Chronic heart failure (CHF) is a major cause of mortality and morbidity in the Western world. The incidence of heart failure is even increasing, because of greater life expectancy and improved medical care of acute cardiac problems. CHF is often associated with exercise intolerance, often with disabling dyspnea and fatigue to such an extent that even activities of daily life can be affected (28). The reduced exercise capacity shows no direct relationship with the pump-function of the heart, and recent studies have indicated that changes in skeletal muscle play a significant role in this exercise intolerance (12,28). Finally, there is evidence that the peripheral limitation of exercise performance even leads to progression of the syndrome via the muscle ergoreflex system and contributes to the risk of sudden death (4).
Skeletal muscle atrophy and consequently muscle weakening occur during CHF (12,13,20). In addition, the fatigue resistant Type I fibers tend to be replaced by the more fatigue-prone Type II fibers (5,12,21). This and the reduced activity of oxidative enzymes contribute to exercise intolerance during chronic heart failure (20,27).
The exchange of nutrients, oxygen, carbon dioxide, metabolites and heat takes place in the capillaries between blood and tissue. Consequently, a decreased capillarization may add to exercise intolerance during CHF. In normal skeletal muscle, the capillary supply to a muscle fiber is coupled to its metabolic profile, the metabolic profile of the surrounding fibers, and its size (8). It is conceivable that the changes in fiber type composition and atrophy during CHF may change this coupling and further aggravate exercise intolerance.
It has been shown in both rats and humans that a reduced activity level cannot completely explain the changes in skeletal muscle during CHF (27,29). The attenuated exercise-induced increase in blood flow in skeletal muscle of rats with CHF (24) likely results in intermittent hypoxia. Indeed, the greatly improved oxygen extraction by muscles of exercising individuals with CHF (16) implies that in particular at the venular site of the capillary network the muscle fibers are exposed to at least a relative hypoxia. Hypoxia has been shown to cause atrophy and changes in myosin heavy chain (MyHC) composition (2,9). In addition, there is evidence that endothelial cell apoptosis and relative ischemia trigger preferential synthesis of fast MyHC isoforms and induce myofiber apoptosis (30). Thus, changes in the microcirculation might play a crucial role in, and even precede, the development of the myopathy accompanying CHF. Therefore, it is important to obtain a better understanding of possible changes in the microcirculation of skeletal muscle.
The aim of the present study was to assess changes in capillary supply in the plantaris muscle and the diaphragm of the rat aorta-caval fistula (ACF) volume overload model for CHF. It was hypothesized that the capillary supply of plantaris and diaphragm muscles would be reduced during volume overload and that the coupling between the capillary supply to a fiber and its size, metabolic type, and metabolic type of surrounding fibers would be compromised. The diaphragm was chosen because CHF is often associated with dyspnea and changes in morphology are larger in the diaphragm than limb muscles (18).
MATERIALS AND METHODS
Approach to the problem and experimental design.
Changes in skeletal muscle capillary supply during CHF may contribute to exercise intolerance. Therefore, the aim of the present study was to determine the effects of cardiac hypertrophy induced by volume overload on the capillary supply of the diaphragm and the m. plantaris, a fast-twitch limb muscle. Cardiac volume overload was induced in rats by creating an aorta-caval fistula, whereas rats undergoing the same surgery but without formation of the anastomosis served as controls. This model has been suggested as a relevant tool for studies of the circulatory effects of heart failure (10,11). Six weeks postsurgery, the hypertrophy of the heart was determined, as well as the capillary supply of the diaphragm and plantaris muscles. It was evaluated whether the relation between fiber type and capillary supply was maintained in skeletal muscle of rats with cardiac volume overload, i.e., numerical capillary density per fiber type and capillary number per fiber of a certain type.
Animals.
The institutional animal use and care committee provided approval for the experiments described below. This approval is consistent with ACSM animal care standards that were followed by adhering to the recommendations for the care of laboratory animals as advocated by the American Association for Accreditation of Laboratory Animal Care (AAALAC) and following the policies and procedures detailed in the Guide for the Care and Use of Laboratory Animals as published by the U.S. Department of Health and Human Services and proclaimed in the Animal Welfare Act (PL89-544, PL91-979, and PL94-279).
Male Sprague-Dawley rats weighing 300–360 g were obtained from Holtzman Co. (Madison, WI). They were kept at a 12-h light:12-h darkness period per 24 h. Food (Purina Rat Chow, Ralston Purina) and water were provided ad libitum.
Induction of chronic heart failure.
