Sage Journals: Discover world-class research

Abstract

Failure of the cartilage canal blood supply leads to ischemic chondronecrosis which causes osteochondrosis, and osteochondral lesions. Osteochondrosis is a disease with a heritable component and usually occurs under aseptic conditions. Because bacteria can bind to growth cartilage and disrupt the blood supply in pigs and chickens, we considered whether this might play a role in development of equine osteochondrosis. The aim of this study was to examine whether bacteria are present in canals in the growth cartilage of foals with septic arthritis/osteomyelitis, and whether this is associated with osteochondrosis. The material consisted of 7 foals aged 9-117 days euthanized because of septic arthritis/osteomyelitis. The 7 cases had 16 lesions in growth cartilage that were evaluated histologically. Bacteria were present in cartilage canals in foals with septic arthritis/osteomyelitis. Portions of necrotic canals adjacent to bacteria frequently contained neutrophils, termed acute septic canals; or granulation tissue with neutrophils, termed chronic septic canals. Acute and chronic septic canals were associated with ischemic chondronecrosis in the articular-epiphyseal cartilage complex (AECC) of 5 cases and in the physis of 2 cases, and ossification was focally delayed in 5 of those 7 cases. Lesions occurred with and without adjacent osteomyelitis. Bacteria were present in cartilage canals and were associated with focal chondronecrosis in both the AECC and the physis. This establishes sepsis as a plausible cause of some osteochondral lesions in horses. It is recommended that horses with sepsis-related osteochondral lesions may be used for breeding without increasing the prevalence of OCD-predisposing genes in the population.

Keywords

bacteria cartilage canals foal growth cartilage histology osteochondral lesions septic arthritis

The long bones of mammals grow by endochondral ossification.²⁹ The growth cartilage has a temporary blood supply that runs in cartilage canals, which usually contain an arteriole, a capillary network and one or more venules organized as anatomic end arteries, and perivascular mesenchymal cells.^5,26 This renders the blood supply vulnerable in case of damage. Cartilage canals regress by 2 mechanisms; chondrification, that is, becoming filled with cartilage,⁴ and incorporation into the advancing ossification front.^37,39 When predilection sites for osteochondrosis in the articular epiphyseal cartilage complex (AECC) were examined in pigs and horses, necrotic cartilage canals surrounded by necrotic chondrocytes were found at intermediate depth.^39,52 Both spontaneous and experimental vascular failure in the AECC result in ischemic chondronecrosis and delayed enchondral ossification, in pigs termed osteochondrosis latens and osteochondrosis manifesta, respectively.^34,48,51,53 Ischemic chondronecrosis and delayed enchondral ossification have the potential to resolve,⁷ or develop into osteochondrosis dissecans (OCD).³³ Secondary responses in the growth cartilage including proliferation of morphologically viable chondrocytes and canals have been observed adjacent to areas of ischemic chondronecrosis.³⁵ After areas of ischemic chondronecrosis are surrounded by the ossification front, repopulation of necrotic canals by cells from subchondral bone marrow has been observed.³⁵

Osteochondrosis is most often a multifocal disease and occurs at specific predilection sites in the joints of various species, including pigs,⁴¹ horses,⁴³ dogs,³¹ and humans.⁸ A heritable predisposition for osteochondrosis has been documented in horses^16,28,49 and pigs.^1,42 Most lesions of osteochondrosis occur under aseptic conditions.^10,21 Cartilage canals have, however, been experimentally established as a target for bacterial infection in pigs⁶ and chickens.⁴⁶ On transmission electron microscopy, bacteria were observed in the distal end of ingrowing cartilage canals in chickens, where the endothelium is discontinuous.^23,46 Necrosis of cartilage canals followed.^6,9,46 This represents a mechanism where bacteria cause ischemic chondronecrosis, with the potential to develop into OCD.³⁴ Septic arthritis has been associated with osteochondral lesions in the femoral condyles of foals.^18,20 Higher numbers of osteochondral lesions were also found in the hock and fetlock joints of horses treated for bacterial infections in their first 6 months of life compared to a control group with unknown infection history.(E.H.S. Hendrickson, personal communication)

As osteochondrosis has a heritable component, individuals with the condition are subject to a variety of breeding restrictions. Differentiating individuals with sepsis-related osteochondrosis from the ones with lesions due to heritable predisposition might allow a larger population to be used for breeding. For this to become reality, it is first necessary to investigate whether bacteria colonize cartilage canals in horses. The aim of the current study was to examine whether bacteria are present in canals in the growth cartilage of foals with septic arthritis/osteomyelitis, as previously documented in pig and chickens, and whether this is associated with ischemic chondronecrosis.

