The micocirculation of bone and marrow in the diaphysis of the rat hemopoietic long bones.код для вставкиСкачать
THE ANATOMICAL RECORD 233169-177 (1992) The Microcirculation of Bone and Marrow in the Diaphysis of the Rat Hemopoietic Long Bones L. DE SAINT-CEORCES AND S. C. MILLER Division of Radbbwlogy, School of Medicine, University of Utah, Salt Lake City, Utah 84112 ABSTRACT The nature of the mimirculation of the diaphyseal portion of long bones and the adjacent bone marrow is poorly understood. The purpose of this study was to describe the blood supply in the diaphyseal cortex and the relationship of the bone vascular circulation to that of the bone marrow in the growing rat. India ink-gelatin was infused in the arterial system of 3-month-old rats and the vascularization was determined from histological sections. In some studies the periosteal circulation was blocked but the nutrient and metaphyseal arteriole systems were left intact. In the growing rat, most of the vascular flow appears to be centripetally through the diaphyseal cortex and this appears to be the primary blood supply for the adjacent bone marrow. The India ink traversed the cortex and entered the marrow through osteal canals at the endocortical surface. At the marrow-endocortical bone surface interface, ink exiting from the osteal canals filled the adjacent marrow sinusoids in what appeared as “bush-like” structures. From the bone marrow the ink appeared to drain into the central vein. Some arterioles from the nutrient system were found to penetrate the inner two thirds of the cortical bone and then reenter the bone marrow. The centripetal flow of blood and the importance of the cortical flow for perfusion of the hemopoietic tissue was further documented when periosteal flow was obstructed. In this situation, the cortical bone and adjacent bone marrow were not perfused while the nutrient system and central vein were filled with ink. These results show that in the growing rat, the diaphyseal circulation is almost entirely centripetal and the adjacent hemopoietic tissues are perfused mostly with blood that must first traverse the cortical bone. 0 1992 Wiley-Lm, Inc. The long bones have three basic blood supplies: 1) the nutrient; 2) the metaphyseal and epiphyseal, which is combined after the growth plates are closed, and 3) the periosteal (Rhinelander, 1972). It is quite clear that the metaphyseal region, especially during endochondral growth, is supplied by the metaphyseal and nutrient system. There is, however, considerable diversity of opinion on the contribution of the afferent systems to the irrigation of the diaphyseal cortical bone as well as the adjacent diaphyseal hemopoietic tissues. In particular, it is not clear if the flow of blood in the diaphyseal cortical bone is centrifugal or centripetal and the exact territories supplied by the afferent nutrient and periosteal systems are not defined. While the microcirculation of most other organ systems is well defined, technical difficulties encountered with processing osseous tissues have likely made the study of the local vascularization of bone somewhat unappealing. The circulation of the middiaphyseal cortical bone in rabbits and rats was described by Brookes (1957, and 1971 for review) as a centrifugal blood flow that could be reversed in certain abnormal conditions, such as blockade of the nutrient afferent system. Later studies conducted by Rhinelander (1972) in dogs generally s u p ported centrifugal flow but noted that a periosteal s u p ply of blood for the outer third or quarter of the com0 1992 WILEY-LISS. INC. pacts only in areas adjacent to fascia1 attachments. De Bruyn et al. (1970) reached a different conclusion in studies with rodents and reported that the marrow and cortical blood flow are in series, that is, that the primary function of the nutrient arterial system was to supply the bone while most of the blood flow in the marrow sinuses was from vessels in the osteal canals. It was suggested that the blood flow through the osteal canals was from the periosteal capillaries and from osteal capillaries arising from the nutrient arterial system. In rats, vascular canals that enter the compacta from the periosteal and endocortical surfaces are usually orientated transverse to the long axis of the bone and the surface plane. This is usually the case in bones from animals, such as shorter-lived rodents, that lack Haversian systems (Ruth, 1953). In addition, the afferent nutrient system arising from the nutrient artery is very conspicuous in histological sections from rats. For Received August 6.1991; accepted November 22,1991. Dr. de Saint-Ceorgee is now at CEN/SCK, Biologie Department, Boeretang 200,B-2400Mol. Belgium. Address reprint ueeta to Dr.Scott C. Miller, Building 586,University of Utah, M%ke City, UT 84112. 