The rats were randomly divided into an aorta caval fistula (ACF) and a sham (S) group. All animals were anesthetized with an i.p. injection of 1 mL·kg−1 equithesin (19.2-mL sodium pentobarbital, 2.125-g chloral hydrate, 21.4-mL propylene glycol, 5.5-mL 95% ethanol, and 3.9-mL 0.9% saline). In the ACF group, volume overload was induced by an end-to-side anastomotic shunt between the abdominal aorta and the inferior vena cava as described by Liu et al. (19). The shunt was approximately 1.5 mm in diameter, as it has been shown that a shunt of this size causes more than a 50% increase in heart weight in 1 month, whereas smaller diameters induce less hypertrophy (19,11). In addition, a shunt of this size results in approximately 50% mortality within 48 h (19), whereas larger diameters will increase the mortality dramatically. In the S group, a similar operation was performed, except for the formation of the arteriovenous shunt. We began with six animals in the S and 10 in the ACF group. Six animals of the ACF group died within 48 h, leaving four animals in this group that were used for further analysis.
Terminal experiment.
After 6 wk, the rats were anesthetized with sodium pentobarbital (70 mg·kg−1 i.p.). The heart, plantaris, and soleus muscles and part of the diaphragm were excised, weighed, and immediately frozen in liquid nitrogen. This way of freezing does not disrupt the morphology to any greater extent than when freezing with isopentane suspended in liquid nitrogen (1). The soleus and plantaris muscles were frozen at slightly stretched lengths to prevent bias in capillary parameters caused by tortuosity of capillaries and effects on fiber cross-sectional areas when the muscle is flaccid. The heart was emptied from excess blood before weighing, by letting it beat in ice-cold saline. The frozen tissues were stored at −80°C until use.
Histochemistry.
Cross-sections (12 μm) of the diaphragm and the mid-belly of the plantaris muscle were cut at −20°C in a cryostat. The sections were stained for myofibrillar ATPase after preincubation at pH 4.55 to classify fibers as Type I (dark), Type IIa (light), and Type IIb (moderate). Serial sections were stained for capillaries with alkaline phosphatase as described previously (8).
Capillarization.
The capillarization was analyzed in the diaphragm and the deep oxidative and superficial glycolytic parts of the plantaris muscle by the method of capillary domains (8,14). In addition to the overall parameters of capillary supply, such as capillary density (CD) and capillary to fiber ratio (C/F), this method allows one to estimate the capillary supply with respect to fiber type and the heterogeneity of capillary spacing. The latter parameter can have a marked effect on the oxygenation of a muscle (6).
By using a digitizing tablet (model MMII 1201, Summagraphics, TX), capillary coordinates and fiber outlines were read from photomicrographs into a computer. The fiber cross-sectional area (FCSA) was derived from complete fiber outlines. The fiber type composition was assessed as number percentage and as area percentage (area occupied by a certain fiber type divided by the total area occupied by each of the fiber types). The amount of noncontractile material was estimated as the percentage area of the photomicrograph not occupied by muscle fibers.
Domains were constructed around the capillaries. A capillary domain is defined as the area surrounding a capillary delineated by equidistant boundaries from adjacent capillaries. Domains at the border of the photomicrograph were excluded from the analysis. The surface area of the domains was calculated. The radius (R) of a domain was calculated from a circle with the same surface area. R gives an indication of the maximal diffusion distance from the capillary to the edge of its domain. R shows a lognormal distribution, and therefore the LogSD of R is a measure for the heterogeneity of capillary spacing (6,8,14).
A common parameter used to assess the capillary supply to individual fibers is the number of capillaries around a fiber (CAF). This parameter does not take into account the fibers that lack direct capillary contact as potentially may occur in glycolytic muscle areas. The number of domains overlapping a fiber (DAF) is comparable to the CAF and does take into account fibers that may lack direct capillary contacts. Yet, these indices of capillary supply, i.e., CAF and DAF, are discontinuous. The local capillary to fiber ratio (LCFR), the sum of the surface fractions of domains overlapping a particular fiber, on the other hand, has a continuous distribution of values and also takes into account fibers that lack direct capillary contact. Therefore, this measure of the number of capillaries supplying a fiber is more sensitive to detect changes that may occur in the capillary to supply to muscle fibers than CAF and DAF. The LCFR divided by the FCSA will provide the capillary density for that fiber and is defined as the capillary fiber density (CFD) (see Fig. 1).