Materials and Methods

Cases

The current study contained 7 euthanized foals, 6 from the Equine Hospital of the Norwegian University of Life Sciences and one from the Swedish University of Agricultural Sciences. Inclusion criteria were treatment for septic arthritis/osteomyelitis, or necropsy diagnosis of septic arthritis/osteomyelitis. The upper age limit was 6 months. Both sexes and any breed of horse or pony were included. Histologic sections that included growth cartilage, with or without bone, had to be available. Cases where necropsy was done more than 3 days after euthanasia and where pathologic changes had resulted in complete loss of anatomical structures were excluded. The 7 cases were given ascending numbers by age, summarized in Supplemental Table 1. The femoral lesions in cases 3 and 6 had been previously described.³⁶

Sample Collection

The third metacarpal bone (2), ilium (1), femur (6), tibia (5), talus (2), and third metatarsal bone (2) were sampled and sawed into approximately 5 mm thick slabs. The tali and ilium were sawed in a sagittal plane, the tibia was sawed in a slightly oblique sagittal plane parallel to the intermediate ridge and other bones were sawed in a transverse plane. All samples of the AECC included bone, whereas for physes, epiphyseal bone was always included but inclusion of metaphyseal bone varied between samples. All samples were fixed in 4% phosphate-buffered formaldehyde for 24-48 hours and decalcified in 10% ethylenediaminetetraacetic acid (EDTA), except for case 3 which was decalcified in formic acid. Sampled regions were chosen based on macroscopically visible changes or taken from standardized slabs, usually the middle slab. The samples were paraffin-embedded, before at least 2 approximately 5 µm-thick sections were cut from each block and stained with hematoxylin and eosin. Based on initial screening, sections with suspicion of bacteria were subjected to Gram staining (Supplemental Methods 1 and 2), provided that there was tissue left to section in the block.

Parameters Observed

Individual cartilage canals were categorized according to criteria published earlier for patent,⁵ chondrifying³ and necrotic canals.³⁷ Chondrocytes were interpreted as necrotic when they were shrunken, had cytoplasmic eosinophilia and nuclear pyknosis or karyolysis.³⁷ When chondrocyte necrosis and matrix change (pallor and relative eosinophilia) were present together, this was referred to as chondronecrosis, and when areas of chondronecrosis were centered on necrotic cartilage canals, they were referred to as areas of ischemic chondronecrosis.³⁷

Results

Details of the 7 foals with septic arthritis/osteomyelitis included in the study are presented in Supplemental Table 1.

The 7 cases included 30 joints with a diagnosis of septic arthritis/osteomyelitis, and one case had osteomyelitis in the growth cartilage of the tuber coxa of the ilium, that is, there were 31 infected sites (Supplemental Table 1). Histologic material was available from 14 of the 31 infected sites. From these sites, a total of 49 blocks from 19 smaller regions were examined and lesions were found in 41 blocks (16 regions) (Supplemental Table 2).

Bacteria were suspected based on examination of hematoxylin and eosin-stained histologic sections in 6 locations from cases 1, 2, 5, and 7 (Supplemental Table 2). Gram staining confirmed presence of gram-positive (cases 1, 5, and 7) or gram-negative (case 2) bacilli or coccobacilli in the 5 locations with adequate available (Supplemental Table 2).

The most striking feature in the AECC of case 1 was that the most superficial (Fig. 1) portions of the cartilage canals contained only intensely eosinophilic, granular material (Fig. 2) comprising remnants of degenerated erythrocytes, fibrin, canal matrix, and a low number of gram-positive bacilli (Fig. 3). In the physis of cases 1 and 2, large numbers of bacteria and granular material were consistently found within portions of the deep (Fig. 4) cartilage canals (ie, those closer to the metaphyseal than the epiphyseal side of the physis). Between the bacteria-containing portions and the diaphysis, canal lumina were reduced to eosinophilic streaks (Fig. 4). In the physis of case 2, bacteria were located within ghost-like remnants of vascular lumina (Fig. 5), as well as in the perivascular connective tissue where they were apparently bound to cartilage matrix at the canal boundary (Fig. 6) and, occasionally within chondrocyte lacunae (Fig. 6). Bacteria and granular material therefore followed the pattern of being distributed superficially within the AECC (Fig. 1) and deep within the physis (Fig. 4).

Figs. 1–6. Bacterial infection of cartilage canals, foal. Superficial is at the top and deep is at the bottom of each image. Fig. 1. Distal tibia, cranial intermediate ridge, articular-epiphyseal cartilage complex (AECC), case 1. Low magnification image to show location of changes. Portions of canals located superficially (between black dashed lines) contain intensely eosinophilic granular material (see Fig. 2). Adjacent canals (between black and white dashed lines), contain fibrin (see Fig. 7). Next to the fibrin there are necrotic canals (between white dashed lines), the majority of which represent acute septic canals (see Fig. 9). Deep to acute septic canals (below white dashed line), there are chronic septic canals (see Fig. 10). The cartilage between the white dashed lines is thickened, representing a focal delay in endochondral ossification, or septic osteochondrosis. Hematoxylin and eosin (HE). Fig. 2. Distal tibia, cranial intermediate ridge, AECC, case 1. Canal located superficially contains ghost-like remnant outlines of necrotic vessels (asterisks) and intensely eosinophilic granular material which comprises remnants of degenerated erythrocytes, disintegrated matrix and fibrin. Chondrocytes on the deep side of the canal are necrotic (arrows). HE. Fig. 3. Distal tibia, cranial intermediate ridge, AECC, case 1. A low number of gram-positive bacilli (arrows) are present within the granular material in a canal superficially. Gram Twort. Fig. 4. Third metacarpal bone, distal physis, case 2. Low magnification image to show location of changes. There are several necrotic canals with suspected bacteria and varying amounts of granular material located deep within the physis (within solid line box, see Figs. 5 and 6). The canals between the 2 boxes (see example, Fig. 7) contain fibrin. Within the dashed box (see example, Fig. 9) are acute septic canals and above the dashed box (see example, Fig. 10) are chronic septic canals. Portions of a canal (arrow) between the bacteria/granular material and the diaphysis is of reduced diameter and contain eosinophilic-staining material, referred to as eosinophilic streaks. HE. Fig. 5. Third metacarpal bone, distal physis, case 2. Numerous gram-negative bacilli (arrows) are present within the remnant outlines of necrotic vessels in canals deep within the physis. Gram fast green (GFG). Fig. 6. Third metacarpal bone, distal physis, case 2. Bacteria are closely apposed and apparently bound to cartilage matrix at the canal boundary (arrows), and also present in chondrocyte lacunae immediately adjacent to canals (arrowheads). GFG.

