Microvascular oxygen transport: impact of a left-shifted dissociation curve J. CHRISTOPHER STEIN AND MARY L. ELLSWORTH Department of Pharmacological and Physiological Science, St. Louis University School of Medicine, St. Louis, Missouri 63104 Stein, J. Christopher, and Mary L. Ellsworth. Microvascular oxygen transport: impact of a left-shifted dissociation curve. Am. J. Physiol. 262 (Heart Circ. Physiol. 31): H517H522, 1992.-The impact of an increased hemoglobin oxygen affinity (decreased P& on oxygen transport was evaluated in capillaries of the retractor muscle under nonhypoxic (FIN, = 0.30 and 0.21) and hypoxic (FIN, = 0.10) conditions in hamsters with normal oxygen affinity [control; PsO = 26.1 t 1.0 (SD) mmHg, n = 121 and in hamsters with an increased oxygen affinity [treated; PsO = 16.2 t 1.6 (SD) mmHg, n = 71 induced by chronic short-term administration of sodium cyanate. Using in vivo video microscopy and computer-aided image analysis, we determined oxygen saturation (SOJ and associated hemodynamic parameters in both arteriolar (n = 30 control, 18 treated) and venular (n = 25 control, 17 treated) capillaries. In response to hypoxia, systemic arterial PO:!decreased to 29.6 t 6.0 (SD) mmHg in control animals and 24.7 t 3.8 (SD) mmHg in treated animals associated with abrupt decreases in systemic arterial blood pressure and increases in respiratory rate. The decrease in So2 across the capillary network during nonhypoxic ventilation was 13.3% So2 for control animals and 11.0% So2 for treated animals. During hypoxic ventilation, the decrease in So2 was 9.1% So2 in control animals and 8.7% So2 in treated animals. Hemodynamic parameters were not significantly different in the two groups during hypoxia. Estimated end-capillary Pop was significantly lower in the treated animals. These data indicate that an increased oxygen affinity does not provide an obvious advantage for oxygen transport during hypoxia at the level of the capillary network in resting striated muscle; however, such an advantage might become apparent in the presence of an increased metabolic rate or a more severe hypoxic challenge. hemoglobin oxygen affinity; sters; hypoxia
capillaries;
striated
muscle; ham-
DELIVERY OF OXYGEN to the tissue requires the convective transport of oxygen to the site of exchange, the release of the oxygen from its carrier molecule (hemoglobin), and its subsequent diffusion down an oxygen tension (PO& gradient to the oxygen-consuming mitochondria. The magnitude of this Paz gradient depends on the amount of oxygen delivered to the site of exchange (determined by the level of blood flow and oxygen content), the level of tissue metabolism (oxygen consumption), and the affinity of the red blood cell hemoglobin for oxygen (i.e., the position of the oxyhemoglobin dissociation curve). Alterations in any or all of these parameters can impact on oxygen transport to cells. Animals native to high altitude have effectively adapted to the hypoxic environment, in part by having a left-shifted oxyhemoglobin dissociation curve. A leftshifted dissociation curve was originally speculated by Barcroft et al. (1) to be of advantage in hypoxic environments due to the increased onloading of oxygen in the pulmonary circulation. A number of studies have subse0363-6135/92 $2.00 Copyright
quently attempted to evaluate the efficacy of a leftshifted dissociation curve for oxygen delivery under various levels of inspired oxygen (FIN,) from measurements of oxygen levels in systemic arterial and mixed venous blood. Turek et al. (19, 20) developed a model of oxygen transport and provided supporting experimental evidence in studies of rats indicating that a left-shifted dissociation curve would only be advantageous under conditions of severe hypoxia (FIN, = 0.08 and 0.056). A similar result was obtained by Teisseire et al. (17, 18). They speculated that this beneficial effect was due to changes in characteristics of the oxyhemoglobin dissociation curve other than the Paz at which 50% of the hemoglobin is saturated with oxygen (P&. Schumacker et al. (15) determined that, during steady-state exercise in dogs, the PsOis not an important factor in determining tissue oxygen extraction during normoxia or mild hypoxia (FIN, = 0.12), but, during extreme hypoxia (FIN, = O.lO), a low PsOmay help maintain tissue Paz by enhancing systemic oxygen delivery. Gutierrez and Andry (7) evaluated the role of P50 in oxygen transport in a study of isolated rabbit hindlimbs perfused with human red blood cells treated to obtain perfusate having a P50of either 32.5 (low affinity) or 22.8 mmHg (high affinity). They determined that, although total oxygen transport was higher in the high-affinity group during hypoxia, venous Paz values were the same in the two groups. Therefore, oxygen extraction was actually lower in the high-affinity group. They attributed this finding to a diffusional limitation to oxygen transport in the tissue in hypoxia due to the kinetics of oxygen off-loading from the red blood cells (6, 7). In all of the studies presented above, the approach has been to evaluate tissue oxygen transport at the whole tissue level and extrapolate the results to the level of the microcirculation. The purpose of this study was to evaluate oxygen transport and the efficacy of a left-shifted dissociation curve at the level of the microcirculation where tissue oxygen transport occurs. We determined oxygen saturation (SOL), red blood cell velocity, lineal density (LD), and red blood cell supply rate (SR) using computer-aided video techniques (2) in both arteriolar and venular capillaries of the cheek pouch retractor muscle. These hemodynamic and oxygenation parameters were evaluated under nonhypoxic (FIN, = 0.3 and 0.21) and hypoxic (FIN, = 0.10) conditions in hamsters with normal PsOand in hamsters with a decreased P50 induced by chronic short-term administration of sodium cyanate. By systematically measuring capillary oxygen transport parameters under conditions of normal and reduced P50, we hope to elucidate the role of oxygen affinity in oxygen transport to the tissue under nonhy-
0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart at Midwestern Univ Lib (132.174.254.157) on February 12, 2019.
H517
H518
OXYGEN
TRANSPORT
poxic and hypoxic conditions. This information should be useful in the formation of a comprehensive model of oxygen transport in the microcirculation. METHODS
Male golden hamsters (48 t 16 days, 98.1 t 21.4 g, n = 19) were separated into two groups, control (n = 12) and treated (n = 7). The treated group received sodium cyanate (0.2 mg/kg ip) five times per week for -2 wk to decrease PsO. Hamsters were initially anesthetized with pentobarbital sodium (6.5 mg/ 100 g body wt ip). The trachea was cannulated to insure a patent airway, and the hamster was allowed to breath spontaneously. The length of the tracheal cannula approximated the distance from the point of insertion into the trachea to the tip of the hamster’s nose to maintain a normal anatomical dead space. A femoral artery was cannulated with PE-10 tubing to enable continuous monitoring of systemic arterial blood pressure (BP) and collection of blood samples for systemic arterial blood gas, pH, hematocrit, and PsO determinations. A femoral vein was likewise cannulated to enable continuous infusion of supplemental anesthetic (pentobarbital sodium, 0.15 mg/min). The right cheek pouch retractor muscle was surgically exteriorized, as described by Sullivan and Pittman (16). Briefly, the muscle was separated from underlying back muscles, and a hemoclip (Edward Week) was used to secure two ligatures to the muscle. The muscle was then cut at its spinal end and placed, ventral side up, at its in situ dimensions on a specially designed Plexiglas platform. The muscle was covered with Saran (Dow Corning) to prevent both desiccation of the tissue and gas exchange with the atmosphere. Deep esophageal and muscle temperatures were monitored and maintained at 37 t 1°C by separate heat exchangers within the animal platform. The muscle was transilluminated with a xenon lamp (75 W) with a stabilized power supply to provide constant light intensity. Observations were made using a Zeiss ACM microscope equipped with a long-working distance objective (UD40/0.65). The objective was used dry to give an equivalent magnification and numerical aperture of x 25.8 and 0.41, respectively. The microscope image was viewed using a dual-camera high-resolution closed-circuit video system consisting of two SIT cameras (model 66; Dage MTI), video monitors (WV-5410; Panasonic), S-VHS video cassette recorders (AG-7500; Panasonic), time-date generators (WJ-810; Panasonic), a video analyzer (model 321; Colorado Video), video field counter (Dept. of Medical Biophysics, University of Western Ontario), and microcomputer (Compaq 386/25) with a 12bit analog-to-digital converter (model RTI-815-A; Analog Devices). The video analyzer superimposed a vertical cursor line on the video image that provided light intensity information (slow scan video output) for the computer analysis of red blood cell So2, velocity, and LD. The gain on both the camera and video cassette recorder was adjusted manually. The output of the video system was linear over the range of light intensities used. Optical parfocalizing devices positioned in front of each video camera were used to ensure that the two images were in equal focus. The vertical and horizontal image sizes and position were adjusted within each camera to make the two images produced by the cameras superimposable. Because the two tapes have to be analyzed separately, an audio signal generated by the video field counter was placed simultaneously on the two tapes at 30s intervals. Individual capillaries, defined as vessels containing single-file red blood cell flow and having a diameter less than that of a red blood cell, were located and designated as arteriolar or venular according to their anatomical relationship with their feed arteriole or collecting venule, respectively. It should be pointed out here that only those portions of the capillaries that
AND OXYGEN
AFFINITY
