. As demonstrated by Honig and colleagues (2, 5, 7), the PO, within the myocyte is uniform, both radially and longitudinally, and very low at maximal work (similar to that conducted in this study). At some points along the longitudinal length of a myocyte frozen while working maximally (7), they showed that the PO, was less than the 0.5Torr value they had previously demonstrated as being necessary for adequate mitochondrial oxygenation (5) for this preparation, and this was during normal 0, delivery conditions. It is quite possible that there is not a unique resolution to this controversy and that, for different people (or muscles), 0, availability to the mitochondria may be sufficient or insufficient at normal VO, max, depending on such variables as muscle blood flow heterogeneity, amount of muscle mass being worked, capillary and muscle fiber morphometry, training level, and muscle enzymatic activity. As demonstrated by Rowe11 et al. (19), 0, supply may be limiting to muscle during whole-body exercise because of the constraints of the cardiac output; however, when a small muscle mass (such as our gastrocnemius preparation) works maximally with this constraint removed, the limiting factors may be different. Concerning VO, m8x during anemic hypoxia, Gregg et al. (8) recently showed reductions in V02mex in running rats and suggested that muscle oxidative capacity did not restrict ire, m8x. Altered [Hb] and 0, diffusion limitation. If the tissue PO, is uniformly low throughout the length of the myocyte (7) and limiting to mitochondrial respiration (especially when 0, delivery is compromised), then Fick’s law
-6
j
1’0
1’5 O2 delivery
2‘0
2’5
io
i5
(ml.100
g-’
.min -‘)
40
45
FIG. 4. Relationship between measured \jo2 mm and O2 delivery for the 4 levels of Hb and Pa,, in the second series of experiments. VO,,, decreased with lower [Hb] and lower Paoz.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
1110
MtJSCLE
MAXIMAL
0,
CONSUMPTION
of diffusion (Q. I), where Vo2max will be determined by the interaction of Dmo, and PCoZ (assuming Ptiog = 0), can be used to model tissue gas exchange. The results of the first set of experiments in this study demonstrate that the relationship found previously [ (12) V,Z M8Xfalling linearly through the origin with PEo,] was not found. According to the relationship shown in Fig. 1, as [Hb] progressively fell, V02 max would be zero even when the driving pressure for 0, diffusion (Pco,) was high (x-intercept = 28 Torr). It is obvious that the relationship between VO 2 maxand PcoZ was not as simple when 0, delivery was reduced with lowered [Hb] as with hypoxemia, as Cain (1) demonstrated in O,-compromised resting muscle. The results of these initial experiments could be interpreted in several ways. If 0, diffusion limitation were a component of the reduced Vozmax with lowered [Hb], then it must have been through changes in DmoZ and not simply changes in PCoz. Alternatively, it could be argued that 0, diffusion limitation was not an important determinant of VO, maxand that the relationship between . vo 2 m8Xand 0, delivery (Fig. 2) was an adequate explanation for the fall in VO,, as other investigators have also demonstrated with lowered [ Hb] (1,4, 15,22). It was the purpose of the second set of experiments to clarify these initial results. If decreasing [Hb] resulted in reductions in the Dmoz, as predicted from the theoretical modeling of Stainsby et al. (2l), then this should be revealed by imposing a hypoxic reduction in 0, delivery along with the anemia. With 0, delivery reduced by hypoxemia for a given [Hb], 0, diffusion limitation would be expected on the basis of our previous work (12, 17) to produce an essentially linear relationship, through the origin as predicted by Fick’s law (Eq. I), between the measured . vo 2 M8Xand the mean capillary driving pressure (calculated PC, ). This relationship was, in fact, demonstrated at each [fib] by the second series of experiments (Fig. 3). These results suggest that 0, diffusion remains an important determinant of Vo2max when 0, delivery is reduced by lowered [Hb]; however, the changes in VO, max during lowered [Hb] are influenced by decreases in Dmop. Once a new Dmq, is established for the given [Hb], then the strong relationship observed previously between V02 In8Xand PC,,, becomes evident as 0, delivery is subsequently lowered by hypoxemia. Anemic hypoxia vs. hypoxic hypoxia. As noted, anemic hypoxia appears quite different from hypoxic hypoxia. Serendipitously, the 0, delivery levels during the three [Hb] conditions of the first anemic experimental series matched three of the 0, delivery levels measured during the hypoxic hypoxia study conducted previously (12), and the relevant information is listed in Table 5. For a given 0, delivery, Vo2 max was not significantly different between anemic and hypoxic hypoxia; however, the estimate of mean capillary driving pressure, PC,, (or even PvoJ, indicated that during hypoxic hypoxia a lower driving pressure could produce the same VO, max. This, in turn, gives rise to the difference in estimated lower in anemic hypoxia. Dmoz 9 which was substantially These results are similar to those found by Cain (1) when resting animal VO, was reduced by lowering 0, supply through either anemia or hypoxemia. He demonstrated that irO, for the anemic and hypoxic conditions fell in a manner more predictable by 0, delivery than by an estimate of mean capillary driving pressure. Because of the weak