Relationship of Bacteria, Cartilage Canal Lesions, and Chondrocyte Necrosis

Adjacent to the bacteria and granular material, there was fibrin (Fig. 7). Next to the fibrin, there were necrotic canals, the minority of which were indistinguishable from necrotic canals earlier described in osteochondral lesions, hereafter called aseptic necrotic canals (Fig. 8). The majority of necrotic canals contained remnants of degenerated neutrophils (Fig. 9) in addition to necrotic endothelial and perivascular mesenchymal cells, and were referred to as acute septic canals (Table 1). Adjacent to the septic canals, and also immediately adjacent to and sometimes visibly bridging the ossification front, there were canal portions that contained morphologically viable cells comprising fibroblast-like cells (Fig. 10), capillaries and neutrophils, interpreted as repopulation of necrotic canals with septic fibrovascular granulation tissue from bone marrow and referred to as chronic septic canals (Table 1). The sequence of changes from canal portions containing bacteria/granular material to the secondary, epiphyseal center of ossification was therefore fibrin (Fig. 7), acute septic canals (Fig. 9) at intermediate depth, and chronic septic canals (Fig. 10) immediately adjacent to the ossification front, both for canals in the AECC and in the physis.

Figs. 7–10. Lesions of cartilage canals in foals with septic arthritis, distal tibia, cranial intermediate ridge, articular-epiphyseal cartilage complex (AECC). Superficial is at the top and deep is at the bottom of each image. Fig. 7. Case 1. Cartilage canal contains fibrillar, eosinophilic material interpreted as fibrin. Hematoxylin and eosin (HE). Fig. 8. Aseptic osteochondrosis, Swedish Warmblood horse, 23 days old. Aseptic necrotic cartilage canal contains necrotic endothelial cells (arrow) and necrotic perivascular mesenchymal cells (between arrowheads). HE. Fig. 9. Case 1. Cartilage canal contains remnants of degenerated neutrophils (inset) in addition to ghost-like remnant outlines of necrotic vessels (asterisk), referred to as an acute septic canal. HE. Fig. 10. Case 1. Cartilage canal contains fibroblast-like cells (arrowheads), interpreted as early repopulation of the canal with fibro-vascular granulation tissue from subchondral bone marrow, referred to as a chronic septic canal. HE.

Table 1.

Definitions for Lesions of Cartilage Canals in Foals with Septic Arthritis/Osteomyelitis.

Lesion category	Endothelium	Vascular luminae	Perivascular mesenchymal cells	Other cells in canal perivascular space	Pericanalicular chondrocytes
Acute septic canal	Necrotic or lysed	Ghost remnant outlines Bacteria	Necrotic or lysed	Neutrophils	AECC^a: intervascular necrosis^b MGP: limited perivascular and intervascular necrosis
Chronic septic canal	Arteriole and venules absent/lysed, capillaries present (angiogenesis)	Capillaries present (angiogenesis)	Absent/lysed	Neutrophils, fibroblasts and capillaries, ie, septic fibro-vascular granulation tissue	AECC: intervascular necrosis MGP: limited perivascular and intervascular necrosis
Inflamed patent canal	Normal	Normal	Normal	Neutrophils	Normal
Inflamed regressing canal	Absent	Absent	Majority: fibroblast-like Minority: chondrocyte-like	Neutrophils	Normal

^a Articular epiphyseal cartilage complex.

^b Necrosis: refers to chondrocytes located at intermediate depth of the growth cartilage; in early, intracartilaginous lesions, the most superficial and the deepest chondrocytes may be morphologically viable.

Chondrocytes around granular portions superficially within the AECC of case 1 tended to be morphologically viable on the superficial side, and necrotic on the deep side of canals (Fig. 2). All chondrocytes around acute septic canals at mid-depth of the AECC were necrotic, similar smaller areas were detected around mid-depth septic canals in the AECC of cases 3, 4 and 6, respectively (Supplemental Table 2). Chondrocytes immediately adjacent to the secondary center of ossification tended to be viable irrespective of canal status. In the physis of case 1, chondrocyte necrosis was limited to occasional cells or a narrow zone surrounding septic canals. Examples of chondrocyte necrosis being extensive within the AECC and limited within the physis of the same foal were also found in cases 1, 4 and 6 (Supplemental Table 2). The only exception to limited chondrocyte necrosis within the physis was that larger areas of necrosis were observed between canals within the resting zone of the physis in one block from the left metatarsus, and at the AECC/physeal junction in another block from the right distal femur of case 2, also at the level of the resting zone.

