The 2 major categories of idiopathic inflammatory arthritis are rheumatoid arthritis and the seronegative spondyloarthropathies. Whilst the synovium is the primary site of joint disease in the former, the primary site in the latter is less well defined. However, it has recently been proposed that enthesitis-associated changes in the spondyloarthropathies are primary and that all other joint manifestations are secondary. Nevertheless, some of the sites of disease localisation have not been adequately explained in terms of enthesitis. This article summarises current knowledge of the structure, function, blood supply, innervation, molecular composition and histopathology of the classic enthesis (i.e. the bony attachment of a tendon or ligament) and introduces the concept of ‘functional’ and articular ‘fibrocartilaginous’ entheses. The former are regions where tendons or ligaments wrap-around bony pulleys, but are not attached to them, and the latter are synovial joints that are lined by fibrocartilage rather than hyaline cartilage. We describe how these 3 types of entheses relate to other, and how all are prone to pathological changes in spondyloarthropathy. We propose that the inflammatory responses characteristic of spondyloarthropathies are triggered at these seemingly diverse sites, in genetically susceptible individuals, by a combination of anatomical factors which lead to higher levels of tissue microtrauma, and the deposition of microbes.

Relative to body size, midsagittal and interpedicular diameters of the cranial and caudal aspects of cervical vertebral foramina (C3–C7) were found to be significantly (P < 0·05) larger in small breeds than in large breeds and Dachshunds, and also larger in Dachshunds (P < 0·05) than in large breeds. This condition increases the risk for spinal cord compression resulting from relative stenosis of the cervical vertebral foramina, especially in large dogs, and this is also exacerbated by the typical shape of the vertebral foramina (i.e. dorsoventrally flattened cranially and bilaterally narrowed caudally). Within large dogs those breeds highly predisposed to cervical spinal cord compression were Great Danes (the breed with the smallest midsagittal vertebral foramen diameters from cranial C6 to cranial T1) and Doberman Pinschers, because of the most strikingly cranially dorsoventrally narrowed cone-shaped vertebral foramina at C6 and C7. The existence of a small midsagittal diameter in the cranial cervical spine was a high risk factor predisposing to spinal cord compression in small breeds and Dachshunds. Remarkable consistency was noted between the spinal level of the maximum enlargement of the spinal cord which previously was reported to be at C6, and the site of maximum enlargement of the vertebral canal currently stated in Dachshunds and small breeds. In large breeds the maximum enlargement of the vertebral canal tended to be located more caudally at the caudal limit of C7. The average age at which large dogs were most susceptible to noxious factors causing abnormal growth of the pedicles was determined to be 16 wk.

The suprascapular ligament converts the suprascapular notch into a foramen separating the vessels and nerve of the same name. It connects 2 regions of the same bone and does not cross any joint, and no mechanical function has yet been attributed to it. Nevertheless, variations in its thickness and length, and its tendency to ossify, suggest that the ligament responds to changes in mechanical load. This should be reflected in the composition of the extracellular matrix. The primary purpose of the present study is to demonstrate that the suprascapular ligament has fibrocartilaginous entheses (i.e. insertion sites), even though there is no obvious change in insertional angle that directly results from joint movement. Such a change is more typical of tendons or ligaments that cross highly mobile joints. The complete ligament (including both entheses) was removed from 7 cadavers shortly after death and fixed in 90% methanol. Cryosections were immunolabelled with a panel of monoclonal antibodies against collagens (types I, II, III, VI), glycosaminoglycans (chondroitin 4 sulphate, chondroitin 6 sulphate, dermatan sulphate and keratan sulphates), proteoglycans (aggrecan and versican) and link protein. Both entheses were strongly fibrocartilaginous, and a moderately fibrocartilaginous matrix was also detected throughout the remainder of the ligament. The extracellular matrix of both entheses labelled strongly for type II collagen, aggrecan and link protein. The fibrocartilaginous character of the entheses suggests that the insertion sites of the ligament are subject to both compressive and tensile loading and are regions of stress concentration. This in turn probably reflects the complex shape of the scapula and the presence of a conspicuous indentation (the suprascapular notch) near the ligament. The loading patterns may reflect either the attachment of muscles and/or the forces transmitted to the suprascapular ligament from the neighbouring coracoclavicular ligament.

