Heart and Circulatory Physiology

Prolonged tissue Po2 reduction after contraction in spinotrapezius muscle of spontaneously hypertensive rats

Lane M. Smith, R. Wayne Barbee, Kevin R. Ward, Roland N. Pittman


We tested the hypothesis that a deficit in oxygen extraction or an increase in oxygen demand after skeletal muscle contraction leads to delayed recovery of tissue oxygen tension (Po2) in the skeletal muscle of hypertensive rats compared with normotensive rats. Blood flow and Po2 recovery at various sites in the spinotrapezius muscle of spontaneously hypertensive rats (SHRs) were evaluated after a 3-min period of muscle contraction and were compared with corresponding values in Wistar-Kyoto rats (WKYs). The recovery of tissue Po2 [75 ± 7 (SHRs) vs. 99 ± 12% (WKYs) of resting values] and venular Po2 [72 ± 13 (SHRs) vs. 104 ± 10% (WKYs) of resting values] were significantly depressed in the SHRs 30 s postcontraction. The delayed recovery persisted for 120 s postcontraction for both tissue [86 ± 11 (SHRs) vs. 119 ± 13% (WKYs) of resting values] and venular [74 ± 2 (SHRs) vs. 100 ± 9% (WKYs) of resting values] Po2 levels. There was no significant difference in the recovery of arteriolar Po2 between the two groups 30 s postcontraction [95 ± 7 (SHRs) vs. 84 ± 8% (WKYs) of resting values]. Values for resting diameter of arcade arterioles in the two groups were not different [52 ± 3 (SHRs) vs. 51 ± 3 μm (WKYs)], but the arteriolar diameter after the 3-min contraction period was greater in the SHRs (71 ± 4 μm) than the WKYs (66 ± 4). Likewise, red blood cell (RBC) velocity [5.8 ± 0.3 (SHRs) vs. 4.7 ± 0.2 mm/s (WKYs)] and blood flow [23.0 ± 0.8 (SHRs) vs. 16.0 ± 1.0 nl/s (WKYs)] measurements were significantly greater in the SHRs at 30 s postcontraction. The delayed recovery of tissue Po2 in the SHRs compared with the WKYs can be explained by a decrease in oxygen diffusion from the rarefied microvascular network due to the increased RBC velocity and the shorter residence time in the microcirculation and the consequent disequilibrium for oxygen between plasma and RBCs. The delayed recovery of venular Po2 in the SHRs is consistent with this explanation, as venular Po2 is slowly restored to baseline by release of oxygen from the RBCs. This leaves the arterioles in the primary role as oxygen suppliers to restore Po2 in the tissue after muscle contraction.

  • hypertension
  • oxygen tension
  • hypoxia
  • microvascular blood flow
  • rarefaction
  • phosphorescence quenching microscopy
  • disequilibrium

the arteriolar dilation that occurs in skeletal muscle in response to contraction is a well-characterized aspect of the active hyperemia that increases oxygen delivery to this tissue (5, 12). However, previous studies have questioned the ability of active hyperemia to compensate for tissue hypoxia in hypertensive individuals immediately after exercise (3). A potential cause of such supply-vs.-demand imbalance of oxygen is the microvascular rarefaction associated with hypertension, which is defined as a reduction of the arteriolar network density in a variety of vascular beds including rat spinotrapezius muscle (4, 7, 19). The significance of this loss of arterioles becomes apparent because oxygen exchange is not limited to the capillaries but occurs at all levels of the microcirculation (7, 16, 17). Moreover, Lash and Bohlen (10, 11) have shown that postexercise vascular reactivity and blood flow at various levels of the arteriolar network in spontaneously hypertensive rats (SHRs) is greater than that observed in Wistar-Kyoto rats (WKYs). Therefore, evidence points to a deficit in oxygen extraction or an increase in oxygen demand in SHRs relative to WKY populations to account for the depressed recovery of tissue oxygen tension (Po2) after contraction of skeletal muscle.