The perichondrium of the majority of cases was normal (Supplemental Table 2), but in cases 1 and 5, it contained linear tracts of strongly eosinophilic material (Fig. 11) consisting mainly of degenerated neutrophils, interpreted as inflammation and destruction of perichondral blood vessels.

Figs. 11-14. Fig. 11. Inflammation of perichondrium, distal tibia, cranial intermediate ridge, foal, case 1. The perichondrium contains linear tracts of strongly eosinophilic material consisting mostly of degenerated neutrophils, interpreted as inflammation of perichondrial blood vessels. Hematoxylin and eosin (HE). Superficial is at the top and deep is at the bottom in Figs. 11-13. Fig. 12. Osteomyelitis with osteochondrosis, distal tibia, cranial intermediate ridge, subchondral bone, foal, case 1. Immediately deep and axially adjacent to the septic osteochondrosis lesion in Fig. 1, within the secondary, epiphyseal center of ossification, there is suppurative osteomyelitis, including necrosis of primary spongiosa and multinucleated giant cells interpreted as osteoclasts (inset). HE. Fig. 13. Nondisplaced pathologic fracture, distal femur, medial condyle, AECC, foal, case 3. The lesion extends through the entire thickness of the AECC (between arrows) as a nondisplaced pathologic fracture. HE. Fig. 14. Osteomyelitis, talus, medial trochlear ridge, foal, case 6. There is a septic osteochondral lesion with a hinged flap in the proximal half of the medial trochlear ridge and a band of septic fibrous tissue in the distal half.

Delayed Ossification, Septic Osteochondrosis, and Osteomyelitis

The area of chondronecrosis containing septic necrotic canals within the AECC of case 1 was associated with marked focal delay in endochondral ossification (Fig. 1), hereafter termed septic osteochondrosis. Delayed ossification was also detected in cases 3, 4, 5, and 6, and described macroscopically for case 7 but not represented in the available histological sections (Supplemental Table 2). In 2 sites (the tuber coxa of case 5 and the medial femoral condyle of case 6), the area of delayed ossification consisted predominantly of necrotic vessels and chondronecrosis. In the majority of cases, however, the area of delayed ossification comprised a mixture of necrotic growth cartilage and granulation tissue containing degenerated neutrophils (Supplemental Table 2).

Immediately deep and axially adjacent to the septic osteochondrosis lesion in case 1, there was evidence of neutrophil invasion, necrosis of primary spongiosa and formation of granulation tissue with multinucleated giant cells interpreted as osteoclasts (Fig. 12), that is, suppurative, necrotizing osteomyelitis. Among all cases, some AECC and physeal lesions had no adjacent osteomyelitis, some had adjacent osteomyelitis in some blocks, and some adjacent osteomyelitis in all blocks (Supplemental Table 2). Within the physis, osteomyelitis occurred within the adjacent epiphyseal center of ossification, within the diaphyseal center of ossification, or both (Supplemental Table 2).

Macroscopic Cyst- and OCD-Like Changes

Part of the lesion in case 3 was encapsulated by fibrous tissue, compatible with formation of a Brodie’s abscess (Supplemental Table 2). Superficial to this, there were vertical linear areas of septic granulation tissue within the AECC that were compatible with, but considerably larger than, most chronic septic canals. In 2/4 blocks, these extended through the entire thickness of the AECC as nondisplaced pathologic fractures (Fig. 13).

Case 6 had a radiographically visible lesion in the distal femur previously diagnosed as a subchondral bone pseudocyst, in which the chondronecrosis was confirmed to be centered on septic canals (Supplemental Table 2). In the caudal distal tibia/hamate process and talus (Fig. 14) of case 6, areas of chondronecrosis were also associated with macroscopically visible wear lines, cartilage loss, protruding bands of fibrous tissue with neutrophils and undermined cartilage margins (tibia) or a hinged cartilage flap (talus). Finally, case 6 had complete pathologic avulsion of the long digital extensor tendon through an area of septic osteochondrosis.

Canal Infection, Inflammation, and Regression Without Necrosis

A low number of patent canals some distance from lesions contained neutrophils outside the circulation, mixed with the perivascular mesenchymal cells, and were surrounded by normal chondrocytes (Fig. 15). Canals with extra-vascular neutrophils were referred to as inflamed patent canals (Table 1). Similarly, in cases 2-7 there were some inflamed canals without evidence of endothelium or vascular luminae that were surrounded by normal chondrocytes and therefore appeared to be regressing (Table 1). Some of the inflamed regressing canals contained chondrocyte-like cells located in lacunar-like structures, and were therefore regressing by chondrification. The majority, however, contained spindle-shaped, fibroblast-like cells (Fig. 16) and were regressing by fibrosis.

