Abstract
A fetal lamb model was developed to investigate the capacity of fetal articular cartilage for repair after the creation of a superficial defect.Superficial defects, 100 micrometers deep, were made in the articular cartilage of the trochlear groove in the distal aspect of the femur in eighteen fetal lambs that were halfway through the 145-day gestational period; the contralateral limb was used as a sham control. The wounds were allowed to heal in utero for three, seven, fourteen, twenty-one, or twenty-eight days. Seven days after the injury, the defects were filled with a hypocellular matrix, which stained lightly with safranin O. At twenty-eight days, the staining of the matrix was similar to that of the sham controls and the chondrocyte density and the architectural arrangement of the cell layers had been restored. An inflammatory response was not elicited, and no fibrous scar tissue was observed.CLINICAL RELEVANCE: An orderly sequence of repair of articular cartilage was observed after the creation of partial-thickness defects in the distal aspect of the femur of mid-gestational fetal lambs. The fetal lamb model may be useful for the investigation of interactions between the chondrocyte and extracellular matrices after mechanical stimulation. Fundamental knowledge of the metabolism of fetal articular cartilage may provide insight into latent reparative processes of mature cartilage.
The study of partial-thickness defects in articular cartilage is of clinical importance because the early pathological changes of osteoarthrosis involve the superficial layers of cartilage. Osteoarthrosis is the most prevalent rheumatic condition, but its etiology and effective treatment remain elusive. Identification of the metabolic processes involved in the repair of cartilage may provide clues to the pathogenesis of osteoarthrosis and may help in the development of new methods for its treatment.
Mature articular cartilage has little inherent capacity to heal superficial defects, which limits the elucidation of reparative processes. Neither animal models nor clinical observation of defects in human articular cartilage have demonstrated an effective reparative response after partial-thickness injuries2-4,6,8,9,13-15.
A fetal articular cartilage model was developed to study whether spontaneous repair of superficial defects in articular cartilage could be observed. To the best of our knowledge, this has not been demonstrated clinically or experimentally. Previous investigators have demonstrated the remarkable regenerative capacity of fetal skin12, fetal cleft palate11, and fetal tendons1. The lamb model was chosen because the fetal limbs were large enough for us to create specific defects in the hyaline cartilage, because there is a relatively long gestational period to permit in utero healing, and because ewes are resistant to preterm labor.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was the Department of Orthopaedic Surgery at the University of California at San Francisco.
†Department of Orthopaedic Surgery, University of California at San Francisco, 500 Parnassus Avenue (MU-320W), San Francisco, California 94143-0728. Please address requests for reprints to Dr. Namba.
‡Department of Surgery, University Children's Hospital, Steinwiesstrasse 75, 8032 Zurich, Switzerland.
§Department of Surgery, Seattle Children's Hospital, 4800 Sand Point Way N.E., Seattle, Washington 98105.
#Department of Pediatric General and Thoracic Surgery, The Children's Hospital of Philadelphia, 34th and Civic Center Boulevard, Philadelphia, Pennsylvania 19104.
This study was approved by our institutional review board for animal research. Eleven time-dated pregnant ewes (one with triplets and five with twins) at seventy-five days of gestation (term, 145 days) had a laparotomy and hysterotomy for manipulation of the fetal hindlimb and knee with use of previously described fetal operative techniques5. In brief, the ewes were not fed for forty-eight hours and then were given general endotracheal anesthesia. An infraumbilical midline laparotomy was performed, followed by exteriorization of the uterus. The position of the fetal lamb was palpated, and a five-centimeter hysterotomy incision was made. The hindlimb of the fetus was delivered into the operative field. Initially, a 1.5-centimeter midline incision was made in the anterior aspect of the knee. However, the fragility of the fetal skin, coupled with active motion of the knee in utero, resulted in wound dehiscence. The arthrotomy was then modified to a lateral longitudinal incision in the mid-sagittal plane over the knee joint. The patella and the extensor mechanism were carefully everted with a cotton-tipped swab moistened with saline solution, in order to expose the trochlear surface of the articular cartilage of the distal aspect of the femur.