were within 130 t 25 pm from their feed arteriole or collecting venule were utilized. In addition, the arteriolar and venular capillaries are not necessarily anatomically related. Scenes of capillary blood flow were initially recorded while the hamster breathed spontaneously either room air [21% 02; normoxia (N)] or a supplemental oxygen gas mixture [30% 02, balance N,; supplemental (S)]. After 5 min of nonhypoxic baseline information was obtained, the hamsters’ inspired gas supply was switched to a hypoxic gas mixture [IO% 02, balance N2; hypoxia (H)]. The same anatomical segment of capillary was recorded for an additional 5 min; therefore, for each located capillary, we obtained 5 min of baseline nonhypoxia information (S or N) followed immediately by 5 min of systemic hypoxia data [H(S) or H(N), respectively]. At the end of the 5-min hypoxic challenge, the hamster was switched back to the prehypoxic gas, either room air or 30% Oz. Blood samples (2-6) were obtained over the course of the experiment during both the baseline and hypoxic periods for determination of systemic arterial pH, POT, Pco~, P 50, and hematocrit. Spontaneous respiratory rate was noted during both hypoxic and prehypoxic conditions. After the hypoxic interval was completed and the hamster was switched back to the prehypoxic gas mixture, sufficient time was allowed for the hemodynamic parameters to return to baseline levels. At this point, another capillary was located, classified as arteriolar or venular, and the protocol was repeated. This protocol was followed for both control and treated hamster groups. All data recorded were subsequently analyzed off-line. So2 was determined using a modified version of the computer-assisted image analysis technique described previously by Ellis et al. (2). This method enables the direct evaluation of So2 of individual red blood cells along a given length of capillary (100 TV lines = 93 pm) for 60 s. In short, digitized light intensities were obtained one time per frame at both wavelengths (see below). The digitized light intensities were redisplayed as a space-time image for both wavelengths simultaneously. The light intensity values corresponding to the amount of light transmitted through the capillary in the presence (I) and absence (I,) of a red blood cell were determined. These values were used to compute an optical density (OD) at each point along the capillary segment, according to the equation OD = loglo(I,/[F(I - I, + 1,/F)]) where F is the glare correction factor described by Pries et al. (13) for both an isosbestic wavelength (420 nm) and a wavelength at which there is a maximum difference between oxy- and deoxyhemoglobin (431 nm). The ratio of the OD values determined from identical time segments at the two wavelengths (OD431/0D420) is linearly related to So2. The local longitudinal So2 gradient (ASo,/Az, where AZ is the distance separating 2 sites) was determined from coincident measurements of So2 at upstream and downstream sites on individual capillaries (ASo,) divided by AZ. Red blood cell SR (cells/s) was determined for each capillary by counting the number of cells crossing a horizontal line on the computergenerated space-time image display divided by the length of time (in s) of that interval. Red blood cell velocity (v; pm/s) and LD (cells/mm) were determined from the light intensity data, as described previously. Supply-rate weighted So2 values (Sf,,) were determined from coincident measurements of So2 and red blood cell SR from arteriolar and venular capillaries as Sfo, = z(SiO, X fi)/zfi, where SiO, and fi are the inlet or outlet So2 values and red blood cell SR values, respectively, in individual capillaries. Capillary diameters were measured directly from the video monitor during both normoxia and hypoxia using a vernier caliper and video-taped micrometer scale. Pso was determined in each animal using the biotonometric method of Neville (10) and corrected to standard conditions
Downloaded from www.physiology.org/journal/ajpheart at Midwestern Univ Lib (132.174.254.157) on February 12, 2019.
OXYGEN
(PH = 7.4, pcoz = 40 mmHg) p50 =
TRANSPORT
using the relationship 0.41(pH-7.40)-0.0610g(40/Pco2)
(PSO)std
x
lo-
Capillary POT values were estimated from the measured capillary So2 values using the appropriate values of standard Pso and Hill’s n (8; unpublished data). The values were subsequently corrected for the appropriate values of PCO~ and pH in the capillaries that we have estimated, based on previous studies (12, 14), to be best approximated by the venous PCO~ (unpublished data) and arterial pH. Data analysis. Data are presented as means t SD. An analysis of variance followed by either a Mann-Whitney U (unpaired data) or Wilcoxon (paired) test was used to evaluate the differences between the different systemic and microcirculatory parameters for the control and treated animals. An unpaired Student’s t test was used to compare the current data with previously published data. All data analysis was done using a statistical software package (Systat, Evanston IL). Significance was assigned at P < 0.05, unless otherwise stated. RESULTS
Systemic effects. The administration of sodium cyanate resulted in a significant decrease in standard P50 from 26.1 t 1.0 to 16.2 t 1.6 mmHg with no significant change in systemic hematocrit (Table 1). While breathing room air, the treated animals tended to have significantly lower systemic arterial BP values (66 t 14 mmHg) than the control animals (85 t 18 mmHg). Because we were concerned that the inclusion of data from animals with low arterial BP values (~70 mmHg) during nonhypoxic conditions might interject an additional complicating factor, we did not include these in the data to be presented here. This criterion precluded the inclusion of a large number of capillaries in the treated animals, especially those whose nonhypoxic gas was room air and a few capillaries from each of the other groups. Systemic arterial blood gases (Pao, and Pace,), arterial pH (pH,), BP, and respiratory rates for the two groups of animals for the different levels of inspired gas are presented in Table 2. In general, the control and treated animals had similar levels of blood gases and pH with the exception that the Pao, of the treated animals during normoxia and hypoxia were significantly lower than those of the control animals. Systemic arterial BP values were not different in the two groups under any conditions; however, the respiratory rates of the treated animals were significantly lower than their corresponding controls. In response to hypoxia, Pao, and Pace, decreased significantly while pH, increased in all but one control group [H(N)] . Arterial BP usually decreased significantly after the induction of hypoxia, whereas respiratory rate Table 1. General information Control (n = 12)
Age, days Weight, g H&s, % PSO, mmHg
49.8t14.7
105.6t23.8 49.71k4.2 26.ltl.O
Treated (n = 7) 45.1*18.9
&5.3-r-5.4* 50.5t4.2 16.2tl.6”
Values are means t SD; n, no. of animals. He&, systemic crit; PsO, 0, tension at which hemoglobin is half saturated * Significantly different from control values at P < 0.05.
hematowith Oz.
AND
OXYGEN
AFFINITY
H519
increased. The decrease in BP was significantly greater in the control animals than in the treated for the transition from supplemental to hypoxia but not for normoxia to hypoxia. The decrease in BP was more rapid in the control animals than in the treated animals. The increase in respiratory rate was significantly greater in the control animals than in the treated in both cases. It has been reported by Walker et al. (21) that hamsters increase frequency and not tidal volume in response to hypoxia; therefore, these changes in respiratory frequency directly reflect changes in total ventilation. Microvascular effects. The level of So2 at the upstream end of the microvasculature (arteriolar, upstream) was not different in those animals breathing room air (N) vs. those breathing 30% O2 (S) for both the control and treated animals. Similar results were found for the downstream end of the network. Switching the level of inspired gas to 10% O2 caused significant decreases in the levels of So2 at both the arteriolar and venular ends of the network, with the hypoxic values not dependent on the level of prehypoxic gas. Because there was no dependence of the level of capillary So2 on the level of nonhypoxic inspired gas, we combined these data into nonhypoxic (supplemental + normoxic) and hypoxic [H(S) + H(N)] groups (Table 3, Fig. 1). In control animals, the arteriolar capillary (upstream) So2 was 59.6 t 10.6% compared with 70.7 t 10.6% in the treated animals. With hypoxic ventilation, these levels decreased significantly to 39.4 t 16.0 and 50.6 t 13.4%, respectively. At the downstream venular capillary, So2 was 46.3 t 14.0% in control animals and 59.7 t 11.0% in treated animals during nonhypoxic ventilation and 33.0 t 12.0 and 41.9 t 9.5% during hypoxic ventilation, respectively. In all cases, th.e levels of So2 in the treated animals were significantly higher than in the controls; however, the levels of PO, were significantly lower in the treated than in the control animals (Table 3, Fig. 1). In control animals during nonhypoxic ventilation, the arteriolar capillary (upstream) Paz was 33.1 t 6.2 mmHg compared with 25.0 t 4.8 mmHg in treated animals. With hypoxic ventilation, these levels decreased significantly to 21.9 t 6+3 and 17.3 t 3.5 mmHg, respectively. At the downstream venular capillary end, PoB was 27.1 t 6.6 mmHg in control animals and 20.7 ~fi3.6 mmHg in treated animals during nonhypoxic ventilation and 18.9 t 5.0 and 14-7 & 2.6 mmHg during hypoxic ventilation, respectively. In all cases there was a significant difference in POT across the network (upstream arteriolar vs. downstream venular). ASOJAZ during nonhypoxic ventilation for the control animals for the arteriolar and venular capillaries combined was 0.067 t 0.352%/pm compared with 0.088 t O.Z53%/pm for the treated animals. During hypoxia, the local longitudinal gradients were unchanged at 0.071 & 0.292 and 0.095 t 0.255%/pm, respectively. Hemodynamically, there were few differences in any of the parameters, as shown in Table 3. Red blood cell SR was significantly lower in the treated animals at the arteriolar end during nonhypoxic ventilation compared with control animals, due almost entirely to a significantly lower LD.
