3. small Anim. Pract. (1970) 11, 621-638. A study of the radiographic and histological changes occurring in Legg-Calve-Perthes Disease (LCP) in the dog R . L E E Department of Veterinary Surgery, University of Glasgow Veterinary Hospital, Bearsden Road, Bearsden, Glasgow A B S T R A C T The radiographic appearance of thirty-six cases of canine Legg-Calve- Perthes Disease and the histological changes in twenty-six of these cases are described. The changes are related to each other, and also to the duration of the clinical condition. The results are discussed in relation to the pathogenesis of the changes observed. R E V I E W OF L I T E R A T U R E The radiographic appearance of the various stages of LCP in the human infant have been well described by Waldenstrom (1922) and most workers seem to follow his classification of the changes. Reports of the histological appearance of the affected femoral head in the human, because of the success of conservative as opposed to surgical treatment, have often been based on small numbers of cases, on biopsy material, or tissue fragments obtained after epiphyseal curettage. Jonsater (1953) has reviewed the literature and presented a detailed study of the histological changes examined in biopsy material obtained with a hollow needle in thirty-four cases in children. The histological findings were related to the stage or duration of the condition and also to the radiographic and arthro- graphic appearance. There have been a limited number of studies of the histology and radiology of LCP in the dog. Moltzen-Nielsen (1938) reported the histological changes in one case in the dog and described the clinical features of a further eighteen cases. Formston & Knight (1942) described fourteen cases of âepiphyseal separation of the femoral headâ ; however, a proportion of these cases had clinical histories and radiographic findings consistent with a diagnosis of LCP. Unfortunately they were not able to obtain any histology from these cases. 62 1 622 R . LEE Hulth, Norberg & Olsson (1962) described the histological and radiographic appearance of five cases of canine LCP and performed angiography on a further two cases. Paatsama, Rissanen & Rokanen (1966, 1967) reported the histological changes observed in twenty-seven dogs, together with reports of the uptake of oxytetra- cycline and radioactive sulphur by the tissues of the femoral head. They did not relate their findings to the duration of the clinical signs. Ljungren (1967) described the histology and radiology of eighty-five cases in the dog and suggested a possible endocrine aetiology to LCP. The histology and radiology were not, however, related to each other nor to the duration of signs. Kemp (1969) gave an account of the histological appearance of the natural disease in both infants and dogs and has also presented the radiographic and histological appearance of changes produced experimentally in the dog. MATERIALS AND M E T H O D S The following study is based on the radiographical appearance of thirty-six cases of LCP in the dog and on the histological appearance of twenty-six of these cases following excision arthroplasty of the femoral head which, as previously reported, appears to be the treatment of choice (Lee & Fry, 1969). The cases were presented to the Surgery Department, Glasgow University Veterinary Hospital, with histories of a hind-leg lameness of gradual onset and of varying duration from 1 week to over 1 year. The additional clinical features noted were pain on manipulation of the hip, particularly on extension and abduction, atrophy of the gluteal and thigh muscles, and apparent shortening of the leg. The radiographic features were studied on the clinical radiographs, taken usually on the conscious dog, the usual view being a ventro-dorsal projection of pelvis and hips, with the hind legs extended as far as the clinical condition would allow. A variety of changes affecting the proximal femoral metaphysis, epiphysis and acetabulum were noted : (a) Flattening and/or irregularity of the articular surface of the femoral head. This ranged from a minimal loss of convexity of the antero-dorsal aspect of the articular surface (Fig. 3) to a marked concave impression on the antero-dorsal aspect (Figs. 8 and 9) with, in some cases, complete disruption and loss of con- tinuity of the articular surface (Fig. 7). (b) Irregular radiographic density of the femoral epiphyseal and metaphyseal regions, sometimes with larger foci of increased or decreased density, or linear radiolucencies. (c) An apparent shortening and increase in width of the femoral neck in the metaphyseal area. (d) An increase in the width of the apparent joint space which, it must be remembered, consists in the radiograph of the articular cartilage of femoral head and