Uniform controlled incisional defects were created on the relatively flat surface of the trochlear articular cartilage with a calibrated ophthalmic diamond knife (manufactured in Biel, Switzerland, by Storz Instrument, St. Louis, Missouri) (Fig. 1). A longitudinal incision was made on the medial and lateral sides of the trochlear groove. Preliminary histological analysis of articular cartilage from the fetal lambs revealed blood vessels more than 400 micrometers below the surface. Therefore, to prevent injuries that were essentially analogous to full-thickness defects, a depth of 100 micrometers was chosen for the incision. To create a sham control, the contralateral knee was treated with arthrotomy and exposure of the surface of the trochlear articular cartilage but without creation of an incisional defect. A defect was made, immediately after procurement of the specimen, in the trochlea of one three-day and one seven-day sham control. An additional control (footpad control) consisted of application of the diamond knife while the blade was retracted. The footpads of the knife, which limit the depth of the incision made by the exposed blade, come into contact with the surface of the cartilage during use of the knife and could potentially injure the fragile fetal tissues.
A single-layer closure of the arthrotomy wound was performed with interrupted 6.0 Prolene (polypropylene) sutures (Davis and Geck, Danbury, Connecticut). The fetus was returned to the uterus, and the amniotic volume was restored with warm sterile saline solution with one million units of penicillin G. The hysterotomy was closed with a TA-90 stapler (United States Surgical, Norwalk, Connecticut), followed by closure of the laparotomy with 3.0 Prolene sutures.
The fetal lambs were removed through laparotomy and hysterotomy with the ewe under general anesthesia. Three lambs were obtained at three days; four, at seven days; four, at fourteen days; four, at twenty-one days; and one, at twenty-eight days after creation of the partial-thickness defects. One ewe, which was intended for the twenty-one-day study group, aborted twin fetal lambs because of fetal death five days after the manipulation. After the femora were procured, the ewe and the fetal lamb were killed with an intracardiac injection of Pentothal (thiopental sodium).
During procurement of the tissue, the integrity of the skin and the degree of wound repair at the site of the incision were determined. An arthrotomy was performed with a scalpel, and the knee was disarticulated. The patella, muscles, and ligaments were removed. The distal aspect of the femur was fixed in 40 per cent formalin, decalcified, and embedded in paraffin. Mid-transverse sections of the trochlea were made to provide samples that contained the two defects in cross section (Fig. 2). The experimental and control specimens were prepared simultaneously and were stained with hematoxylin and eosin and counterstained with safranin O for histological analysis.
Histological analysis consisted of blinded review of the surfaces of the cartilage for assessment of the repair response. Each trochlear section was considered to have a medial and a lateral half since a defect was made on either side of the trochlear groove. A modified semiquantitative scale17 was used to assess the histological appearance of the reparative response, with a composite score based on the sum of three observed features. For filling of the defect, a score of 0 points indicated a continuous surface without depression; a score of 1 point, no or slight depression but a non-continuous surface; a score of 2 points, depression of less than 50 per cent of the original defect; and a score of 3 points, depression of more than 50 per cent of the original defect. Matrix staining was given a score of 0 points if it was the same as that of adjacent tissue, 1 point if it was slightly decreased, 2 points if it was moderately decreased, and 3 points if there was no repair tissue. In the assessment of cell morphology and density, a score of 0 points indicated chondrocytes of normal appearance and density, including the superficial layer; 1 point, chondrocytes that appeared normal but were hypocellular; 2 points, abnormal cells; and 3 points, an absence of cells. A composite score of 0 points represented normal tissue, while a score of 9 points indicated the absence of any reparative tissues within the defect.