Downloaded from www.physiology.org/journal/ajpheart at Midwestern Univ Lib (132.174.254.157) on February 12, 2019.
H520 Table
OXYGEN
TRANSPORT
AND
OXYGEN
AFFINITY
2. Systemic parameters Pa0,,
Pace,,
mmHg
Control S N H(S) H(N) Treated S N H(S) H(N)
mmHg
PHa
Systemic BP, mmHg
79.2t18.3 58.2t8.5 27.9t5.7* 32.3t5.9*
42.0k9.6
34.3t5.1 29.4t4.2* 27.7-+2.3*
7.321t0.035 7.387t0.051 7.417*0.042* 7.428t0.038
86.8k17.9 85.4t13.4 61.8t18.6* 59.7*16.9*
70.7t9.2
39.5t4.5 36.0t9.3 31.8t2.5" 30.5t4.7*
7.327t0.045 7.365t0.078 7.418zko.o45* 7.434&0.105*
81.9t7.5 84.4t9.4 67.2&13.3* 63.8&11.0*
46.lk9.O.f 25.5*3.4* 24.5*5.2*"r
Respiration, breaths/min
73.8d8.5 59.3t12.9 129.9tl2.5* 103.8&20.0* 54.5+11.7$ 50.2k9.0 87.0+23.1*$ 70.5+7.4*$
Values are means t SD; supplemental gas oxygen mixture (S) = 0.3 inspired oxygen (FIN,); normoxia (N) = 0.21 FIN,; hypoxia supplemental gas oxygen mixture [H(S)] and hypoxia normoxia [H(N)] = 0.1 FIN, associated with S and N groups, respectively. For control animals, n for arterial blood gases were S and H(S) = 11 and N and H(N) = 8; for systemic blood pressure and respiratory rate, S and H(S) = 32 and N and H(N) = 26. For treated animals, n for arterial blood gases were S and H(S) = 7 and N and H(N) = 7; for systemic blood pressure and respiratory rate, S and H(S) = 27 and N and H(N) = 8. * Significantly different from corresponding nonhypoxic values at P < 0.05. t Significantly different from control values at P < 0.05. $ Significantly different from control values at P < 0.01.
Table 3. Oxygen transport parameters in arteriolar and venular capillaries Control
Treated
Nonhypoxia
Arteriolar v, Pm/s LD, cells/mm SR, cells/s so,, % Po2, mmHg Sfo,, 5% ASo,/Az, %/pm Venular v, w/s LD, cells/mm SR, cells/s SO& % Po2, mmHg Sf,,, 5% ASo,/Az, %/pm
119.6t58.7 57.2519.1 7.0t4.6 59.6klO.6 33.1k6.2 59.3 0.142~10.334 102.6k56.9 62.6kl6.2 6.723.0 46.3214.0 27.1t6.6 50.8 - .0.043t0.356
Hypoxia
Nonhypoxia
Hypoxia
122.9t67.6 58.8t23.7 7.6rt4.6 39.4+16.Ot 21.9+6.3-t 42.1 - .0.002stO.306
105.6k56.8 45.6tl7.3" 4.4&1.5* 70.7+10.6$ 25.0*4.8$ 70.0 0.063t0.229
116.9k67.3 52.2k20.8 6.4t4.2 50.6+13.4*-f 17.3+3.5'f$ 48.6 0.055t0.269
102.9k65.3 63.8218.6 6.7t3.8 30.3+12.0t 18.9+5.O"f 33.6 0.158t0.253
122.7261.0 60.3t22.9 6.9t3.0 59.7+11.0$ 20.7k3.61 60.5 0.114t0.282
105.lk68.0 67.Ok22.0 7.723.8 41.9+9.5t$ 14.7+2.6"f$ 43.8 0.137t0.240
Values are means t SD; v, velocity; LD, lineal density; SR, supply rate; Soz, oxygen saturation; Po2, oxygen tension; Sf,,, supply-rate weighted ASO, values; AZ, distance separating 2 sites; A, change. For arteriolar control group, n = 30 animals each for nonhypoxia and hypoxia, and for arteriolar treated, n = 18 each for nonhypoxia and hypoxia groups. For venular control, n = 25 animals in each group, and for venular treated, n = 17 each. * Significantly different from control at P c 0.05. t Significantly different from nonhypoxic values at P < 0.05. $ Significantly different from control at P < 0.01. DISCUSSION
The data obtained in this study suggest that, in resting striated muscle, an increase in hemoglobin oxygen affinity (decreased P& does not affect the amount of oxygen lost across the capillary network. We found that, in capillaries of the hamster retractor muscle, the change in So2 across the network during nonhypoxic ventilation was 13.3% So2 for control animals and 11.0% SO:! for treated animals. This value of 13.3% So2 for the control animals is slightly smaller but not significantly different from the value obtained previously for capillaries of the hamster retractor muscle (5). During hypoxic ventilation, the overall changes in So2 were somewhat smaller for both control and treated animals (9.1 and 8.7% So2, respectively) compared with their nonhypoxic values. Because, during hypoxia, the red blood cell SR values were not different in the two groups, this would imply that oxygen extraction during hypoxia was not significantly different in the two groups, even though the
computation of the oxygen extraction ratio would suggest otherwise (see below). In both normoxia and hypoxia, the end-capillary Paz values were significantly lower in treated animals, which might be considered detrimental. Therefore, it seemsclear that, based on the information available in this study, a left shift of the oxyhemoglobin dissociation curve does not provide any significant benefit for capillary oxygen exchange in resting striated muscle at this level of hypoxia. Turek and his colleagues (19, 20) came to a different conclusion in their study of rats. There are a number of differences between their study and ours that might help explain the discrepancy. First, there are differences between the oxyhemoglobin dissociation curves for rats and hamsters. Their control rat oxyhemoglobin dissociation curve had a PsOof 35.6 mmHg compared with the hamster’s of 26.1 mmHg. In their treated animals, the P50was decreased to 22.3 mmHg, whereas ours was 16.2 mmHg. Second, they used levels of FIN, that were some-
Downloaded from www.physiology.org/journal/ajpheart at Midwestern Univ Lib (132.174.254.157) on February 12, 2019.
OXYGEN
80 E
TRANSPORT
AND
Treated
70
d
3 60 cd 25 50
Control
0
10
20 30 Oxygen Tension
40 (mmHg)
50
60
Fig. 1. Estimated oxygen tension vs. oxygen saturation at upstream arteriolar capillary (a) and downstream venular capillary (u) sites in control and treated animals during nonhypoxic (open symbols) and hypoxic (closed symbols) ventilation and corresponding estimated oxyhemoglobin dissociation curves. Error bars indicate 1 SD.