The purpose of this study was to measure changes in microvascular blood flow and Po2 in the spinotrapezius muscle of SHRs and WKYs after a 3-min contraction period to test whether significant differences existed between the responses of SHRs and WKYs in these two variables. Phosphorescence quenching microscopy (PQM) was used to determine microvascular and tissue Po2 values (18, 2224), whereas fluorescent labeling of red blood cells (RBCs) was used to determine RBC velocity and arteriolar blood flow rates (13, 15, 20). We propose that alterations in the ability of SHRs to extract oxygen from microvessels due to structural changes in the microvasculature (i.e., rarefaction) relative to the WKYs lead to depressed recovery in tissue Po2 after muscle contraction.


Surgical Preparation

Male SHR (n = 13) and WKY (n = 15) controls (Taconic Farms; Germantown, NY) weighing 240–280 g were selected for study. One group from each strain (SHR, n = 4; WKY, n = 5) was used to measure tissue Po2 in the spinotrapezius muscle before and after contraction (22). A second group (SHR, n = 4; WKY, n = 5) was used to assess recovery of arteriolar and venular Po2, and a third group (SHR, n = 5; WKY, n = 5) was used to determine RBC velocity and arteriolar blood flow rates. All animals in the study were initially anesthetized with pentobarbital sodium (65 mg/kg ip). Additional doses of pentobarbital were administered as needed to maintain adequate anesthesia. After hair removal (shaving and depilatory agent) over the neck and back, the animals were intubated with polyethylene (PE)-240 tubing via tracheostomy to facilitate respiration. The left carotid artery and left jugular vein were cannulated with PE-50 tubing. Mean arterial blood pressure (MABP) and heart rate (HR) values were recorded continuously through the carotid artery cannula using an Acknowledge data-acquisition system (model MP100; Biopac; Santa Barbara, CA), while for measurement of central venous pressure and intravascular injections, the jugular vein catheter was used. Ringer solution was infused through the jugular catheter at a rate of 1 ml·kg−1·h−1 to prevent dehydration. Body temperature was maintained at 37°C by a warm-water heating pad set at 105°F.

The left spinotrapezius muscle was partially exteriorized using a modification of the procedure originally described by Gray (6). The animals were transferred to a Plexiglas viewing platform, and rectal temperature was maintained at 37°C via warm circulating water within the platform. The excised muscle flap was moistened with Ringer solution. The muscle was placed dorsal side up on the platform and spread to its physiological length via sutures at the lateral margins. Finally, the muscle was covered with Saran plastic film (Dow Corning) to minimize desiccation and atmospheric gas exchange. Platinum electrodes connected to a Grass stimulator (model S9; Quincy, MA) were placed at the ends of the muscle with the proximal stimulating electrode placed within 1 mm of the spinal nerve that supplied the muscle. Muscle temperature was maintained at 37°C by warm water circulating under the transparent viewing region.

Approximately 95 μl of blood was removed from the carotid catheter for arterial blood analysis (model 700; Radiometer; Copenhagen) of the following parameters: Po2, Pco2, pH, lactate, base excess, Na+, and K+. Those animals with systemic arterial Po2 < 80 mmHg were considered to be abnormally hypoxic and were excluded from the study.

Animal use in this study followed the National Institutes of Health guidelines for the humane treatment of laboratory animals. The procedures used in this study were approved by the Institutional Animal Care and Use Committee of Virginia Commonwealth University.

RBC Labeling and Injection

RBCs were labeled with fluorescein isothiocyanate (FITC) using a modified procedure described by Lipowsky et al. (13). RBCs from donor rats were washed in Tris-buffered Ringer solution (pH 7.4) and incubated for 2 h in a solution of 5 mg of FITC (Research Organics; Cleveland, OH) dissolved in 10 ml of Tris-buffered Ringer solution containing 0.5% BSA (TRA). The labeled RBCs (FRBCs) were washed several times in TRA to remove excess dye and were resuspended in TRA to a hematocrit of 50%. The FRBCs were injected through the jugular catheter to achieve a labeled fraction (LF) of FRBCs that was ∼1% of the animal's total RBC volume (assuming a blood volume of 6% body wt and a systemic hematocrit of 50%; Refs. 15, 20). After FRBC injection, the LF was determined by counting labeled and unlabeled RBCs in a drop of blood (diluted 1:50 in saline) placed on a microscope slide.