All of the ewes survived the operative manipulation. Complications included wound dehiscence, which was discovered in five fetal knees at the time of procurement. There were no additional complications related to the wound after we began to use a lateral incision rather than an anterior one. Another set of twin fetal lambs was found to have arthrogrypotic knees with abnormal articular cartilage and fibrous tissue filling the joint.
Twenty-one specimens with the defect and twenty-one sham controls were available for semiquantitative analysis (Table I). In addition, four footpad controls were created: two of the specimens were procured immediately after application of the footpad, one was procured three days after application of the footpad, and one was procured seven days after application of the footpad.
Histologically, the sham controls were all identical in appearance. The surface layer was intact and consisted of transversely oriented cells with spindle-shaped nuclei, three to five cell layers thick. Deep to the superficial layer, the remainder of the cartilaginous tissue consisted of evenly spaced, rounded chondrocytes residing in a homogeneous matrix. Approximately 400 micrometers from the surface, vascular channels were found scattered in the matrix with loosely associated connective tissues. Application of the footpads of the ophthalmic knife to two of the sham controls at the time of procurement did not result in any observable abnormality. Three and seven days after application of the footpad, necrosis of the transversely oriented superficial cells was identified, corresponding to the sites of application.
The cartilage examined immediately after creation of the defects demonstrated sharply demarcated defects with parallel sides limited to the superficial layers of the cartilage (Fig. 3). No other apparent histopathological changes were observed on adjacent surface tissue or in the deeper layers.
After three days of healing in utero, the superficial defect was v-shaped and the newly formed matrix filled the depths of the cleft. Dead chondrocytes were seen in the tissue immediately adjacent to, and along each side of, the incisional defect, corresponding to sites where the footpads came into contact with the surface of the cartilage (Fig. 4). Lighter safranin-O staining of the extracellular matrix was observed surrounding the dead chondrocytes with pyknotic nuclei. No inflammatory infiltrate was evident, and no vascular channels were seen around the zone of injury.
Seven days after the injury, the defect was completely filled with matrix that stained lightly with safranin O. The tissue was hypocellular, consisting mainly of dead chondrocytes with fading, crenated nuclei. Sparsely scattered viable cells with rounded nuclei were visible in the deeper layer of the defect but not in the surface layer (Fig. 5). There was no visible demarcation between the hypochromatic, hypocellular repair tissue and the surrounding matrix.
At fourteen days, there was increasing cellularity of the deeper layer although the superficial layer remained hypocellular. The extracellular matrix stained lightly with safranin O. The source of the cells appeared to be the surrounding cartilage, as increased numbers of mitotic figures were visible in the normally staining matrix adjacent to the zone of injury. There was no evidence of cell migration from the surface to the margin of the injury site.
Twenty-one days after the injury, there was increased safranin-O staining and cellularity, with the appearance of many cells in the superficial layer (Fig. 6). The repopulation of the surface of the articular cartilage with cells occurred as the final phase of repair. Increased numbers of superficial cells were not observed at the margins of the repair site.
At twenty-eight days, the superficial layer of transversely oriented cells had been reconstituted. The cellularity of the matrix in the region of the defect and the intensity of safranin-O staining were the same as those of the surrounding tissue (Fig. 7).
The average rating with use of the semiquantitative scale was 8.3 points at three days, 5.0 points at seven days, 4.2 points at fourteen days, 2.3 points at twenty-one days, and 0.3 point at twenty-eight days (Fig. 8). The sham controls had a score of 0 points at all time-intervals.
In the lambs in which a lateral arthrotomy had been used, the incision healed without the formation of an apparent scar. Three and seven days after the injury, the incision was visible but healed. At later time-points, the incision was completely healed and was identifiable only by the location of the sutures. There were no adhesions within the knee joints with healed incisions except in the pair of arthrogrypotic knees.