acetabulum as well as the true joint space. L E G G - C A L V E - P E R T H E S D I S E A S E I N T H E D O G 623 An attempt has been made to categorize the severity of these changes and to relate this to the duration of the condition. Any assessment of severity is necessarily somewhat subjective, and, to a certain extent, dependent on the quality of the radiographs. The excised femoral heads were fixed in formol-saline, decalcified, and em- bedded in paraffin wax. Sections were obtained in a mid-sagittal plane and stained with haematoxylin and eosin, Van Gieson and Toluidene Blue. The histological appearance was studied on representative sections from each specimen, and the following features were observed and their extent assessed : (a) The presence of necrotic bone trabeculae and marrow tissue. (b) The presence of invasive, vascular âgranulationâ tissue. (c) The degree of deformity of the articular surface of the femoral head. (d) The presence and extent of clefts, or cavities, immediately under the (e) Areas of thick, irregular, trabecular bone and foci of fibrous tissue or fibro- subchondral bone. cartilage. RESULTS The severity of the various changes assessed on the radiographs are presented in Table 1. The cases are listed according to increasing duration of clinical signs noted by the owner. The radiographs of eight cases are presented in association with the histological appearance (Figs. 1-9). Widening of the joint space was a constant feature in all but two cases, both of which were of less than 1 month duration. Increase in width of the femoral neck was a virtually constant feature. The remaining features observed tended to show an increase in severity with increasing duration, although approximately 30% of the early cases seen also showed fairly severe changes. Another feature that was occasionally noticed on the radiographs was the presence of linear radiolucencies parallel to the articular surface and deep to the subchondral bone. It is assumed that these correspond to the subchondral cavities that were observed histologically (vide iizfra) and to the so-called âFreunds linesâ described in the radiographs of affected children. These linear radiolucencies were only infre- quently seen and obviously the likelihood of recognizing these on a radiograph depends on the position of the femoral head, as they will be only readily visible when seen tangentially. Histological sections of the complete resected femoral heads of a selection of cases of varying duration, tissue diagrams to show the relative distribution of the features described, together with the clinical radiographs, are presented in Figs. 1-9. On histological examination necrotic bone and marrow tissue was present in all but one of the specimens, although the amount of necrotic tissue varied con- siderably (Figs. 1-9). Only in cases of relatively short duration were large areas of bone necrosis observed. 624 R. LEE TABLE 1. The relative severity of the radiographic features described, related to the duration of the clinical signs Flattening and Case irregularity of Irregular density Irregular Width of Width of No. the articular of epiphysis density of femoral joint surface metaphysis neck space 36992 35439 36921 (r) 37572 (r) 34469 34412 39707 37401 (r) 37401 (1) 38563 39578 40222 38084 34745 36907 (r) 36907 (1) 393 14 Duration of clinical lameness of 1 month or less ?+ + + ++++ ++ + + + + ++ + + + + + ++ + + +++ +++ + ++ ++ ++ + ++++ ++++ +++ - - - Duration of clinical lameness 1-2 months +++ ++ + + + ++ + ++ + +++ + + + + + ++ + ++ ++ + +++ +++ + + +++ +++ +++ - Duration of clinical lameness 2-3 months - 37786 + +++ 37703 (1) ++ +++ + + 37703 (r) ++++ ++++ +++ 37900 + + + 3600 1 + + +++ ++ 34748 - 35676 - +++ + + - ++ Duration of clinical lameness 3 4 months 35752 (r) +++ +++ + 33733 + + + + Duration of clinical lameness 4-5 months 37420 (1) +++ ++ ++ 37420 (r) ++++ +++ ++ 38510 (1) +++ +++ +++ Nor. Inc. Inc. Nor. Inc. Inc. Inc. Inc. Inc. Nor. Inc. Inc. Inc. Nor. Inc. and irr. S1. inc. Inc. Inc. and irr. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Irr. Nor. Inc. Inc. ? Inc. Inc. Inc. Inc. Inc. Inc. Inc. and irr. Inc. ? Inc. Inc. Inc. Inc. Inc. and irr. Inc. Inc. and irr. Table 1 (contd) Flattening and Case irregularity of Irregular density Irregular Width of Width of No. the articular of epiphysis density of femoral joint surface metaphysis neck space Duration of clinical lameness 5-6 months 35752 (1 ) +++ +++ + + Inc. Inc. 38510 (r) +++ +++ + + + Inc. Inc. and irr. 4031 1 + + + + ++++ ++++ Inc. Inc. and irr. Duration of clinical lameness 6-12 months 3 1859 + + + + ++++ + + + + Inc. Inc. and irr. 39595 +++ + + ++ Inc. Inc. Duration of clinical lameness-over 12 months 37484 + + + + + ++ Inc. Inc. 36246 ++++ + + + ++ Inc. Inc. and irr. +, Slight; + +, moderate; + + +, severe; + + + +, very severe; Inc., increased; Nor., normal; Irr., irregular. FIG. 1. Case No. 36993. Poodle. Female. Lame for 1 week before excision of right femoral head. Filled areas, cavity formation deep to subchondral bone; hatched areas, necrotic bone and marrow; areas filled with crosses, vascular, cellular reactive tissue showing osteoblastic and osteoclastic activity. 