;)
FIGURE 1: Illustration of the interaction between domains and fibers. For each capillary (
crosses), only that fraction of the domain that actually overlaps the fiber (
dashed) is counted. All such fractions together for a fiber provide the local capillary to fiber ratio (LCFR). For the
dashed fiber, the LCFR is 0.96 (i.e., 0.34 of domain 1 + 0.34 of domain 2 + 0.28 of domain 3 = 0.96). The capillary fiber density (CFD) is the LCFR of a fiber divided by its fiber cross-sectional area (FCSA), i.e., 0.96/fiber area. (From [
8] with permission from Cambridge University Press).
Myofibrillar protein content and myosin heavy chain composition.
The dry/wet weight ratio, protein content, and myofibrillar protein content were determined in the heart, plantaris, soleus, and diaphragm muscles as described previously (7). In short, the tissue was speed freeze-dried for 1 h, and the dry weight determined. The tissue was homogenized and the total protein content determined by using a bicinchoninic acid (BCA) reagent (Pierce, Rockford, IL). The samples were subsequently used for the preparation of washed myofibrils, and the myofibrillar protein content was determined by using BCA reagent. The washed myofibrils were dissolved in a SDS sample buffer. About 0.5 μg of the washed myofibrils were used to determine the myosin heavy chain composition by SDS-PAGE. The acrylamide-bisacrylamide (37.5:1) concentrations were 4% (w/v) and 7% (w/v) in the stacking and separating gel, respectively. The samples were run for 25 h, 120 V at 15°C (7).
Statistics.
All data are presented as mean ± SEM. To test for the differences in protein content, MyHC composition of the muscles, and CD, C/F, LogSD and fiber type composition of the diaphragm, we used a two-sample t-test to assess the effect of ACF. To evaluate the effects of ACF and fiber type, and interactions between ACF and fiber type on the DAF, LCFR, and CFD of the diaphragm, we applied a repeated-measures ANOVA. The within subject factor was “fiber,” and it was assigned three levels (Type I, IIa, and IIb). The between-subject factor for the repeated measures ANOVA was ACF. In the plantaris muscle, we also used a repeated measures ANOVA; the within-subjects factor “area” had two levels (deep or superficial region), and the second within-subjects factor was, if applicable, “fiber,” which had three levels. The between subjects variable was ACF. No post hoc test was applied for ACF. A Tukey post hoc test was applied to assess differences between fiber types only if a fiber effect was detected. Where assumptions were violated, data were log - transformed before applying the tests. Effects, interactions, and differences were considered significant at P < 0.05.
RESULTS
Animal characteristics.
Body weights did not differ significantly between ACF and S rats. Hearts of ACF rats developed a significant hypertrophy (P < 0.002), a 96% and 72% higher heart weight and heart weight to body weight ratio than that of the S rats, respectively (Table 1). The lower increase in heart to body weight ratio than absolute heart weight may be related to the formation of edema, as evidenced by a puffy appearance of the rats, and in one particular case the presence of about 100 mL of fluid in the thorax and abdomen. The muscle weights did not differ significantly between the S and ACF rats (Table 1).
Table 1: Body (BW), heart (HW), and skeletal muscle weights of sham-operated (S) rats and rats with an aorta-caval fistula (ACF).
Protein content and myosin heavy chain composition.
The dry/wet weight ratio was decreased in the soleus and plantaris muscles (P < 0.01) of the ACF rats (Table 2), indicating that there was edema in those muscles. In line with this is the reduced total protein content in the soleus and the diaphragm muscles (P < 0.05) of the ACF rats.
Table 2: Dry/wet (D/W) weight ratio, total protein content (Tprot/W), myofibrillar protein content (Myo/W), and myofibrillar protein/total protein ratio (Myo/Tprot) of the heart, soleus, plantaris, and diaphragm muscles of sham-operated (S) rats and rats with an aorta-caval fistula (ACF).
Although the ratio myofibrillar protein/total protein content was not significantly affected in muscles of ACF rats (Table 2), Type IIb MyHC content was increased (P < 0.015) whereas Type IIa MyHC content was decreased (P < 0.05) in the plantaris muscles (Fig. 2). No significant shift in MyHC composition was observed in the soleus and diaphragm muscles (Fig. 2). Interestingly, in one soleus muscle of the ACF rats, a, α cardiac-like MyHC isoform (14%) was found. This isoform has also been observed occasionally in soleus muscles of rats treated with T3 and is thought to represent a transitional step in the slow to fast transition of MyHC isoforms or vice versa(32). The β MyHC content of the hearts did not differ significantly between the S (22.8 ± 7.4%, N = 6) and ACF (31.7 ± 7.3%, N = 3) rats.