Palladium Phosphor Preparation and Injection

Palladium-meso-tetra(4-carboxyphenyl)porphyrin (concentration, 10 mg/ml; Oxygen Enterprises; Philadelphia, PA) was bound to BSA (A-9647 fraction V; Sigma), and the pH of the solution was adjusted to 7.4 (24). For the group of rats designated for tissue Po2 analysis, five microinjections (30 μl) of the phosphor-BSA conjugate were made directly into the tissue adjacent to the peripheral sutures using a 32-gauge needle (Popper and Sons; New Hyde Park, NY) and a 1-ml syringe (22). Each injection site was <2 mm from the edge of the spinotrapezius muscle to minimize trauma and alterations in microvascular blood flow to the central portions of the muscle, where measurements would subsequently be made. After injection, a 30-min equilibration period was allowed for uniform distribution of the phosphor near the injection sites.

The group of rats assigned for microvascular Po2 measurements received an intravascular bolus of phosphor (concentration, 10 mg/ml; dose, 30 mg/kg body wt) through the jugular catheter. The animals were allowed a 15-min equilibration period to insure homogeneous distribution of phosphor in the intravascular compartment.

Experimental Protocol

Phosphorescence quenching microscopy.

For the purpose of local Po2 determination by PQM, the Plexiglas platform was placed on the stage of a Zeiss microscope. The adequacy of regional blood flow was qualitatively assessed using a ×10 objective with 0.25 numerical aperture (Zeiss), and a Neofluar ×40 objective with 0.75 numerical aperture (Zeiss) was employed to excite an area measuring 10 × 10 μm over a selected tissue or microvascular region. A xenon flash lamp (model FX-249; EG&G Optoelectronics; Salem, MA) with 10-Hz frequency, 0.5-J flash energy, and 3-μs flash duration was used to excite the injected phosphor. The excitation wavelength was selected by passing the incident light through a 430-nm band-pass excitation filter (430BP100; Omega Optical; Brattleboro, VT). A dichroic mirror reflected the light through the objective and onto the tissue sample. The resulting phosphorescence signal passed through the dichroic mirror and a 650-nm cuton filter (Oriel; Stratford, CT) before being detected by a photomultiplier tube (model R1617; Hamamatsu; Middlesex, NJ). The output from the photomultiplier was input to a low-noise current-to-voltage converter (model OP27EP; Analog Devices; Norwood, MA) and then passed to an analog-to-digital converter (AT MIO/6F5; National Instruments; Austin, TX) that was housed in a personal computer. The photomultiplier output was also monitored with an oscilloscope (model 72-3060; Tenma Test Equipment; Springboro, OH). A minimum signal amplitude of 1 V and a signal-to-noise ratio of 100 were established to insure accurate determination of tissue Po2 (9). Other data-acquisition parameters included a sampling rate of 200 kHz with 200 points per decay curve. Twenty curves were averaged and saved before analysis via the curve-fitting procedure below. The LabVIEW 4.01 software package (National Instruments) was used for data acquisition and preliminary online processing (curve averaging, normalization, and fitting by the Marquardt-Levenburg procedure) to determine Po2 according to the rectangular model of Po2 distribution (23) Math(1) where k0 = 1/τ0, τ0 (lifetime in the absence of oxygen, 546 μs) and kq (quenching coefficient, 3.06 × 10−4·μs−1·mmHg−1) were determined from in vitro calibration experiments, and δ is the half-width of the rectangular distribution (24). Final Po2 values were determined offline using nonlinear analysis of the phosphorescence decay curves according to Eq. 2 that includes the above expression for a rectangular distribution of Po2 (24) Math(2) where τF is the lifetime of the decay of the excitation flash, a is the fraction of the signal contributed by the phosphorescence decay, and (1 − a) is the fraction contributed by the tail of the excitation flash. Averaged and normalized decay curves were fit to Eq. 2 using the nonlinear least-squares Marquardt-Levenburg algorithm in Microcal Origin 6.0 software (Microcal Software; Northampton, MA).

Determination of postcontraction microvessel Po2 recovery.