In the knee joints in which there was wound dehiscence, extensive fibrosis was grossly evident throughout. There was direct communication between the joint and the external amniotic fluid, and fibrous tissue filled the joint space. Histologically, the partial-thickness defects had not healed at all and disorganized loose connective tissue was attached to the superficial layer of the cartilage. Despite the lack of healing of the skin and cartilage incisions, the width of the distal aspect of the femur increased as it did in the knees in which the wound had healed.
Spontaneous repair of partial-thickness defects in the articular cartilage was demonstrated in this fetal lamb model. Early filling with a hypocellular matrix was identified seven days after creation of the incisional defects. Rapid filling with matrix occurred without an identifiable interface between the defect and the surrounding tissue. Repopulation with normal-appearing fetal chondrocytes and restoration of the layered tissue architecture was observed at twenty-eight days. The fetal articular cartilage was repaired in the absence of directly derived blood products, without formation of scar tissue, and without an associated inflammatory reaction.
The cartilage defects healed spontaneously despite the fact that the area of injured tissue was larger than intended. Areas of necrosis of superficial chondrocytes on either side of the defect corresponded to the regions where the footpads of the ophthalmic knife had come into contact with the surface of the tissue. The smooth horizontal metal flanges of the ophthalmic knife limit projection of the diamond blade to a desired depth. The expanded zone of superficial injury was consistent with the fragility of fetal tissues in general.
Healing of tissue defects without scarring in mid-gestational fetal skin12, cleft palate11, and tendon1 has been observed previously. Fetal tissue is relatively avascular, hypoxic, and hypocellular. Articular cartilage is also hypovascular, hypoxic, and hypocellular. Lack of scar formation in selected fetal tissues has been partly attributed to the absence of any substantial inflammatory response following injury, which prevents fibrosis. Some fetal tissues, including diaphragmatic16 and gastric10 tissues, however, have demonstrated scarring following injury.
It is recognized that fetal chondrocytes possess metabolic properties that are lost as development and aging occur. These properties include growth of the joint surface, which occurs naturally, in the absence of injury. However, in the present study, the appearance of the knee joints with a wound dehiscence indicated that the repair of defects even in fetal articular cartilage has limitations. Disorganized fibrous tissue adherent to the surface of the cartilage and a lack of any observable healing of the incisional defects was seen despite apparently unaltered growth of the joint surface. Locally produced factors formed at the site of the injury, and synovial tissue may have been lost through the wound dehiscence. Hunziker and Rosenberg found that the synovial membrane was the source of mesenchymal cells that led to the healing of partial-thickness defects in rabbit cartilage that had been treated with chondroitinase. Although growth of the fetal limbs may have contributed to the observed healing responses in the present study, an intact joint was apparently necessary for spontaneous repair of the cartilage to occur.
While spontaneous regeneration of partial-thickness defects in cartilage may be unique to fetal tissues, some of the processes by which fetal chondrocytes interact with the extracellular matrix may persist in postnatal cartilage. The described model permits observation of cartilage metabolism in response to mechanical injury, the fundamental processes of which may be latent in mature tissues. Previous animal models failed to reveal any observable response to lacerative superficial defects2,4,8. The spatial and temporal reparative response after the creation of discrete superficial defects in the cartilage offers a unique opportunity to study the interactions between the chondrocyte and the extracellular matrix on an organ level.
In conclusion, complete spontaneous repair of partial-thickness defects of the articular cartilage was demonstrated in a fetal lamb model, which as far as we know has not been described previously. Repair of articular cartilage proceeded without an inflammatory response and without creation of scar tissue. Additional studies with this model may provide insight into the metabolism of repair in fetal articular cartilage. Detailed knowledge of these processes may ultimately provide clues for enhancing the reparative capacity of postnatal cartilage.