AND
HB CONCENTRATION
relationship between VO, and the mixed venous PO, below which VO, began to decrease, Cain (1) concluded that Vo2 was not limited by PO, gradients (diffusion limitation) during these conditions of reduced 0, delivery. However, we are suggesting that the PO, driving gradient remains a primary component determining VO, m8Xduring anemic hypoxia, with this relationship being affected by the changes in Dmo2. Why is Dmoz reduced with lowered [Hb]? In a recent model analysis of tissue oxygenation, Stainsby et al. (21) considered the origin of reduced Dm,, during anemic hypoxia “an enigma.” There may be several ways in which the lower [Hb] may decrease DmoZ. Dmoz is a complex parameter that encompasses a broad spectrum of variables. Some of the “[Hb] -dependent” components of Dmo, include the capillary hematocrit (and its distribution), erythrocyte spacing, 0, off-loading kinetics from Hb, and erythrocyte transit time, while some of the “[ Hb] -independent” components include muscle capillarity and the degree of capillary recruitment, diffusion distances and solubilities in the various compartments from Hb to mitochondria, heterogeneity, and myoglobinfacilitated 0, transport within the cell. A change in any one or more of these components could lower the effective Dmo2 during reduced [Hb]. With the equal muscle blood flows used in this study, a change in flow distribution and the diffusion distances involved is possible but seems unlikely; however, the inevitable increase in mean erythrocyte spacing with reduced [Hb] could also reduce Dmo , as proposed by Federspiel and Pope1 (3). Another possible explanation is that, simply due to the lower [Hb], the term “@Vc” (see Eq. 2) of Roughton and Forster (18) would be less, and this would reduce overall Dmoz accordingly. Whether any of these mechanisms or others not considered are important remains to be determined. Roughton and Forster (US), analyzing the diffusion pathway in the lung, divided the total resistance to 0, diffusion (l/Do,) into resistance imposed between the plasma and alveolus (1 /Dperice iiiav) and the resistance to 0, on-loading (l/@Vc), where B is the 0, on-loading reaction rate from Hb and Vc is the capillary blood volume. Thus l/D02
(2)
= 1 /Dpericapiiiary + I/OVC
A similar kind of analysis can be proposed in the muscle but formulated more generally as follows: the total resistance (1 /DmoJ is divided into a [Hb] -independent component (l/D) and a [Hb]-dependent component (11 K[Hb]). The assumption is made that the [Hbl-dependent component is directly proportional to the [Hb] and D is constant (at constant blood flow in a given muscle) by definition. Equation 2 can then be used to partition the diffusion resistance into [Hb]-dependent and [Hb]independent parts (3) 1/Dm,:, = l/D + l/K[Hb] In the second series of experiments, the calculated Dm,, at the normal [Hb] of 14.1 g/l00 ml was 0.36 ml. 100 g -l. min-l Torr-1 and 0.25 ml 100 g-’ mine1 Torr-’ at 48% of that [ Hb]. Equation 3 can be used for each of the two [Hb] conditions to simultaneously solve for the unknown values of D and K. When this was done, it was found that, at normal [ Hb], roughly 60% of the total resisl
l
l
l
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
MUSCLE
[Hb], g/100 ml Paoz, Torr Caoz, ml/100 ml Muscle blood flow, ml. 100 g-’ I)z delivery, ml 100 g-’ min-’ vo 2max, ml- 100 g-‘=min-’ Pvo,, Torr Pco2, Torr DO,, ml 100 g-’ min-’ Torr-’ l
l
l
min-’
l
l
l
MAXIMAL
0,
CONSUMPTION
5.8+0.3 77+3 7.4kO.5 122+11 9.OkO.8 6.1kO.7 23+2 35+2 0.20~0.02
13.3kO.6 29&l 7.9k0.6 118+15 9.3k1.6 7.2k1.3 14rtl 22&l 0.32kO.06
Values are means + SE; n = 8 during HH and n = 6 during previous study (12). Condition 3 represents control conditions
We gratefully acknowledge Dr. Steve Hempleman for helpful discussions concerning the analyses of some of these data. This research was supported by National Heart, Lung, and Blood Institute Grant HL-17731. Address for reprint requests: M. C. Hogan, Dept. of Medicine M-023A, University of California, San Diego, La Jolla, CA 92093-0623. 26 March
1990; accepted
in final
form
4 October
1990.