A baseline set of Po2 measurements was taken at the junction of primary and secondary arterioles and venules in the animals that received intravenous injections of the palladium porphyrin compound. All selected sites for venular Po2 analysis were at least 100 μm from arterioles that might serve as diffusive sources of oxygen. Then, muscle contraction was induced for a period of 3 min as described below. Because of tissue movement during muscle contraction, no measurements of microvessel Po2 could be made during the period of contraction. After contraction, Po2 measurements in both vessels were repeated at 30-s intervals for 3 min. Because of the time required to locate and refocus the microvessel after contractions ceased, the first measurements are reported at 30 s postcontraction. To reduce animal-to-animal variability in Po2 values, results for each preparation during the recovery period were expressed as a fraction of the Po2 measured during the precontraction control period. The normalized values that appear in Figs. 1 and 2 represent means (±SD) of the different preparations.

Fig. 1.

Time course of the recovery of arteriolar oxygen tension (Po2) values after a 3-min contraction period. Measurements are reported from 30 s into the recovery period. Values represent means ± SD of Po2 values normalized to precontraction control Po2 levels. There were no significant differences in the fraction of resting Po2 at the various time points between the Wistar-Kyoto rat (WKY; n = 5) and spontaneously hypertensive rat (SHR; n = 4) groups.

Fig. 2.

Time course of the recovery of venular Po2 after a 3-min contraction period. Measurements are reported from 30 s into the recovery period. *P < 0.05, significant difference between the fraction of resting Po2 in WKY (n = 5) and SHR (n = 4) groups.

Determination of postcontraction tissue Po2 recovery.

A baseline set of Po2 measurements was taken in a 15 × 15-μm area of tissue that was at least 500 μm from arcade venules and arterioles. Muscle contraction was induced with the stimulator for a period of 3 min at a frequency of 8 Hz. The stimulation voltage was 5 V, and the pulse duration was 0.2 ms. Tissue Po2 measurements were obtained at 30-s intervals during contraction and for 3 min after contraction. Because Po2 could be measured in the tissue during contraction, recovery Po2 is reported starting at the end of the contraction (t = 0 s). The Po2 values in Fig. 3 were normalized in the same way as described above for the microvessel values.

Fig. 3.

Time course of the recovery of tissue Po2 after a 3-min contraction period. Measurements are reported from the beginning of the recovery period. *P < 0.05, significant difference between the fraction of resting Po2 in WKY (n = 5) and SHR (n = 4) groups.

Determination of RBC velocity and arteriolar blood flow.

The Plexiglas platform was attached to a Zeiss microscope equipped for epifluorescence with a high-pressure xenon arc lamp (XBO 75 W/2), a 470-nm band-pass excitation filter, a 515-nm cuton emission filter, and a dichroic mirror (Chroma Technology). A long working distance objective (Leitz UMK 50; ×50 with 0.60 numerical aperture) was used for fluorescence measurements. Images were captured using a closed-circuit video system consisting of an intensifier (GenIISys), video camera (model CCD-72; Dage-MTI), video monitor (model WV-5410; Panasonic), and S-VHS recorder (model AG-7350; Panasonic). Electronic shuttering of the intensifier was used to produce multiple images of a single RBC in the same video frame as a means to accommodate RBC velocities >2 mm/s (15). The pulse signal for shuttering was provided by a signal generator (model CFG 250; Tektronix), and the frequency was controlled at 120 Hz (4 images/video frame) by a frequency counter (model 1823 universal counter; BK Precision). Analysis of the video frames was used to determine the velocity of a single RBC (VRBC) according to the equation Math(3) where Δz is the distance between the center of the leading and trailing FRBC images in a single frame (i.e., 3 times the average distance traveled by an FRBC between shutter openings), Δt is the change in time (= fs−1), and fs is the shuttering frequency (120 Hz). Measurements of RBC velocity were obtained for 10 RBCs traveling in the center of an arcade arteriole according to the criteria of Bishop et al. (2) and were averaged to yield a mean RBC velocity on the axis. As determined by Bishop et al. (2), νmax/2 is the average luminal velocity. Arteriolar blood flow was calculated according to the equation Math(4) where F is blood flow (in nl/s), νmax is the mean maximum velocity (in mm/s), and r is the radius of the arteriolar lumen (in μm). Measurements of arcade arteriole diameter and maximum RBC velocity were repeated at 30-s intervals after the 3-min contraction period.