FIGURE 2: Myosin heavy chain (MyHC) composition in soleus (A), plantaris (B), and diaphragm (D) muscles of sham-operated (white bars, N = 6) rats and rats with an aorta-caval fistula (black bars, N = 4). Values are mean ± SEM; * different from sham at P < 0.05.
Connective tissue content, fiber type composition, and FCSA.
The amount of connective tissue was not significantly affected by the ACF in either the diaphragm or the plantaris muscles (Table 3). The fiber type composition of the plantaris muscles did not differ significantly between S and ACF rats (Table 3). In the diaphragm muscle, however, the percentage of Type I and IIb fibers was increased (P < 0.05) in the ACF rats, at the expense of Type IIa fibers (P < 0.005) (Table 3). This coincided with a significant increase in the area percentage of Type I fibers and a decrease in the area percentage of Type IIa fibers (P < 0.005) (Table 3). ACF resulted in an increase in FCSA (P < 0.05) irrespective of fiber type but only in the deep part of the plantaris muscle (Fig. 3A). In the diaphragm, a significant interaction (P = 0.003) between fiber type and ACF was found for FCSA, indicating that fibers of different type responded differently to the experimental treatment (Fig. 3D).
Table 3: Connective tissue (%CT) content, number fiber type composition (%N), and area fiber type composition (%A) in the plantaris (plant), and diaphragm muscles of sham (S) rats and rats with an aorta-caval fistula (ACF).
FIGURE 3: The fiber cross-sectional area (FCSA) in deep (A) and superficial (B) part of the plantaris, the overall plantaris (C), and diaphragm (D) muscles of sham-operated (white bars, N = 6) rats and rats with an aorta-caval fistula (black bars, N = 4) Values are mean ± SEM; between-brackets N if different from N = 6 or N = 4; * fiber type effect, post hoc test indicates different from IIb at P < 0.001; interactions for plantaris: area·ACF P = 0.043, area·fiber type P = 0.008; interaction for diaphragm: fiber·ACF P = 0.003.
Capillarization.
The overall capillary supply was not significantly affected by ACF-induced volume overload in either the diaphragm or plantaris muscle (Table 4). Also, the heterogeneity of capillary spacing in the plantaris and diaphragm muscles was not significantly affected by ACF (Table 4). No effects of ACF were observed on the capillary supply of the plantaris muscle, as reflected by the absence of any significant effect on DAF, LCFR, and CFD (Table 4, Figs. 4 and 5). In the diaphragm, however, a significant interaction between fiber type and ACF (P = 0.002) was found for DAF, which indicates that fibers of a different type reacted differently to the ACF (Table 4). However, for LCFR no interaction between fiber type and ACF was found.
Table 4: Indices of overall capillary supply and domains overlapping a fiber (DAF) in the plantaris (plant) and diaphragm muscles of sham (S) rats and rats with an aorta-caval fistula (ACF).
FIGURE 4: The local capillary to fiber ratio (LCFR) in deep (A) and superficial (B) part of the plantaris, the overall plantaris (C) and diaphragm (D) muscles of sham-operated (white bars, N = 6) rats and rats with an aorta-caval fistula (black bars, N = 4) Values are mean ± SEM; between-brackets N if different from N = 6 or N = 4; * fiber type effect, post hoc test indicates different from IIb at P < 0.05; # area effect, deep larger than superficial at P = 0.001.
FIGURE 5: Capillary fiber density (CFD) in deep (A) and superficial (B) part of the plantaris, the overall plantaris (C) and diaphragm (D) muscles of sham-operated (white bars, N = 6) rats and rats with an aorta-caval fistula (black bars, N = 4). Values are mean ± SEM; between-brackets N if different from N = 6 or N = 4; fiber type effect, post hoc test indicates different from IIb at P < 0.01; # area effect, deep larger than superficial at P < 0.019; interaction for plantaris: area·fiber type P = 0.03.
The deep oxidative part of the plantaris muscle had a significantly higher C/F (P < 0.001) than the superficial glycolytic part (Table 4). This was also reflected in a significantly larger R (P = 0.007) and a lower CD (P = 0.01) in the glycolytic superficial part than the oxidative deep part of the plantaris muscle (Table 4). The absence of any significant effect of ACF or interaction between ACF and area indicates that this applied to muscles from both S and ACF rats. These observations confirm that the capillarization of a muscle or muscle region is at least partly determined by its fiber type composition. The significantly higher LCFR for each fiber type (Fig. 4, A and B) in the deep oxidative than the superficial glycolytic part of the plantaris muscle (P = 0.001) indicates that this held true at the level of the muscle fiber. The same was found for the CFD (P = 0.019;Fig. 5, A and B). This confirms that the metabolic surrounding of the fibers plays an important role also at the fiber level, which held true in the ACF rats as indicated by the absence of interactions between fiber type and ACF.