what lower (FIEF = 0.08 and 0.056) than that used in the present study (FIN, = 0.10). However, the levels of sys-
temic arterial Paz in their control animals at both 0.08 or 0.056 FIN, were not different from ours at 0.10 FIN,. In the treated animals, our level of systemic Pao, was intermediate between their group that inspired 8.0% O2 and their group that inspired 5.6% 02. Thus it is unlikely that the level of FIN, is responsible for the difference in the results. Third, and possibly most importantly, their study evaluated systemic blood gas values obtained from the carotid artery and right atrium, which assessed oxygen exchange in the whole animal comprised of tissues having both a high and low metabolic rate, whereas we evaluated oxygen transport at the capi llary-tissue interface in a tissue with a low metabolic rate. Previous investigators have noted that, after a left shift of the oxyhemoglobin dissociation curve, blood flow shifts away from tissues with a low metabolic demand toward tissues with a high metabolic demand (23). However, as noted above, the capillary red blood cell SR is lower at the arteriolar end of the network in the treated animals during nonhypoxic ventilation but otherwise is not significantly different in the two groups of animals (control and treated). Therefore, it seems unlikely that there is a significant diversion of blood away from the retractor muscle. We have seen prev iously (5) that, when one compares th .e locally computed value for the longitudinal So2 gradient (local ASoz/Az = 0.113 t O.l96%/pm) with the average longitudinal gradient obtained by dividing the difference in mean SOBdetermined at the upstream ends of arteriolar capillaries and the downstream ends of venular capillaries (60%39.9%) by the average anatomical capillary length (412 pm), the latter quantity was smaller than the local ASOJAZ by a factor of two. We were interested to know if a similar comparison of the
OXYGEN
AFFINITY
HZ1
present data would result in a similar discrepancy. In the control animals during nonhypoxic ventilation, the two values differed by a factor of two, which is consistent with the previous result; however, for the treated animals, the average gradient was smaller than the local gradient by a factor of three, which suggests an even greater degree of diffusional interaction occurs among capillaries in the treated animals. In hypoxia, the two gradients differed by a factor of 3 for the control animals and by a factor of 4.5 for the treated animals. Again, this degree of discrepancy would imply increased diffusional oxygen exchange in hypoxia most likely due to the presence of local areas of hypoxia and consequent increased local POT gradients. Gutierrez and Andry (7) have postulated the presence of a diffusional limitation to oxygen transport at the venular end of the capillary network in hypoxia based on data from isolated perfused rabbit hindlimbs. They based this on their findings of a lower oxygen extraction but unchanged venous Po2 values in the presence of the high-affinity perfusate compared with the low-affinity perfusate. In the present study, we found that the change in SO, across the network during both normoxia and hypoxia were not significantly different; however, if one computes the oxygen extraction ratio { [ (Sfo,), (Sf&]/(SfoJ,) for the control and treated animals, one obtains values of 14 and 13.6%, respectively, during nonhypoxic ventilation and 20.2 and 9.9%, respectively, during hypoxic ventilation, indicating an impaired oxygen extraction in the treated animals during hypoxia. These results are consistent with those of Gutierrez and Andry (7). However, the presence of a significant difference in the red blood cell SR at the arteriolar end during nonhypoxic ventilation and the fact that this calculation neglects the contribution of changes in Po2 may affect this conclusion. However, we did observe a significantly lower end-capillary Po2 in the treated animals during both nonhypoxic and hypoxic ventilation. These data certainly do not exclude the possibility that a diffusional limitation does exist, since no measurements of tissue Paz were made coincident with these capillary data. However, they do lend some support for the idea also put forward by Gutierrez (6) that mixed venous Paz may not be a good indicator of end-capillary Po2 in hypoxia. We were concerned that the fall in systemic arterial BP that usually occurs after the introduction of hypoxia might be adversely affecting our results. In 15 of the 90 capillaries included in this study, the advent of hypoxia was not associated with a fall in arterial BP of >5 mmHg. When we compared the data from these capillaries with those from the remaining 75 capillaries, we found no significant differences in any of the systemic or microcirculatory parameters, other than the level of systemic arterial BP during hypoxia. There was a tendency for red blood cell SR to increase during hypoxia in those animals in which there was no fall in BP, but the change was not significant. However, as we have shown previously, there is no correlation between red blood cell SR and oxygen loss in the capillary network due to diffusional exchange (4). Therefore, it is not surprising that we saw no differences in SO, values in these two groups. Mian and Marshall (9) have reported the lack of a
Downloaded from www.physiology.org/journal/ajpheart at Midwestern Univ Lib (132.174.254.157) on February 12, 2019.
H522
OXYGEN
TRANSPORT
relationship between the hypoxia-induced fall in systemic arterial BP and arteriolar diameter in rats, a result that is consistent with our findings. One result of this study that we found intriguing and somewhat unexpected was the apparent impairment in the responsiveness of the peripheral chemoreceptors to the advent of hypoxic ventilation in the treated animals. In all cases, the levels of Pao, during hypoxia in the treated animals were significantly lower than values for the control animals, which should result in respiratory rates that are higher; however, the reverse was found to be true. In all instances, the respiratory rate was significantly lower in the treated animals. In addition, the fall in BP that generally occurs after the initiation of hypoxic ventilation occurred much more slowly in the treated animals, with the magnitude of the drop also attenuated. The explanation for these results is unclear but suggests some change in the sensitivity of the peripheral chemoreceptors possibly due to some fairly rapid adaptation to chronic hypoxia associated with the left shift of the dissociation curve (21). Hamsters, being fossorial animals, are commonly exposed to hypoxic and hypercapnic environments and, as a result, have become genetically adapted. They have been shown to have an impaired peripheral chemoreceptor response to hypoxia that may be related to these adaptations, which include an unexpectedly low PSO, high systemic hematocrit, and unexpectedly low levels of Pao, (11, 21). In summary, we found that, in hamsters, a left shift of the oxyhemoglobin dissociation curve resulting from the chronic short-term administration of sodium cyanate does not provide an advantage for resting striated muscle oxygen transport during hypoxic ventilation at an FIN, of 0.10. However, it is possible that, under more severe hypoxic conditions or in the presence of an increased metabolic rate, such an advantage might become apparent. We thank Steven Huang for technical assistance throughout the latter portions of this investigation. This study was supported by National Heart, Lung, and Blood Institute Grant HL-39226. Address for reprint requests: M. L. Ellsworth, Dept. of Pharmacological and Physiological Science, St. Louis University School of Medicine, 1402 S. Grand Boulevard, St. Louis, MO 63104. Received
18 January
1991; accepted
in final
form
13 September
1991.
REFERENCES 1. Barcroft, J., C. A. Binger, A. V. Bock, J. H. Goddart, H. S. Forbes, G. Harrop, J. C. Meakins, and A. C. Redfield. Observations upon the effect of high altitude on the physiological processes of the human body carried out in the Peruvian Andes chiefly at Cerro de Pasco. Philos. Trans. R. Sot. Lond. B Biol. Sci. 211: 325- 354,1922. 2. Ellis, C. G., M. L. Ellsworth, and R. N. Pittman. Determination of red blood cell oxygenation in vivo by dual video densitometric image analysis. Am. J. Physiol. 258 (Heart Circ. Physiol.
AND
OXYGEN
AFFINITY
27): Hl216-H1223, 1990. 3. Ellis, C. G., M. L. Ellsworth, and R. N. Pittman. Determination of red blood cell dynamics in capillaries using computer assisted image analysis (Abstract). FASEB J. 4: 1251, 1990. 4. Ellsworth, M. L., and R. N. Pittman. Arterioles supply oxygen to capillaries by diffusion as well as by convection. Am. J. Physiol. 258 (Heart Circ. Physiol. 27): Hl240-H1243, 1990. 5. Ellsworth, M. L., A. S. Popel, and R. N. Pittman. Assessment and impact of heterogeneities of convective oxygen transport parameters in capillaries of striated muscle: Experimental and theoretical. Microvasc. Res. 35: 341-362, 1988. 6. Gutierrez, G. The rate of oxygen release and its effect on capillary O2 tension: a mathematical analysis. Respir. Physiol. 63: 79-96, 1986. 7. Gutierrez, G., and J. M. Andry. Increased hemoglobin O2 affinity does not improve O2 consumption in hypoxemia. J. Appl. Physiol. 66: 837-843, 1989. 8. Hill, A. V. The possible effects of the aggregation of the molecules of haemoglobin on its dissociation curve. J. Physiol. Lond. 40: ivvii, 1910. R., and J. M. Marshall. Responses observed in individual 9. Mian, arterioles and venules of rat skeletal muscle during systemic hypoxia. J. Physiol. Lond. 436: 485-497, 1991. 10. Neville, J. R. Hemoglobin oxygen affinity measurement using biotonometry. J. AppZ. Physiol. 37: 967-971, 1974. 11. O’Brien, J. J., E. C. Lucey, and G. L. Snider. Arterial blood gases in normal hamsters at rest and during exercise. J. Appl. Physiol. 46: 806-810, 1979. 12. Pittman, R. N., and B. R. Duling. Effects of altered carbon dioxide tension on hemoglobin oxygenation in hamster cheek pouch microvessels. Microvasc. Res. 13: 211-224, 1977. 13. Pries, A. R., G. Kanzow, and P. Gaehtgens. Microphotometric determination of hematocrit in small vessels. Am. J. Physiol. 245 (Heart Circ. Physiol. 14): Hl67-H177, 1983. 14. Roth, A. C., and K. Wade. Effects of transmural transport in microcirculation: a two gas species model. Microvasc. Res. 32: 6483, 1986. 15. Schumacker, P. T., A. J. Suggett, P. D. Wagner, and J. B. West. Role of hemoglobin PsO in O2 transport during normoxic and hypoxic exercise in the dog. J. Appl. Physiol. 59: 749-757, 1985. 16. Sullivan, S. M., and R. N. Pittman. Hamster retractor muscle: a new preparation for intravital microscopy. Microvasc. Res. 23: 329-335,1982. 17. Teisseire, B. P., C. D. Soulard, R. A. Herigault, L. F. Leclerc, and 191. B. Laver. Effects of chronic changes in hemoglobin-O2 affinity in rats. J. Appl. Physiol. 46: 816-822, 1979. B. P., C. D. Soulard, L. J. Teisseire, R. A. Heri18. Teisseire, gault, and D. N. Laurent. Induced low PsO in anesthetized rats: blood gas, circulatory and metabolic adjustments. Respir. Physiol. 58: 335-344, 1984. 19. Turek, Z., F. Kreuzer, and B. E. M. Ringnalda. Blood gases at several levels of oxygenation in rats with a left-shifted dissociation curve. Pfluegers Arch. 376: 7-13, 1978. 20. Turek, Z., F. Kreuzer, M. Turek-Maischeider, and B. E. M. Ringnalda. Blood O2 content, cardiac output, and flow to organs at several levels of oxygenation in rats with a left-shifted blood oxygen dissociation curve. Pfluegers Arch. 376: 201-207, 1978. 21. Walker, B. R., E. M. Adams, and N. F. Voelkel. Ventilatory responses of hamsters and rats to hypoxia and hypercapnia. J. Appl. Physiol. 59: 1955-1960, 1985. 22. Walker, B. R., N. F. Voelkel, I. F. McMurtry, and E. M. Adams. Evidence for diminished sensitivity of the hamster pulmonary vasculature to hypoxia. J. Appl. Physiol. 52: 1571-1574, 1982. 23. Woodson, R. D., and S. Auerbach. Effect of increased oxygen affinity and anemia on cardiac output and its distribution. J. Appl. Physiol. 53: 1299-1306, 1982.