170 L. DE SAINT-GEORGES AND S. C. MILLER Fig. 1. A plastic hose clamp (a) and a piece of latex tubing (b) ensure the compression of the periosteal vessels around the diaphysis in the position shown on this isolated rat femur. x 3. Fig. 2. India-ink-gelatin perfusion of the femoral diaphyseal shaft illustrating the cortical bone canals (cc), the marrow vascular spaces (MI, and the surrounding muscle vessels (ms). x 12. these reasons, the rat model should be a convenient system to study the respective contributions of blood supply from the periosteal vessels and nutrient arterial blood supply for the irrigation of the cortex and the bone marrow. This study describes the relative contribution of the nutrient and periosteal blood supplies to the irrigation of the bone cortex and the marrow in the diaphyseal regions of the long bone by using a n India ink-gelatin perfusion method with a periosteal blockage technique. MATERIALS AND METHODS Three-month-old male Sprague-Dawley rats (Simonsen Labs, Gilroy, CAI were used in this study. The animals were maintained and used according to the principles in the NIH Guide for the Care and Use of Laboratory Animals (1985) and guidelines established by the Institutional Animal Care and Use Committee. For anesthesia, the rats were given a n intraperitoneal injection of 1.5 ml of chloral hydrate (0.7 g/10 ml). In some rats, the periosteal circulation was blocked by local compression of the periosteum prior to perfusion with India ink-gelatin. For this, the muscles of the femoral shaft, just distal to the third trochanter, were carefully displaced. A piece of latex tubing was placed around the shaft of the bone and clamped with a small plastic hose clamp to ensure compression of the periosteal vessels (Fig. l). All rats were then perfused through the abdominal aorta with a mixture of polyvinylpyrrolidone, heparin, and procaine hydrochloride in saline to clear the blood from the vascular network (Forssmann et al., 1977). The perfusate was heated to BONE-BONE MARROW MICROCIRCULATION IN T H E RAT 40°C by transit through a plastic coil immersed in a water bath. The perfusion pressure was determined by gravity at 160 cm of H,O (14 cm Hg). When the blood cleared from the perfusate, the animals were then perfused with a India ink-gelatin mixture (Jee and Arnold, 1960). This consisted of 28 g purified calfskin gelatin (Eastman Kodak, Rochester, NY), 200 ml distilled water, and 200 ml India-ink. This was also kept at 40°C but was administered by manual pressure to achieve a flow rate of 0.5 mlhec. During the perfusion procedure, the animal was kept warm with a medicinal heating lamp. When the perfusion was completed, the abdominal aorta and vein were clamped and the limbs were placed on ice to solidify the gelatin in the perfusate. The femurs and tibias were removed and cut either at the diaphyseal-metaphysealjunctions or at the center of the diaphysis and placed in fixative. The bones were immersion fixed for 3 days in 4% glutaraldehyde and 1%formaldehyde in 0.1 M phosphate buffer a t pH 7.2. The bones were then decalcified, dehydrated, and embedded in epoxy resin. The bone pieces were sectioned longitudinally using a low speed diamond saw (Isomet, Buehler, Inc., U.S.A.), mounted on plastic slides, and ground to about 30 pm in thickness. Serial sections were collected and were necessary for the interpretation of some of the material. Because we were using histological and perfusion methods similar to that described by De Bruyn (1970), we found it appropriate to use the same terminology for the vessels as described in detail in his work. RESULTS The perfusion method employed in this study was very effective in filling the vascular network in bone and bone marrow (Fig. 2). While complete vascular perfusions were frequently obtained, most of our results were obtained from series of slides with incomplete perfusion areas, permitting the sequence of filling of the vascular network to be determined. The extensive vascular constitution of the marrow and bone cortex was clearly