Figs. 15–20. Fig. 15. Inflamed patent canal, third metatarsal bone, distal physis, foal, case 2. A cartilage canal contains neutrophils (arrows), referred to as an inflamed patent canal. The canal and surrounding chondrocytes are morphologically viable. Hematoxylin and eosin (HE). Fig. 16. Distal femur, medial condyle, articular-epiphyseal cartilage complex (AECC), foal, case 3. A cartilage canal contains no vessels and is surrounded by viable chondrocytes. This was interpreted as regression of an inflamed canal. The canal contains some spindle-shaped, fibroblast-like cells (arrows and inset) and therefore appears to be regressing by fibrosis rather than chondrification. HE. Fig. 17. Distal tibia, caudal intermediate ridge/hamate process, AECC, foal, case 7. Only a low number of regressing or patent canals (arrow and inset) with functional vascular lumina (asterisk) remain deep in the cartilage. HE. Fig. 18. Distal tibia, caudal intermediate ridge/hamate process, AECC, foal, case 7. Numerous gram-positive coccobacilli (arrows) compatible with the Rhodococcus equi (which was cultured from this case) are present within macrophages within one of the patent cartilage canals shown in Fig. 17. Gram fast green. Fig. 19. Standardbred foal, distal tibia, 5 mm thick sagittal slab through cranial intermediate ridge, AECC. The arterial side of the circulation has been perfused with barium, the slab has been decalcified and cleared by the modified Spalteholz technique. Dashed line: Chondro-osseous junction. Anatomical end arteries enter epiphyseal growth cartilage canals from the perichondrium. One perichondrial arteriole courses within the cartilage (arrow), and an adjacent perichondrial arteriole (arrowheads) courses predominantly within subchondral bone, before turning in the distal tip of the canal, located superficially within the growth cartilage (asterisks). Fig. 20. Tracing of arteries in Fig. 19 superimposed on Figs. 1 and 12. Dashed green line: Joint cartilage surface. Dashed blue line: Chondro-osseous junction. The location of portions of canals containing eosinophilic granular material (see Fig. 2) in the histological section (between black dashed lines, Fig. 1) corresponds to the distal termini of perfused canal arterioles in cleared cartilage (asterisks). The location of the osteomyelitis in Fig. 12 corresponds to where one perichondrial arteriole courses predominantly within subchondral bone (arrowheads).

In case 7, the physis of the hamate process was virtually completely replaced by granulation tissue containing degenerated neutrophils and there was delayed ossification, osteomyelitis and a subcutaneous closed abscess. The AECC contained only patent or regressing canals located deep in the cartilage (Fig. 17), several of which were inflamed. Numerous gram-positive coccobacilli compatible with the Rhodococcus equi cultured from the abscess in case 7 (Supplemental Table 1) were also observed within the osteomyelitic area and the patent or regressing cartilage canals (Fig. 18). The coccobacilli were, however, located within macrophages (Fig. 18) and the surrounding chondrocytes were normal.

Discussion

In this study, bacteria were present in canals in the growth cartilage and bone of the AECC and physis of foals with septic arthritis/osteomyelitis. The typical location of bacteria in cartilage canals was superficially in the AECC and deep in the physis. Anatomical studies have confirmed that a major part of blood supply to the epiphyses in the forelimb of foals originates from perichondral arteries.^13,14 Branches of these vessels enter both the epiphyseal cartilage and the physis.^14,15 A similar arrangement has been identified in the AECC of hind limb joints (Fig. 19), although the physis remains to be investigated.^38,39 Vessels superficial in the AECC and deep in the physis thus are the most distal part of this vasculature (Fig. 20). The location of bacteria here is in accordance with experimental studies in chickens and pigs, where bacteria were first found in the distal tips of metaphyseal vessels.^2,9,24,46

Discontinuities in the endothelium of epiphyseal chondrifying cartilage canals of foals aged 1-122 days were recently described.²² Intravenously injected carbon particles accumulated outside growing distal ends of vessels in cartilage canals in noninfected rats.¹⁹ Preexisting fenestrations and gaps in the endothelium could allow small numbers of bacteria to adhere and trigger an inflammation-mediated disruption of endothelium, leading to more widespread colonization and infection.⁴⁰ In this cross-sectional study, it was not possible to determine if fenestrations or gaps were present before bacterial adhesion as the endothelium of the cartilage canals was for the most part absent in the lesions examined.

Bacteremia and the accompanying inflammatory response have an impact on the integrity of the endothelium and exacerbate the permeability to bacteria.^40,50 An inflammation-related increase in the permeability of vessel walls has been suggested to allow bacteria to cross where they otherwise would not be able to cross.⁶ A more recent study of intravenous inoculation of Staphylococcus aureus in 8-week-old pigs found that destruction of endothelium in the vasculature of the physis did not occur until influx of neutrophils at 48 hours.²⁴ In the same study, micro abscesses in the lungs were resolved within 48 hours, while in the cartilage they persisted and increased in size. The combination of the special vascular anatomy in growth cartilage and a systemic inflammation-induced impact because of bacteremia is a likely reason for the persistence of infections in the physis.^15,38
–40

Neutrophils are not a feature of necrotic canals in studies of osteochondral lesions in heritably predisposed breeds (Fig. 8) or in surgically induced lesions.^3,35,37 Increased numbers of neutrophils in and around cartilage canals were demonstrated in experimental studies 24 hours after inoculation of bacteria in pigs and 12 hours after inoculation in chickens.^9,24 Due to their larger size and more common presence, neutrophils are easier to detect in histologic sections than bacteria.⁴⁷ In the absence of bacteria, neutrophils were therefore used as a criterion for identification of septic canals.