Downloaded from www.physiology.org/journal/ajpheart at Midwestern Univ Lib (132.174.254.157) on February 12, 2019.
capillaries;
striated
muscle; ham-
DELIVERY OF OXYGEN to the tissue requires the convective transport of oxygen to the site of exchange, the release of the oxygen from its carrier molecule (hemoglobin), and its subsequent diffusion down an oxygen tension (PO& gradient to the oxygen-consuming mitochondria. The magnitude of this Paz gradient depends on the amount of oxygen delivered to the site of exchange (determined by the level of blood flow and oxygen content), the level of tissue metabolism (oxygen consumption), and the affinity of the red blood cell hemoglobin for oxygen (i.e., the position of the oxyhemoglobin dissociation curve). Alterations in any or all of these parameters can impact on oxygen transport to cells. Animals native to high altitude have effectively adapted to the hypoxic environment, in part by having a left-shifted oxyhemoglobin dissociation curve. A leftshifted dissociation curve was originally speculated by Barcroft et al. (1) to be of advantage in hypoxic environments due to the increased onloading of oxygen in the pulmonary circulation. A number of studies have subse0363-6135/92 $2.00 Copyright
quently attempted to evaluate the efficacy of a leftshifted dissociation curve for oxygen delivery under various levels of inspired oxygen (FIN,) from measurements of oxygen levels in systemic arterial and mixed venous blood. Turek et al. (19, 20) developed a model of oxygen transport and provided supporting experimental evidence in studies of rats indicating that a left-shifted dissociation curve would only be advantageous under conditions of severe hypoxia (FIN, = 0.08 and 0.056). A similar result was obtained by Teisseire et al. (17, 18). They speculated that this beneficial effect was due to changes in characteristics of the oxyhemoglobin dissociation curve other than the Paz at which 50% of the hemoglobin is saturated with oxygen (P&. Schumacker et al. (15) determined that, during steady-state exercise in dogs, the PsOis not an important factor in determining tissue oxygen extraction during normoxia or mild hypoxia (FIN, = 0.12), but, during extreme hypoxia (FIN, = O.lO), a low PsOmay help maintain tissue Paz by enhancing systemic oxygen delivery. Gutierrez and Andry (7) evaluated the role of P50 in oxygen transport in a study of isolated rabbit hindlimbs perfused with human red blood cells treated to obtain perfusate having a P50of either 32.5 (low affinity) or 22.8 mmHg (high affinity). They determined that, although total oxygen transport was higher in the high-affinity group during hypoxia, venous Paz values were the same in the two groups. Therefore, oxygen extraction was actually lower in the high-affinity group. They attributed this finding to a diffusional limitation to oxygen transport in the tissue in hypoxia due to the kinetics of oxygen off-loading from the red blood cells (6, 7). In all of the studies presented above, the approach has been to evaluate tissue oxygen transport at the whole tissue level and extrapolate the results to the level of the microcirculation. The purpose of this study was to evaluate oxygen transport and the efficacy of a left-shifted dissociation curve at the level of the microcirculation where tissue oxygen transport occurs. We determined oxygen saturation (SOL), red blood cell velocity, lineal density (LD), and red blood cell supply rate (SR) using computer-aided video techniques (2) in both arteriolar and venular capillaries of the cheek pouch retractor muscle. These hemodynamic and oxygenation parameters were evaluated under nonhypoxic (FIN, = 0.3 and 0.21) and hypoxic (FIN, = 0.10) conditions in hamsters with normal PsOand in hamsters with a decreased P50 induced by chronic short-term administration of sodium cyanate. By systematically measuring capillary oxygen transport parameters under conditions of normal and reduced P50, we hope to elucidate the role of oxygen affinity in oxygen transport to the tissue under nonhy-
0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart at Midwestern Univ Lib (132.174.254.157) on February 12, 2019.
H517
H518
OXYGEN
TRANSPORT
poxic and hypoxic conditions. This information should be useful in the formation of a comprehensive model of oxygen transport in the microcirculation. METHODS
Male golden hamsters (48 t 16 days, 98.1 t 21.4 g, n = 19) were separated into two groups, control (n = 12) and treated (n = 7). The treated group received sodium cyanate (0.2 mg/kg ip) five times per week for -2 wk to decrease PsO. Hamsters were initially anesthetized with pentobarbital sodium (6.5 mg/ 100 g body wt ip). The trachea was cannulated to insure a patent airway, and the hamster was allowed to breath spontaneously. The length of the tracheal cannula approximated the distance from the point of insertion into the trachea to the tip of the hamster’s nose to maintain a normal anatomical dead space. A femoral artery was cannulated with PE-10 tubing to enable continuous monitoring of systemic arterial blood pressure (BP) and collection of blood samples for systemic arterial blood gas, pH, hematocrit, and PsO determinations. A femoral vein was likewise cannulated to enable continuous infusion of supplemental anesthetic (pentobarbital sodium, 0.15 mg/min). The right cheek pouch retractor muscle was surgically exteriorized, as described by Sullivan and Pittman (16). Briefly, the muscle was separated from underlying back muscles, and a hemoclip (Edward Week) was used to secure two ligatures to the muscle. The muscle was then cut at its spinal end and placed, ventral side up, at its in situ dimensions on a specially designed Plexiglas platform. The muscle was covered with Saran (Dow Corning) to prevent both desiccation of the tissue and gas exchange with the atmosphere. Deep esophageal and muscle temperatures were monitored and maintained at 37 t 1°C by separate heat exchangers within the animal platform. The muscle was transilluminated with a xenon lamp (75 W) with a stabilized power supply to provide constant light intensity. Observations were made using a Zeiss ACM microscope equipped with a long-working distance objective (UD40/0.65). The objective was used dry to give an equivalent magnification and numerical aperture of x 25.8 and 0.41, respectively. The microscope image was viewed using a dual-camera high-resolution closed-circuit video system consisting of two SIT cameras (model 66; Dage MTI), video monitors (WV-5410; Panasonic), S-VHS video cassette recorders (AG-7500; Panasonic), time-date generators (WJ-810; Panasonic), a video analyzer (model 321; Colorado Video), video field counter (Dept. of Medical Biophysics, University of Western Ontario), and microcomputer (Compaq 386/25) with a 12bit analog-to-digital converter (model RTI-815-A; Analog Devices). The video analyzer superimposed a vertical cursor line on the video image that provided light intensity information (slow scan video output) for the computer analysis of red blood cell So2, velocity, and LD. The gain on both the camera and video cassette recorder was adjusted manually. The output of the video system was linear over the range of light intensities used. Optical parfocalizing devices positioned in front of each video camera were used to ensure that the two images were in equal focus. The vertical and horizontal image sizes and position were adjusted within each camera to make the two images produced by the cameras superimposable. Because the two tapes have to be analyzed separately, an audio signal generated by the video field counter was placed simultaneously on the two tapes at 30s intervals. Individual capillaries, defined as vessels containing single-file red blood cell flow and having a diameter less than that of a red blood cell, were located and designated as arteriolar or venular according to their anatomical relationship with their feed arteriole or collecting venule, respectively. It should be pointed out here that only those portions of the capillaries that
AND OXYGEN
AFFINITY
were within 130 t 25 pm from their feed arteriole or collecting venule were utilized. In addition, the arteriolar and venular capillaries are not necessarily anatomically related. Scenes of capillary blood flow were initially recorded while the hamster breathed spontaneously either room air [21% 02; normoxia (N)] or a supplemental oxygen gas mixture [30% 02, balance N,; supplemental (S)]. After 5 min of nonhypoxic baseline information was obtained, the hamsters’ inspired gas supply was switched to a hypoxic gas mixture [IO% 02, balance N2; hypoxia (H)]. The same anatomical segment of capillary was recorded for an additional 5 min; therefore, for each located capillary, we obtained 5 min of baseline nonhypoxia information (S or N) followed immediately by 5 min of systemic hypoxia data [H(S) or H(N), respectively]. At the end of the 5-min hypoxic challenge, the hamster was switched back to the prehypoxic gas, either room air or 30% Oz. Blood samples (2-6) were obtained over the course of the experiment during both the baseline and hypoxic periods for determination of systemic arterial pH, POT, Pco~, P 50, and hematocrit. Spontaneous respiratory rate was noted during both hypoxic and prehypoxic conditions. After the hypoxic interval was completed and the hamster was switched back to the prehypoxic gas mixture, sufficient time was allowed for the hemodynamic parameters to return to baseline levels. At this point, another capillary was located, classified as arteriolar or venular, and the protocol was repeated. This protocol was followed for both control and treated hamster groups. All data recorded were subsequently analyzed off-line. So2 was determined using a modified version of the computer-assisted image analysis technique described previously by Ellis et al. (2). This method enables the direct evaluation of So2 of individual red blood cells along a given length of capillary (100 TV lines = 93 pm) for 60 s. In short, digitized light intensities were obtained one time per frame at both wavelengths (see below). The digitized light intensities were redisplayed as a space-time image for both wavelengths simultaneously. The light intensity values corresponding to the amount of light transmitted through the capillary in the presence (I) and absence (I,) of a red blood cell were determined. These values were used to compute an optical density (OD) at each point along the capillary segment, according to the equation OD = loglo(I,/[F(I - I, + 1,/F)]) where F is the glare correction factor described by Pries et al. (13) for both an isosbestic wavelength (420 nm) and a wavelength at which there is a maximum difference between oxy- and deoxyhemoglobin (431 nm). The ratio of the OD values determined from identical time segments at the two wavelengths (OD431/0D420) is linearly related to So2. The local longitudinal So2 gradient (ASo,/Az, where AZ is the distance separating 2 sites) was determined from coincident measurements of So2 at upstream and downstream sites on individual capillaries (ASo,) divided by AZ. Red blood cell SR (cells/s) was determined for each capillary by counting the number of cells crossing a horizontal line on the computergenerated space-time image display divided by the length of time (in s) of that interval. Red blood cell velocity (v; pm/s) and LD (cells/mm) were determined from the light intensity data, as described previously. Supply-rate weighted So2 values (Sf,,) were determined from coincident measurements of So2 and red blood cell SR from arteriolar and venular capillaries as Sfo, = z(SiO, X fi)/zfi, where SiO, and fi are the inlet or outlet So2 values and red blood cell SR values, respectively, in individual capillaries. Capillary diameters were measured directly from the video monitor during both normoxia and hypoxia using a vernier caliper and video-taped micrometer scale. Pso was determined in each animal using the biotonometric method of Neville (10) and corrected to standard conditions
Downloaded from www.physiology.org/journal/ajpheart at Midwestern Univ Lib (132.174.254.157) on February 12, 2019.