All data are expressed as means ± SD, and comparisons between two independent means (i.e., between WKYs and SHRs or between precontraction control and either WKYs or SHRs at a given time point) were made using Student's t-test. One and two-way ANOVA followed by Tukey's multiple comparison testing were used for multiple comparisons. Significance was taken at the P < 0.05 level.


A total of 15 WKYs and 13 SHRs were evaluated in this study with a mean body weight of 261 ± 8 g for the WKYs and 260 ± 12 g for the SHRs. The MABP of the WKY group was 110 ± 11 mmHg, and the MABP for the SHR group was 164 ± 15 mmHg.

Arteriolar, Venular, and Tissue Po2 during Recovery

The arteriolar Po2 values (values normalized to precontraction control Po2; see Table 1) were not statistically different between the two groups at any time during the 180-s recovery period (see Fig. 1). In both groups, the arteriolar Po2 values progressively increased at a near-linear rate and exceeded control values over most of the observed recovery period.

View this table:
Table 1.

Po2 measurements at tissue, arteriolar, and venular sites under baseline conditions for SHRs and WKYs

The recovery of venular Po2 values (normalized to precontraction control Po2; see Table 1) after a 3-min contraction period is shown in Fig. 2. Venular Po2 in the WKY group fully recovered within 60 s after contraction and even exceeded resting values (see Fig. 2; P < 0.05). In contrast, venular Po2 values remained significantly depressed below resting values in the SHR group for the entire 180-s recovery period (P < 0.05). In this instance, recovery in venular Po2 values for the WKY group was significantly greater than that of the SHR group for each of the measured time points in the 3-min recovery interval.

Resting tissue Po2 values were not significantly different between WKYs and SHRs (26 ± 5 vs. 30 ± 1 mmHg, respectively; P = 0.39; Table 1). The response of tissue Po2 values (normalized to precontraction control Po2 values; see Table 1) after a 3-min period of muscle contraction is illustrated (see Fig. 3). At the end of the contraction period, tissue Po2 values had dropped to 32 ± 10 and 29 ± 4% of the resting values in the WKY and SHR groups, respectively. Although the differences in tissue Po2 values for the WKY and SHR groups were not significant (P = 0.65) immediately after the 180-s contraction period, Fig. 3 shows that the recovery of Po2 occurred at a slower rate in the SHR compared with the WKY group. At 30 s postcontraction, tissue Po2 values in the WKYs had recovered to 99 ± 12% of the resting value and were statistically different (P < 0.05) from the 75 ± 7% recovery observed in the SHRs. Additional inspection of Fig. 3 reveals that, although tissue Po2 values in the WKYs fully recovered 30 s postcontraction, the SHRs required 120 s to fully recover.

Response of Arteriolar Diameter to Contraction

As displayed in Table 2, the mean diameters of arcade arterioles in the resting spinotrapezius were not different for the two groups (P = 0.90). However, the SHR group experienced greater arteriolar dilation relative to the WKY group in response to the 3-min contraction period (P < 0.05; Fig. 4).

Fig. 4.

Arteriolar diameter in WKY (n = 5) and SHR (n = 5) groups before contraction and during the 180-s recovery period. *P < 0.05, significant difference between WKY and SHR groups. R, rest.

View this table:
Table 2.

Baseline (resting muscle) vessel diameter, RBC velocity, and blood flow in arcade arterioles for SHRs and WKYs

Responses of RBC Velocity and Blood Flow to Contraction

Table 2 compares the baseline RBC velocity and blood flow measurements for the WKY and SHR groups in arcade arterioles. There were no significant differences in the measurements of resting RBC velocity (P = 0.06) or blood flow (P = 0.10) between the WKY and SHR groups. Figures 5 and 6 illustrate the responses of arcade arteriole RBC velocity and blood flow, respectively, to the 3-min contraction period. RBC velocity was greater in the SHR group for the first 90 s after cessation of contraction (P < 0.05) but was not significantly different from the RBC velocity in the WKY group during the last half of the recovery period (see Fig. 5). Furthermore, blood flow values were significantly greater in the SHRs for the entire 3-min recovery period (P < 0.05; Fig. 6).

Fig. 5.