626 R. LEE T h e necrotic tissue could be recognized with relative ease by the absence of osteocytes in the bone lacunae and by a complete lack of cellular detail in the intertrabecular spaces (Fig. 10). The site at which areas of necrosis persisted, even in cases of considerable duration, was the subchondral bone that was separated by cavitation from the main part of the epiphysis (Figs. 4-8). FIG. 2. Case No. 39707. Yorkshire Terrier. Female. Lame for 3 weeks before right femoral head excised. For key to shading, see legend to Fig. 1. Invasive, highly vascular, tissue, with an abundance of cells-which for con- venience will be referred to as âgranulation tissueâ-was observed to some degree in all the material examined (Fig. 11). I n the very early cases this tissue was minimal in amount, and appeared first at the angle formed between the epiphy- seal line and articular cartilage (Figs. 1 and 2). In later cases, where the epiphyseal line was not usually present, the zone of granulation tissue varied in width but was usually narrow and was seen at the junction between the necrotic and living bone (Fig. 11 and Figs. 3-9). When examined in more detail, this tissue appeared to be involved in the removal of necrotic bone and in the formation of new bone. FIG. 3. Case No. 40222. Poodle. Male. Clinical lameness for 2 months. Left femora1 head excised. For key to shading, see legend to Fig. I. FIG. 4. Case No. 34046. West Highland White Terrier. Male. Lame for 2 months before left femoral head excised. Unfortunately, no radiograph is available. For key to shading, see legend to Fig. 1. FIG. 5. Case No. 35676. Cairn Terrier. Female. Lame for 24 months. Right femoral head excised. For key to shading, see legend to Fig. 1. FIG. 6. Case No. 33733. Poodle. Male. Lame for 4 months before left femoral head L E G G - C A L V E - P E R T H E S DISEASE I N T H E D O G 629 At some points the necrotic trabeculae were the site of osteoclastic activity (Fig. 12), but at others the necrotic trabeculae appeared to persist and act as a âscaffoldâ upon which new bone was being laid down by osteoblastic activity (Fig. 13). These two processes could often be seen occurring in close proximity to each other and occasionally the same trabeculum would be the site of both types of activity. Encapsulation of necrotic trabeculae by new bone occurred FIG. 7. Case No. 38510. West Highland White Terrier. Female. Lame for 5 months. Bilateral involvement, histology shown is of left femoral head. For key to shading, see legend to Fig. 1. frequently and these encapsulated trabeculae could often be observed a t some distance from the main area of necrotic tissue. The degree of deformity of the histological section of the femoral head varied 630 R. L E E considerably but appeared to correlate well with the degree of deformity observed on the radiograph. T h e presence of clefts, or cavities, immediately underlying the subchondral bone has been mentioned and these were a very common finding (Figs. 2, 4-8). They appeared to be associated, particularly in the earlier cases, with a degree of trabecular fragmentation, the fractured trabeculae becoming impacted into the intertrabecular spaces (Fig. 14). The articular cartilage did not appear to be necrotic even in those parts of the articular surface that overlaid subchrondral cavities. The presence of these cavities could also be readily appreciated on the gross specimen observed at surgery. FIG. 8. Case No. 40311. Cairn Terrier. Female. Clinical lameness for 6 months. Left femoral head excised. For key to shading, see legend to Fig. 1. I n the cases of moderate, or long duration, where the total amount of necrotic tissue tended to be least, the trabeculae were often thick and irregular as com- pared with the trabecular pattern of normal femoral heads (Figs. 15 and 16) ; in addition, foci of fibrous tissue (Fig. l l ) , and, in some cases, of fibrocartilage (Fig. 13b), were observed in the epiphysis. Occasionally, fragments of necrotic bone were found within these foci. L E G G - C A L V E - P E R T H E S D I S E A S E I N T H E D O G 63 1 FIG. 9. Case No. 36246. Pekingese. Male. Clinical lameness for 2 years. Left femoral head excised. For key to shading, see legend to Fig. 1. D I S C U S S I O N These results indicate that the initial change occurring in LCP is that the proximal femoral epiphysis, still separated from the