evident. The marrow vascular space was distinctly demarcated from the surrounding hemopoietic tissues as no leakage of the ink from the vessels into the domains of hemopoietic cells was observed. The metaphyseal region, which in these animals is a region of rapid growth and remodeling, generally appeared to be the first part of the bone to be perfused (Fig. 3), followed by the large central sinusoid in the diaphysis (Fig. 4).The central sinus was filled with ink even when the adjacent bone cortex and hemopoietic tissues showed little evidence of perfusion. These observations were frequently noted and indicated the principal role of the nutrient system in supplying the growing region of the bone. These observations also emphasized the importance of the central sinus as the drainage of the metaphysis and nutrient system. The blood supply from the nutrient system in the metaphysis appeared to reach the metaphyseal side of the epiphyseal growth plate and then loop back (Fig. 5) towards the metaphyseal sinusoids. Typically, the perfusion of the marrow hemopoietic tissue was observed only when the diaphyseal cortex vascular channels were filled. Thus, the perfusion appeared to be in a centripetal direction. At the exits of the transverse cortical canals on the endocortical sur- 171 face, the vessels formed “bush-like” patterns as they entered the marrow (Fig. 6). In perfusions that were more complete, the individual “bush-like” vascular patterns in the hemopoietic tissue were progressively obscured, suggesting lateral connections between these vascular domains in the marrow (Fig. 7). The perfusion of the marrow sinusoids and the disappearance of the bush-like formations progressively delineated the collecting sinusoids (Fig. 7). Very narrow dividing branches of the nutrient artery (Figs. 8, 9) were seen in the marrow but were never observed with a sinusoid perfusion without simultaneous perfusion of the cortex. In some preparations, some cortical vessels appeared to arise from the nutrient artery and usually curved back towards the bone marrow (Fig. 8). These vessels entered at the endocortical surface and appeared to penetrate into the inner two thirds of the cortex before exiting at the same surface. It was not possible to determine if these vessels also formed a distinct marrow supply or if they communicated with the transcortical vessels before reaching the marrow sinusoids. The serial disposition of the blood supply to the cortical bone and the adjacent marrow was further investigated in the bones where periosteal blood flow was blocked by a clamp. Not only did the periosteal blockade prevent the underlying cortical bone from being perfused but also prevented the perfusion of the underlying bone marrow (Fig. 10). The nutrient arterial system appeared therefore to be incapable of filling the marrow in the diaphyseal region where the periosteal flow was obstructed (Fig. 11).The perfusion of the nonblocked regions of the diaphysis and the filling of the central sinusoid appear to exclude a lack of ink filling due to a n exit blockade. DISCUSSION In general, our data supports the results of De Bruyn et al. (19701, who suggested that the flow of blood in rat, guinea pig, and rabbit bone is centripetal in the cortex. The histological images obtained in the perfused material indicates that the blood supply to the bone marrow in the diaphyseal regions of the long bones in rats is transosteal. The “bush-like’’ vascular structures of the marrow that arose from the ink exiting the osteal canals at the endocortical surface appeared to be progressively filled and were often observed not yet connected to the central sinus. This provides evidence for a centripetal flow of blood but does not support the suggestion that in the rat a substantial centrifugal flow might occur, even in the outer portion of the cortex (Rhinelander, 1972). The centripetal flow of blood was further demonstrated in the animals where the periosteal blood flow was blocked. In these animals, not only was the cortical flow restricted but flow in the adjacent bone marrow was also restricted. In these same sections, however, the nutrient vessels were filled suggesting that the afferent nutrient system does not perfuse most of the diaphyseal marrow or cortex. The lack of the hemopoietic tissue perfusion in the absence of cortical bone perfusion emphasizes the serial connection of the cortex and marrow with the obliged transit of