Arteriolar red blood cell (RBC) velocity in WKY (n = 5) and SHR (n = 5) groups before contraction and during the 180-s recovery period. *P < 0.05, significant difference between WKY and SHR groups.

Fig. 6.

Arteriolar blood flow in WKY (n = 5) and SHR (n = 5) groups before contraction and during the 180-s recovery period. *P < 0.05, significant difference between WKY and SHR groups.


Under physiological conditions, the structural elements that serve the transport of oxygen in striated muscle appear to be optimized to provide rapid and efficient adjustments of the oxygen supply to contracting muscle. In the case of the spinotrapezius muscle of WKYs, the robust hyperemic response to muscle contraction and the rapid (within 30 s) recovery of venular and tissue Po2 values to baseline levels are impressive confirmations of this belief. For the SHRs, however, this is not the case. Although the hyperemic response to muscle contraction in the SHR group even exceeds that of the WKY group, the recovery of tissue Po2 values in the SHRs is retarded relative to the WKYs and venular Po2 is even more so.

What differences between the WKY and SHR groups could account for this dramatic difference in behavior of oxygenation? It is known that the structural defect of rarefaction in the microvascular network is present in the SHRs relative to the WKYs (7). Two obvious consequences that affect oxygen transport are 1) a reduced surface area for the exchange of oxygen, and 2) a reduced transit time of RBCs through the exchange vessels of the SHRs. Gutierrez (8) has shown that under conditions of elevated oxygen demand (muscle contraction) and reduced transit time, a significant disequilibrium for oxygen can develop between the plasma and RBC hemoglobin. At normal hematocrit levels, almost all of the oxygen in blood is bound to the hemoglobin inside RBCs. Furthermore, it is generally assumed that in blood flowing through the microcirculation, a near-equilibrium situation exists for oxygen between the plasma and RBC phases. For muscle contraction in normal tissues, the oxygen transport system appears to be designed to provide adequate oxygenation of the muscle despite the elevated demand and decreased transit time. However, when the oxygen exchange parameters (e.g., surface area and transit time) are compromised as in the SHRs because of microvascular rarefaction, the consequences can be manifested in prolonged recovery of tissue oxygenation. Our findings are consistent with previous observations that the spinotrapezius muscle of the SHRs, relative to that of the WKYs, exhibits significantly slower recovery of tissue Po2 values after contraction despite having greater blood flow during the hyperemic recovery period (3, 10, 11).

Tissue oxygenation at rest (low V̇o2) in this muscle derives from oxygen exchange across the walls of arterioles and capillaries. The substantial longitudinal gradient in Po2 reported by Lash and Bohlen (12) shows that most of the decrease occurs in the arteriolar network. In the transition from rest to contraction (high V̇o2), the primary site for oxygen exchange in muscle shifts from the arterioles to the capillaries (17) because the functional hyperemia leads to decreased contact time for blood in the arterioles (less precapillary oxygen diffusion), and under this condition the capillaries become relatively more efficient oxygen exchangers compared with the arterioles.

Arteriolar response.

In the postcontraction recovery period, Po2 values in arterioles of SHRs had recovered to baseline and continued an upward trend through the end (180 s) of the postcontraction observation period. The Po2 values in the arterioles of the WKYs increased in parallel to the SHRs. Although we found no significant difference in blood flow before contraction between the SHR and WKY groups, the hyperemic response observed in the arterioles was larger in the SHRs than in the WKYs (see Fig. 6). Immediately after the 3-min contraction period, RBC velocity (see Fig. 5) was increased ∼20% in the SHRs with respect to the WKYs (increase of ∼40% from baseline in both groups). This higher velocity resulted in less precapillary oxygen diffusion and allowed the SHRs to maintain their arteriolar Po2 values despite having lower tissue Po2 values. We expect that lower RBC velocities would not have allowed the SHRs to maintain arteriolar Po2, as the more hypoxic tissues in the SHR would cause more precapillary diffusion due to the greater blood-to-tissue Po2 difference. RBC velocity remained elevated above resting levels throughout the 3-min recovery period. The increased flow (see Fig. 6) in the SHRs was primarily due to greater vasodilatation in the SHRs (∼30% greater increase in cross-sectional area compared with the WKY group; see Fig. 4). Blood flow in the SHRs began to decline at a greater rate 120 s after contraction (see Fig. 6), which is near the same time that tissue Po2 values began to approach baseline levels (see Fig. 3); this suggests a connection between tissue oxygenation and the arteriolar vasomotor response. Because arteriolar blood flow was elevated in the SHRs relative to the WKYs and arteriolar Po2 values were similar between the two groups, a decrement in the convective flow of oxygen cannot be the explanation for the slower recovery of tissue and venular Po2 in the SHRs.