REFERENCES 1. CAIN, S. Oxygen delivery and uptake in dogs during anemic and hypoxic hypoxia. J. Appl. Physiol. 42: 228-234, 1977. 2. CONNEIT, R. J., T. E. J. GAYESKI, AND C. R. HONIG. An upper bound on the minimum PO, for 0, consumption in red muscle. Adv. Exp. Med. Biol. 191: 291-300, 1985.
1111
HB CONCENTRATION
9.420.1 79+3 12.OkO.2 121+10 14.5k1.4 9.Wl.O 27+2 40+2 0.25-tO.02
AH. AH, anemic for the 2 studies.
tance was associated with the [ Hb] -independent component and thus 40% with the [Hb]-dependent component. At the lower [Hb], there was an approximate reversal of these relative resistances, with the [ Hb] -dependent component now becoming more important. Intuitively, if Dmoz was determined wholly by [Hb] -dependent factors, then the percent change in Dmoz would have been equal to the percent change in [Hb]. The 50% reduction in [Hb] did not result in a 50% drop in DmoP, which fell by only 33%, indicating that [Hb] is not the only factor affecting DmOz. Fatigue. The time before fatigue set in and the lactate output followed the same pattern with lowered [Hb] as found previously in hypoxemia (12). Fatigue development occurred significantly sooner with the lower 0, deliveries, but lactate output was not different. In fact, a comparison of the anemic hypoxia (4.3 t 0.2, 5.3 t 0.2, and 6.3 t 0.2 min; lowest 0, delivery to highest) and the hypoxic hypoxia fatigue times (4.7 t 0.2, 5.3 t 0.2, and 6.2 + 0.5; lowest OS delivery to highest) demonstrates lit&difference in fatigue times for the two methods of 0, delivery. This suggests that the development of fatigue was dependent only on 0, delivery and not on the manner in which it was altered. Conclusion. This study suggests that reducing the [Hb] as a means of lowering the 0, delivery to working muscle decreases the effective Dmo2. These results suggest that . vo 2max during anemia is reduced not only by the expected effect from the decreased convective 0, delivery to the muscle but also by these reductions in Dmq,. Whether this reduction in Drn,? relates to changes in oxyhemoglobin off-loading kinetics, altered erythrocyte spacing, or other effects of altering [Hb] remains to be determined.
Received
AND
conditions
13.5k0.6 38&l 11.6t0.6 122-tl5 14.lrt2.0 10.2-tl.2 20+1 29&l 0.31kO.05 of this study;
HH,
14.4t0.6 74+4 17.9t0.8 120+13 2O.lk2.9 12.2U.O 3Ok2 4122 0.32kO.03 hypoxic
hypoxia
13.5+0.8 79+4 16.9kO.8 124+16 21.Ok2.9 14.Ok1.5 28k2 44+1 0.30~0.05 conditions
of the
3. FEDERSPIEL, W. J., AND A. S. POPEL. A theoretical analysis of the effect of the particulate nature of blood on oxygen release in capillaries. Microvasc. Rex 32: 164-189, 1986. 4. GAEHTGENS, P., F. KREUTZ, AND K. H. AI~BRECHT. Optimal hematocrit for canine skeletal muscle during rhythmic isotonic exercise. Eur. J. Appl. Ph.ysiol. Occup. Physiol. 41: 27-39, 1979. 5. GAYESKI, T. E. J., R. J. CONNETT, AND C. R. HONE. Minimum intracellular PO, for maximum cytochrome turnover in red muscle in situ. Am. J. Yhysiol. 252 (Heart Circ. Physiol. 21): H906-H915, 1987. 6. GAYESKI, T. E. J., W. J. FEDERSPIEL, AND C. R. HONIG. A graphical analysis of the influence of red cell transit time, carrier-free layer thickness, and intracellular PO, on blood-tissue 0, transport. Adv. Exp. Med. Viol. 222: 25-35, 1988. 7. GAYESKI, T. E. J., AND C. R. HONIG. Intracellular PO, in long axis of individual fibers in working dog gracilis muscle. Am. J. Physiol. 