metaphysis by an intact cartilagenous growth plate, for some reason becomes totally necrotic, except for the articular cartilage. There is probably, at this stage, little, or no evidence of radiographic abnormality and, indeed, there may also be a little or no clinical manifestation of lameness. In fact, very few cases-if any-will be referred for radiography at this stage. Continued weight-bearing of the affected leg results in microfractures of necrotic trabeculae, with associated cavity formation in the subchondral zone. In addition, in occasional cases, microfractures of necrotic trabeculae occurred throughout the affected epiphysis with complete disintegration of the head. The most vulnerable area would appear to be the antero-dorsal aspect of the articular surface where it is in contact with the dorsal acetabular rim. The formation of subchondral cavities can be accounted for by the resilience of the articular cartilage which can spring back into a more normal shape after the underlying trabeculae have fractured. These changes would explain the initial appearance 632 R. L E E FIG. 10. Photomicrograph to show the appearance of necrotic trabeculae and marrow. Note empty lacunae and absence of cellular detail in marrow. FIG. 1 1. Photomicrograph to show the junction between necrotic and re-vascularized zone. (a) Necrotic bone and marrow; (b) re-vascularized zone; (c) advancing highly vascularized and cellular âgranulationâ tissue. L E G G - C A L V E - P E R T H E S DISEASE I N T H E D O G 633 of irregular density on the radiograph due partly to cavity formation and, partly, to impaction of the fragmented trabeculae. Concurrent with these changes there is probably the initial stage of a vascular reaction, with hyperaemia of the metaphysis and perforation of the epiphyseal plate by vascular reactive âgranulation tissueâ. This vascular response probably accounts for the advent of irregular radiographic density of the metaphysis. I t has been suggested that the growth plate may be damaged by the ischaemia of the epiphysis and it is also possible that in some cases parts of the juxtaepiphyseal regions of the metaphysis may also undergo necrosis and repair (Kemp, 1969). FIG. 12. Osteoclastic resorption of a necrotic bony trabeculum. The vascular reaction then progressively invades the epiphysis, replacing the necrotic tissue with new bone, the necrotic trabeculae often persisting for a period within the centres of new-formed trabeculae until remodelling occurs (Fig. 13a and b). This is similar to the so-called âcreeping substitutionâ described by Waldenstrom (1922). The thicker and more irregular arrangement of these new trabeculae (Fig. 15), together with the persistence of foci of fibrous tissue and fibrocartilage, would explain the persistence in the later stages of irregularity of the radiographic density of the femoral head and neck. I t is usually difficult or impossible to identify the site of the growth plate in specimens where this âwaveâ of re-vascularization has occurred. I n the normal femoral head this area is usually recognized as a line of thickened trabeculae. I t may be that the site of the growth plate is masked by an overall thickening of the metaphyseal and epiphyseal trabeculae as a result of the re-vascularization process. E FIG. 13. (a) Necrotic trabeculae with re-vascularized intertrabecular spaces and osteo- blastic activity and deposition of new lamellae on the surfaces of the dead trabeculae. (b) High power view of osteoblastic activity on the surface of a dead trabeculum. (c) As in (a) but in addition an island of cartilage has formed in the marrow space. L E G G - C A L V E - P E R T H E S D I S E A S E I N T H E D O G FIG. 14. Fragmentation of necrotic trabeculae bordering a âsubchondral cavityâ. FIG. 15. Thickened irregular trabecular arrangement in a re-vascularized epiphysis. Cement lines between the layers of bone in the trabecular can be seen and areas of bone with empty lacunae presumably representing persistent fragments of necrotic bone can also be seen. 635 636 R. L E E FIG. 16. Normal trabeculation and live bone and marrow from the normal femoral head. T h e presence of subchondral clefts appears to act as a barrier to the process of re-vascularization, with the result that in long standing cases, when the vast bulk of the epiphysis has been reformed, there is the persistence of necrotic bone on the deep face of the flap of cartilage overlying the cavity. This may give an appear- ance similar to that expected in Osteochondritis dissecans. T h e widening of the joint-space observed radiographically and the widening of the femoral neck are more difficult to explain. Kemp (1969) has suggested that increase in joint space could be due to inflammatory effusion into the hip- joint and/or to inflammatory changes in the round ligament displacing the femoral head laterally. Another possible explanation of the increase in joint-space is that enlargement of the epiphysis occurs by growth of the articular cartilage, whose deeper zone becomes ossified to contribute to the increasing size of the epiphysis; if avascular necrosis of the epiphysis occurs, then the articular cartilage will continue to increase in thickness, as it will possibly receive adequate nutrition from the synovial fluid, whereas the deep zone will no longer be ossified and there will be a resultant increase in width of the joint space, as visualized on the radio- graph. The increase in width of the femoral neck may be associated with the progressive re-vascularization involving the metaphyseal area, and on the radio- graphs this feature is somewhat accentuated by the partial collapse of the femoral head. The reactive changes occurring in the acetabulum are presumably reflec- tions of secondary arthritic change resulting from the grossly disturbed joint function. There would therefore appear to be good correlation between the radiographic L E G G - C A L V E - P E R T H E S D I S E A S E I N T H E D O G 637 and histological appearances. However, it is difficult from the radiographic picture alone to determine the duration of the pathological lesion, except that marked irregularity of shape and density are more likely to be seen in cases of longer duration. It may be difficult to demonstrate sufficient radiographic change in the early case to confirm a clinical diagnosis, although by the time lameness manifests itself the epiphysis will be probably totally necrotic and infraction will be beginning. The appearance of an overall increase in radiographic density, described as an early feature of LCP in the human infant, has not been observed in any of the cases described. The findings presented here are in close agreement with the biopsy results obtained by Jonsater (1953) in the human, with the limited number of cases presented by Hulth et al. (1962), and with the appearance of both human and canine material reported by Kemp ( 1969). The description of the radiographical features agrees with the various âgradesâ described by Ljungren (1967) although many of the histological appearances and the proposed sequence of events are at variance with her results. Ljungren describes the earliest change as a thickening of the trabeculae which then âcollideâ, with a resultant occlusion of the blood supply, necrosis and trabecular fracture. In addition, she describes the common site of osteonecrosis as being the central portions of otherwise viable trabeculae, and the trabeculae immediately underlying the articular surface. The results presented above, however, suggest that all of these features are consistent with extensive re-vascularization of the necrotic epiphyseal bone and do not indicate the mechanism of the necrotic process. A C K N O W L E D G M E N T S I should like to thank the members of the Surgery Department, in particular Professor D. D. Lawson, for their continued help and encouragement, M r Bert McLaren for the preparation of the histological sections and Miss M. Connell for typing the manuscript. R E F E R E N C E S FORMSTON, C . & KNIGHT, G.C. (1942) Vet. Rec. 54,481. HULTH, A., NORBERG, I . & OLSONN, S-E. (1962) J . BoneJt Surg. 44A, 918. JONASTER, S. (1953) Acta orthop. scand. (suppl.), XII. KEMP, H.B.S. (1969) M.S. thesis, University of London. LEE, R. & FRY, P.D. (1969) 3. small Anin. Pract. 10, 309. LJUNGREN, G. ( I 967) Acta orthop. scand. (suppl.), 95. MOLTZEN-NIELSEN, H. (1938) Arch wiss. prakt. Tierhailk. 72, 91. PAATSAMA, S., RISSANAN, P. & ROKKANEN, P. (1966) 3. small Anim. Pract. 7, 483. PAATSAMA, S., RISSAKEN, P. & ROKKANEN, P. (1969) J. small Anim. Pract. 8, 215. WALDENSTROM, H. (1922) Acta radiol. 1, 393. 638 R. L E E Risum6. Description des images radiologiques de trente-six cas de syndrome de Legg-Calve- Perthes chez le chien. Le tableau clinique et les ltsions histologiques observbes dans 26 de ces cas sont present&. I1 y a une correlation entre les symptbmes et les modifications histologiques, d e meme quâavec la durte de Iâaffection. Les rtsultats de lâttude sont discutts en fonction de la pathogtnie des altkrations observtes. Zusammenfassung. Das radiographische Erscheinungsbild von 36 Fallen von Legg-Calvt- Perthesscher Krankheit beim Hund und die histologischen Veranderungen in 26 dieser Falle werden beschrieben. Die Veranderungen stehen in Beziehung zu einander und auch zu der Dauer des klinischen Zustands. Die Ergebnisse werden in bezug auf die Pathogenese der beobachteten hdeungren besprochen.
Comments
Report "A study of the radiographic and histological changes occurring in Legg-Calve-Perthes Disease (LCP) in the dog"