Venular response.

Venular Po2 values in the WKYs had already returned to baseline 30 s after the end of contractions, and this is consistent with the rapid restoration of tissue Po2 values in these animals. Venular Po2 values for the SHRs did not return to baseline and remained depressed throughout the 3-min postcontraction observation period. This finding is in apparent conflict with that of Lash and Bohlen (11), who found that venular oxygen saturation in the SHRs was elevated after contraction (in WKYs, it fell from 43% at rest to 41% with 8 Hz contraction; in SHRs, it rose from 31 to 51% under the same conditions). It appears that this discrepancy can be resolved by considering the locations of the two measurements and the possibility of substantial disequilibrium for oxygen between plasma and RBCs. The venular oxygen saturation (So2) measurement was close to the downstream end of the capillaries (11), so that a substantial disequilibrium between plasma and RBCs would predict high So2 and low Po2 values. The Po2 measurement we report was made further downstream in larger venules, and the disequilibrium hypothesis would predict that Po2 could remain depressed until sufficient oxygen was released from the RBCs to restore equilibrium between the plasma and RBC phases. Thus our findings in SHRs of reduced plasma Po2 from the arcade microvessels coupled with Lash and Bohlen's observations of elevated RBC So2 at postcapillary venules are consistent with the disequilibrium hypothesis that is a consequence of the elevated oxygen demand, reduced surface area for oxygen exchange, and reduced time for oxygen release from RBC hemoglobin.

Capillary response.

The above considerations about arterioles and venules lead to the question, Why are not the capillaries the major site of oxygen exchange in the SHRs during the postcontraction period? Greene et al. (7) found that the microvascular rarefaction observed in the SHRs relative to the WKYs resulted in a reduction of ∼14% in capillary density in the SHRs. In addition, Skalak and Schmid-Schonbein (21) found that there were not as many anastomotic connections in the capillary network of the SHRs vs. WKYs, so that the RBC pathways in the capillary networks of SHRs would be less tortuous and shorter than in the WKYs. The increased hyperemic response in the arterioles of the SHRs vs. WKYs together with the reduced capillarity of the SHRs would lead to more flow passing through a smaller number of vessels. The resulting increase in RBC velocity in capillaries of the SHRs would lead to reduced residence time of the RBCs in the capillaries and hence less time for the hemoglobin to release its oxygen, possibly creating a situation in which there was disequilibrium in oxygen between the erythrocytic hemoglobin and the plasma. Under these adverse conditions for oxygen exchange, one would expect the plasma to be almost depleted of oxygen, because the oxygen-carrying capacity of the plasma is so low. If the capillaries are not the major site of oxygen exchange in the SHRs during the postcontraction period, then this leaves the arterioles and venules as the remaining microvascular exchange sites. One would then expect that oxygen would diffuse more from the arterioles, because they have higher Po2 value and larger surface area (greater arteriolar dilation and longer duration in the SHRs compared with the WKYs; see Fig. 4) throughout the recovery period during which tissue oxygenation was gradually restored. Although we did not measure postcapillary venular Po2 values, we would expect that Po2 in these areas would be significantly lower in the SHRs due to the extraction of most of the oxygen in the plasma from the SHR capillaries. Also, at 180 s postcontraction, in the SHR tissue, Po2 had returned to 28 mmHg (baseline was 30 mmHg), whereas venular Po2 had only returned to 24 mmHg (baseline was 33 mmHg), which indicates that the venules were not a source of oxygen for the tissue during recovery. The higher arteriolar Po2 throughout recovery suggests that the arterioles were the primary source of oxygen during this time.