blood through the bone prior to perfusing the marrow. While it might be argued that the India-ink may not have perfused the 172 L. DE SAINT-GEORGES AND S. C. MILLER Fig. 3. Distal region of a partially perfused femur. The highly vascular and metabolically active metaphyseal (Met) region of the long bone was readily perfused. Portions of the diaphyseal (Dial marrow were generally less perfusable than the metaphyseal areas. x 6. Fig. 4. Diaphysis from an incompletely perfused femur. India-ink filled the central sinus (CS) but incompletely filled the marrow sinusoids (MS). X 18. Fig. 5. Distal metaphysis of the femur illustrating nutrient arterial branches (ab) beneath the growth plate (gp). x 18. BONE-BONE MARROW MICROCIRCULATION IN THE RAT Fig. 6. Femoral diaphyseal shaft illustrating the “bush-like” marrow sinusoids formations (bf) in the marrow (M) where the vessels (Vc) exit the bone cortex at the endocortical surface. The marrow is perfused only where the cortical canals (cc) are also perfused. The central sinus (CS) is perfused in the absence of complete perfusion of the marrow sinusoids. x 12, 173 Fig. 7. Femoral diaphyseal shaft with more filling of the marrow sinusoids than illustrated in Figure 6. Many of the bush-like formations become indistinguishable but the collecting sinusoids (c0.s) become more apparent. x 12. 174 L. DE SAINT-GEORGES AND S. C. MILLER Fig. 8. Femur diaphysis. Occasionally some arterial branches (ab) of the nutrient system enter the cortex and curve back to the marrow. The thickness of the sections (30 pm) makes it difficult to determine the connectivity between arterial branches and other cortical vessels. x 12. Fig. 9. Femur diaphysis. Some small nutrient arterial branches (ab) can be found entering the inner two thirds of the cortex (C). The marrow is perfused only where cortical canals (cc) are also perfused. x 20. BONE-BONE MARROW MICROCIRCULATION IN THE RAT Fig. 10. A plastic clamp was placed on the periosteal surface of the diaphyseal shaft, in the area indicated (A). The cortical bone under this clamp was not perfused. The marrow (M) adjacent to these nonperfused cortical areas ( C ) were also not perfused, but the central sinus was fully perfused. X 10. 175 Fig. 1 1, A plastic clamp placed on a portion of the periosteal surface area (A) prevented the perfusion of the underlying cortex (C) and marrow (MI. Branches of the nutrient system (arrows) were well perfused with ink but the marrow around these vessels was poorly perfused. x6. 176 L. DE SAINT-GEORGES AND S. C. MILLER very small arterial branches, it did, however, perfuse the cortical vessels which are much smaller than the sinusoids that we propose they supply. The role of the small branches of the nutrient artery that were observed entering and leaving the inner cortex is not clear. Such vessels were mostly observed in or near the metaphyseal region. Nutrient and metaphyseal arterial branches were often difficult to distinguish from each other in sections that were partially perfused and thus they are referred to here collectively as the nutrient system, except if stated otherwise. The scarcity of these nutrient vessels entering the cortex and inability to supply blood in the area of the periosteal circulation blockade argue for a non-essential role of the nutrient system in the diaphysis. They could subsist as a remnant of the vascular network present during earlier growth (endochondral osteogenesis) when this region was formihg in the metaphysis of the growing bone. This vascular network in the cortical bone and bone marrow may serve as a collateral circulation when blood flow is reduced at the periosteum (i.e., bone fracture). A parallel circulation pattern between marrow and cortical bone was suggested from blood flow studies in the dog by Kelly (1973) and Morris et al. (1980). There are two main systems of afferent blood supply in the diaphyseal shaft: the nutrient and periosteal, and only one major exit, the comitant vein. Thus the afferent systems may appear to be in a parallel configuration and in blood flow studies it was possible to obtain regional flow values of cortical bone separate from that of marrow. This would not be the case if the afferent supply to the cortex and marrow was serially arranged. It is possible, however, that there are changes in the contributions of the afferent systems to different tissue compartments a s bone develops and ages and thus differences