OXYGEN
(PH = 7.4, pcoz = 40 mmHg) p50 =
TRANSPORT
using the relationship 0.41(pH-7.40)-0.0610g(40/Pco2)
(PSO)std
x
lo-
Capillary POT values were estimated from the measured capillary So2 values using the appropriate values of standard Pso and Hill’s n (8; unpublished data). The values were subsequently corrected for the appropriate values of PCO~ and pH in the capillaries that we have estimated, based on previous studies (12, 14), to be best approximated by the venous PCO~ (unpublished data) and arterial pH. Data analysis. Data are presented as means t SD. An analysis of variance followed by either a Mann-Whitney U (unpaired data) or Wilcoxon (paired) test was used to evaluate the differences between the different systemic and microcirculatory parameters for the control and treated animals. An unpaired Student’s t test was used to compare the current data with previously published data. All data analysis was done using a statistical software package (Systat, Evanston IL). Significance was assigned at P < 0.05, unless otherwise stated. RESULTS
Systemic effects. The administration of sodium cyanate resulted in a significant decrease in standard P50 from 26.1 t 1.0 to 16.2 t 1.6 mmHg with no significant change in systemic hematocrit (Table 1). While breathing room air, the treated animals tended to have significantly lower systemic arterial BP values (66 t 14 mmHg) than the control animals (85 t 18 mmHg). Because we were concerned that the inclusion of data from animals with low arterial BP values (~70 mmHg) during nonhypoxic conditions might interject an additional complicating factor, we did not include these in the data to be presented here. This criterion precluded the inclusion of a large number of capillaries in the treated animals, especially those whose nonhypoxic gas was room air and a few capillaries from each of the other groups. Systemic arterial blood gases (Pao, and Pace,), arterial pH (pH,), BP, and respiratory rates for the two groups of animals for the different levels of inspired gas are presented in Table 2. In general, the control and treated animals had similar levels of blood gases and pH with the exception that the Pao, of the treated animals during normoxia and hypoxia were significantly lower than those of the control animals. Systemic arterial BP values were not different in the two groups under any conditions; however, the respiratory rates of the treated animals were significantly lower than their corresponding controls. In response to hypoxia, Pao, and Pace, decreased significantly while pH, increased in all but one control group [H(N)] . Arterial BP usually decreased significantly after the induction of hypoxia, whereas respiratory rate Table 1. General information Control (n = 12)
Age, days Weight, g H&s, % PSO, mmHg
49.8t14.7
105.6t23.8 49.71k4.2 26.ltl.O
Treated (n = 7) 45.1*18.9
&5.3-r-5.4* 50.5t4.2 16.2tl.6”
Values are means t SD; n, no. of animals. He&, systemic crit; PsO, 0, tension at which hemoglobin is half saturated * Significantly different from control values at P < 0.05.
hematowith Oz.
AND
OXYGEN
AFFINITY
H519
increased. The decrease in BP was significantly greater in the control animals than in the treated for the transition from supplemental to hypoxia but not for normoxia to hypoxia. The decrease in BP was more rapid in the control animals than in the treated animals. The increase in respiratory rate was significantly greater in the control animals than in the treated in both cases. It has been reported by Walker et al. (21) that hamsters increase frequency and not tidal volume in response to hypoxia; therefore, these changes in respiratory frequency directly reflect changes in total ventilation. Microvascular effects. The level of So2 at the upstream end of the microvasculature (arteriolar, upstream) was not different in those animals breathing room air (N) vs. those breathing 30% O2 (S) for both the control and treated animals. Similar results were found for the downstream end of the network. Switching the level of inspired gas to 10% O2 caused significant decreases in the levels of So2 at both the arteriolar and venular ends of the network, with the hypoxic values not dependent on the level of prehypoxic gas. Because there was no dependence of the level of capillary So2 on the level of nonhypoxic inspired gas, we combined these data into nonhypoxic (supplemental + normoxic) and hypoxic [H(S) + H(N)] groups (Table 3, Fig. 1). In control animals, the arteriolar capillary (upstream) So2 was 59.6 t 10.6% compared with 70.7 t 10.6% in the treated animals. With hypoxic ventilation, these levels decreased significantly to 39.4 t 16.0 and 50.6 t 13.4%, respectively. At the downstream venular capillary, So2 was 46.3 t 14.0% in control animals and 59.7 t 11.0% in treated animals during nonhypoxic ventilation and 33.0 t 12.0 and 41.9 t 9.5% during hypoxic ventilation, respectively. In all cases, th.e levels of So2 in the treated animals were significantly higher than in the controls; however, the levels of PO, were significantly lower in the treated than in the control animals (Table 3, Fig. 1). In control animals during nonhypoxic ventilation, the arteriolar capillary (upstream) Paz was 33.1 t 6.2 mmHg compared with 25.0 t 4.8 mmHg in treated animals. With hypoxic ventilation, these levels decreased significantly to 21.9 t 6+3 and 17.3 t 3.5 mmHg, respectively. At the downstream venular capillary end, PoB was 27.1 t 6.6 mmHg in control animals and 20.7 ~fi3.6 mmHg in treated animals during nonhypoxic ventilation and 18.9 t 5.0 and 14-7 & 2.6 mmHg during hypoxic ventilation, respectively. In all cases there was a significant difference in POT across the network (upstream arteriolar vs. downstream venular). ASOJAZ during nonhypoxic ventilation for the control animals for the arteriolar and venular capillaries combined was 0.067 t 0.352%/pm compared with 0.088 t O.Z53%/pm for the treated animals. During hypoxia, the local longitudinal gradients were unchanged at 0.071 & 0.292 and 0.095 t 0.255%/pm, respectively. Hemodynamically, there were few differences in any of the parameters, as shown in Table 3. Red blood cell SR was significantly lower in the treated animals at the arteriolar end during nonhypoxic ventilation compared with control animals, due almost entirely to a significantly lower LD.
Downloaded from www.physiology.org/journal/ajpheart at Midwestern Univ Lib (132.174.254.157) on February 12, 2019.