Chondronecrosis around septic canals was observed in 6 out of 7 foals in this study, both in the AECC and the physis. The areas of chondronecrosis were consistently more extensive in the AECC than in the physis. Osteochondral lesions have been suggested to be a consequence of septic processes in earlier case reports.^18,20 Hendrickson et al found up to 18% higher frequency of osteochondral lesions in yearlings treated for bacterial infections before 6 months of age than controls with unknown infection history.(E.H.S. Hendrickson, personal communication) Bacterial infection and destruction of cartilage canals may therefore account for a small portion of all developmental osteochondral lesions. Causes of delayed ossification other than failure of incorporation of canals were recently described.¹¹ It therefore makes sense to add a prefix to osteochondrosis in lesions where the cause of delayed ossification is known. Because bacterial infection can lead to ischemic chondronecrosis and focally delayed ossification—the very definition of osteochondrosis^32,33—we propose that this condition is termed septic osteochondrosis.

The lesions in cases 3 and 6 (Figs. 13, 14) contained macroscopically visible physical disruptions of the articular cartilage. Whether the disruptions should be termed OCD is debatable. The pathologic fractures observed extended through septic granulation tissue rather than areas of ischemic chondronecrosis, which is the defining criterion for OCD.³³ There is evidence that sepsis-survivors have more OCD lesions than animals with unknown status.(E.H.S. Hendrickson, personal communication) The lesions in these cases were in the caudal part of the distal intermediate of the tibia (hamate process), which is not a recognized predilection site for OCD. None of the sepsis-survivors had lesions at the hamate process, (E.H.S. Hendrickson, personal communication) and this may be because lesions resolved or because they were so extensive that the horses did not survive to OCD-screening age (12 months).²⁷ Both sepsis-survivors (E.H.S. Hendrickson, personal communication) and many of the present cases had lesions at recognized predilection sites for OCD, making it likely that some lesions do persist to screening age. This could be determined through frequent radiography of survivors from initial sepsis, until screening age.

Cases 1 and 2 had areas in the AECC and physis, or just the physis, respectively, with both early chondronecrosis and early osteomyelitis in the same lesion. Osteomyelitis in foals is usually found in the junction between bone and cartilage.¹² Various mechanisms for localization of bacteria in this anatomic site have been in different species: sluggish blood flow in sinusoidal loops in the bone marrow,¹² transphyseal vessels allowing migration of bacteria from the diaphysis to the epiphysis,^2,12,30 or endothelial gaps in growing vessels which bacteria may migrate through.⁴⁶ In a previous study, it was considered whether infection spread to the growth cartilage from subchondral bone, or the other way around.³⁶ Given the common observation of chondronecrosis and osteomyelitis in this material harvested from foals, especially in the very early epiphyseal and metaphyseal lesions in the young cases 1 and 2, it is likely that these processes can take place simultaneously. A significant portion of the blood supply to the epiphysis in foals originates from the perichondrium in the form of transverse vessels with perpendicular branches toward the articular surface and the physis.^13,38,39 The branches enter bone, cartilage or both, depending on location in the epiphysis (Fig. 19).^13,38,39 Whether bacterial destruction of these vessels results in chondronecrosis/osteochondrosis or osteomyelitis could be related to the location of the vessel affected at the time of bacteremia (Figs. 19, 20). The distribution of epiphyseal vessels originating in the perichondrium varies depending on bone and age,¹³ which could explain the age-related variations of locations of septic arthritis/osteomyelitis in foals.¹² This potentially also explains why osteomyelitis and hematogenous septic arthritis are such rare events in adult horses, as they do not have perichondral or cartilage canal vessels.

The canals categorized as inflamed patent canals (Fig. 15) had morphologically functional vasculature and were surrounded by normal cartilage, and regressed by chondrification. This illustrates that lesser grades of inflammation can be resolved without lasting damage. The majority of inflamed regressing canals (Fig. 16) seemed to regress by fibrosis instead of chondrification. Whether bacteremia results in septic cartilage lesions likely depend on factors like immuno-competence of the foal⁴⁴ combined with infective dose^17,25 and pathogenicity⁴⁵ of the bacteria involved.

In this study, bacteria were present in cartilage canals in foals and were associated with ischemic chondronecrosis in both the AECC and the physis. Together with evidence from experimental studies of osteochondrosis and higher incidence of OCD in sepsis-survivors, this establishes sepsis as a plausible cause of some osteochondral lesions in horses. It is recommended that horses with sepsis-related osteochondral lesions may be used for breeding without increasing the prevalence of heritable OCD in the population.

Supplemental Material

Supplemental Material, DS1_VET_10.1177_0300985818777786 - Septic Arthritis/Osteomyelitis May Lead to Osteochondrosis-Like Lesions in Foals

Supplemental Material, DS1_VET_10.1177_0300985818777786 for Septic Arthritis/Osteomyelitis May Lead to Osteochondrosis-Like Lesions in Foals by Bjørn Wormstrand, Liv Østevik, Stina Ekman, and Kristin Olstad in Veterinary Pathology

Footnotes

Authors’ Note

The manuscript has been prepared in the Uniform Requirements format.

Acknowledgements

The authors are grateful to the following pathologists for permission to include material from 2 foals as follows: Dr Gjermund Gunnes (case 5) and Prof Arild Espenes (case 7). The authors also are grateful to the Imaging Centre at NMBU for their assistance with the figures.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The study was funded by grant H1147117/218962 from the Swedish-Norwegian Foundation for Equine Research/Research Council of Norway, with contributions from Norsk Rikstoto and Jordbruksavtalen.