between shorter- and longer-lived animals. For example, Haversian systems (osteons) develop in bones from longer-lived animals, whereas they are not normally found in bone from shorter-lived animals, except in very old animals or under unusual experimental conditions (Ruth, 1953). Thus the circulatory pathways of bone from longer-lived species may differ and/ or change as these bones develop a n osteonal structure. Many years ago surgeons noted that when the periosteum was stripped from a bone, blood would ooze from the bone surface (Brookes, 1971). This was taken as evidence for a centrifugal flow of blood from the cortex. While it is possible that in humans and other long-lived animals, the flow of blood in the diaphysis is centrifugal (Rhinelander 19721, as discussed above, it could also be explained by a simple leakage effect caused by the surgical removal of the periosteal supply. The backflow of blood, in this case, could be due to the high pressures that appears to exist in the bone marrow (Petrakis, 1954). The sinusoid network of the hemopoietic marrow of the young rat appears to have a slow centripetal flow toward the central sinus. In the marrow, a light centrifugal pressure gradient should exist between the cortex and the central sinus. If we consider the corticoendosteal surface as a cylindrical isobaric higher zone and the central sinus surface as a isobaric lower zone, intermediate isobaric surfaces could be traced as concentric cylinders of intermediate diameters. Along such a n isobaric surface there should be no flow. When some of the cortical canals were obstructed, such as by a brace around the bone, perfusion of the underlying marrow was prevented, emphasizing the centripetal flow and the inability of blood from the neighboring zone to spread in the sinusoids along the bone axis. The endocortical surface could be considered as a n isobaric distribution surface with each distribution point corresponding to a n osteal canal aperture. The flow issued from each canal will then generate the bush-like appearance of the sinusoids that may be interpreted as a functional manifestation of a centripetal blood flow rather than a n actual anatomic sinusoid structure. The sinusoids in the marrow may be interconnected and networked, but because the point source of the blood flow is the canal aperture, there would be local sinusoidal regions, or domains. The possible physiological significance, if any, of such “territories” is not known. When in traumatic situations the normal pressure gradient is perturbed, such as after a fracture, a completely different situation may develop which might allow blood flow to change directions, depending on the nature of change in the pressure gradient. For example, disruption of the periosteum, which is much different from a periosteal blockade, as used in the present study, might result in a leakage effect that would divert some sinusoidal blood from the comitant vein exit or increase a centrifugal flow from the collateral arterioles from the nutrient system that enter the cortex. These alternative, or collateral circulatory patterns could help prevent tissue necrosis and promote healing following trauma to the skeleton. Based on our observations of extensive vascularization of the cancellous bone in the metaphyseal region from the nutrient arterioles and metaphyseal vessels, it is evident that blood flow to the metaphyseal cancellous bone is greater than to the cortical bone (Morris et al., 1980). The afferent nutrient system is likely essential for the perfusion of the growth plate and the highly active metaphyseal bone in more rapidly growing younger animals. This area of the bone would not be as readily accessible to the periosteal circulation. Metaphyseal veins have been observed (Brookes, 1971; Trueta, 1968; Rhinelander, 1972; Danckwardt-Lilliestrom, 1969) and it is likely that this venous system drains the cancellous bone circulation into the comitant vein. From the observations made in this study of the rodent long bone diaphysis, it appears that the bone marrow is more dependent on the periosteal, trans-osseous blood flow than on the nutrient system. This is in agreement with the conclusion of Sorrel (1988), who found in embryonic chick bone that arterial development does not appear as a prerequisite for marrow hemopoiesis. In the metaphyseal region, however, the hemopoietic tissue would be dependent on the nutrient supply. Assuming