The higher CFD of Type I and IIa fibers than that of Type IIb fibers in both the plantaris and the diaphragm muscle (P < 0.01) indicates that the metabolic type of the fiber determines its capillary supply to a large extent.
That FCSA may even be the most important determinant of capillary supply is reflected by the significantly higher DAF and LCFR (P = 0.01) for Type IIb than Type I and IIa fibers in both the diaphragm and plantaris muscle (Table 4). In both muscles, the Type IIb (glycolytic) fibers have a larger (P < 0.001) FCSA (Fig. 3) than Type IIa and I fibers. In the ACF rats, this held true, as reflected by the absence of any significant interaction between volume overload and fiber type for DAF and LCFR (Table 3, Fig. 4). In summary, 1) the coupling between capillary supply to a muscle, muscle region, or fiber and the size, metabolic surrounding, and metabolic character of the fiber that exists in control plantaris muscles is also present in the diaphragm, and 2) this coupling is maintained in the muscles of the ACF rats.
DISCUSSION
The main finding of the present study is that the capillarization of both the fast-twitch plantaris and diaphragm muscles is both quantitatively and qualitatively similar in healthy rats and rats with an ACF. Not only was the overall capillarization of the muscles maintained but also the relation between fiber size, metabolic type of the fiber, and metabolic type of surrounding fibers on the one hand and the capillary supply to a fiber on the other.
In spite of the small number of rats that survived the fistula surgery, statistical assumptions were not violated. In addition, power analysis indicated that four animals were adequate to detect a 25% difference at a power of 0.80. To detect differences of 10% between the groups, we would have had to examine another 20–40 animals in each group (depending on the parameter tested). For example, if we had examined another 10 animals in each group, based on the power calculations, the outcome of the study would not have been affected, and our conclusions would not have changed.
The hearts developed a 72% hypertrophy 6 wk after induction of volume overload, similar or even larger than what has been obtained by others (3,10,11,19). In female rats, the hypertrophy may lead to compensated CHF, as evidenced by near normal heart function after 1 month (19). However, the occurrence of heart failure is greater in male than female rats with this model (22). Indeed, the plasma norepinephrine and epinephrine levels were elevated 2 months after formation of an ACF in male rats that induced a less-pronounced cardiac hypertrophy (∼20%) than we obtained (10,11). Furthermore, in a similar male rat model with about 30% cardiac hypertrophy the plasma atrial natriuretic peptide was significantly elevated and correlated with myocardial weight (3). In addition, the rate of phosphocreatine depletion in contracting skeletal muscle was increased (3), and skeletal muscle blood flow is reduced both at rest (11) and during exercise (10). Based on their observations, Flaim et al. (11) concluded that the ACF model is a relevant tool for studies of the circulatory effects of heart failure. Therefore, the model seems adequate to study changes in the microcirculation of skeletal muscle as a consequence of CHF in male rats.
In contrast to the loss of muscle bulk observed by others during CHF (13,20) we did not find a decrease in muscle weight, or atrophy of muscle fibers. The increase in fiber size in the deep part of the plantaris muscle is most likely related to fiber swelling caused by edema, as reflected by the decreased dry/wet weight ratio in the limb muscles. Also the protein content was decreased in the soleus muscle and the diaphragm in rats with an ACF. This is in line with the muscle weakening that was greater than predicted from the decrease in muscle cross-sectional area in individuals with CHF, indicating both quantitative and qualitative abnormalities of skeletal muscle (13), such as edema or a reduced myofibrillar protein content.
The amount of myofibrillar protein relative to the total amount of protein in the muscles was not significantly affected by ACF. Yet, in the fast-twitch plantaris muscle, there was a significant increase in the Type IIb MyHC content at the expense of Type IIa MyHC. Other investigators have also observed slow-to-fast transitions in skeletal muscle during CHF (27,30). In the diaphragm and soleus muscles, no significant transition was observed. If ischemia and/or hypoxia are involved in the MyHC transitions (2,9) during CHF, then the elevated diaphragmatic blood flow during submaximal exercise (23) may have attenuated such a MyHC shift in the diaphragm.