ORCID iD

Bjørn Wormstrand

The Supplemental Material is available at .

References

Aasmundstad

Gjerlaug-Enger

Grindflek

, et al. Genetic trends of conformation traits and genetic correlations to osteochondrosis in boars. Animal. 2014;8(7):1045–1052.

Alderson

Speers

Emslie

, et al. Acute haematogenous osteomyelitis and septic arthritis—a single disease. J Bone and Joint Surg. 1986;68B(2):268–274.

Carlson

Cullins

Meuten

. Osteochondrosis of the articular-epiphyseal cartilage complex in young horses: evidence for a defect in cartilage canal blood supply. Vet Pathol. 1995;32(6):641–647.

Carlson

Hilley

Meuten

. Degeneration of cartilage canal vessels associated with lesions of osteochondrosis in swine. Vet Pathol. 1989;26:47–54.

Carlson

Meuten

Richardson

. Ischemic necrosis of cartilage in spontaneous and experimental lesions of osteochondrosis. J Orthop Res. 1991;9(3):317–329.

Denecke

Trautwein

Kaup

. The role of cartilage canals in the pathogenesis of experimentally induced polyarthritis. Rheumatol Int. 1986;6(6):239–243.

Dik

Enzerink

van Weeren

. Radiographic development of osteochondral abnormalities in the hock and stifle of Dutch Warmblood foals, from age 1 to 11 months. Equine Vet J Suppl. 1999;(31):9–15.

Edmonds

Polousky

. A review of knowledge in osteochondritis dissecans: 123 years of minimal evolution from Konig to the ROCK study group. Clin Orthop Relat Res. 2013;471(4):1118–1126.

Emslie

Nade

. Acute hematogenous staphylococcal osteomyelitis. A description of the natural history in an avian model. Am J Pathol. 1983;110(3):333–345.

10.

Etterlin

Morrison

Osterberg

, et al. Osteochondrosis, but not lameness, is more frequent among free-range pigs than confined herd-mates. Acta Vet Scand. 2015;57:63.

11.

Finnoy

Olstad

Lilledahl

. Non-linear optical microscopy of cartilage canals in the distal femur of young pigs may reveal the cause of articular osteochondrosis. BMC Vet Res. 2017;13(1):270.

12.

Firth

Goedegebuure

. The site of focal osteomyelitis lesions in foals. Vet Q. 1988;10(2):99–108.

13.

Firth

Poulos

. Blood vessels in the developing growth plate of the equine distal radius and metacarpus. Res Vet Sci. 1982;33(2):159–166.

14.

Firth

Poulos

. Microangiographic studies of metaphyseal vessels in young foals. Res Vet Sci. 1983;34(2):231–235.

15.

Firth

Poulos

. Vascular characteristics of the cartilage and subchondral bone of the distal radial epiphysis of the young foal. N Z Vet J. 1993;41(2):73–77.

16.

Grøndahl

Dolvik

. Heritability estimations of osteochondrosis in the tibiotarsal joint and of bony fragments in the palmar/plantar portion of the metacarpo- and metatarsophalangeal joints of horses. J Am Vet Med Assoc. 1993;203(1):101–104.

17.

Hackett

Lunn

Ferris

, et al. Detection of bacteraemia and host response in healthy neonatal foals. Equine Vet J. 2015;47(4):405–409.

18.

Haggett

Foote

Head

, et al. Necrosis of the femoral condyles in a four-week-old foal: clinical, imaging and histopathological features. Equine Vet J Suppl. 2012(41):91–95.

19.

Ham

Hurley

Ryan

, et al. Localization of particulate carbon in metaphyseal vessels of growing rats. Aust J Exp Biol Med Sci. 1965;43(5):625–638.

20.

Hance

Schneider

Embertson

, et al. Lesions of the caudal aspect of the femoral condyles in foals: 20 cases (1980-1990). J Am Vet Med Assoc. 1993;202(4):637–646.

21.

Heinonen

Hakala

Hameenoja

, et al. Case-control study of factors associated with arthritis detected at slaughter in pigs from 49 farms. Vet Rec. 2007;160(17):573–578.

22.

Hellings

Ekman

Hultenby

, et al. Discontinuities in the endothelium of epiphyseal cartilage canals and relevance to joint disease in foals. J Anat. 2016;228(1):162–175.

23.

Howlett

. The fine structure of the proximal growth plate and metaphysis of the avian tibia: endochondral osteogenesis. J Anat. 1980;130(pt 4):745–768.

24.

Jensen

Nielsen

Agerholm

, et al. A non-traumatic Staphylococcus aureus osteomyelitis model in pigs. In Vivo. 2010;24(3):257–264.

25.

Johansen

Frees

Aalbaek

, et al. A porcine model of acute, haematogenous, localized osteomyelitis due to Staphylococcus aureus: a pathomorphological study. Apmis. 2011;119(2):111–118.

26.

Lutfi

. Mode of growth, fate and function of cartilage canals. J Anat. 1970;106(1):135–145.

27.

Lykkjen

Dolvik

McCue

, et al. Genome-wide association analysis of osteochondrosis of the tibiotarsal joint in Norwegian Standardbred trotters. Anim Genet. 2010;41(suppl 2):111–120.

28.