that our perfusion model of the long bone is correct, it is interesting that the distribution of hemopoietic tissues appears to occur in the regions of higher vascular supply while fatty marrow appears to occur in downstream regions. For example, fatty marrow deposits are usually seen in the central marrow areas while red marrow is observed adjacent to endocortical surfaces. In some long bones in older dogs, red marrow is also seen in metaphyseal areas while the BONE-BONE MARROW MICR(ICIRCULATION IN THE RAT adjacent epiphyseal region may contain yellow marrow (Miller et al., 1980). In dogs, the perfusable vascular space in red marrow is much greater than in fatty marrow (Miller and Jee, 1980; Smith et al., 1984). Similar conclusions have been made in studies of human bone (Trueta and Harrison, 1953) and other animals including rats, monkeys, and an opossum (Van Dyke, 1967). In addition, osseous tissues in red marrow sites have a greater metabolic activity than those found in yellow marrow sites (Smith et al., 1984).A positive correlation between the occurrence of yellow marrow and osteoporosis has been noted in humans (Meunier et al., 1971) and the relationship between bone marrow composition and bone mass might be mediated, a t least in part, by changes in the vascularization and/or blood flow in bone. ACKNOWLEDGMENTS This study was supported by U S . Public Health Service grant DE-07006 from the National Institutes of Health and US. Department of Energy Contract DEAC0276EV00119. LITERATURE CITED Brookes, M. 1971 The Blood Supply of Bone: An Approach t o Bone Biology. Butterworths Inc., London. Brookes, M., and R.G. Harrison 1957 The vascularisation of the Rabbit femur and tibio-fibula. J . Anat., 91:61-71. Danckwardt-Lillestrom, G. 1969 Reaming of the medullary cavity and its effect on diaphyseal bone: A fluorochromic, microradiographic and histological study on the rabbit tibia and dog femur. Acta Orthop. Scand., Suppl. 128r1-153. De Bruyn, P.P.H., P.C. Breen, and T.B. Thomas 1970 The microcirculation of the bone marrow. Anat. Rec., 168:55-58. Forssmann, W.G., S. Ito, E. Weihe, A. Aoki, M. Dym, and D.W. Faw- 177 cett 1977 In improved perfusion fixation method for the testis. Anat. Rec., 188:307-314. Guide for the Care and Use of Laboratory Animals. 1985 In: NIH Guide Supplement for Grants and Contracts. U S . Department of Health and Human Services, Vol. 14, No. 8, pp. 1-83. Jee, W.S.S., and J.S. Arnold 1960 India-ink vascular injection of skeletal tissues. Stain Technol., 35r59-65. Kelly, P.J. 1973 Comparison of marrow and cortical bone blood flow by 125-I-labelled 4-iodoantipyrine (I-Ap) washout. J. Lab. Clin. Med., 81:497-505. Meunier, P., J . Aaron, C. Edouard, and G. Vignon 1971 Osteoporosis and the replacement of cell populations of the marrow by adipose tissue. Clin. Orthop. Rel. Res., 8Or147-154. Morris, M.A., and P.J. Kelly 1980 Use of tracer microspheres to measure bone blood flow in conscious dogs. Calcif. Tissue Int., 32: 69-76. Miller, S.C., B.M. Bowman, J.M. Smith, and W.S.S. Jee 1980 Characterization of endosteal bone-lining cells from fatty marrow bone sites in adult beagles. Anat. Rec., 198:163-175. Miller, S.C., and W.S.S. Jee 1980 The microvascular bed of fatty bone marrow in the adult beagle. Metab. Bone Dis. Rel. Res., 2:239246. Petrakis, N.L. 1954 Bone marrow pressure in leukemic and non-leukemic patients. J . Clin. Invest., 33:27-34. Rhinelander F.W. 1972 Circulation in Bone. In: The Biochemistry and Physiology of Bone. G.H. Bourne, ed. Academic Press, New York, pp. 1-76. Ruth, E.B. 1953 Bone studies. 11. An experimental study of the Haversian-type vascular channels. Am. J. Anat., 93:429-455. Sorrel, J.M. 1988 Development of arteries in embryonic chick bone marrow with special reference to the appearance of periarterial granulopoietic sheaths. Anat. Rec., 221,730-736. Smith, J.M., S.C. Miller, and W.S.S. Jee 1984 The relationship of bone marrow type and microvasculature to the microdistribution and local dosimetry of plutonium in the adult skeleton. Radiat. Res., 99:324 -335. Trueta, J . 1968 Studies of the Development and Decay of the Human Frame. W.B. Saunders Go., Philadelphia, pp. 389. Trueta, J., and M.H.M. Harrison 1953 The normal vascular anatomy of the femoral head in adult man. J. Bone Jt. Surg. 35Bt442-461. Van Dyke, D. 1967 Similarity in distribution of skeletal blood flow and erythropoietic marrow. Clin. Orthop. Rel. Res., 52:37-51.