A total of 192 embalmed cadavers were examined in order to present a detailed study of arterial variations in the upper limb and a meta-analysis of them. The variable terminology previously used was unified into a homogenous and complete classification, with 12 categories covering all the previously reported variant patterns of the arm and forearm.

The three-dimensional (3D) microanatomy of the cornea is the major determinant of its optical and mechanical properties. Scanning electron microscopy (SEM) is the most commonly used method to obtain information on the overall 3D microanatomy of organs. However, SEM has not been successful in revealing the 3D microanatomy of the cornea, because the interior of the cornea is too compact to be explored by the electron beam. In this study, the 3D organisation of the cells and extracellular materials of human and rabbit corneas was examined after exposure by HCl and NaOH digestion, and by microdissection by the adhesive tape method. In the cornea of both species, all epithelial cells exhibited microplicae regardless of their location. This raises doubts about the tear film-holding role assigned to the microplicae of the superficial cells. Human and rabbit corneas differed in the collagen fibre patterns of the epithelial basement membranes. The 3D organisation of the stromal lamellae was similar in both species. In humans and rabbits, the keratocytes showed similar 3D features. However, the surface of human keratocytes located near Descemet's membrane exhibited small fenestrations that were not present in the rabbit keratocytes. The pattern of keratocyte innervation by the stromal neural plexus and 3D keratocyte microanatomy confirms that keratocytes form a large intercommunicating network within the corneal stroma. Two morphologically discrete subpopulations of keratocytes located at different stromal levels were identified in both human and rabbit corneas, suggesting that keratocytes are not functionally homogeneous. In addition, the density of the stromal neural plexus appeared to be greater in rabbits than in humans. Clear differences between human and rabbit corneas were observed in the collagen arrangement in Descemet's membrane, which may reflect their different biomechanical requirements.

This study investigated the relationships between the density of Meissner's corpuscles and the dermatoglyphic characteristics of human fingers. Dermatoglyphic prints and tissue samples were collected from the index (II) and ring (IV) fingers of 28 cadavers from the Medical School, University of Adelaide. Pattern types, pattern intensity, total ridge count and ridge breadth, were determined for each sample and the density of Meissner's corpuscles established by recording the mean number of Meissner's corpuscles underlying a 10 mm long line on the skin surface. No relationship was found between the density of Meissner's corpuscles and pattern type, pattern intensity or total ridge count. Negative correlations were found for the density of Meissner's corpuscles with both ridge width and size of fingers. Males were found to have significantly larger fingers, larger ridge breadth, and a lower density of Meissner's corpuscles per 10 mm compared with females. However, there was no difference between the total number of Meissner's corpuscles per finger in males and females. No significant differences were found in either the density of Meissner's corpuscles between antimeres or between the index and ring fingers. These results suggest that for both the left and right hand in males and females, there is a similar number of Meissner's corpuscles for both the index and ring fingers.

Splanchnic neurectomy is of value in the management of chronic abdominal pain. It is postulated that the inconsistent results of splanchnicectomies may be due to anatomical variations in the pattern of splanchnic nerves. The advent of minimally invasive and video-assisted surgery has rekindled interest in the frequency of variations of the splanchnic nerves. The aims of this study were to investigate the incidence, origin and pattern of the splanchnic nerves in order to establish a predictable pattern of splanchnic neural anatomy that may be of surgical relevance. Six adult and 14 fetal cadavers were dissected (n = 38). The origin of the splanchnic nerve was bilaterally asymmetrical in all cases. The greater splanchnic nerve (GSN) was always present, whereas the lesser splanchnic nerve (LSN) and least splanchnic nerve (lSN) were inconsistent (LSN, 35 of 38 sides (92%); LSN, 21 of 38 sides (55%). The splanchnic nerves were observed most frequently over the following ranges: GSN, T6–9: 28 of 38 sides (73%); LSN, when present, T10–11: (10 of 35 sides (29%); and lSN, T11–12: 3 of 21 sides (14%). The number of ganglionic roots of the GSN varied between 3 and 10 (widest T4–11; narrowest, T5–7). Intermediate splanchnic ganglia, when present, were observed only on the GSN main trunk with an incidence of 6 of 10 sides (60%) in the adult and 11 of 28 sides (39%) in the fetus. The higher incidence of the origin of GSN above T5 has clinical implications, given the widely discussed technique of undertaking splanchnicectomy from the T5 ganglion distally. This approach overlooks important nerve contributions and thereby may compromise clinical outcome. In the light of these variations, a reappraisal of current surgical techniques used in thoracoscopic splanchnicectomy is warranted.

