Poole, David C., Timothy. I. Musch, and Casey A. Kindig. In vivo microvascular structural and functional consequences of muscle length changes. Am. J. Physiol.
In vivo microvascular structural and functional consequences of muscle length changes DAVID C. POOLE, TIMOTHY I. MUSCH, AND CASEY A. KINDIG Departments of Kinesiology and Anatomy I Physiology, Kansas State University, Manhattan, Kansas 66506-5602 Poole, David C., Timothy I. Musch, and Casey A. Kindig. In vivo microvascular structural and functional consequences of muscle length changes. Am. J. Physiol. 272 (Heart Circ. PhysioZ. 41): H2107-H2114, 1997. -As muscles are stretched, blood flow and oxygen delivery are compromised, and consequently muscle function is impaired. We tested the hypothesis that the structural microvascular sequellae associated with muscle extension in vivo would impair capillary red blood cell hemodynamics. We developed an intravital spinotrapezius preparation that facilitated direct on-line measurement and alteration of sarcomere length simultaneously with determination of capillary geometry and red blood cell flow dynamics. The range of spinotrapezius sarcomere lengths achievable in vivo was 2.17 t_ 0.05 to 3.13 t 0.11 pm. Capillary tortuosity decreased systematically with increases of sarcomere length up to 2.6 pm, at which point most capillaries appeared to be highly oriented along the fiber longitudinal axis. Further increases in sarcomere length above this value reduced mean capillary diameter from 5.61 t 0.03 pm at 2.4-2.6 pm sarcomere length to 4.12 t 0.05 pm at 3.2-3.4 pm sarcomere length. Over the range of physiological sarcomere lengths, bulk blood flow (radioactive microspheres) decreased -40% from 24.3 t 7.5 to 14.5 t 4.6 ml* 100 g 1 emin-? The proportion of continuously perfused capillaries, i.e., those with continuous flow throughout the 60-s observation period, decreased from 95.9 ? 0.6% at the shortest sarcomere lengths to 56.5 2 0.7% at the longest sarcomere lengths and was correlated significantly with the reduced capillary diameter (r = 0.711, P < 0.01; n = 18). We conclude that alterations in capillary geometry and luminal diameter consequent to increased muscle sarcomere length are associated with a reduction in mean capillary red blood cell velocity and a greater proportion of capillaries in which red blood cell flow is stopped or intermittent. Thus not only does muscle stretching reduce bulk blood (and oxygen) delivery, it also alters capillary red blood cell flow dynamics, which may further impair blood-tissue oxygen exchange. spinotrapezius length; muscle velocity
muscle; intravital blood flow; capillary
microscopy; sarcomere geometry; red blood cell
MUSCLE PERFORMANCE depends critically on adequate blood flow and its microvascular distribution. Factors that compromise blood flow invariably reduce muscle oxygen uptake and increase fatigability. Given that muscle blood flow is controlled relative to metabolic demands by a complex array of neural, humoral, and mechanical influences, blood flow is decreased when muscles are stretched toward and beyond that length at which peak force production is achieved (1, 18, 39, 40, 42). It has been speculated that this effect results from increased microvascular impedance (22, 33,40).
OPTIMAL
0363-6135/97
$5.00
Copyright
o 1997
In 1983, Ishikawa et al. (19) and Potter and Groom (35) used microvascular corrosion casts to demonstrate that skeletal muscle capillaries are not straight, unbranched structures but rather may exhibit considerable tortuosity and branching. Subsequently, in perfusion-fixed muscles, Mathieu-Costello (25) determined that the degree of tortuosity of the capillary bed is a function of muscle sarcomere length and not a consequence of fiber type or muscle architecture (see also Ref. 27). Thus, at short muscle sarcomere lengths of -- 1.6 pm, capillaries are extremely tortuous. As the muscle is extended or stretched, this tortuosity decreases in a curvilinear fashion until at mean sarcomere lengths of -2.2 pm the majority of capillaries are highly anisotropic, i.e., closely aligned with the longitudinal axis of the muscle fibers (25). Further sarcomerelength increases beyond this point actually stretch the capillaries (10,29,33). Morphometric analysis of perfusion-fixed muscle reveals that this stretching is accompanied by a narrowing of capillary luminal diameter (25,29,33). Decreased capillary diameter will increase microvascular resistance and may potentially impede blood flow and/or increase the heterogeneity of that blood flow distribution (22, 33). Changes of muscle length also affect microvascular geometry at the arteriolar and venular level. Specifically, in the hamster retractor muscle, vessel curvature, bifurcation angle, and diameter all change with alterations in muscle length (31). To date, morphometric evaluation of capillary geometry has been performed almost exclusively on ex vivo, perfusion-fixed tissue. However, Ledvina and Segal (22) have recently corroborated that the proportion of tortuous capillaries is reduced at increased sarcomere lengths in slow- and fast-twitch rat skeletal muscle in vivo. The purpose of the present investigation was to 1) confirm quantitatively whether the contribution of capillary tortuosity and branching to total capillary length is reduced at extended sarcomere lengths in vivo, 2) determine the relationship between sarcomere length and capillary diameter in vivo, and 3) establish whether structural changes that accompany muscle stretching are associated with alterations in capillary red blood cell flow dynamics, i.e., red blood cell velocities and the proportion of capillaries in which flow is stopped or intermittent. Specifically, we tested the hypothesis that the stretch-induced reduction in mean capillary diameter at sarcomere lengths of ~2.3 pm in vivo would be associated with an increased proportion of capillaries in which flow was either stopped or intermittent. the American
Physiological
Society
H2107
HZ108
MUSCLE
LENGTH
AND
MICROVASCULAR
METHODS
Experimental
Procedures
A total of 26 female Sprague-Dawley rats (236.4 t_ 7.1 g) were used in this investigation. All procedures were approved by Kansas State University animal-handling guidelines. All surgical interventions were conducted under general anesthesia (ketamine/xylazine; 44/16 mg/kg im), which was supplemented as necessary before the animal was set on the observation platform. During the intravital observation period, no animals required additional anesthesia. Initially, either the carotid or femoral artery was cannulated using polyethylene (PE)-50 (Intra-Medic polyethylene tubing; Clay Adams, Sparks, MD) to monitor arterial blood pressure (Digi-Med BPA model 200, Louisville, KY) and to facilitate fluid replacement. MuscZe preparation. The spinotrapezius muscle is a thin, straplike muscle that lies longitudinally anterior to the spine in the thoracic and upper lumbar regions, where its primary action is drawing the scapula anteriorly. The muscle fibers originate from the spines of the fourth thoracic to the third lumbar vertebrae and converge to their insertion on the spine of the scapula (14). Spinotrapezius muscles were prepared according to previously described methods (14) with minimal fascial disturbance, thereby limiting tissue damage and its associated microcirculatory consequences. This preparation is particularly valuable because the origin of the muscle fibers and all vascular and nervous connections remain intact (14, 28). Preparations were placed on a circulation-heated (38°C) platform and continuously superfused (Julabo Microcirculator U3-7A, Schwarzwald, Germany) with a Krebs-Henseleit bicarbonate-buffered solution equilibrated with a 95% N&j% CO2 gas mixture (41). The exposed surrounding tissue was protected with Saran Wrap (Dow, Indianapolis, IN), and the muscle was sutured (6.0 silk, Ethicon, Somerville, NJ) at five equidistant positions around the caudal periphery to a thin wire horseshoe manifold (41) attached with a swivel to a muscle-stretching apparatus specifically designed for this procedure. This apparatus was attached to the platform and permitted precise and systematic stretching of the whole spinotrapezius muscle along the principal fiber longitudinal axis. IntrauitaZ video microscopy. The microcirculation images were acquired using an intravital video microscope (Olympus OVM-lOOONM, Tokyo, Japan) equipped with a noncontact, illuminated lens and viewed on a high-resolution color monitor (Sony Trinitron PVM-1954Q, Ichinoniya, Japan) at a final magnification of x 1,500. The system was calibrated by photographing a stage micrometer (MA285, Meiji Techno). Images were time-referenced by frame, and fields were stored via videocassette recorder (JVC BR-S822U, Elmwood Park, NJ) on Super VHS cassettes (JVC S-VHS Master XG) for subsequent off-line analysis. This microscopic viewing field (270 x 210 pm) typically contained five to seven muscle fibers with their associated capillaries. The muscle was transilluminated using a fiberoptic light source (Fiber-Lite, DolanJenner, Lawrence, MA), which provided increased visibility and clarity without compromising the hemodynamic behavior of the preparation. By angling the incident light, we were able to visualize clearly the A bands within 30-60% of the fiber length. Thus simultaneous measurements of sarcomere length, capillary geometry, and flow dynamics could be obtained. Determination of sarcomere length range in situ. In four animals, the entire spinotrapezius was exposed and all overlying fatty tissue was removed. With the origin and insertion intact, two sutures were placed along the muscle fiber axis near the cranial and caudal ends of the muscle. The
STRUCTURE
AND
FUNCTION
ipsilateral shoulder of the animal and body position were manipulated to obtain the physiological extremes of spinotrapezius lengthening and shortening. Care was taken not to place the muscle in any orientations that might have been considered nonfeasible under normal conditions of movement. The extremes of shortening and lengthening were determined using calipers accurate to within 0.25 mm. The muscle was then prepared as previously described (41) (with the sutures in place) and systematically set at several points ranging from the shortest up to the longest muscle length as determined previously in situ with the sutures as reference points. At each length, sarcomere length values were measured on-line using techniques described in CapiZZary and fiber structural data. Experimental design. In 18 animals, the spinotrapezius was prepared for intravital microscopy; initial measurements were made at the shortest muscle and sarcomere length was achieved spontaneously. Typically, up to four fields were observed for a period of -3 min each. Subsequently, the muscle was systematically stretched up to 3.2-3.4 pm in two to three increments with similar recordings made at each increment in approximately the same location within the muscle. After these measurements, muscle length was returned to the initial length in one to two increments, and the recordings were repeated. In those instances in which the muscle either would not shorten spontaneously to its initial length or, alternatively, the original flow profile, i.e., the percentage of capillaries in which flow was maintained, was not recovered, this was taken as evidence that the muscle had been overstretched and/or lost its viability. In these instances, the preparation was discarded, and the data were not analyzed further. In all preparations in which the normal flow profile was sustained, topical adenosine (loo4 M) induced a profound arteriolar vasoactive response. After surgery, experimental duration was typically l-l.5 h, during which time up to 1.5 ml of isotonic saline was infused intra-arterially to counteract dehydration. Of 18 animals used for the intravital preparation, 2 died under anesthesia before the experiment; in 2 animals their mean blood pressure dropped to ~70 mmHg and the film records for these animals were discarded; and in 6 animals the preparation was judged unsuitable for analysis because of the lack of image clarity or failure of flow profiles to return to control levels when the stretch condition was relieved. Data Acquisition
and Subsequent
Analysis
In all, 182 capillaries within 20 muscle fields were analyzed. Capillary and fiber structural data. Each viewing field was traced directly from the video monitor screen onto acetate paper. The details traced were as follows: 1) muscle fiber boundaries together with -25 A-bands within each fiber, and 2) the lower margin of the capillary endothelium where it was continuously visible and in focus. From these tracings, mean sarcomere length was determined from sets of 10 consecutive in-register sarcomeres (distance between 11 consecutive Abands) measured parallel to the muscle fiber axis. This procedure was performed several times on each muscle fiber and on every muscle fiber within the viewing field, where sarcomeres were visible to obtain a mean sarcomere length for each viewing field. Where the capillary endothelium was clearly visible on both sides of the capillary lumen, capillary lumen diameter was measured at several random locations per capillary with calipers accurate to ~0.25 mm, which at x 1,500 final magnification corresponds to approximately t 0.17 pm; the mean value was recorded. Red blood cell shape was also traced on acetate paper for subsequent inspection.
MUSCLE
LENGTH
AND
MICROVASCULAR
CapiZZary geometry. Capillary geometry was determined using a variation of the technique described by Batra et al. (2) in heart and provided a value analagous to the capillary anisotropy coefficient [c(K,O)] (24,25). From the tracing, each capillary was labeled (for organizational purposes), and beginning and end points of the visible length of each capillary were established. For each capillary, these points encompassed as much of the length as possible with the prerequisite that the capillary remained in focus for the entire length measured. A thin, flexible wire was then used to trace the actual path length of each capillary along the lower endothelium from beginning to end points. The wire was then straightened, and its length was measured. In those capillaries where capillary interconnections were encountered, onehalf of that capillary interconnection length was added to the measured capillary length. This enabled measurement of the length of a given capillary independent of other capillaries to which it may be connected. This provided a measurement of absolute capillary length within the screen analyzed and facilitated calculation of a value for c(K,O) that was analagous to that obtained by other morphometric methods (e.g., that in Ref. 25). Next, the corresponding length along the muscle fiber longitudinal axis was obtained by measuring the straightline distance strictly parallel to the muscle fiber axis between the previously established visible beginning and end points of each capillary. To obtain the proportion of the additional capillary length resulting from capillary tortuosity, branching, and deviations of alignment with the fiber axis [i.e., c(K,O)l, the actual capillary path length and length of the associated capillary interconnections were divided by the straight-line length along the muscle fiber longitudinal axis. FunctionaL data collection. Red blood cell flow was observed in real time and also using playback and frame-by-frame techniques. On the basis of the following criteria, each capillary visible in a given field was observed for a random, continuous 60-s period and placed into one of two categories: 1) flowing: 60 s with continuous flow; 2) partiaZZy impeded: stopped flow or sluggish flow [red blood cell velocity (VR& < VRBC was 50 pm/s] for 210 s out of 60 s. In addition, determined in -80 capillaries at a sarcomere length of >2.8 pm and a similar number (2.8 pm. These capillaries were selected at random under the condition that the red blood cell and capillary endothelium were clearly visible over several frames. In those capillaries in which flow was periodically arrested, V RBc was determined when flow resumed. Where no flow was evident over the 60-s observation period, VRBC was recorded as 0 pm/s. Determination of spinotrapezius bZood flow. In an additional four animals, we determined the effect of alterations in muscle length on absolute blood flow in the spinotrapezius muscle. Under xylazineketamine anesthesia, catheters (PE-10 connected to PE-50) were placed in the ascending aorta via the right carotid artery and caudal (tail) artery as previously described (30). The animal was placed on the heated platform, and arterial pressure was measured via the carotid artery catheter. After blood pressure attained a steady state, the tail artery catheter was connected to a 5-ml glass syringe and Harvard withdrawal pump (model 907). With umbilical tape, one forelimb was then positioned in either the flexed or the extended position to lengthen or shorten the ipsilateral spinotrapezius muscle. To obtain the reference sample for regional blood flow measurements with radiolabeled microspheres, blood withdrawal from the tail artery catheter was initiated at a rate of 0.25 ml/min. The carotid artery catheter was then disconnected from the pressure transducer, and 0.7-0.8 x lo6 15-pm-diameter microspheres (isotopes used: 46Sc, 85Sr, or l13Sn, in random order; New m England Nuclear
STRUCTURE
AND
HZ109
FUNCTION
Research Products, Boston, MA) were injected into the aortic arch of the animal for the determination of tissue blood flows. The carotid artery catheter was reattached to the pressure transducer, and arterial pressure was measured and recorded. After -10 min, while the forelimb was in the neutral position, hemodynamic measurements were checked, the forelimb was placed at the opposite extreme of the range of motion, and a second microsphere infusion was then performed using the same procedures as described above. Animals were killed after the second microsphere infusion by means of KC1 infusion. The thorax was opened, and placement of the carotid artery catheter into the aortic arch was verified. The lungs, kidneys, and ipsilateral spinotrapezius muscle were removed. Dissections were performed by a skilled research technician to ensure uniformity of the dissection process. Tissues were weighed and placed immediately into counting vials. The radioactivity of each tissue was determined on a gamma scintillation counter (Packard Cobra II Auto-Gamma Spectrometer, Downers Grove, IL) set to record the peak energy activity of each isotope for 2 min. Absolute blood flow to each tissue was determined using the reference sample method (12,17) and expressed as milliliters per minute per 100 g of tissue. Adequate mixing of the microspheres was verified for each microsphere infusion by demonstrating that the blood flows to the right and left kidney were similar to one another under both conditions. StatisticaL anaZysis. AI1 data are presented as means * SE. Differences in bulk blood flow and VRBC at different muscle lengths were analyzed by unpaired t-test. Data were regressed using standard least-squares regression techniques. A significance level of P 5 0.05 was accepted. RESULTS
There was no significant change in mean arterial blood pressure over the duration of the experiment. Mean arterial blood pressure averaged 89 t 6 and 90 t 8 mmHg at the beginning and end of the experiments, respectively. The in vivo range of sarcomere length excursion was from 2.17 t 0.05 to 3.13 t 0.11 pm. As demonstrated in Fig. 1, gross muscle length changes were indicative of alterations of sarcomere length. However, this was not a direct relationship, because sarcomere length changes averaged approximately two-thirds that calculated on 4
1
I = 0.80 Y=1.18X+1.11
tl
0 •I A v 1 1
RAT 1 RAT2 RAT3 RAT4
I 1.5
1.0
NORMALIZED
INTERMARKER
I 2.0
SPACING
Fig. 1. Relationship between muscle sarcomere length measured by intravital microscopy and muscle length measured between 2 sutures embedded in muscle along fiber longitudinal axis -1.0 cm apart (at 2.17 pm sarcomere length) and normalized for initial length. Range of muscle lengths shown replicate those found in situ (n = 4).
H2110
MUSCLE
LENGTH
AND
MICROVASCULAR
the basis that muscle and sarcomere length changes would be proportionally equivalent. As sarcomere length was increased >2.1 pm, capillaries became systematically less tortuous. This is depicted in Fig. -2, where at increased sarcomere lengths, a curvilinear decrease was found in the proportion of the additional capillary length that arose from capillary tortuosity and branching. At sarcomere lengths of ~2.6 pm, capillaries appeared highly oriented (anisotropic) with respect to the fiber longitudinal axis. The only principal nonanisotropic structures remaining under these conditions were capillary branches or interconnections. Thus further increases in sarcomere length beyond 2.6 pm would be expected to result in capillary stretchi .ng and a consequent reduc tion in mean luminal diameter. Note that this relationship (Fig. 2, closed circles) was similar to (although right-shifted from) that published previously (solid curve) for perfusionfixed rat hindlimb muscles (26). Figure 3 indicates that mean capillary luminal diameter -in the spinotrapezius does- indeed decrease at extended sarcomere lengths. Furthermore, it appears to be a biphasic process, with diameter decreasing modestly up to sarcomere lengths of -2.9 pm and precipitously beyond that. This behavior probably reflects that between 2.6 and 3.0 pm sarcomere length some capillaries are still not absolutely straight. Thus the mean capillary diameter measurement averages these capillaries with those that are straight and in which diameter is decreasing with increased muscle length. From the extreme of muscle shortening up to the longest length achievable in vivo, spinotrapezius muscle blood flow decreased by 40% from 24.3 t 7.5 to 14.5 t 4.6 ml 100 g- l min -l (P < 0.05). The blood flows to the left and right kidneys were highly correlated under both conditions (r =-0.94), indicative of good microsphere mixing. l
l
l 0 0.
P I 1.8
I 2.1
I 2.4
SARCOMERE
I 2.7
LENGTH
I 3.0
I 3.3
(pm)
Fig. 2. Relationship between spinotrapezius sarcomere length and additional capillary length that derives from capillary tortuosity and branching in 20 viewing fields (0; n = 8). Curved line: capillary anisotropy coefficient [c(K,O)l for perfusion-fixed rat hindlimb muscles
(26).
STRUCTURE
zs s7 IW
26 a
AND
FUNCTION
r=0.897
I
1
r=0.764 Y=9.65-7.47X for sarcomere 2.0
lengths
(2.9
2.4
SARCOMERE
pm 2.8
LENGTH
3.2
3.6
(pm)
Fig. 3. Relationship between spinotrapezius sarcomere length and mean capillary luminal diameter in each of 18 fields analyzed over the range of muscle lengths depicted in Fig. 1 (n = 8). Note precipitous fall in capillary diameter at sarcomere lengths of >2.9 pm.
Red blood cell flow profiles were also altered profoundly by increases in sarcomere length. Specifically, there was a significant negative correlation (r = -0.774, P < 0.01, n = 18) b et ween sarcomere length and the proportion of capillaries supporting unimpeded flow (Fig. 4A). As shown in Fig. 4B, this decreased capillary flow was associated proportionally (r = 0.711, P < 0.01, n = 18) with the reduced capillary diameter found at increased sarcomere lengths. Above sarcomere lengths of 2.9 pm, the average diameter of those capillaries in which flow was impeded was 4.19 2 0.23 pm. However, there were a few instances in which red blood cells moved sporadically through capillaries with luminal diameters as small as 2.5 pm, in which case the capillary lumen appeared to widen to permit cell passage. Figure 5 demonstrates the frequency distribution of capillary VRBCvalues over both sarcomere length ranges below and above 2.8 pm. Mean VRRcaveraged 251 t 14 J&S at sarcomere lengths of ~2.8 pm and was significantly (P < 0.05) slowed to 170 t 13 pm/s at sarcomere lengths of >2.8 pm. This reflected a marked redistribution of capillary VRBc values such that at ~2.8 pm only 24% of capillaries had a I&c value of ~200 pm/s; however, in the stretched muscle, this increased to 65%. The coefficient of variation of the VRBCincreased from 45% at the shorter sarcomere lengths to 72% at the longer sarcomere lengths. The shape of red blood cell profiles in individual capillaries was determined in large part by capillary diameter. Figure 6 illustrates the altered red blood cell shape over most of the range of capillary diameters encountered in the spinotrapezius. Thus, as capillary diameter decreased, the red blood cell shape underwent marked distortion until at a. capil lary diameter of ~4.0 pm red blood cells assumed an elongated “sausage” appearance. Although most red blood cells can squeeze through glass tubes of 2.45 pm diameter, and the minimum estimated diameter they can achieve without increases of membrane surface area is 1.8 pm (16), the
MUSCLE
(3 z
120
27 a?
1
LENGTH
A
AND
MICROVASCULAR
AND
STRUCTURE
B
r 120
r=0.711 Y=77.4X-16.8
loo-
tr TO rii cod
H2111
FUNCTION
Fig. 4. Relationship of spinotrapezius sarcomere length (A) and capillary luminal diameter (B) with the proportion of capillaries maintaining unimpeded flow in each viewing field (n = 8). Solid curves, linear regressions; dashed curves, 95% confidence intervals.
80-
wa r=0.774 Y=l69.6-32.1X I
6 I
2.0
I
2.4
I
2.8
SARCOMERE
3.2
LENGTH
(pm)
1
/
3.6
3
CAPILLARY
40
4
5
LUMEN
minimum red blood cell diameter that we observed did not decrease below 3.3 pm even at the extremes of sarcomere length, i.e., >3.2 pm. DISCUSSION
This investigation reports for the first time the simultaneous quantitative relationships among muscle sarcomere length, capillary geometry, and flow dynamics in vivo. These data confirm the effects of altered
ti = 251 + 14
pm/s
15-
t5 6
lo-
2
5--
O-
cn i
7
DIAMETER
(pm)
sarcomere length on capillary geometry and the contribution of capillary tortuosity and branching to total capillary length demonstrated previously in perfusionfixed tissue (25, 26). A necessary consequence of capillary tortuosity decreasing to its nadir at a sarcomere length of -2.6 pm is that further length increases beyond this stretch the capillaries (lo), thereby reducing their diameter (33). The present investigation demonstrates that this occurs in vivo and that decreased capillary luminal diameter is correlated significantly with the proportion of capillaries in which red blood cell flow is either stopped or intermittent. Furthermore, as muscle sarcomere length increases beyond -2.8 pm, mean VRBcis reduced, and the coefficient of variation of VRBcis increased from 45% to 72%. Range of Sarcomere Lengths
a 0
E 3
6
30- B 25-
170+13pm/s
v,=
9
i 20 - cv = 72% E
a
The ramifications of the present findings depend largely on whether sarcomere lengths of greater than -2.8 pm are achieved in vivo. In this regard, the available literature indicates that many mammalian skeletal muscles may operate at sarcomere lengths well beyond this point. These include the rat soleus, extensor digitorum longus, and diaphragm (10,22,25,32,33) and the human carpi brevis (23), as well as the lower limb flexors and extensors (5). Thus these muscles may show sufficient lengthening such that capillary diameter is reduced and capillary red blood cell dynamics are altered as described herein. Under certain circumstances, muscle function is improved by decreasing stretching. Specifically, in the human extensor carpi radialis brevis (ECRB) muscle, surgical lengthening of
CAPILLARY
3.3 0
25
75
4.0
5.0
(pm)
DIAMETER
6.0
6.3
6.5
8.0
125 175 225 275 325 375 A00
RBC VELOCITY (pm/s) Fig. 5. Frequency histograms depicting red blood cell velocities (B) spinotrapezius sarcomere WI& at short (A) and extended lengths. Mean V RBC was significantly (P < 0.05) reduced at sarcomere lengths of >2.8 pm. Coefficient of variation (CV) is calculated as mean/SD.
t
4
Fig. 6. Schematic representation of red blood cell profiles traced directly from the video monitor at capillary diameters ranging from 3.3 to 8.0 pm (bar = 10 pm). Even at capillary diameters as low as 2.5 urn. no further detectable deformation of red blood cells was evident.
H2112
MUSCLE
LENGTH
AND
MICROVASCULAR
the muscle-tendon junction acts to shorten muscle sarcomere length (13), which improves function and reduces pain. These effects have been attributed to decreasing insertional tension and allowing the ECRB to operate over a more advantageous portion of its length-tension relationship (13). The present investigation suggests that enhanced muscle blood flow and its microvascular distribution might also be important benefits of surgically shortening that muscle. Capillary
Geometry
The relationship between capillary geometry and sarcomere length has been defined for many different skeletal muscles using morphometric analysis of perfusion-fixed tissue (10,24-26,33). This technique has the advantage of sampling many hundreds of capillaries in any given muscle and providing tissue suitable for high-resolution microscopy. One disadvantage is that only one sarcomere length can be examined per muscle. Despite this concern, the curvilinear relationship obtained between c( K,O) (capillary anisotropy coefficient, i.e., additional capillary length provided by capillary tortuosity and branching) and sarcomere length is similar for all terrestrial mammalian muscles studied irrespective of fiber type or oxidative capacity (10, 26, 27, 34). The present investigation represents the first attempt to quantify this relationship in vivo. Figure 2 demonstrates that our results in the spinotrapezius suggest a similar (although right-shifted) curvilinear relationship between c(K,O) and sarcomere length to that described in perfusion-fixed rodent hindlimb muscles (26). This modest discrepancy probably relates to the presence of some shrinkage along the fiber longitudinal axis, which is an unavoidable consequence of processing tissue for electron microscopy (33). Indeed, it can be appreciated that correction of hindlimb muscle sarcomere length by 6-10% upward would bring the two data sets into very close quantitative agreement. Capillary
Luminal
Diameter
Capillary diameters measured in vivo typically average 5.5-6.5 pm (36-38). This is somewhat greater than those determined from corrosion casts (e.g., Ref. 35) or in perfusion-fixed tissue (e.g., Ref. 33). The capillary diameter values we measured at sarcomere lengths of ~2.6 pm averaged 6.36 t 0.05 pm and are therefore in close agreement with previously reported in vivo values. However, at greater sarcomere lengths, muscle stretch significantly (P < 0.01) reduced mean capillary diameter to values approaching 3.0-4.0 pm. Proportion
of Capillaries
Supporting
Flow
There is some controversy as to whether all muscle capillaries are perfused at rest (20). Thus there is evidence that, in resting muscle, flow may be stationary in some capillaries (3, 6, 15, 21). Alternatively, there is strong evidence that the vast majority of capillaries maintain flow at rest; however, there may be a large disparity in flow rates and transit times between capillaries (4,11,38). Using fluorescent dye techniques,
STRUCTURE
AND
FUNCTION
Kayar and Banchero (20) found all capillaries analyzed in the rat gastrocnemius to be labeled within 30 s of dye injection into the aortic arch. Because of the rapidity with which red blood cell oxygen suppl .ies m ay be depleted, capillaries wi .th very long transit times of the order of several seconds or more would still be counted as perfused. However, their contribution to overall muscle oxygen delivery would be small. In the present investigation, at sarcomere lengths of ~2.4 pm, >90% of capillaries sustained unimpeded red blood cell flow during the observation period (i.e., 60 s; Fig. 4A). At greater sarcomere lengths, the proportion of impeded vessels increased sharply, with the proportion of unimpeded vessels falling to -55% at a sarcomere length of 3.3 pm. The probability that differences exist in the proportion of perfused capillaries in muscle at rest based on species or fiber type variations must be acknowledged (20). However, the present investigation demonstrates that muscle sarcomere length, a heretofore largely uncontrolled or unquantified variable in in vivo microcirculation studies, may have a major impact on capillary perfusion. In this regard, it is-pertinent that in a previous study with preparations that demonstrated very straight, n.arrow capillari .es, there was a substanof nonflowing tial proportion capillaries (7). With respect to the present investigation, this scenario could be produced simply by stretchin .g the muscle to a sarcomere length of ~3.0 pm. Methodological
Considerations
In many instances, it was not possible to view preci .sely the same mu .scle field after alterin .g sarcomere length. Therefore, our measurements reflect biological variability both within and between muscles. In addition, there is evidence that capillaries increase in width -1 pm from their arteriolar to th .eir venular end (11). It was not pos #sible to visualize the entire length of each capillary from arteriole to venule in the present investigation. Therefore, diameter measurements were made at random locations along the capillary, and this source of variability adds some noise to the mean capillary diameter values reported herein. The measurement and interpretation of the capillary diameter data presumes that capillaries in vivo are circular. We are not aware of any compelling evidence for this. However, in the microvascular corrosion casts of Potter and Groom (35) made from rat cardiac and skeletal muscles, capillaries were close to circular. In addition, Mathieu-Costello (25) verified the circularity of capillaries morphometrically in perfusion-fixed rat hindlimb muscle. It may be argued that nonphysiological perfusion conditions may have been responsible for those findings. However, in the latter study, perfusion pressure was maintained within the physiological range. With respect to the quantification of the contribution of capillary tortuosity and branching to capillary length, it is possible that a minor degree of tortuosity in the z-axis was not measured. Thus, as discussed by Batra and Rakusan (2), the planar geometry quantified in this investigation probably u .nderestim ated the tr ue
MUSCLE
LENGTH
AND
MICROVASCULAR
contribution of capillary tortuosity by a small amount. Given that each capillary measured was in focus over the entire length measured, combined with the narrow focal depth of the OVM 1OOONM videomicroscope, we consider that this effect should have been minimal. The work of Henquell et al. (16) demonstrates that the minimum tube diameter through which rat red blood cells can pass in vitro is 2.45 pm. In vivo, however, there are structural elements associated with the endothelial surface that may effectively reduce that portion of the capillary lumen available for red blood cell flow (8, 9). These include endothelial cell projections and highly charged proteoglycans, which are not visible via intravital microscopy. These factors are held to produce a slow-moving plasma layer of -0.5-1.0 pm thickness in close association with the capillary luminal wall (8, 9). Thus it is likely that the lower limit of capillary luminal diameter that will permit red blood cell flux in vivo is somewhat higher than that found in vitro. We observed instances of red blood cell flow in capillaries as narrow as 2.5 pm in luminal diameter. However, in these, the velocity was very low and passage of the cell appeared to entail distension of the capillary lumen to -3.3 pm. Immediately after the passage of each cell, the capillary luminal diameter returned to its previous dimensions. In summary, we did not find that muscle stretching reduced capillary diameter to such an extent that there was a population of capillaries so narrow that they precluded red blood cell flow. However, the reduced capillary diameter probably increased resistance within individual capillaries, thereby exacerbating the heterogeneity of V RBc (increased coefficient of variation) through the capillary bed. From these results, it cannot be resolved whether the increased proportion of capillaries with impeded flow was due to effects in the capillaries themselves or alternatively within larger vessels. However, the interspersion of capillaries with vigorous flow and those impeded does suggest that some flow impedance may occur within the capillary itself. Irrespective of the precise site, the net effect of this behavior is that mean capillary VRBc is reduced and the proportion of capillaries with stopped or intermittent flow is increased when muscles are stretched. This condition must ultimately impair oxygen delivery and also metabolite removal. These processes occur at a time when bulk muscle blood flow is reduced and, in combination with the impaired capillary hemodynamits, this scenario will be expected to severely impair muscle performance and enhance fatigability. NOTE
ADDED
IN PROOF
While this manuscript was in press, Welsh and Segal (Circ. Res. 79: 551459, 1996) reported that stretching of the hamster retractor muscle elicited an arteriolar vasoconstriction (transduced via periarteriolar sympathetic nerves) and reduced blood flow. Treatment with tetrodotoxin, phentolamine, or prazosin abolished this vasoconstrictor response and restored blood flow to -70 to 85% of control values. These experiments suggest that active arteriolar vasoconstriction can account for approximately two-thirds of the stretchinduced reduction in muscle blood flow. Presumablv. the
STRUCTURE
AND
HZ113
FUNCTION
remaining one-third may be due to mechanical the arteriolar and capillary levels.
alterations
at
We thank Karen Sue Hageman for assistance with the measurement of spinotrapezius blood flow. In addition, we are grateful to Dr. J. L. Kilgore for helpful comments during the preparation of the manuscript and Kristen Schweitzer and T. W. Funk for technical assistance. This work was supported, in part, by grants from the National Institutes of Health (HL-50306, HL-17731, and AG-11535) and a Kansas State Veterinary Medical Sciences Dean’s Fund grant (96607). Address reprint requests to D. C. Poole. Received
24 April
1996;
accepted
in final
form
25 September
1996.
REFERENCES 1. Ameredes, B. T., W. F. Brechue, and W. N. Stainsby. Preload release increases blood flow and decreases fatigue during repetitive isotonic muscle contractions. J. Appl. Physiol. 77: 26412647,1994. 2. Batra, S., and K. Rakusan. Capillary length, tortuosity, and spacing in rat myocardium during cardiac cycle. Am. J. PhysioZ. 263 (Heart Circ. Physiol. 32): H1369-H1376,1992. 3. Bourdeau-Martini, J., C. L. Odoroff, and C. R. Honig. Dual effect of oxygen on magnitude and uniformity of coronary intercapillary distance. Am. J. PhysioZ. 226: 800-810, 1974. 4. Burton, K. S., and P. C. Johnson. Reactive hyperemia in individual capillaries of skeletal muscle. Am. J. PhysioZ. 223: 517-524,1972. 5. Cutts, A. Sarcomere length changes in muscles of the human thigh during walking. J. Anat. 166: 77-84,1989. 6. Damon, D. H., and B. R. Duling. Distribution of capillary blood flow in the microcirculation of the hamster: an in vivo study using epifluorescent microscopy. Microuasc. Res. 27: 81-95, 1984. 7. Dawson, J. M., K. R. Tyler, and 0. Hudlicka. A comparison of the microcirculation in rat fast glycolytic and slow oxidative muscles at rest and during contractions. Microuasc. Res. 33: 167-182,1987. 8. Desjardins, C., and B. R. Duling. Heparinase treatment suggests a role for the endothelial cell glycocalyx in regulation of capillary hematocrit. Am. J. PhysioZ. 258 (Heart Circ. Physiol. 27): H647-H654,1990. 9. Duling, B. R., I. H. Sarelius, and W. F. Jackson. Acomparison of microvascular estimates of capillary blood flow with direct measurements of total striated muscle flow. Znt. J. Microcirc. Clin. Exp. 1: 409-424,1982. 10. Ellis, C. G., 0. Mathieu-Costello, R. F. Potter, I. C. MacDonald, and A. C. Groom. Effect of sarcomere length on total capillary length in skeletal muscle: in vivo evidence for longitudinal stretching of capillaries. Microuasc. Res. 40: 63-72,199O. 11. Eriksson, E., and R. Myrhage. Microvascular dimensions and blood flow in skeletal muscle. Acta PhysioZ. Stand. 86: 211-222, 1972. 12. Flaim, S. F., S. H. Nellis, E. J. Toggart, H. Drexler, K. Kanda, E. D. Newman. Multiple simultaneous determinations of hemodynamics and flow distribution in conscious rat. J. Pharmacol. Methods 11: l-39,1984. 13. Friden, J., and R. L. Lieber. Physiologic consequences of surgical lengthening of extensor carpi radialis brevis muscletendon junction for tennis elbow. J. Hand. Surg. 19A: 269-274, 1994. 14. Gray, S. D. Rat spinotrapezius muscle preparation for microscopic observation of the terminal vascular bed. Microuasc. Res. 5: 395-400,1973. 15. Gray, S. D., P. F. McDonagh, and R. W. Gore. Comparison of functional and total capillary densities in fast and slow muscles of the chicken. Pfliigers Arch. 397: 209-213,1983. 16. Henquell, L., P. L. LaCelle, and C. R. Honig. Capillary diameter in rat heart in situ: Relation to erythrocyte deformability, 02 transport, and transmural 02 gradients. Microuasc. Res. 12: 259-274,1976.
H2114
MUSCLE
LENGTH
AND
MICROVASCULAR
17. Heyman, M. A., B. D. Payne, J. I. E. Hoffman, and A. M. Rudolph. Blood flow measurements with radionuclide-labeled particles. Prog. Cardiovasc. Dis. 20: 55-79, 1977. 18. Hirche, H. J., W. K. Raff, and D. Grun. The resistance to blood flow in the gastrocnemius of the dog during sustained and rhythmical isometric and isotonic contractions. PfZiigers Arch. 314: 97-112,197o. 19. Ishikawa, H., H. Sawada, and E. Yamada. Surface and internal morphometry of skeletal muscle. In: Handbook ofphysiology. SkeZetaZ MuscZe. Bethesda, MD: Am. Physiol. Sot. 1983, sect. 10, chapt. 1, p. 1-21. 20. Kayar, S. R., and N. Banchero. Sequential perfusion of skeletal muscle capillaries. Microvasc. Res. 30: 298-305, 1985. 21. Krogh,A. The supply of oxygen to the tissues and the regulation of the capillary circulation. J. Physiol. (Land.) 52: 457-474,1919. 22. Ledvina, M. A., and S. S. Segal. Sarcomere length and capillary curvature of rat hindlimb muscles in vivo. J. AppZ. PhysioZ. 78: 2047-2051, 1995. 23. Lieber, R. L., G. J. Loren, and J. Friden. In vivo measurement of human wrist extensor muscle sarcomere length changes. J. NeurophysioZ. 71: l-8, 1994. 24. Mathieu, O., L.-M. Cruz-Orive, H. Hoppeler, and E. R. Weibel. Estimating length density and quantifying anisotropy in skeletal muscle capillaries. J. Microsc. 131: 131-146,1983. 25. Mathieu-Costello, 0. Capillary tortuosity and degree of contraction or extension of skeletal muscles. Microvasc. Res. 33: 98-117, 1987. 26. Mathieu-Costello, O., C. G. Ellis, R. F. Potter, I. C. MacDonald, andA. C. Groom. Muscle capillary-to-fiber perimeter ratio: morphometry. Am. J. Physiol. 261 (Heart Circ. PhysioZ. 30): H1617-H1625,1991. 27. Mathieu-Costello, O., H. Hoppeler, and E. R. Weibel. Capillary tortuosity in skeletal muscles of mammals depends on muscle contraction. J. AppZ. Physiol. 66: 1436-1442, 1989. 28. Mazzoni, M. C., T. C. Skalak, and G. W. Schmid-Schiinbein. Effects of skeletal muscle fiber deformation on lymphatic volumes. Am. J. PhysioZ. 259 (Heart Circ. PhysioZ. 28): H1860H1868,1990. 29. Mermod, L., H. Hoppeler, S. R. Kayar, R. Straub, and E. R. Weibel. Variability of fiber size, capillary density, and capillary length related to horse muscle fixation processes. Acta Anat. 133: 89-95,1988.
STRUCTURE 30.
31.
32. 33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
AND
FUNCTION
Musch, T. I., and J. A. Terrell. Skeletal muscle blood flow abnormalities in rats with a chronic myocardial infarction: rest and exercise. Am. J. Physiol. 262 (Heart Circ. PhysioZ. 31): H411-H419,1992. Nakao, M., and S. S. Segal. Muscle length alters geometry of arterioles and venules of hamster retractor. Am. J. PhysioZ. 268 (Heart Circ. PhysioZ. 37): H336-H344,1995. Pollack, G. H. The cross-bridge theory. Physiol. Rev. 63: 10491113,1983. Poole, D. C., and 0. Mathieu-Costello. Capillary and fiber geometry in rat diaphragm perfusion fixed in situ at different sarcomere lengths. J. AppZ. Physiol. 73: 151-159,1992. Poole, D. C., 0. Mathieu-Costello, and J. B. West. Capillary tortuosity in rat soleus muscle is not affected by endurance training. Am. J. PhysioZ. 256 (Heart Circ. Physiol. 25): HlllOH1116,1989. Potter, R. F., and A. C. Groom. Capillary diameter and geometry in cardiac and skeletal muscle studied by means of corrosion casts. Microvasc. Res. 25: 68-84,1983. S&mid-Schiinbein, G. W., B. W. Zweifach, and S. Kovalcheck. The application of stereological principles to morphometry of the microcirculation in different tissues. Microvasc. Res. 14: 303-317,1977. Skalak, T. C., and G. W. S&mid-Schonbein. The microvasculature in skeletal muscle IV A model of the capillary network. Microvasc. Res. 32: 333-347,1986. Smaje, L., B. W. Zweifach, and M. Intaglietta. Micropressures and capillary filtration coefficients in single vessels of the cremaster muscle of the rat. Microvasc. Res. 2: 96-110,197O. Stainsby, W. N. Oxygen uptake for negative work, stretching contractions by in situ dog skeletal muscle. Am. J. Physiol. 230: 1013-1017,1976. Supinski, G. S., H. Bark, A. Buanciale, and S. G. Kelsen. Effect of alterations in muscle fiber length on diaphragm blood flow. J. AppZ. Physiol. 60: 1789-1796, 1986. Suzuki, H., D. C. Poole, B. W. Zweifach, and G. W. SchmidSchiinbein. Temporal correlation between maximum tetanic force and cell death in postischemic rat skeletal muscle. J. CZin. Invest. 96: 2892-2897,1995. Wisnes, A., and A. Kirkebo. Regional distribution of blood flow in calf muscles of rat during passive stretch and sustained contraction. Acta PhysioZ. Stand. 96: 256-266, 1976.