Despite the shift in MyHC composition, we did not observe a significant shift in fiber type composition in the plantaris and soleus muscles during CHF. In the diaphragm, however, there was a significant increase in Type I and IIb fibers at the expense of Type IIa fibers. Similarly, Howell and colleagues (15) observed that the fiber type composition of the latissimus dorsi muscle remained unchanged, whereas the percentage of type I fibers increased at the expense of Type IIa fibers in the diaphragm of minipigs with congested heart failure. Interestingly, in cardiomyopathic hamsters, the oxidative and glycolytic enzyme activities were affected only in the proximal but not in the distal muscles (26). These data clearly indicate that muscles do not react uniformly to CHF.
Although the histochemical data do seem to indicate some deviation from the MyHC isoform composition data, it should be kept in mind that the classification of fiber types and MyHC isoforms are not completely parallel; e.g., with the histochemical method we employed, we were able to distinguish only Type I, IIa, and IIb fibers, whereas with gel electrophoresis we could distinguish Type I, IIa, IIx/d, and Type IIb MyHC isoforms. In addition, muscles very often contain hybrid fibers, coexpressing more than one MyHC isoform (7), whose number is likely to be increased during CHF. However, we did not find intermediately stained fibers, indicating that if hybrid fibers are present, they contain significant amounts of one MyHC isoform and only a minor amount of an additional MyHC isoform.
During CHF, peripheral muscles (12,20,21) undergo a shift from oxidative toward glycolytic metabolism, whereas the energy metabolism of the diaphragm shows the opposite (12). Based on these observations, one would expect a decrease in the capillarization of peripheral muscles and an increase in that of the diaphragm, as there is generally a coupling between the metabolic type of a fiber and its capillary supply (8). Yet, the literature is equivocal concerning changes in skeletal muscle capillary supply during CHF. C/F has been reported to be unchanged (5,21) or to be reduced (31) in CHF. Similarly, CD is decreased (20) or unchanged (31), and linear capillary density even increased (17). We did not find any change in CD or C/F in the diaphragm and plantaris muscles with an ACF. In addition, the heterogeneity of capillary spacing, which may affect the oxygenation of a muscle (6), remained unaffected. Obviously, the degree of atrophy influences the data on CD and may explain part of the contradictory observations. Nevertheless, the unchanged indices of global capillary supply indicate that the diffusion distances and surface area for the exchange of, e.g., oxygen between the capillaries and skeletal muscle tissue is maintained with volume overload CHF.
The C/F may not be sensitive enough to detect changes in capillary supply. In addition, indices of overall capillary supply do not provide information about the local capillary supply to fibers of different types. Therefore, we used a more sensitive method to estimate the capillary supply to a fiber. The interaction between volume overload and fiber type for DAF (similar to capillaries around a fiber) in the diaphragm suggested a change in capillary supply to myocytes in this muscle. However, to calculate DAF, a capillary is completely assigned to a particular fiber, whereas to calculate LCFR (number of capillaries supplying a particular fiber), one takes into account that a capillary supplies more than one fiber. Therefore, we consider the LCFR a more reliable parameter of capillary supply to a fiber, and the absence of any interaction between volume overload and fiber type for LCFR indicated that changes in capillary supply, if they did occur, were only minimal. This corresponds with the absence of any significant change in parameters of overall capillary supply to the diaphragm. An additional explanation is that the change in fiber type composition in the diaphragm, a Type IIa fiber transformed to IIb, would lower the DAF for Type IIb fibers, without any change in the microvasculature. No further significant effects of volume overload were observed in either the diaphragm or plantaris muscle.
The maintained LCFR in both the diaphragm and plantaris muscle indicates that capillary loss did not occur. Furthermore, the absence of any significant interaction between fiber type and volume overload, either in terms of LCFR or CFD, indicates that the coupling between LCFR and CFD with fiber size, metabolic type, and metabolic type of surrounding fibers (8) was maintained during CHF in both plantaris and diaphragm muscles, despite changes in MyHC and fiber type composition. However, it is possible that in proximal muscles changes in capillarization might occur, because enzyme activities have been shown to respond to CHF in proximal but not distal muscles in the cardiomyopathic hamster (26). Nonetheless, the pattern emerges that the capillarization of skeletal muscle is only minimally affected, if at all, during CHF. The maintained capillarization during CHF would contribute to the near maximal oxygen extraction that has been observed in exercising patients with CHF (16). The maintained capillary number and coupling between fiber type and size with capillary supply in the plantaris muscle of the ACF rats is in line with preliminary data that indicate that the isometric contractile properties and fatigability of the plantaris muscle are unaffected in ACF rats (unpublished observations).