al-Qattan, M. M.; Posnick, J. C.; Lin, K. Y.; and Thorner, P.: Fetal tendon healing: development of an experimental model. Plast. and Reconstr. Surg.,92: 1155-1160, 1993.921155
1993
Cheung, H. S.; Cottrell, W. H.; Stephenson, K.; and Nimni, M. E.: In vitro collagen biosynthesis in healing and normal rabbit articular cartilage. J. Bone and Joint Surg.,60-A: 1076-1081, Dec. 1978.60-A1076
1978
Fuller, J. A., and Ghadially, F. N.: Ultrastructural observations on surgically produced partial-thickness defects in articular cartilage. Clin. Orthop.,86: 193-205, 1972.86193
1972
[PubMed]
Ghadially, F. N.; Thomas, I.; Oryschak, A. F.; and Lalonde, J. M.: Long-term results of superficial defects in articular cartilage: a scanning electron-microscope study. J. Pathol.,121: 213-217, 1977.121213
1977
[PubMed]
Harrison, M. R.; Jester, J. A.; and Ross, N. A.: Correction of congenital diaphragmatic hernia in utero. I. The model: intrathoracic balloon produces fatal pulmonary hypoplasia. Surgery,88: 174-182, 1980.88174
1980
[PubMed]
Hunter, W.: On the structure and disease of articulating cartilages. Philos. Trans. Roy. Soc. London, B,42: 514-521, 1743.42514
1743
Hunziker, E. B., and Rosenberg, L. C.: Repair of partial-thickness defects in articular cartilage: cell recruitment from the synovial membrane. J. Bone and Joint Surg.,78-A: 721-733, May 1996.78-A721
1996
Kim, H. K. W.; Moran, M. E.; and Salter, R. B.: The potential for regeneration of articular cartilage in defects created by chondral shaving and subchondral abrasion. J. Bone and Joint Surg.,73-A: 1301-1315, Oct. 1991.73-A1301
1991
Landells, J. W.: The reactions of injured human articular cartilage. J. Bone and Joint Surg.,39-B(3): 548-562, 1957.39-B(3)548
1957
Longaker, M. T.; Whitby, D. J.; Jennings, R. W.; Duncan, B. W.; Ferguson, M. W.; Harrison, M. R.; and Adzick, N. S.: Fetal diaphragmatic wounds heal with scar formation. J. Surg. Res.,50: 375-385, 1991.50375
1991
[PubMed]
Longaker, M. T.; Stern, M.; Lorenz, P.; Whitby, D. J.; Dodson, T. B.; Harrison, M. R.; Adzick, N. S.; and Kaban, L. B.: A model for fetal cleft lip repair in lambs. Plast. and Reconstr. Surg.,90: 750-756, 1992.90750
1992
Longaker, M. T.; Whitby, D. J.; Adzick, N. S.; Crombleholme, T. M.; Langer, J. C.; Duncan, B. W.; Bradley, S. M.; Stern, R.; Ferguson, M. W.; and Harrison, M. R.: Studies in fetal wound healing, VI. Second and early third trimester fetal wounds demonstrate rapid collagen deposition without scar formation. J. Pediat. Surg.,25: 63-68, 1990.2563
1990
[PubMed]
Mankin, H. J.: The reaction of articular cartilage to injury and osteoarthritis (first of two parts). New England J. Med.,291: 1285-1292, 1974.2911285
1974
Mankin, H. J.: The reaction of articular cartilage to injury and osteoarthritis (second of two parts). New England J. Med.,291: 1335-1340, 1974.2911335
1974
Meachim, G.: The effect of scarification on articular cartilage in the rabbit. J. Bone and Joint Surg.,45-B(1): 150-161, 1963.45-B(1)150
1963
Meuli, M.; Lorenz, H. P.; Hedrick, M. H.; Sullivan, K. M.; Harrison, M. R.; and Adzick, N. S.: Scar formation in the fetal alimentary tract. J. Pediat. Surg.,30: 392-395, 1995.30392
1995
[PubMed]
Pineda, S.; Pollack, A.; Stevenson, S.; Goldberg, V.; and Caplan, A.: A semiquantitative scale for histologic grading of articular cartilage repair. Acta Anat.,143: 335-340, 1992.143335
1992
[PubMed]