A possible reconciliation of the results of this study with those reported by Lash and Bohlen (11) has to do with the fact that our measurements of Po2 using the phosphorescence quenching method reported Po2 in the plasma phase, whereas Lash and Bohlen's measurement of venular blood oxygenation reported hemoglobin So2 within erythrocytes. Given that the immediate source of molecular oxygen for diffusion from the blood to the tissue is that meager amount dissolved in the plasma, and the much larger amount of oxygen bound to the hemoglobin might not be readily available due to the rapid transit of the RBCs through the microvessels, the likely result is a plasma phase depleted of oxygen moving with a RBC phase with most of its oxygen still bound to hemoglobin. There would be a large sink for oxygen in the postcontraction state of the recovering muscle in the SHRs and an adequate convective supply of oxygen in the microvessels but an inadequate amount of time for that oxygen to be released from the RBCs until a longer contact time were available for the blood in the venules.

Oxygen consumption.

The tissue Po2 values under resting conditions for the WKYs and SHRs were 26 and 30 mmHg, respectively. In addition, the corresponding arteriolar and venular Po2 values were 37 and 31 mmHg, respectively, for the WKYs and 34 and 33 mmHg, respectively, for the SHRs. These sets of Po2 values suggest that resting oxygen consumption was smaller for the SHRs than for the WKYs. For both WKY and SHR groups, tissue Po2 in the spinotrapezius muscle decreased to ∼8 mmHg (30% of the resting value) after 3 min of contraction at 8 Hz. Although Boegehold and Bohlen found that tissue Po2 immediately after the contraction (t = 0 s), was significantly lower in the SHRs (3), the present study found no difference in tissue Po2 values between the two groups at this time (see Fig. 3). The evidence presented in this study is consistent with the premise that SHRs experience impaired oxygen extraction from RBCs but not plasma in microvessels during recovery from muscle contraction, because it takes tissue Po2 longer to return to baseline values in SHRs. However, a greater increase in contraction-induced oxygen consumption in SHRs relative to WKYs might yield similar results. Such a phenomenon is documented in human subjects with chronic congestive heart failure (14) and is shown to be independent of blood flow. Although histochemical analysis of the soleus muscle in young SHRs has shown a significant increase in oxidative and anaerobic enzymatic activity (1), the SHR spinotrapezius muscle does not show the increased number of slow-twitch fibers that one might expect if oxygen consumption were increased more than in the WKYs after muscle contraction. Furthermore, Lash and Bohlen (11) argued that the oxygen consumption associated with muscle contraction for WKYs and SHRs was similar, so it is unclear whether differences in V̇o2 are the cause of the prolonged depression in tissue and venular Po2 values. A larger increase in oxygen consumption by SHRs during contraction would exacerbate the disequilibrium caused by decreased transit time of RBCs through the oxygen-exchange vessels.

In conclusion, the results of this study support the view that structural alterations such as rarefaction in the SHR microcirculation provide a mechanism for slower recovery of tissue Po2 values after muscle contraction compared with WKYs. The consequences of these geometric alterations are a decrease in arteriolar and capillary surface area, an increase in RBC velocity that impairs diffusive transport of oxygen to the tissues from arterioles and capillaries due to disequilibrium of oxygen between plasma and RBCs, and a prolongation in oxygen transfer from RBCs to plasma to restore equilibrium of oxygen between these two phases. These considerations suggest the value of future studies that focus on the role of the capillary and venular networks in muscle contraction for WKYs and SHRs. Measurements of pertinent variables such as capillary network geometry (degree of rarefaction, RBC flow pathways), RBC dynamics (velocity, transit time, supply rate, longitudinal spacing, capillary RBC content), and blood (RBC So2 and plasma Po2) and tissue oxygenation before, during (if possible), and after contraction would provide valuable data to delineate the potentially different behaviors of the capillary network in WKYs and SHRs. Greene et al. (7) have made calculations relevant to this situation and their predictions can be tested by direct microvascular measurements. In particular, direct evidence for or against disequilibrium in oxygen between the RBCs and plasma can be obtained from simultaneous blood oxygen measurements (in RBCs and plasma) in capillary and venular networks.


This work was supported in part by National Heart, Lung, and Blood Institute Grant HL-18292 and the American Foundation for Aging Research.


The authors are grateful to Dr. Aleksander S. Golub and Dr. Mohammad Tiba for technical assistance.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract