- Email: [email protected]
Differential diagnosis of an unusual shoulder articular lesion in an ancient domestic dog (Canis lupus familiaris L., 1758)
International Journal of Paleopathology 3 (2013) 282–287
Contents lists available at ScienceDirect
International Journal of Paleopathology journal homepage: www.elsevier.com/locate/ijpp
Case study
Differential diagnosis of an unusual shoulder articular lesion in an ancient domestic dog (Canis lupus familiaris L., 1758) D.F. Lawler a,∗ , D.A. Rubin b , R.H. Evans c , C.F. Hildebolt b , K.E. Smith b , C. Widga a , T.J. Martin a , M. Siegel b , J.E. Sackman d , G.K. Smith e , T.K. Patel f a
Illinois State Museum, Research and Collections Center, 1011 East Ash St., Springfield, IL 62703-3500, USA Malinckrodt Institute of Radiology (Box 8131), Washington University School of Medicine, 510 S. Kingshighway Blvd, St Louis, MO 63110, USA c Pacific Marine Mammal Center, 20612 Laguna Canyon Road, Laguna Beach, CA 92651, USA d Numerof & Associates, Inc., Four City Place Drive, Suite 430, St. Louis, MO 63141, USA e School of Veterinary Medicine, Department of Clinical Studies, University of Pennsylvania, 3850 Spruce St., Philadelphia, PA 19104, USA f Department of Orthopaedic Surgery (Box 8233), Washington University School of Medicine, 660 S. Euclid Ave, St Louis, MO 63110, USA b
a r t i c l e
i n f o
Article history: Received 21 January 2013 Received in revised form 7 June 2013 Accepted 10 June 2013 Keywords: Micro-computed tomography Dog Paleopathology Shoulder Prehistoric Native American archeological site
a b s t r a c t A proximal humeral articular surface from an ancient domestic dog deliberate burial was examined during survey of small mammal bones from a prehistoric early Late Woodland archeological site. An unusual lesion on the caudolateral articular surface prompted micro-computed tomography to define detailed structure. Results indicate cortical or immature woven bone arising subchondrally, replacing normal trabeculae, extending through a breach in the cortical surface, and having sharp transition with surrounding normal bone. Organized bone within the lesion indicates that the dog lived for monthsto-years following insult. Differential diagnoses initially included: sharp penetrating trauma; intrinsic or extrinsic blunt fracturing force; osteochondrosis or complication of an osteochondral lesion; unusual osteoarthritis; and neoplasia. Computed tomography ruled out normal or unusual osteoarthritis, and neoplasia. The nature and small size of the lesion, relatively small size of the dog, and lack of evidence for complicating infection, suggest against sharp penetrating trauma as a sole cause. The most plausible differential diagnoses include: uncommon fracture-producing force in a companion animal, and blunt intrinsic or extrinsic force causing fracture at a weak point, such as an early osteochondral lesion, that was obliterated by healing. Combined gross examination, micro-computed tomography, and archeological-anthropological influences facilitated refinement of differential diagnosis. © 2013 Elsevier Inc. All rights reserved.
1. Introduction Paleopathology of ancient domestic and non-domestic animals presents diagnostic challenges that seldom are encountered in the living or recently deceased. Whole animal skeletons are uncommon finds in archeology, although deliberate burial, environments favoring preservation, mummification, or rapid postmortem freezing, occasionally provide more complete remains. Taphonomic damage may complicate diagnostic evaluation because most
∗ Corresponding author at: 1105 Woodleaf Dr, O’Fallon, IL 62269, USA. Tel.: +1 618 632 4950. E-mail addresses: [email protected], [email protected] (D.F. Lawler), [email protected] (D.A. Rubin), revans@pacificmmc.org (R.H. Evans), [email protected] (C.F. Hildebolt), [email protected] (K.E. Smith), [email protected] (C. Widga), [email protected] (T.J. Martin), [email protected] (M. Siegel), [email protected] (J.E. Sackman), [email protected] (G.K. Smith), [email protected] (T.K. Patel). 1879-9817/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ijpp.2013.06.001
archeological specimens are unprotected by burial containment. Unfused epiphyses can be lost from skeletal components of immature animals, and evidence of gnawing or partial consumption by other animals is observed frequently. Thus, the tissues that most frequently are available for study are cartilage-free bones that can be complete, but more often are fragmentary. We evaluate an unusual bony mass on the proximal humeral articular surface of an ancient domestic dog (Canis lupus familiaris L., 1758). This lesion occurs near the usual location of osteochondritis dissecans in the domestic dog shoulder, but it has an atypical appearance. Knowledge of the prehistoric human culture at the archeological site was used, along with micro-computed tomography, to develop and refine differential diagnoses. 2. Materials and methods An unusual linear, raised mass was observed on the proximal caudolateral articular surface of a complete left humerus, from a
D.F. Lawler et al. / International Journal of Paleopathology 3 (2013) 282–287
283
Fig. 1. Map showing the Weaver Native American archeological site in the Illinois River Valley, Fulton County (larger inset), Illinois (smaller inset), USA.
deliberate burial of an ancient domestic dog. The specimen is from one of three known dog burials at the Weaver archeological site (11F229), a prehistoric Middle Woodland Native American habitation, located along the central Illinois River valley in Fulton County, Illinois (Wray and MacNeish, 1961) (Fig. 1). Skeletal remains from this site are curated at the Illinois State Museum in Springfield, Illinois. Proximal and distal components of the humerus were evaluated for pathology of articular and periarticular structures, with magnification (2.5× or 20×) as needed. Photographs were taken, and observations were recorded. The specimen was transported to Washington University School of Medicine (St. Louis, MO) for noninvasive imaging studies using micro-computed tomography with a vivaCT 40 (Scanco Medical, Switzerland) made available for the project through the musculoskeletal structure and strength core. Imaging parameters were 70 kV, 114 A, 1024 × 1024 matrix, 500 projections, and 38-m isotropic voxels.
reveals a mild periarticular rim. The caudal articular surface reveals mildly prominent articular margins (Fig. 5); the cranial articular surface is near-normal, exhibiting some post-deposition wear. A linear crack in the humeral shaft reflects post-mortem sub-aerial weathering (Stage 1; Fig. 6) (Behrensmeyer, 1978). A corresponding scapular articular surface is not available, thus preventing evaluation of the entire bony shoulder joint. Initial differential diagnosis includes (Cordy and Wind, 1969; Olsson and Reiland, 1978; Milton, 1983; Ekman and Carlson, 1998; Olive et al., 2009): (1) sharp penetrating trauma; (2) extrinsic or intrinsic blunt fracturing force; (3) complication of an osteochondral lesion (OCL); (4) unusual manifestation of osteoarthritis; and (5) neoplasia. Multiple simultaneous causes also were considered to be possible. Micro-computed tomography images confirm the size of the exophytic, ossified mass that protrudes from the articular surface of the bone (Fig. 7a). Cross-sectional images (Fig. 7b–d) show that the lesion is composed
3. Results The dog’s chronological ages at the times of injury and subsequent death are unknown. However, based on closed proximal epiphyses, its age would have been older than about one and one-half years at the time of death (Sisson, 1975). The humerus length is 14.6 cm, and mid-shaft breadth is 10.08 mm. Bone dimensions and closed epiphyses indicate an adult of rather moderate size, with living shoulder height approximating 23.5 cm (Harcourt, 1974), and body mass estimated at 11.0 kg (Anyonge, 1993). Gross examination reveals mild, periarticular, osteoarthritic, rim formation around morphologically intact proximal articular surfaces (Fig. 2). The floor of the intertubercular groove is smooth, while its superior-lateral aspect is slightly rough and irregular. A linear extraction artifact also is visible (Fig. 3). The caudolateral articular surface contains a roughly linear, exophytic ridge, measuring 9 mm × 1 mm × 1 mm, emanating from subchondral bone, extending upward toward the original joint space, and oriented approximately with the long axis of the articulation (Fig. 4). The medial aspect of the distal humerus is normal. The lateral aspect
Fig. 2. Left proximal humerus, caudomedial view. Rough lateral margin of intertubercular groove (one arrow); raised mineralized lesion (two arrows); periarticular osteophytic rim (three arrows).
284
D.F. Lawler et al. / International Journal of Paleopathology 3 (2013) 282–287
of compact or immature woven bone, arising from the subchondral bone, where it replaces normal trabeculae. The lesion extends through a breach in the cortical surface (subchondral plate), thereby penetrating into the area that was occupied by articular cartilage, and potentially into the joint space, depending on thickness of the hyaline cartilage at the time of occurrence. There is a sharp interface between the osseous lesion and the surrounding trabecular bone and cortical bone, the latter appearing normal. The sharp zone of transition between the surrounding normal bone and the lesion, and the full-thickness defect in the bony articular surface, suggest traumatic etiology. The organized bone within the lesion indicates that the dog lived for months-to-years after the insult. 4. Discussion
Fig. 3. Left proximal humerus, intertubercular groove area. Normal intertubercular groove floor (one arrow); cut-like artifact of extraction from ground (two arrows); rough superior-lateral margin (three arrows).
The boney articular surface lesion would preclude many activities if it caused symptoms, but whether pain and accompanying degrees of debilitation occurred is unknown. Penetration of such a lesion through articular cartilage and into a joint space causes a symptomatic inflammatory response. However, a lesion of this nature, remaining within articular cartilage, could be minimally symptomatic in the absence of secondary complications (Ekman and Carlson, 1998). Lacking definitive information on the clinical status of this dog, we further examined differential diagnoses in both pathological and cultural contexts. 4.1. Sharp trauma
Fig. 4. Left proximal humerus, caudal aspect. Raised linear, mineralized lesion (three arrows).
Fig. 5. Left distal humerus, posterior aspect. Prominent articular margins, lateral more blunt (2 arrows), medial more sharp (3 arrows).
A narrow, sharp object, penetrating a joint space with sufficient force and a proper angle, conceivably could produce a linear laceration of articular cartilage and fracture of subchondral bone. In the dog, the articular surface of the proximal humerus has about twice the area of the articular surface of the corresponding scapular glenoid fossa (Olmstead, 1993). As a result, portions of the humeral articular surface are not shielded by bone at some points during rotation, thus conceptually favoring sharp penetrating trauma. However, a deep penetrating injury to a small or moderate-sized dog should produce considerable soft tissue damage. The high risk of subsequent secondary infection in a subsistence environment, with no access to surgery and medically appropriate supportive care, would greatly reduce the possibility of the long-term survival that the specimen suggests. Furthermore, the tool kit of Late Woodland Native American groups did not contain many instruments that were compatible with inflicting the injury observed in this specimen. Chipped stone projectile points were too large, as were smaller-diameter stone tools (bladelets, drills) that also are present in Late Woodland assemblages (Fortier, 2000). The Weaver assemblage does contain bone and antler fragments modified into pointed awls, shuttles, and perforating tools (Wray and MacNeish, 1961). Could these tools penetrate skin, muscle, tendon, ligament, and joint capsule, then lacerate the hard, moist, smooth articular cartilage, fracture underlying bone, and still cause minimal secondary complications in a relatively moderate-sized and subsistence-living dog? Whether any of the smaller and sharper aforementioned tools could be applied with sufficient force to produce the observed lesion, is questionable. Finally, although violent human conflict occurred during the Late Woodland period, it was less common than during subsequent Mississippian times (Milner, 1999). Nonetheless, despite these caveats, an isolated, deliberate human act cannot be ruled out. Findings of only minor accompanying osteoarthritis, evident lengthy post-injury survival, and relatively few potentially causal objects, are important factors with respect to penetrating sharp trauma. They suggest that the soft tissue trauma and subsequent infection that would have accompanied a penetrating injury to the
D.F. Lawler et al. / International Journal of Paleopathology 3 (2013) 282–287
285
Fig. 6. Left humeral shaft, longitudinal crack, post-mortem artifact.
shoulder of this dog are unlikely. Thus, accidental sharp penetration, while possible, seems an implausible sole explanation. 4.2. Blunt fracturing trauma A full-thickness, linear, articular surface fracture, extending through the cartilage and subchondral plate, is a possible
originating insult. A sufficiently severe blunt trauma, of extrinsic or intrinsic origin, might produce the force vectors necessary to cause such a fracture, particularly to a smaller dog or to an immature skeleton. Our evaluation indicates that the dog was mature at the time of death. However, age at the time of the originating insult is unknown. It could have occurred either before or after the time of skeletal maturity. Search of pertinent clinical and orthopedic
Fig. 7. Micro-CT images. (a) 3D surface reconstruction shows approximately rectangular exophytic lesion with rough surface arising from articular surface of proximal humerus; (b) 2D section shows lesion (*) composed of compact bone, extending through defect in subchondral plate (arrowheads). Note sharp demarcation between normal trabecular bone and base of lesion (arrows); (c) and (d) 2D sections orthogonal to (b).
286
D.F. Lawler et al. / International Journal of Paleopathology 3 (2013) 282–287
literature did not reveal other instances of fractures of this nature in modern dogs, thus supporting the possibility of an unusual causal component, such as a pre-existing susceptibility or focal weakness, compounded by blunt trauma. While the possibility remains that intrinsic or extrinsic blunt trauma alone caused the lesion, the physiological likelihood seems low. Dogs were important in hunting during the Woodland period, and hunting-related accidents are risks of subsistence life. For dogs, such injuries could include being thrown or kicked by a large, strong opponent, such as a black bear, elk, or white-tailed deer. However, a 9 mm isolated trauma to an articular surface in a smaller dog, sparing the remainder of the bone, suggests a much less violent insult. 4.3. Complications of an osteochondral lesion (OCL) OCL is a separation of bone and cartilage from an articular surface, usually resulting from chronic repetitive injury and most often initiating from focal loss of vascular support to growth cartilage (Probst and Johnston, 1993; Ytrehus et al., 2007). OCL represents a mechanically weakened focus that can heal spontaneously, remain in situ, or dislodge and leave an empty crater behind to heal by fibrocartilage deposition (Olsson, 1982; Yoshida et al., 1998). Secondary and significant osteoarthritis complicates the latter. In any event, OCLs where a fragment detaches, remains attached, or is partly attached typically do not heal with raised bony prominences (Yoshida et al., 1998; Mihara et al., 2009). While the location of the lesion could be compatible with OCL, another problematic aspect is the shape. Insofar as we are aware, OCLs at the observed site are circular, not linear (Olsson and Reiland, 1978; Olsson, 1982; Milton, 1983; Probst and Johnston, 1993; Ekman and Carlson, 1998). Search of pertinent literature revealed no report of linear OCL on the caudolateral articular surface of the canine proximal humerus. The lesion we evaluated thus would represent an unreported outcome of uncomplicated OCL in the dog shoulder. The probability of fortuitously encountering an evidently rare lesion at a huntergatherer-horticultural archeological site should be considered. One additional trauma-associated diagnostic possibility could be a focal tissue weakness associated with an early OCL and subjected to concentrated stress forces that became complicated by force-induced fracture. OCL at this location is a common event in the dog (Olsson and Reiland, 1978; Olsson, 1982; Milton, 1983; Probst and Johnston, 1993; Ekman and Carlson, 1998), and early OCL would be obscured by a healing minor fracture over the site. This possibility is speculative, but is compatible with the morphological evidence. 4.4. Osteoarthritis Adult humans with degenerative arthritis occasionally experience full-thickness articular cartilage defects that heal by an intra-cartilaginous osteophyte that grows into the gap left behind by the defect (Kindynis et al., 1990; McCauley et al., 2001). However, CT scanning reveals that the latter are composed of trabecular bone, not compact bone, and are thus unlike our specimen. Thoroughbred and standardbred racehorses with osteoarthritis of the metacarpophalangeal joint develop the expected periarticular osteophytes, and also may develop osteophytes at central subchondral sites (Olive et al., 2009). Magnetic Resonance Imaging (MRI) identifies these as smooth-surfaced, dome-shaped subchondral plate lesions that develop into the articular cartilage (Olive et al., 2009). While metacarpophalangeal central subchondral osteophytes in horses resemble similarly placed lesions that occur in the human knee joint (Abrahim-Zadeh et al., 1994; McCauley et al., 2001; Gold et al., 2003; Olive et al., 2009), the lesion in this dog specimen is neither smooth-surfaced nor dome-shaped.
4.5. Neoplasia Neoplasia was ruled out of the differential diagnosis for several reasons. Aggressive sarcomas, like osteosarcoma and chondrosarcoma, destroy bone cortex and trabeculae, extending into soft tissues. Osteochondromas are exophytic, benign tumors that emanate from bone surfaces, are composed of trabecular bone with cartilaginous caps, and involve metaphyses of long bones but not articular surfaces (Murphey et al., 2000). Osteomas are surface lesions, composed of compact bone, that originate from the periosteum of long bone shafts, but not from articular surfaces (Bertoni et al., 1995). Enostoses are benign lesions composed of compact bone. They occur within bone and do not extend through the subchondral plate (Greenspan, 1995). The lesion we evaluated resembles none of these bone neoplasms. 4.6. Anthropological-archeological observations The Weaver archeological site is one of a series of known Native American sites in the central part of the Illinois River valley, spanning the last 12,000 years. At the time of occupation, the geographical area was biologically diverse. The Weaver community had a subsistence base that consisted primarily of hunted and gathered resources, with minimal reliance on domesticated plants (Wray and MacNeish, 1961). There are no direct dates for the Weaver site assemblage, but cultural materials found at the site indicate affinity to other early Late Woodland, Weaver ‘variant’ sites in central and western Illinois. These latter sites date between AD 250 and AD 600 (Green and Nolan, 2000). Discovering this unusual lesion in an archeological setting, in a context of deliberate burial in a hunter-gatherer-horticultural community, is a subject for discussion of probability, and suggests that known mechanistic possibilities should be given the strongest consideration. The role of the domestic dog in prehistoric Native American cultures of what is now the Midwestern United States is not entirely clear. There is little evidence of deliberate trauma on dog remains from Weaver or other Late Woodland period assemblages. However, Warren (2000) suggested abuse of dogs in the North American Southeast during the earlier Archaic period, opening the possibility of varying cultural influences. Historical accounts of Native American societies suggest that they kept dogs for protection, as components of transport and hunting technology systems, and as participants in their ceremonial sphere (Schwartz, 1997; Morey, 2010). They also were companion animals, although perhaps not in the modern sense. It is impossible to extrapolate to general cultural patterns based on a single specimen, but the dog that is the subject of this study offers a degree of insight into cultural roles of dogs of the time. Specifically, the supportive care and the protection that would have been necessary to reach the state of healing present in this individual suggests a close human–dog bond, regardless of whether this bond was based in affection or the practical considerations of work and defense (Thomas, 2005). Deliberate burial suggests significant regard (Wray and MacNeish, 1961; Morey, 2006). Dog burials are (comparatively) common in Woodland cultures of Midwestern North America. They occur in singles and multiples, inside and outside of living areas, and at times, they are associated with human mortuary settings. Their prevalence in the archeological record, combined with their close associations with human living and mortuary spaces, indicates that canine companions were a vital part of Native American communities in this region and during this time. 5. Conclusion We evaluated and synthesized relevant information from several scientific disciplines, to identify differential diagnoses that
D.F. Lawler et al. / International Journal of Paleopathology 3 (2013) 282–287
might be more plausible (uncommon fracture-producing accident in a companion; accident resulting from normally non-violent activity such as sentry work; fracture-producing complication of early OCL) and less plausible (sharp external penetration; independent OCL; accident resulting from violent work, such as warfare or hunting; complication of osteoarthritis). While a definitive diagnosis was not firmly established in the given circumstance, the field of differential diagnoses was narrowed considerably by interdisciplinary synthesis of morphology, micro-computed tomography scanning, and anthropological information. References Abrahim-Zadeh, R., Yu, J.S., Resnick, D., 1994. Central (interior) osteophytes of the distal femur. Imaging and pathologic findings. Invest. Radiol. 29, 1001–1005. Anyonge, W., 1993. Body mass in large extant and extinct carnivores. J. Zool. 231 (2), 339–350. Behrensmeyer, A.K., 1978. Taphonomic and ecologic information from bone weathering. Paleobiology 4 (2), 150–162. Bertoni, F., Unni, K.K., Beabout, J.W., Sim, F.H., 1995. Parosteal osteoma of bones other than of the skull and face. Cancer 75 (10), 2466–2473. Cordy, D.R., Wind, A.P., 1969. Transverse fracture of the proximal humeral articular cartilage in dogs (so called osteochondritis dissecans). Pathol. Vet. 6 (5), 424–436. Ekman, S., Carlson, C.S., 1998. The pathophysiology of osteochondrosis. Vet. Clin. North Am. 28 (1), 17–52. Fortier, A.C., 2000. The emergence and demise of the middle Woodland small-tool tradition in the American Bottom. Midcont. J. Archaeol. 25 (2), 191–213. Gold, G.E., McCauley, T.R., Gray, M.L., Disler, D.G., 2003. What’s new in cartilage? Radiographics 23, 1227–1242. Green, W., Nolan, D.J., 2000. Late Woodland peoples in West-Central Illinois. In: Emerson, T.E., McElrath, E.L., Fortier, A.C. (Eds.), Late Woodland Societies: Tradition and Transformation across the Midcontinent. University of Nebraska Press, Lincoln, pp. 345–386. Greenspan, A., 1995. Bone island (enostosis): current concept – a review. Skeletal Radiol. 24 (2), 111–115. Harcourt, R.A., 1974. The dog in prehistoric and early historic Britain. J. Archaeol. Sci. 1 (2), 151–175. Kindynis, P., Haller, J., Kang, H.S., Resnick, D., Sartoris, D.J., Trudell, D., Tyson, R., 1990. Osteophytosis of the knee: anatomic, radiologic, and pathologic investigation. Radiology 174, 841–846. McCauley, T.R., Kornaat, P.R., Jee, W.H., 2001. Central osteophytes in the knee: prevalence and association with cartilage defects on MR imaging. Am. J. Roentgenol. 176 (2), 359–364.
287
Mihara, K., Tsutsui, H., Nishinaka, N., Yamaguchi, K., 2009. Nonoperative treatment for osteochondritis dissecans of the capitellum. Am. J. Sports Med. 37 (2), 298–304. Milner, G.R., 1999. Warfare in prehistoric and early historic eastern North America. J. Archaeol. Res. 7 (2), 105–151. Milton, J.L., 1983. Osteochondritis dissecans in the dog. Vet. Clin. North Am. 13, 117–134. Morey, D.F., 2006. Burying key evidence: the social bond between dogs and people. J. Archaeol. Sci. 33 (2), 158–175. Morey, D.F., 2010. Dogs: Domestication and the Development of a Social Bond. Cambridge University Press, New York, NY. Murphey, M.D., Choi, J.J., Kransdorf, M.J., Flemming, D.J., Gannon, F.H., 2000. Imaging of osteochondroma: variants and complications with radiologic-pathologic correlation. Radiographics 20 (5), 1407–1434. Olive, J., D’Anjou, M.A., Girard, C., Laverty, S., 2009. Imaging and histological features of central subchondral osteophytes in racehorses with metacarpophalangeal joint osteoarthritis. Equine Vet. J. 41 (9), 859–864. Olmstead, M.L., 1993. Fractures of the humerus. In: Slatter, D. (Ed.), Textbook of Small Animal Surgery, vol. II, 2nd ed. W.B. Saunders, Philadelphia, pp. 1716–1728. Olsson, S.-E., 1982. Morphology and physiology of the growth cartilage under normal and pathologic conditions. In: Sumner-Smith, G. (Ed.), Bone in Clinical Orthopaedics. W.B. Saunders, Philadelphia, pp. 159–196. Olsson, S.-E., Reiland, S., 1978. The nature of osteochondrosis in animals. Summary and conclusions with comparative aspects on osteochondrosis dissecans in man. Acta Radiol. 358, 299–306. Probst, C.W., Johnston, S.A., 1993. Osteochondrosis. In: Slater, D. (Ed.), Textbook of Small Animal Surgery. , 2nd ed. W.B. Saunders, Philadelphia, pp. 1944–1966. Schwartz, M., 1997. A History of Dogs in the Early Americas. Yale University Press, New Haven. Sisson, S., 1975. Carnivore osteology. In: Getty, R. (Ed.), Sisson and Grossman’s The Anatomy of the Domestic Animals. , 5th ed. W.B. Saunders, Philadelphia, pp. 1427–1503. Thomas, R., 2005. Perceptions versus reality: changing attitudes towards pets in medieval and post-medieval England. In: Pluskowski, A. (Ed.), Just Skin and Bones? New Perspectives on Human–Animal Relations in the Historical Past. BAR International Series, No. 1410. Archaeopress, Oxford, England, pp. 95–104. Warren, D.M., 2000. Paleopathology of Archaic period dogs from the North American Southeast. In: Crockford, S. (Ed.), Dogs Through Time. An Archaeological Perspective. BAR International Series No. 889. Archaeopress, Oxford, England, pp. 105–114. Wray, D.E., MacNeish, R.S., 1961. The Hopewellian and Weaver Occupations of the Weaver Site, Fulton County, Illinois. Scientific Papers 7(2), Springfield, Illinois State Museum. Yoshida, S., Ikata, T., Takai, H., Kashiwaguchi, S., Katoh, S., Takeda, Y., 1998. Osteochondritis dissecans of the femoral condyle in the growth stage. Clin. Orthop. Relat. Res. 346, 162–170. Ytrehus, B., Carlson, C.S., Eckman, S., 2007. Etiology and pathogenesis of osteochondrosis. Vet. Pathol. 44, 429–448.
Contents lists available at ScienceDirect
International Journal of Paleopathology journal homepage: www.elsevier.com/locate/ijpp
Case study
Differential diagnosis of an unusual shoulder articular lesion in an ancient domestic dog (Canis lupus familiaris L., 1758) D.F. Lawler a,∗ , D.A. Rubin b , R.H. Evans c , C.F. Hildebolt b , K.E. Smith b , C. Widga a , T.J. Martin a , M. Siegel b , J.E. Sackman d , G.K. Smith e , T.K. Patel f a
Illinois State Museum, Research and Collections Center, 1011 East Ash St., Springfield, IL 62703-3500, USA Malinckrodt Institute of Radiology (Box 8131), Washington University School of Medicine, 510 S. Kingshighway Blvd, St Louis, MO 63110, USA c Pacific Marine Mammal Center, 20612 Laguna Canyon Road, Laguna Beach, CA 92651, USA d Numerof & Associates, Inc., Four City Place Drive, Suite 430, St. Louis, MO 63141, USA e School of Veterinary Medicine, Department of Clinical Studies, University of Pennsylvania, 3850 Spruce St., Philadelphia, PA 19104, USA f Department of Orthopaedic Surgery (Box 8233), Washington University School of Medicine, 660 S. Euclid Ave, St Louis, MO 63110, USA b
a r t i c l e
i n f o
Article history: Received 21 January 2013 Received in revised form 7 June 2013 Accepted 10 June 2013 Keywords: Micro-computed tomography Dog Paleopathology Shoulder Prehistoric Native American archeological site
a b s t r a c t A proximal humeral articular surface from an ancient domestic dog deliberate burial was examined during survey of small mammal bones from a prehistoric early Late Woodland archeological site. An unusual lesion on the caudolateral articular surface prompted micro-computed tomography to define detailed structure. Results indicate cortical or immature woven bone arising subchondrally, replacing normal trabeculae, extending through a breach in the cortical surface, and having sharp transition with surrounding normal bone. Organized bone within the lesion indicates that the dog lived for monthsto-years following insult. Differential diagnoses initially included: sharp penetrating trauma; intrinsic or extrinsic blunt fracturing force; osteochondrosis or complication of an osteochondral lesion; unusual osteoarthritis; and neoplasia. Computed tomography ruled out normal or unusual osteoarthritis, and neoplasia. The nature and small size of the lesion, relatively small size of the dog, and lack of evidence for complicating infection, suggest against sharp penetrating trauma as a sole cause. The most plausible differential diagnoses include: uncommon fracture-producing force in a companion animal, and blunt intrinsic or extrinsic force causing fracture at a weak point, such as an early osteochondral lesion, that was obliterated by healing. Combined gross examination, micro-computed tomography, and archeological-anthropological influences facilitated refinement of differential diagnosis. © 2013 Elsevier Inc. All rights reserved.
1. Introduction Paleopathology of ancient domestic and non-domestic animals presents diagnostic challenges that seldom are encountered in the living or recently deceased. Whole animal skeletons are uncommon finds in archeology, although deliberate burial, environments favoring preservation, mummification, or rapid postmortem freezing, occasionally provide more complete remains. Taphonomic damage may complicate diagnostic evaluation because most
∗ Corresponding author at: 1105 Woodleaf Dr, O’Fallon, IL 62269, USA. Tel.: +1 618 632 4950. E-mail addresses: [email protected], [email protected] (D.F. Lawler), [email protected] (D.A. Rubin), revans@pacificmmc.org (R.H. Evans), [email protected] (C.F. Hildebolt), [email protected] (K.E. Smith), [email protected] (C. Widga), [email protected] (T.J. Martin), [email protected] (M. Siegel), [email protected] (J.E. Sackman), [email protected] (G.K. Smith), [email protected] (T.K. Patel). 1879-9817/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ijpp.2013.06.001
archeological specimens are unprotected by burial containment. Unfused epiphyses can be lost from skeletal components of immature animals, and evidence of gnawing or partial consumption by other animals is observed frequently. Thus, the tissues that most frequently are available for study are cartilage-free bones that can be complete, but more often are fragmentary. We evaluate an unusual bony mass on the proximal humeral articular surface of an ancient domestic dog (Canis lupus familiaris L., 1758). This lesion occurs near the usual location of osteochondritis dissecans in the domestic dog shoulder, but it has an atypical appearance. Knowledge of the prehistoric human culture at the archeological site was used, along with micro-computed tomography, to develop and refine differential diagnoses. 2. Materials and methods An unusual linear, raised mass was observed on the proximal caudolateral articular surface of a complete left humerus, from a
D.F. Lawler et al. / International Journal of Paleopathology 3 (2013) 282–287
283
Fig. 1. Map showing the Weaver Native American archeological site in the Illinois River Valley, Fulton County (larger inset), Illinois (smaller inset), USA.
deliberate burial of an ancient domestic dog. The specimen is from one of three known dog burials at the Weaver archeological site (11F229), a prehistoric Middle Woodland Native American habitation, located along the central Illinois River valley in Fulton County, Illinois (Wray and MacNeish, 1961) (Fig. 1). Skeletal remains from this site are curated at the Illinois State Museum in Springfield, Illinois. Proximal and distal components of the humerus were evaluated for pathology of articular and periarticular structures, with magnification (2.5× or 20×) as needed. Photographs were taken, and observations were recorded. The specimen was transported to Washington University School of Medicine (St. Louis, MO) for noninvasive imaging studies using micro-computed tomography with a vivaCT 40 (Scanco Medical, Switzerland) made available for the project through the musculoskeletal structure and strength core. Imaging parameters were 70 kV, 114 A, 1024 × 1024 matrix, 500 projections, and 38-m isotropic voxels.
reveals a mild periarticular rim. The caudal articular surface reveals mildly prominent articular margins (Fig. 5); the cranial articular surface is near-normal, exhibiting some post-deposition wear. A linear crack in the humeral shaft reflects post-mortem sub-aerial weathering (Stage 1; Fig. 6) (Behrensmeyer, 1978). A corresponding scapular articular surface is not available, thus preventing evaluation of the entire bony shoulder joint. Initial differential diagnosis includes (Cordy and Wind, 1969; Olsson and Reiland, 1978; Milton, 1983; Ekman and Carlson, 1998; Olive et al., 2009): (1) sharp penetrating trauma; (2) extrinsic or intrinsic blunt fracturing force; (3) complication of an osteochondral lesion (OCL); (4) unusual manifestation of osteoarthritis; and (5) neoplasia. Multiple simultaneous causes also were considered to be possible. Micro-computed tomography images confirm the size of the exophytic, ossified mass that protrudes from the articular surface of the bone (Fig. 7a). Cross-sectional images (Fig. 7b–d) show that the lesion is composed
3. Results The dog’s chronological ages at the times of injury and subsequent death are unknown. However, based on closed proximal epiphyses, its age would have been older than about one and one-half years at the time of death (Sisson, 1975). The humerus length is 14.6 cm, and mid-shaft breadth is 10.08 mm. Bone dimensions and closed epiphyses indicate an adult of rather moderate size, with living shoulder height approximating 23.5 cm (Harcourt, 1974), and body mass estimated at 11.0 kg (Anyonge, 1993). Gross examination reveals mild, periarticular, osteoarthritic, rim formation around morphologically intact proximal articular surfaces (Fig. 2). The floor of the intertubercular groove is smooth, while its superior-lateral aspect is slightly rough and irregular. A linear extraction artifact also is visible (Fig. 3). The caudolateral articular surface contains a roughly linear, exophytic ridge, measuring 9 mm × 1 mm × 1 mm, emanating from subchondral bone, extending upward toward the original joint space, and oriented approximately with the long axis of the articulation (Fig. 4). The medial aspect of the distal humerus is normal. The lateral aspect
Fig. 2. Left proximal humerus, caudomedial view. Rough lateral margin of intertubercular groove (one arrow); raised mineralized lesion (two arrows); periarticular osteophytic rim (three arrows).
284
D.F. Lawler et al. / International Journal of Paleopathology 3 (2013) 282–287
of compact or immature woven bone, arising from the subchondral bone, where it replaces normal trabeculae. The lesion extends through a breach in the cortical surface (subchondral plate), thereby penetrating into the area that was occupied by articular cartilage, and potentially into the joint space, depending on thickness of the hyaline cartilage at the time of occurrence. There is a sharp interface between the osseous lesion and the surrounding trabecular bone and cortical bone, the latter appearing normal. The sharp zone of transition between the surrounding normal bone and the lesion, and the full-thickness defect in the bony articular surface, suggest traumatic etiology. The organized bone within the lesion indicates that the dog lived for months-to-years after the insult. 4. Discussion
Fig. 3. Left proximal humerus, intertubercular groove area. Normal intertubercular groove floor (one arrow); cut-like artifact of extraction from ground (two arrows); rough superior-lateral margin (three arrows).
The boney articular surface lesion would preclude many activities if it caused symptoms, but whether pain and accompanying degrees of debilitation occurred is unknown. Penetration of such a lesion through articular cartilage and into a joint space causes a symptomatic inflammatory response. However, a lesion of this nature, remaining within articular cartilage, could be minimally symptomatic in the absence of secondary complications (Ekman and Carlson, 1998). Lacking definitive information on the clinical status of this dog, we further examined differential diagnoses in both pathological and cultural contexts. 4.1. Sharp trauma
Fig. 4. Left proximal humerus, caudal aspect. Raised linear, mineralized lesion (three arrows).
Fig. 5. Left distal humerus, posterior aspect. Prominent articular margins, lateral more blunt (2 arrows), medial more sharp (3 arrows).
A narrow, sharp object, penetrating a joint space with sufficient force and a proper angle, conceivably could produce a linear laceration of articular cartilage and fracture of subchondral bone. In the dog, the articular surface of the proximal humerus has about twice the area of the articular surface of the corresponding scapular glenoid fossa (Olmstead, 1993). As a result, portions of the humeral articular surface are not shielded by bone at some points during rotation, thus conceptually favoring sharp penetrating trauma. However, a deep penetrating injury to a small or moderate-sized dog should produce considerable soft tissue damage. The high risk of subsequent secondary infection in a subsistence environment, with no access to surgery and medically appropriate supportive care, would greatly reduce the possibility of the long-term survival that the specimen suggests. Furthermore, the tool kit of Late Woodland Native American groups did not contain many instruments that were compatible with inflicting the injury observed in this specimen. Chipped stone projectile points were too large, as were smaller-diameter stone tools (bladelets, drills) that also are present in Late Woodland assemblages (Fortier, 2000). The Weaver assemblage does contain bone and antler fragments modified into pointed awls, shuttles, and perforating tools (Wray and MacNeish, 1961). Could these tools penetrate skin, muscle, tendon, ligament, and joint capsule, then lacerate the hard, moist, smooth articular cartilage, fracture underlying bone, and still cause minimal secondary complications in a relatively moderate-sized and subsistence-living dog? Whether any of the smaller and sharper aforementioned tools could be applied with sufficient force to produce the observed lesion, is questionable. Finally, although violent human conflict occurred during the Late Woodland period, it was less common than during subsequent Mississippian times (Milner, 1999). Nonetheless, despite these caveats, an isolated, deliberate human act cannot be ruled out. Findings of only minor accompanying osteoarthritis, evident lengthy post-injury survival, and relatively few potentially causal objects, are important factors with respect to penetrating sharp trauma. They suggest that the soft tissue trauma and subsequent infection that would have accompanied a penetrating injury to the
D.F. Lawler et al. / International Journal of Paleopathology 3 (2013) 282–287
285
Fig. 6. Left humeral shaft, longitudinal crack, post-mortem artifact.
shoulder of this dog are unlikely. Thus, accidental sharp penetration, while possible, seems an implausible sole explanation. 4.2. Blunt fracturing trauma A full-thickness, linear, articular surface fracture, extending through the cartilage and subchondral plate, is a possible
originating insult. A sufficiently severe blunt trauma, of extrinsic or intrinsic origin, might produce the force vectors necessary to cause such a fracture, particularly to a smaller dog or to an immature skeleton. Our evaluation indicates that the dog was mature at the time of death. However, age at the time of the originating insult is unknown. It could have occurred either before or after the time of skeletal maturity. Search of pertinent clinical and orthopedic
Fig. 7. Micro-CT images. (a) 3D surface reconstruction shows approximately rectangular exophytic lesion with rough surface arising from articular surface of proximal humerus; (b) 2D section shows lesion (*) composed of compact bone, extending through defect in subchondral plate (arrowheads). Note sharp demarcation between normal trabecular bone and base of lesion (arrows); (c) and (d) 2D sections orthogonal to (b).
286
D.F. Lawler et al. / International Journal of Paleopathology 3 (2013) 282–287
literature did not reveal other instances of fractures of this nature in modern dogs, thus supporting the possibility of an unusual causal component, such as a pre-existing susceptibility or focal weakness, compounded by blunt trauma. While the possibility remains that intrinsic or extrinsic blunt trauma alone caused the lesion, the physiological likelihood seems low. Dogs were important in hunting during the Woodland period, and hunting-related accidents are risks of subsistence life. For dogs, such injuries could include being thrown or kicked by a large, strong opponent, such as a black bear, elk, or white-tailed deer. However, a 9 mm isolated trauma to an articular surface in a smaller dog, sparing the remainder of the bone, suggests a much less violent insult. 4.3. Complications of an osteochondral lesion (OCL) OCL is a separation of bone and cartilage from an articular surface, usually resulting from chronic repetitive injury and most often initiating from focal loss of vascular support to growth cartilage (Probst and Johnston, 1993; Ytrehus et al., 2007). OCL represents a mechanically weakened focus that can heal spontaneously, remain in situ, or dislodge and leave an empty crater behind to heal by fibrocartilage deposition (Olsson, 1982; Yoshida et al., 1998). Secondary and significant osteoarthritis complicates the latter. In any event, OCLs where a fragment detaches, remains attached, or is partly attached typically do not heal with raised bony prominences (Yoshida et al., 1998; Mihara et al., 2009). While the location of the lesion could be compatible with OCL, another problematic aspect is the shape. Insofar as we are aware, OCLs at the observed site are circular, not linear (Olsson and Reiland, 1978; Olsson, 1982; Milton, 1983; Probst and Johnston, 1993; Ekman and Carlson, 1998). Search of pertinent literature revealed no report of linear OCL on the caudolateral articular surface of the canine proximal humerus. The lesion we evaluated thus would represent an unreported outcome of uncomplicated OCL in the dog shoulder. The probability of fortuitously encountering an evidently rare lesion at a huntergatherer-horticultural archeological site should be considered. One additional trauma-associated diagnostic possibility could be a focal tissue weakness associated with an early OCL and subjected to concentrated stress forces that became complicated by force-induced fracture. OCL at this location is a common event in the dog (Olsson and Reiland, 1978; Olsson, 1982; Milton, 1983; Probst and Johnston, 1993; Ekman and Carlson, 1998), and early OCL would be obscured by a healing minor fracture over the site. This possibility is speculative, but is compatible with the morphological evidence. 4.4. Osteoarthritis Adult humans with degenerative arthritis occasionally experience full-thickness articular cartilage defects that heal by an intra-cartilaginous osteophyte that grows into the gap left behind by the defect (Kindynis et al., 1990; McCauley et al., 2001). However, CT scanning reveals that the latter are composed of trabecular bone, not compact bone, and are thus unlike our specimen. Thoroughbred and standardbred racehorses with osteoarthritis of the metacarpophalangeal joint develop the expected periarticular osteophytes, and also may develop osteophytes at central subchondral sites (Olive et al., 2009). Magnetic Resonance Imaging (MRI) identifies these as smooth-surfaced, dome-shaped subchondral plate lesions that develop into the articular cartilage (Olive et al., 2009). While metacarpophalangeal central subchondral osteophytes in horses resemble similarly placed lesions that occur in the human knee joint (Abrahim-Zadeh et al., 1994; McCauley et al., 2001; Gold et al., 2003; Olive et al., 2009), the lesion in this dog specimen is neither smooth-surfaced nor dome-shaped.
4.5. Neoplasia Neoplasia was ruled out of the differential diagnosis for several reasons. Aggressive sarcomas, like osteosarcoma and chondrosarcoma, destroy bone cortex and trabeculae, extending into soft tissues. Osteochondromas are exophytic, benign tumors that emanate from bone surfaces, are composed of trabecular bone with cartilaginous caps, and involve metaphyses of long bones but not articular surfaces (Murphey et al., 2000). Osteomas are surface lesions, composed of compact bone, that originate from the periosteum of long bone shafts, but not from articular surfaces (Bertoni et al., 1995). Enostoses are benign lesions composed of compact bone. They occur within bone and do not extend through the subchondral plate (Greenspan, 1995). The lesion we evaluated resembles none of these bone neoplasms. 4.6. Anthropological-archeological observations The Weaver archeological site is one of a series of known Native American sites in the central part of the Illinois River valley, spanning the last 12,000 years. At the time of occupation, the geographical area was biologically diverse. The Weaver community had a subsistence base that consisted primarily of hunted and gathered resources, with minimal reliance on domesticated plants (Wray and MacNeish, 1961). There are no direct dates for the Weaver site assemblage, but cultural materials found at the site indicate affinity to other early Late Woodland, Weaver ‘variant’ sites in central and western Illinois. These latter sites date between AD 250 and AD 600 (Green and Nolan, 2000). Discovering this unusual lesion in an archeological setting, in a context of deliberate burial in a hunter-gatherer-horticultural community, is a subject for discussion of probability, and suggests that known mechanistic possibilities should be given the strongest consideration. The role of the domestic dog in prehistoric Native American cultures of what is now the Midwestern United States is not entirely clear. There is little evidence of deliberate trauma on dog remains from Weaver or other Late Woodland period assemblages. However, Warren (2000) suggested abuse of dogs in the North American Southeast during the earlier Archaic period, opening the possibility of varying cultural influences. Historical accounts of Native American societies suggest that they kept dogs for protection, as components of transport and hunting technology systems, and as participants in their ceremonial sphere (Schwartz, 1997; Morey, 2010). They also were companion animals, although perhaps not in the modern sense. It is impossible to extrapolate to general cultural patterns based on a single specimen, but the dog that is the subject of this study offers a degree of insight into cultural roles of dogs of the time. Specifically, the supportive care and the protection that would have been necessary to reach the state of healing present in this individual suggests a close human–dog bond, regardless of whether this bond was based in affection or the practical considerations of work and defense (Thomas, 2005). Deliberate burial suggests significant regard (Wray and MacNeish, 1961; Morey, 2006). Dog burials are (comparatively) common in Woodland cultures of Midwestern North America. They occur in singles and multiples, inside and outside of living areas, and at times, they are associated with human mortuary settings. Their prevalence in the archeological record, combined with their close associations with human living and mortuary spaces, indicates that canine companions were a vital part of Native American communities in this region and during this time. 5. Conclusion We evaluated and synthesized relevant information from several scientific disciplines, to identify differential diagnoses that
D.F. Lawler et al. / International Journal of Paleopathology 3 (2013) 282–287
might be more plausible (uncommon fracture-producing accident in a companion; accident resulting from normally non-violent activity such as sentry work; fracture-producing complication of early OCL) and less plausible (sharp external penetration; independent OCL; accident resulting from violent work, such as warfare or hunting; complication of osteoarthritis). While a definitive diagnosis was not firmly established in the given circumstance, the field of differential diagnoses was narrowed considerably by interdisciplinary synthesis of morphology, micro-computed tomography scanning, and anthropological information. References Abrahim-Zadeh, R., Yu, J.S., Resnick, D., 1994. Central (interior) osteophytes of the distal femur. Imaging and pathologic findings. Invest. Radiol. 29, 1001–1005. Anyonge, W., 1993. Body mass in large extant and extinct carnivores. J. Zool. 231 (2), 339–350. Behrensmeyer, A.K., 1978. Taphonomic and ecologic information from bone weathering. Paleobiology 4 (2), 150–162. Bertoni, F., Unni, K.K., Beabout, J.W., Sim, F.H., 1995. Parosteal osteoma of bones other than of the skull and face. Cancer 75 (10), 2466–2473. Cordy, D.R., Wind, A.P., 1969. Transverse fracture of the proximal humeral articular cartilage in dogs (so called osteochondritis dissecans). Pathol. Vet. 6 (5), 424–436. Ekman, S., Carlson, C.S., 1998. The pathophysiology of osteochondrosis. Vet. Clin. North Am. 28 (1), 17–52. Fortier, A.C., 2000. The emergence and demise of the middle Woodland small-tool tradition in the American Bottom. Midcont. J. Archaeol. 25 (2), 191–213. Gold, G.E., McCauley, T.R., Gray, M.L., Disler, D.G., 2003. What’s new in cartilage? Radiographics 23, 1227–1242. Green, W., Nolan, D.J., 2000. Late Woodland peoples in West-Central Illinois. In: Emerson, T.E., McElrath, E.L., Fortier, A.C. (Eds.), Late Woodland Societies: Tradition and Transformation across the Midcontinent. University of Nebraska Press, Lincoln, pp. 345–386. Greenspan, A., 1995. Bone island (enostosis): current concept – a review. Skeletal Radiol. 24 (2), 111–115. Harcourt, R.A., 1974. The dog in prehistoric and early historic Britain. J. Archaeol. Sci. 1 (2), 151–175. Kindynis, P., Haller, J., Kang, H.S., Resnick, D., Sartoris, D.J., Trudell, D., Tyson, R., 1990. Osteophytosis of the knee: anatomic, radiologic, and pathologic investigation. Radiology 174, 841–846. McCauley, T.R., Kornaat, P.R., Jee, W.H., 2001. Central osteophytes in the knee: prevalence and association with cartilage defects on MR imaging. Am. J. Roentgenol. 176 (2), 359–364.
287
Mihara, K., Tsutsui, H., Nishinaka, N., Yamaguchi, K., 2009. Nonoperative treatment for osteochondritis dissecans of the capitellum. Am. J. Sports Med. 37 (2), 298–304. Milner, G.R., 1999. Warfare in prehistoric and early historic eastern North America. J. Archaeol. Res. 7 (2), 105–151. Milton, J.L., 1983. Osteochondritis dissecans in the dog. Vet. Clin. North Am. 13, 117–134. Morey, D.F., 2006. Burying key evidence: the social bond between dogs and people. J. Archaeol. Sci. 33 (2), 158–175. Morey, D.F., 2010. Dogs: Domestication and the Development of a Social Bond. Cambridge University Press, New York, NY. Murphey, M.D., Choi, J.J., Kransdorf, M.J., Flemming, D.J., Gannon, F.H., 2000. Imaging of osteochondroma: variants and complications with radiologic-pathologic correlation. Radiographics 20 (5), 1407–1434. Olive, J., D’Anjou, M.A., Girard, C., Laverty, S., 2009. Imaging and histological features of central subchondral osteophytes in racehorses with metacarpophalangeal joint osteoarthritis. Equine Vet. J. 41 (9), 859–864. Olmstead, M.L., 1993. Fractures of the humerus. In: Slatter, D. (Ed.), Textbook of Small Animal Surgery, vol. II, 2nd ed. W.B. Saunders, Philadelphia, pp. 1716–1728. Olsson, S.-E., 1982. Morphology and physiology of the growth cartilage under normal and pathologic conditions. In: Sumner-Smith, G. (Ed.), Bone in Clinical Orthopaedics. W.B. Saunders, Philadelphia, pp. 159–196. Olsson, S.-E., Reiland, S., 1978. The nature of osteochondrosis in animals. Summary and conclusions with comparative aspects on osteochondrosis dissecans in man. Acta Radiol. 358, 299–306. Probst, C.W., Johnston, S.A., 1993. Osteochondrosis. In: Slater, D. (Ed.), Textbook of Small Animal Surgery. , 2nd ed. W.B. Saunders, Philadelphia, pp. 1944–1966. Schwartz, M., 1997. A History of Dogs in the Early Americas. Yale University Press, New Haven. Sisson, S., 1975. Carnivore osteology. In: Getty, R. (Ed.), Sisson and Grossman’s The Anatomy of the Domestic Animals. , 5th ed. W.B. Saunders, Philadelphia, pp. 1427–1503. Thomas, R., 2005. Perceptions versus reality: changing attitudes towards pets in medieval and post-medieval England. In: Pluskowski, A. (Ed.), Just Skin and Bones? New Perspectives on Human–Animal Relations in the Historical Past. BAR International Series, No. 1410. Archaeopress, Oxford, England, pp. 95–104. Warren, D.M., 2000. Paleopathology of Archaic period dogs from the North American Southeast. In: Crockford, S. (Ed.), Dogs Through Time. An Archaeological Perspective. BAR International Series No. 889. Archaeopress, Oxford, England, pp. 105–114. Wray, D.E., MacNeish, R.S., 1961. The Hopewellian and Weaver Occupations of the Weaver Site, Fulton County, Illinois. Scientific Papers 7(2), Springfield, Illinois State Museum. Yoshida, S., Ikata, T., Takai, H., Kashiwaguchi, S., Katoh, S., Takeda, Y., 1998. Osteochondritis dissecans of the femoral condyle in the growth stage. Clin. Orthop. Relat. Res. 346, 162–170. Ytrehus, B., Carlson, C.S., Eckman, S., 2007. Etiology and pathogenesis of osteochondrosis. Vet. Pathol. 44, 429–448.