254 (Heart Circ. Physiol. 23): H1179-H1186, 1988. 8. GREGG, S. G., R. S. MAZZEO, T. F. BUDINGER, AND G. A. BROOKS. Acute anemia increases lactate production and decreases clearance during exercise. J. Appl. Physiol. 67: 756-764, 1989. G., AND ,J. M. ANDRY. Increased hemoglobin 0, affin9. GUTIERREZ, ity does not improve 0, consumption in hypoxemia. J. Appl. Physiol. 66: 837-843, 1989. 10. GUTIERREZ, G., C. MARINI, A. L. ACERO, AND N. LUND. Skeletal muscle PO, during hypoxemia and isovolemic anemia. J. Appl. Physiol. 68: 2047-2053, 1990. 11. GUTIERREZ, G., R. J. POHIL, AND R. STRONG. Effect, of flow on 0, consumption during progressive hypoxemia. J. Appl. Physiol. 65: 601-607, 1988. 12. HOGAN, M. C., D. E. BEBOUT, P. D. WAGNER, AND J. R. WEST. Maximal 0, uptake of in situ dog muscle during acute hypoxemia with constant perfusion. J. Appl. Physiol. 69: 570-576, 1990. 13. HOGAN, M. C., ,J. ROCA, P. D. WAGNER, AND J. B. WEST. Limitation of maximal 0, uptake and performance by acute hypoxia in dog muscle in situ. J. Appl. Physiol. 65: 815-821, 1988. 14. HOGAN, M. C., J. ROCA, J. B. WEST, AND P. D. WAGNER. Dissociation of maximal 0, uptake from 0, delivery in canine gastrocnemius in situ. J. Appl. Physiol. 66: 12 19-1226, 1989. 15. HOGAN, M. C., AND H. G. WELCH. Effect of altered arterial 0, tensions on muscle metabolism in dog skeletal muscle during fatiguing work. Am. J. Physiol. 251 (Cell Physiol. 20): C216-C222, 1986. 16. HORSTMAN, D. H., M. GLESER, D. WOLFE, T: TRYON, AND J. DELEHUNT. Effects of hemoglobin reduction on VO, max and related hemodynamics in exercising dogs. J. Appl. Physiol. 37: 97-102, 1974. 17. ROCA, J., M. C. HOGAN, D. STORY, D. E. BEBOUT, P. HAAB, R. GONZALEZ, 0. UENO, AND P. D. WAGNER. Tissue 0, diffusion limitation of maximal 0, uptake in man. J. Appl. Physiol. 67: 291-299, 1989. 18. ROUGHTON, F. J. W., AND R. E. FORSTER. Relative importance of diffusion and chemical reaction rates in determining rate of exchange of gases in the human lung, with special reference to true diffusing capacity of pulmonary membrane and volume of blood in the lung capillaries. J. Appl. Physiol. 11: 290-302, 1957. 19. ROWELL, L. B., B. SALTIN, B. KIENS, AND N. J. CHRISTENSEN. IS peak quadriceps flow in humans even higher during exercise with hypoxemia? Am. J. Physiol. 251 (Heart Circ. Physiol. 20): H 1038H1044, 1986. 20. STAINSBY, W. N., AND A. B. OTIS. Blood flow, blood oxygen ten-
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
1112
MUSCLE
MAXIMAL
0,
CONSUMPTION
sion, oxygen uptake, and oxygen transport in skeletal muscle. Am. J. Physiol. 206: 858-866, 1964. 21. STAINSBY, W. N., B. SYNDER, AND H. G. WELCH. A pictographic essay on blood and tissue oxygen transport. Med. Sci. Sports Exercise 20: 213-221,198s. 22. WAGNER, P. D. An integrated view of the determinants of maxi-
AND
HB
CONCENTRATION
mum oxygen uptake. In: Oxygen Transfer from Atmosphere to Tissues, edited by N. C. Gonzalez and M. R. Fedde. New York: Plenum, 1988. 23. WOODSEN, R. D., R. E. WILLS, AND C. LENFANT. Effect of acute and established anemia on 0, transport at rest, submaximal and maximal work. J. Appl. Physiol. 44: 36-43, 1978.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (128.252.067.066) on August 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