Finally, models of CHF differ in their effect on heart function, and this may provide an additional explanation for the conflicting data on skeletal muscle capillarization; e.g., left coronary artery ligation results in a decreased cardiac output (25), whereas volume overload is accompanied by an increase in cardiac output (11,19). However, the aorta-caval shunt equals about 50% of the total cardiac output (10,11), and taking the shunting into account, cardiac output is decreased (10). Consequently, the increase in skeletal muscle blood flow during exercise is not only attenuated in a rat model of CHF induced by left coronary artery ligation (24) but also in CHF induced by an ACF (10). Further attenuation of increase in skeletal muscle blood flow may be related to the development of edema, as reflected by the reduced dry/wet weight ratio in the ACF rats, resulting in increased intramuscular pressure impeding blood flow.
In conclusion, the capillary to fiber ratio, capillary density, and heterogeneity of capillary spacing to the plantaris and diaphragm muscles are not significantly affected by CHF induced by volume overload. The unchanged capillary to fiber ratio indicates that no capillary loss occurred. In addition, the coupling between capillary supply to a fiber on the one hand, and its metabolic characteristics, metabolic characteristics of surrounding fibers and size on the other hand, is maintained in both diaphragm and plantaris muscles during CHF induced by volume overload. These data indicate that changes in the morphology of the vascular network only play a minor role in exercise intolerance during CHF.
The authors are grateful for the help of Linda Clark with the surgery. Jo Ann Moore contributed considerably in digitizing. Dr. Wesely has kindly provided the rats for this study, which were used for sampling small amounts of blood in another study.
Address for correspondence: S. E. Alway, Division of Exercise Physiology, West Virginia University School of Medicine, P.O. Box 9227, Robert C. Byrd Health Science Center, Morgantown, WV 26506-9227.
REFERENCES
1. Alway, S. E. Overload-induced c-Myc oncoprotein is reduced in aged skeletal muscle. J. Gerontol. 52A: B203–B211, 1997.
2. Bigard, A. X., H. Sanchez, O. Birot, and B. Serrurier.
Myosin heavy chain composition of skeletal muscles in young rats growing under hypobaric hypoxia conditions. J. Appl. Physiol. 88: 479–486, 2000.
3. Chati, Z., F. Zannad, C. Michel, et al. Skeletal muscle phosphate metabolism abnormalities in volume-overload experimental heart failure. Am. J. Physiol. 267: H2186–H2192, 1994.
4. Coats, A. J. S. The “muscle hypothesis” of chronic heart failure. J. Mol. Cardiol. 28: 2255–2262, 1996.
5. Davidse, J. H. L., C. M. H. B. Lucas, M. J. A. P. Daemen, O. C. K. M. Penn, and F. H. van der Veen. Skeletal muscle adaptation in patients with severe heart failure: implications for aorta- and cardiomyoplasty. Basic Appl. Myol. 7: 23–30, 1997.
6. Degens, H., B. E. M. Ringnalda, and L. J. C. Hoofd. Capillarisation, fibre types and myoglobin content of the dog gracilis muscle. Adv. Exp. Med. Biol. 361: 533–539, 1994.
7. Degens, H., M. Soop, P. Höök, O. Ljungqvist, and L. Larsson. Post-operative effects on insulin resistance and specific tension of single human skeletal muscle fibres. Clin. Sci. 97: 449–455, 1999.
8. Degens, H., Z. Turek, L. J. C. Hoofd, M. A. van‘t Hof, and R. A. Binkhorst. The relationship between capillarisation and fibre types during compensatory hypertrophy of the plantaris muscle in the rat. J. Anat. 180: 455–463, 1992.
9. Dodd, S. L., I. S. Vrabas, and D. S. Stetson. Effects of intermittent ischemia on contractile properties and myosin isoforms in skeletal muscle. Med. Sci. Sports Exerc. 30: 850–855, 1998.
10. Flaim, S. F., and W. J. Minteer. Ventricular
volume overload alters cardiac output distribution in rats during exercise. J. Appl. Physiol. 49: 482–490, 1980.
11. Flaim, S. F., W. J. Minteer, S. N. Nellis, and D. P. Clark. Chronic arteriovenous shunt: evaluation of a model for chronic heart failure in rat. Am. J. Physiol. 236: H698–H704, 1979.
12. Gosker, H. R., E. F. M. Wouters, G. J. van der Vusse, and A. M. W. J. Schols. Skeletal muscle dysfunction in chronic obstructive pulmonary disease and chronic heart failure: underlying mechanisms and therapy prospectives. Am. J. Clin. Nutr. 71: 1033–1047, 2000.
13. Harrington, D., S. D. Anker, T. P. Chua, et al. Skeletal muscle function and its relation to exercise intolerance in chronic heart failure. J. Am. Coll. Cardiol. 30: 1758–1764, 1997.
14. Hoofd, L. J. C., Z. Turek, K. Kubat, B. E. M. Ringnalda, and S. Kazda. Variability of intercapillary distance estimated on histological sections of rat heart. Adv. Exp. Med. Biol. 191: 239–247, 1985.
15. Howell, S., J-M. I. Maarek, M. Fournier, K. Sullivan, W-Z. Zhan, and G. C. Sieck. Congestive heart failure: differential adaptation of the diaphragm and latissimus dorsi. J. Appl. Physiol. 79: 389–397, 1995.
16. Katz, S. D., C. Maskin, G. Jondeau, T. Cocke, R. Berkowitz, and T. Lejemtel. Near-maximal fractional oxygen extraction by active skeletal muscle in patients with chronic heart failure. J. Appl. Physiol. 88: 2138–2142, 2000.
17. Kindig, C. A., T. I. Musch, R. J. Basaraba, and D. C. Poole. Impaired capillary hemodynamics in skeletal muscle of rats in chronic heart failure. J. Appl. Physiol. 87: 652–660, 1999.
18. Lindsay, D. C., C. A. Lovegrove, M. J. Dunn, et al. Histological abnormalities of muscle from limb, thorax and diaphragm in chronic heart failure. Eur. Heart J. 17: 1239–1250, 1996.
19. Liu, Z., D. R. Hilbelink, W. B. Crockett, and A. M. Gerdes. Regional changes in hemodynamic and cardiac myocyte size in rats with aortacaval fistulas. 1. Developing and established hypertrophy. Cardiovasc. Res. 69: 52–58, 1991.
20. Magnusson, G., L. Kaijser, H. Rong, B. Isberg, C. Sylven, and B. Saltin. Exercise capacity in heart failure patients: relative importance of heart and skeletal muscle. Clin. Physiol. 16: 183–195, 1996.
21. Massie, B. M., A. Simonini, P. Sahgal, L. Wells, and G. A. Dudley. Relation of systemic and local muscle exercise capacity to skeletal muscle characteristics in men with congestive heart failure. J. Am. Coll. Cardiol. 27: 140–145, 1996.
22. Muffly, K. E., A. Thomas, J. A. Moore, L. Clark, and D. Hilbelink. Comparison of cardiac hypertrophy and failure in volume overloaded male and female rats. FASEB J. 13: A440, 1999.
23. Musch, T. I. Elevated diaphragmatic blood flow during submaximal exercise in rats with chronic heart failure. Am. J. Physiol. 265: H1721–H1726, 1993.
24. 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: H411–H419, 1992.
25. Pfeffer, M. A., J. M. Pfeffer, M. C. Fishbein, et al. Myocardial infarct size and ventricular function in rats. Circ. Res. 44: 503–512, 1979.
26. Punkt, K., and I. Erzen. Changes of enzyme activities in the myocardium and skeletal muscle fibres of cardiomyopathic hamsters: a cytophotometrical study. Exp. Toxicol. Pathol. 52: 103–110, 2000.
27. Simonini, A., C. S. Long, G. A. Dudley, P. Yue, J. McElhinny, and B. M. Massie. Heart failure in rats causes changes in skeletal muscle morphology and gene expression that are not explained by reduced activity. Circ. Res. 79: 128–136, 1996.
28. Sullivan, M. J., and M. H. Hawthorne. Exercise intolerance in patients with chronic heart failure. Prog. Cardiovasc. Dis. 38: 1–22, 1995.
29. Vescovo, G., F. Serafini, L. Facchin, et al. Specific changes in skeletal muscle
myosin heavy chain composition in cardiac failure: differences compared with disuse atrophy as assessed on microbiopsies by high resolution electrophoresis. Heart 76: 337–343, 1996.
30. Vescovo, G., R. Zennaro, M. Sandri, et al. Apoptosis of skeletal muscle myofibers and interstitial cells in experimental heart failure. J. Mol. Cell Cardiol. 30: 2449–2459, 1998.
31. Xu, L., D. C. Poole, and T. I. Musch. Effect of heart failure on muscle capillary geometry: implications for O
2 exchange. Med. Sci. Sports Exerc. 30: 1230–1237, 1998.
32. Yu, F., H. Degens, X. Li, and L. Larsson. Gender- and age-related differences in the regulatory influence of thyroid hormone on the contractility and myosin composition of single rat soleus muscle fibres. Pflügers Arch. 437: 21–30, 1998.