H520 Table
OXYGEN
TRANSPORT
AND
OXYGEN
AFFINITY
2. Systemic parameters Pa0,,
Pace,,
mmHg
Control S N H(S) H(N) Treated S N H(S) H(N)
mmHg
PHa
Systemic BP, mmHg
79.2t18.3 58.2t8.5 27.9t5.7* 32.3t5.9*
42.0k9.6
34.3t5.1 29.4t4.2* 27.7-+2.3*
7.321t0.035 7.387t0.051 7.417*0.042* 7.428t0.038
86.8k17.9 85.4t13.4 61.8t18.6* 59.7*16.9*
70.7t9.2
39.5t4.5 36.0t9.3 31.8t2.5" 30.5t4.7*
7.327t0.045 7.365t0.078 7.418zko.o45* 7.434&0.105*
81.9t7.5 84.4t9.4 67.2&13.3* 63.8&11.0*
46.lk9.O.f 25.5*3.4* 24.5*5.2*"r
Respiration, breaths/min
73.8d8.5 59.3t12.9 129.9tl2.5* 103.8&20.0* 54.5+11.7$ 50.2k9.0 87.0+23.1*$ 70.5+7.4*$
Values are means t SD; supplemental gas oxygen mixture (S) = 0.3 inspired oxygen (FIN,); normoxia (N) = 0.21 FIN,; hypoxia supplemental gas oxygen mixture [H(S)] and hypoxia normoxia [H(N)] = 0.1 FIN, associated with S and N groups, respectively. For control animals, n for arterial blood gases were S and H(S) = 11 and N and H(N) = 8; for systemic blood pressure and respiratory rate, S and H(S) = 32 and N and H(N) = 26. For treated animals, n for arterial blood gases were S and H(S) = 7 and N and H(N) = 7; for systemic blood pressure and respiratory rate, S and H(S) = 27 and N and H(N) = 8. * Significantly different from corresponding nonhypoxic values at P < 0.05. t Significantly different from control values at P < 0.05. $ Significantly different from control values at P < 0.01.
Table 3. Oxygen transport parameters in arteriolar and venular capillaries Control
Treated
Nonhypoxia
Arteriolar v, Pm/s LD, cells/mm SR, cells/s so,, % Po2, mmHg Sfo,, 5% ASo,/Az, %/pm Venular v, w/s LD, cells/mm SR, cells/s SO& % Po2, mmHg Sf,,, 5% ASo,/Az, %/pm
119.6t58.7 57.2519.1 7.0t4.6 59.6klO.6 33.1k6.2 59.3 0.142~10.334 102.6k56.9 62.6kl6.2 6.723.0 46.3214.0 27.1t6.6 50.8 - .0.043t0.356
Hypoxia
Nonhypoxia
Hypoxia
122.9t67.6 58.8t23.7 7.6rt4.6 39.4+16.Ot 21.9+6.3-t 42.1 - .0.002stO.306
105.6k56.8 45.6tl7.3" 4.4&1.5* 70.7+10.6$ 25.0*4.8$ 70.0 0.063t0.229
116.9k67.3 52.2k20.8 6.4t4.2 50.6+13.4*-f 17.3+3.5'f$ 48.6 0.055t0.269
102.9k65.3 63.8218.6 6.7t3.8 30.3+12.0t 18.9+5.O"f 33.6 0.158t0.253
122.7261.0 60.3t22.9 6.9t3.0 59.7+11.0$ 20.7k3.61 60.5 0.114t0.282
105.lk68.0 67.Ok22.0 7.723.8 41.9+9.5t$ 14.7+2.6"f$ 43.8 0.137t0.240
Values are means t SD; v, velocity; LD, lineal density; SR, supply rate; Soz, oxygen saturation; Po2, oxygen tension; Sf,,, supply-rate weighted ASO, values; AZ, distance separating 2 sites; A, change. For arteriolar control group, n = 30 animals each for nonhypoxia and hypoxia, and for arteriolar treated, n = 18 each for nonhypoxia and hypoxia groups. For venular control, n = 25 animals in each group, and for venular treated, n = 17 each. * Significantly different from control at P c 0.05. t Significantly different from nonhypoxic values at P < 0.05. $ Significantly different from control at P < 0.01. DISCUSSION
The data obtained in this study suggest that, in resting striated muscle, an increase in hemoglobin oxygen affinity (decreased P& does not affect the amount of oxygen lost across the capillary network. We found that, in capillaries of the hamster retractor muscle, the change in So2 across the network during nonhypoxic ventilation was 13.3% So2 for control animals and 11.0% SO:! for treated animals. This value of 13.3% So2 for the control animals is slightly smaller but not significantly different from the value obtained previously for capillaries of the hamster retractor muscle (5). During hypoxic ventilation, the overall changes in So2 were somewhat smaller for both control and treated animals (9.1 and 8.7% So2, respectively) compared with their nonhypoxic values. Because, during hypoxia, the red blood cell SR values were not different in the two groups, this would imply that oxygen extraction during hypoxia was not significantly different in the two groups, even though the
computation of the oxygen extraction ratio would suggest otherwise (see below). In both normoxia and hypoxia, the end-capillary Paz values were significantly lower in treated animals, which might be considered detrimental. Therefore, it seemsclear that, based on the information available in this study, a left shift of the oxyhemoglobin dissociation curve does not provide any significant benefit for capillary oxygen exchange in resting striated muscle at this level of hypoxia. Turek and his colleagues (19, 20) came to a different conclusion in their study of rats. There are a number of differences between their study and ours that might help explain the discrepancy. First, there are differences between the oxyhemoglobin dissociation curves for rats and hamsters. Their control rat oxyhemoglobin dissociation curve had a PsOof 35.6 mmHg compared with the hamster’s of 26.1 mmHg. In their treated animals, the P50was decreased to 22.3 mmHg, whereas ours was 16.2 mmHg. Second, they used levels of FIN, that were some-
Downloaded from www.physiology.org/journal/ajpheart at Midwestern Univ Lib (132.174.254.157) on February 12, 2019.
OXYGEN
80 E
TRANSPORT
AND
Treated
70
d
3 60 cd 25 50
Control
0
10
20 30 Oxygen Tension
40 (mmHg)
50
60
Fig. 1. Estimated oxygen tension vs. oxygen saturation at upstream arteriolar capillary (a) and downstream venular capillary (u) sites in control and treated animals during nonhypoxic (open symbols) and hypoxic (closed symbols) ventilation and corresponding estimated oxyhemoglobin dissociation curves. Error bars indicate 1 SD.
what lower (FIEF = 0.08 and 0.056) than that used in the present study (FIN, = 0.10). However, the levels of sys-
temic arterial Paz in their control animals at both 0.08 or 0.056 FIN, were not different from ours at 0.10 FIN,. In the treated animals, our level of systemic Pao, was intermediate between their group that inspired 8.0% O2 and their group that inspired 5.6% 02. Thus it is unlikely that the level of FIN, is responsible for the difference in the results. Third, and possibly most importantly, their study evaluated systemic blood gas values obtained from the carotid artery and right atrium, which assessed oxygen exchange in the whole animal comprised of tissues having both a high and low metabolic rate, whereas we evaluated oxygen transport at the capi llary-tissue interface in a tissue with a low metabolic rate. Previous investigators have noted that, after a left shift of the oxyhemoglobin dissociation curve, blood flow shifts away from tissues with a low metabolic demand toward tissues with a high metabolic demand (23). However, as noted above, the capillary red blood cell SR is lower at the arteriolar end of the network in the treated animals during nonhypoxic ventilation but otherwise is not significantly different in the two groups of animals (control and treated). Therefore, it seems unlikely that there is a significant diversion of blood away from the retractor muscle. We have seen prev iously (5) that, when one compares th .e locally computed value for the longitudinal So2 gradient (local ASoz/Az = 0.113 t O.l96%/pm) with the average longitudinal gradient obtained by dividing the difference in mean SOBdetermined at the upstream ends of arteriolar capillaries and the downstream ends of venular capillaries (60%39.9%) by the average anatomical capillary length (412 pm), the latter quantity was smaller than the local ASOJAZ by a factor of two. We were interested to know if a similar comparison of the
OXYGEN
AFFINITY
HZ1
present data would result in a similar discrepancy. In the control animals during nonhypoxic ventilation, the two values differed by a factor of two, which is consistent with the previous result; however, for the treated animals, the average gradient was smaller than the local gradient by a factor of three, which suggests an even greater degree of diffusional interaction occurs among capillaries in the treated animals. In hypoxia, the two gradients differed by a factor of 3 for the control animals and by a factor of 4.5 for the treated animals. Again, this degree of discrepancy would imply increased diffusional oxygen exchange in hypoxia most likely due to the presence of local areas of hypoxia and consequent increased local POT gradients. Gutierrez and Andry (7) have postulated the presence of a diffusional limitation to oxygen transport at the venular end of the capillary network in hypoxia based on data from isolated perfused rabbit hindlimbs. They based this on their findings of a lower oxygen extraction but unchanged venous Po2 values in the presence of the high-affinity perfusate compared with the low-affinity perfusate. In the present study, we found that the change in SO, across the network during both normoxia and hypoxia were not significantly different; however, if one computes the oxygen extraction ratio { [ (Sfo,), (Sf&]/(SfoJ,) for the control and treated animals, one obtains values of 14 and 13.6%, respectively, during nonhypoxic ventilation and 20.2 and 9.9%, respectively, during hypoxic ventilation, indicating an impaired oxygen extraction in the treated animals during hypoxia. These results are consistent with those of Gutierrez and Andry (7). However, the presence of a significant difference in the red blood cell SR at the arteriolar end during nonhypoxic ventilation and the fact that this calculation neglects the contribution of changes in Po2 may affect this conclusion. However, we did observe a significantly lower end-capillary Po2 in the treated animals during both nonhypoxic and hypoxic ventilation. These data certainly do not exclude the possibility that a diffusional limitation does exist, since no measurements of tissue Paz were made coincident with these capillary data. However, they do lend some support for the idea also put forward by Gutierrez (6) that mixed venous Paz may not be a good indicator of end-capillary Po2 in hypoxia. We were concerned that the fall in systemic arterial BP that usually occurs after the introduction of hypoxia might be adversely affecting our results. In 15 of the 90 capillaries included in this study, the advent of hypoxia was not associated with a fall in arterial BP of >5 mmHg. When we compared the data from these capillaries with those from the remaining 75 capillaries, we found no significant differences in any of the systemic or microcirculatory parameters, other than the level of systemic arterial BP during hypoxia. There was a tendency for red blood cell SR to increase during hypoxia in those animals in which there was no fall in BP, but the change was not significant. However, as we have shown previously, there is no correlation between red blood cell SR and oxygen loss in the capillary network due to diffusional exchange (4). Therefore, it is not surprising that we saw no differences in SO, values in these two groups. Mian and Marshall (9) have reported the lack of a
Downloaded from www.physiology.org/journal/ajpheart at Midwestern Univ Lib (132.174.254.157) on February 12, 2019.