Blind mole-rats (Spalax ehrenbergi) are fossorial solitary rodents that present striking morphological, physiological and behavioural adaptations to the subterranean environment in which they live. Previous studies have shown that mole-rats are specialised in tooth-digging. The rapid eruption-rate of their incisors has evolved to compensate for their excessive wear by excavation. Males use their incisors more than females for digging and fighting, and their rate of incisor eruption is significantly more rapid than in females. Since mole-rats use their incisors for digging throughout the year, we suggest that continuous mechanical pressure on their oral tissues concentrated at the apical sites of the upper incisors leads to cell and tissue fatigue. We provide evidence for 5 stages of palatal perforation by the upper incisors at their apical sites, with maximum perforation characterising aged males. Interspecies comparisons with 7 other fossorial and semi-fossorial rodent species, and with beavers, which expose their incisors to enormous mechanical pressure, revealed that this palatal perforation is unique to the male mole-rat. We suggest that while the fast eruption rate of incisors in the mole-rat compensates for the rapid wear resulting from digging, evolutionary adaptation to continuous tooth-digging is still ongoing, since the physical pressure of digging at the apical sites of the upper incisors leads to tissue destruction, breakage of the palatal bone and possibly to death, as a result of maxillary inflammation.

The muscle changes related to pelvic floor disorders are poorly understood. We conducted an anatomical and histological study of the perineum of the normal mouse and of a transgenic mouse strain deficient in urokinase-type plasminogen activator (uPA−/−) that was previously reported to develop a high incidence of rectal prolapse. We could clearly identify the iliococcygeus (ILC) and pubococcygeus (PC) muscles and anal (SPA) and urethral (SPU) sphincters in male and female mice. The bulbocavernosus (BC), ischiocavernosus (ISC) and levator ani (LA) muscles could be found only in male mice. Histochemical analysis of the pelvic floor muscles revealed a majority of type IIA fibres. Rectal prolapses were observed only in male uPA−/− mice. The most obvious finding was an irreducible evagination of the rectal mucosa and a swelling of the entire perineal region corresponding to an irreducible hernia of the seminal vesicles through the pelvic outlet. The hernia caused stretching and thinning of the ISC, BC and LA. Myopathic damage, with degenerated and centronucleated myofibres, were observed in these muscles. The PC, ILC, SPA and SPU were not affected. This study provides an original description of a model of pelvic floor disorder and illustrates the differences existing between the perineum of humans and that of a quadruped species. In spite of these differences, the histopathologic changes observed in the pelvic floor muscles of uPA−/− mice with rectal prolapse suggest that prolonged muscular stretching causes a primary myopathic injury. This should be taken into account in the evaluation of pelvic floor disorders.

Perichondral bone, the circumferential grooves of Ranvier and cartilage canals are features of endochondral bone development. Cartilage canals containing connective tissue and blood vessels are found in the epiphysis of long bones and in cartilaginous anlagen of small and irregular bones. The pattern of cartilage canals seems to be integral to bone development and ossification. The canals may be concerned with the nourishment of large masses of cartilage, but neither their role in the formation of ossification centres nor their interaction with the circumferential grooves of Ranvier has been established. The relationships between cartilage canals, perichondral bone and the ossification centre were studied in the calcaneus of 9 to 38-wk-old human fetuses, by use of epoxy resin embedding, three-dimensional computer reconstructions and immunhistochemistry on paraffin sections. We found that cartilage canals are regularly arranged in shells surrounding the ossification centre. Whereas most of the shell canals might be involved in the nourishment of the cartilage, the inner shell is directly connected with the perichondral ossification groove of Ranvier and with large vessels from outside. In this way the inner shell canal imports extracellular matrix, cells and vessels into the cartilage. With the so-called communicating canals it is also connected to the endochondral ossification centre to which it delivers extracellular matrix, cells and vessels. The communicating canals can be considered as inverted ‘internal’ ossification grooves. They seem to be responsible for both build up intramembranous osteoid and for the direction of growth and thereby for orientation of the ossication centre.

We analysed a large series of axillary lymph nodes, with and without metastases following radical mastectomy for breast cancer. We found left/right asymmetry in numbers of lymph nodes, and also asymmetry of lymph node dimensions, which could have been the caused by tumoral antigenic stimulation. The distribution of hyperplastic node patterns differed significantly.