Lykkjen

Olsen

Dolvik

, et al. Heritability estimates of tarsocrural osteochondrosis and palmar/plantar first phalanx osteochondral fragments in Standardbred trotters. Equine Vet J. 2014;46(1):32–37.

29.

Mackie

Ahmed

Tatarczuch

, et al. Endochondral ossification: how cartilage is converted into bone in the developing skeleton. Int J Biochem Cell Biol. 2008;40(1):46–62.

30.

Ogden

. Pediatric osteomyelitis and septic arthritis: the pathology of neonatal disease. Yale J Biol Med. 1979;52(5):423–448.

31.

Olsson

SE.

Osteochondrosis dissecans in the dog. A study of pathogenesis, clinical signs, pathologic changes, natural course and sequelae. J Amer Vet Radiol Soc. 1973;14(4):8.

32.

Olsson

Reiland

. The nature of osteochondrosis in animals. Summary and conclusions with comparative aspects on osteochondritis dissecans in man. Acta Radiol Suppl. 1978;358:299–306.

33.

Olstad

Ekman

Carlson

. An update on the pathogenesis of osteochondrosis. Vet Pathol. 2015;52(5):785–802.

34.

Olstad

Hendrickson

EHS

Carlson

, et al. Transection of vessels in epiphyseal cartilage canals leads to osteochondrosis and osteochondrosis dissecans in the femoro-patellar joint of foals; a potential model of juvenile osteochondritis dissecans. Osteoarthritis Cartilage. 2013;21:730–738.

35.

Olstad

Hendrickson

EHS

Ekman

, et al. Local morphological response of the distal femoral articular-epiphyseal cartilage complex of young foals to surgical stab incision, and potential relevance to cartilage injury and repair in children. Cartilage. 2013;4(3):239–248.

36.

Olstad

Ytrehus

Carlson

, et al. Early lesions of articular osteochondrosis in the distal femur of foals. Vet Pathol. 2011;48(6):1165–1175.

37.

Olstad

Ytrehus

Ekman

, et al. Early lesions of osteochondrosis in the distal tibia of foals. J Orthop Res. 2007;25(8):1094–1105.

38.

Olstad

Ytrehus

Ekman

, et al. Epiphyseal cartilage canal blood supply to the distal femur of foals. Equine Vet J. 2008;40(5):433–439.

39.

Olstad

Ytrehus

Ekman

, et al. Epiphyseal cartilage canal blood supply to the tarsus of foals and relationship to osteochondrosis. Equine Vet J. 2008;40(1):30–39.

40.

Razakandrainibe

Combes

Grau

, et al. Crossing the wall: the opening of endothelial cell junctions during infectious diseases. Int J Biochem Cell Biol. 2013;45(7):1165–1173.

41.

Reiland

. Morphology of osteochondrosis and sequelae in pigs. Acta Radiol Suppl. 1978;358:45–90.

42.

Reiland

Ordell

Lundeheim

, et al. Heredity of osteochondrosis, body constitution and leg weakness in the pig. A correlative investigation using progeny testing. Acta Radiol Suppl. 1978;358:123–137.

43.

Rejnö

Strömberg

. Osteochondrosis in the horse. II. Pathology. Acta Radiol Suppl. 1978;358:153–178.

44.

Robinson

Allen

Green

, et al. A prospective study of septicaemia in colostrum-deprived foals. Equine Vet J. 1993;25(3):214–219.

45.

Russell

Axon

Blishen

, et al. Blood culture isolates and antimicrobial sensitivities from 427 critically ill neonatal foals. Aust Vet J. 2008;86(7):266–271.

46.

Speers

Nade

. Ultrastructural studies of adherence of Staphylococcus aureus in experimental acute hematogenous osteomyelitis. Infect Immun 1985;49(2):443–446.

47.

Tiemann

Hofmann

Krukemeyer

, et al. Histopathological osteomyelitis evaluation score (HOES)—an innovative approach to histopathological diagnostics and scoring of osteomyelitis. GMS Interdiscip Plast Reconstr Surg DGPW. 2014;3:Doc08.

48.

Toth

Nissi

Wang

, et al. Surgical induction, histological evaluation, and MRI identification of cartilage necrosis in the distal femur in goats to model early lesions of osteochondrosis. Osteoarthritis Cartilage. 2015;23(2):300–307.

49.

van Grevenhof

Ducro

Van Weeren

, et al. Prevalence of various radiographic manifestations of osteochondrosis and their correlations between and within joints in Dutch warmblood horses. Equine Vet J. 2009;41(1):11–16.

50.

Weiss

Rashid

. The sepsis-coagulant axis: a review. J Vet Intern Med. 1998;12(5):317–324.

51.

Ytrehus

Andreas Haga

Mellum

, et al. Experimental ischemia of porcine growth cartilage produces lesions of osteochondrosis. J Orthop Res. 2004;22(6):1201–1209.

52.

Ytrehus

Ekman

Carlson

, et al. Focal changes in blood supply during normal epiphyseal growth are central in the pathogenesis of osteochondrosis in pigs. Bone. 2004;35(6):1294–1306.

53.

Ytrehus

Grindflek

Teige

, et al. The effect of parentage on the prevalence, severity and location of lesions of osteochondrosis in swine. J Vet Med A Physiol Pathol Clin Med. 2004;51(4):188–195.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.14 MB