H522
OXYGEN
TRANSPORT
relationship between the hypoxia-induced fall in systemic arterial BP and arteriolar diameter in rats, a result that is consistent with our findings. One result of this study that we found intriguing and somewhat unexpected was the apparent impairment in the responsiveness of the peripheral chemoreceptors to the advent of hypoxic ventilation in the treated animals. In all cases, the levels of Pao, during hypoxia in the treated animals were significantly lower than values for the control animals, which should result in respiratory rates that are higher; however, the reverse was found to be true. In all instances, the respiratory rate was significantly lower in the treated animals. In addition, the fall in BP that generally occurs after the initiation of hypoxic ventilation occurred much more slowly in the treated animals, with the magnitude of the drop also attenuated. The explanation for these results is unclear but suggests some change in the sensitivity of the peripheral chemoreceptors possibly due to some fairly rapid adaptation to chronic hypoxia associated with the left shift of the dissociation curve (21). Hamsters, being fossorial animals, are commonly exposed to hypoxic and hypercapnic environments and, as a result, have become genetically adapted. They have been shown to have an impaired peripheral chemoreceptor response to hypoxia that may be related to these adaptations, which include an unexpectedly low PSO, high systemic hematocrit, and unexpectedly low levels of Pao, (11, 21). In summary, we found that, in hamsters, a left shift of the oxyhemoglobin dissociation curve resulting from the chronic short-term administration of sodium cyanate does not provide an advantage for resting striated muscle oxygen transport during hypoxic ventilation at an FIN, of 0.10. However, it is possible that, under more severe hypoxic conditions or in the presence of an increased metabolic rate, such an advantage might become apparent. We thank Steven Huang for technical assistance throughout the latter portions of this investigation. This study was supported by National Heart, Lung, and Blood Institute Grant HL-39226. Address for reprint requests: M. L. Ellsworth, Dept. of Pharmacological and Physiological Science, St. Louis University School of Medicine, 1402 S. Grand Boulevard, St. Louis, MO 63104. Received
18 January
1991; accepted
in final
form
13 September
1991.
REFERENCES 1. Barcroft, J., C. A. Binger, A. V. Bock, J. H. Goddart, H. S. Forbes, G. Harrop, J. C. Meakins, and A. C. Redfield. Observations upon the effect of high altitude on the physiological processes of the human body carried out in the Peruvian Andes chiefly at Cerro de Pasco. Philos. Trans. R. Sot. Lond. B Biol. Sci. 211: 325- 354,1922. 2. Ellis, C. G., M. L. Ellsworth, and R. N. Pittman. Determination of red blood cell oxygenation in vivo by dual video densitometric image analysis. Am. J. Physiol. 258 (Heart Circ. Physiol.
AND
OXYGEN
AFFINITY
27): Hl216-H1223, 1990. 3. Ellis, C. G., M. L. Ellsworth, and R. N. Pittman. Determination of red blood cell dynamics in capillaries using computer assisted image analysis (Abstract). FASEB J. 4: 1251, 1990. 4. Ellsworth, M. L., and R. N. Pittman. Arterioles supply oxygen to capillaries by diffusion as well as by convection. Am. J. Physiol. 258 (Heart Circ. Physiol. 27): Hl240-H1243, 1990. 5. Ellsworth, M. L., A. S. Popel, and R. N. Pittman. Assessment and impact of heterogeneities of convective oxygen transport parameters in capillaries of striated muscle: Experimental and theoretical. Microvasc. Res. 35: 341-362, 1988. 6. Gutierrez, G. The rate of oxygen release and its effect on capillary O2 tension: a mathematical analysis. Respir. Physiol. 63: 79-96, 1986. 7. Gutierrez, G., and J. M. Andry. Increased hemoglobin O2 affinity does not improve O2 consumption in hypoxemia. J. Appl. Physiol. 66: 837-843, 1989. 8. Hill, A. V. The possible effects of the aggregation of the molecules of haemoglobin on its dissociation curve. J. Physiol. Lond. 40: ivvii, 1910. R., and J. M. Marshall. Responses observed in individual 9. Mian, arterioles and venules of rat skeletal muscle during systemic hypoxia. J. Physiol. Lond. 436: 485-497, 1991. 10. Neville, J. R. Hemoglobin oxygen affinity measurement using biotonometry. J. AppZ. Physiol. 37: 967-971, 1974. 11. O’Brien, J. J., E. C. Lucey, and G. L. Snider. Arterial blood gases in normal hamsters at rest and during exercise. J. Appl. Physiol. 46: 806-810, 1979. 12. Pittman, R. N., and B. R. Duling. Effects of altered carbon dioxide tension on hemoglobin oxygenation in hamster cheek pouch microvessels. Microvasc. Res. 13: 211-224, 1977. 13. Pries, A. R., G. Kanzow, and P. Gaehtgens. Microphotometric determination of hematocrit in small vessels. Am. J. Physiol. 245 (Heart Circ. Physiol. 14): Hl67-H177, 1983. 14. Roth, A. C., and K. Wade. Effects of transmural transport in microcirculation: a two gas species model. Microvasc. Res. 32: 6483, 1986. 15. Schumacker, P. T., A. J. Suggett, P. D. Wagner, and J. B. West. Role of hemoglobin PsO in O2 transport during normoxic and hypoxic exercise in the dog. J. Appl. Physiol. 59: 749-757, 1985. 16. Sullivan, S. M., and R. N. Pittman. Hamster retractor muscle: a new preparation for intravital microscopy. Microvasc. Res. 23: 329-335,1982. 17. Teisseire, B. P., C. D. Soulard, R. A. Herigault, L. F. Leclerc, and 191. B. Laver. Effects of chronic changes in hemoglobin-O2 affinity in rats. J. Appl. Physiol. 46: 816-822, 1979. B. P., C. D. Soulard, L. J. Teisseire, R. A. Heri18. Teisseire, gault, and D. N. Laurent. Induced low PsO in anesthetized rats: blood gas, circulatory and metabolic adjustments. Respir. Physiol. 58: 335-344, 1984. 19. Turek, Z., F. Kreuzer, and B. E. M. Ringnalda. Blood gases at several levels of oxygenation in rats with a left-shifted dissociation curve. Pfluegers Arch. 376: 7-13, 1978. 20. Turek, Z., F. Kreuzer, M. Turek-Maischeider, and B. E. M. Ringnalda. Blood O2 content, cardiac output, and flow to organs at several levels of oxygenation in rats with a left-shifted blood oxygen dissociation curve. Pfluegers Arch. 376: 201-207, 1978. 21. Walker, B. R., E. M. Adams, and N. F. Voelkel. Ventilatory responses of hamsters and rats to hypoxia and hypercapnia. J. Appl. Physiol. 59: 1955-1960, 1985. 22. Walker, B. R., N. F. Voelkel, I. F. McMurtry, and E. M. Adams. Evidence for diminished sensitivity of the hamster pulmonary vasculature to hypoxia. J. Appl. Physiol. 52: 1571-1574, 1982. 23. Woodson, R. D., and S. Auerbach. Effect of increased oxygen affinity and anemia on cardiac output and its distribution. J. Appl. Physiol. 53: 1299-1306, 1982.
Downloaded from www.physiology.org/journal/ajpheart at Midwestern Univ Lib (132.174.254.157) on February 12, 2019.
