Tendon Segments Research Articles

A comparative dose–response study of cartilage-derived morphogenetic protein (CDMP)-1, -2 and -3 for tendon healing in rats

Cartilage-derived morphogenetic proteins (CDMPs), belonging to the bone morphogenetic protein (BMP) family, are known to be cartilage and bone inducers as well as to induce tendon and ligament-like tissue. In this study we investigated the influence of CDMP-1, -2 or -3 at four different doses (0, 0.4, 2 and 10 μg) on tendon healing in a rat model, as well as differences in osteogenesis between the different CDMPs and doses. In 110 rats, a 3 mm segment of the Achilles tendon was removed via a 2 mm skin incision. CDMP-1, -2 or -3 was injected into the defect 6 h postoperative. The rats were killed 8 days after operation. The tendon regenerates were tested biomechanically. There was a significant dose-related increase in strength and stiffness with all three CDMPs, but no difference between the CDMPs was found. Another 50 rats were used to compare the highest dose of the CDMPs with controls and osteogenic protein 1 (OP-1), as regards cartilage or bone formation after 4 weeks. Cartilage occurred in all groups, including the controls. Some specimens in all groups contained bone, except the controls. No difference between the CDMPs could be demonstrated. The CDMP-1, CDMP-3 and OP-1 groups contained significantly more calcium than controls. Only the CDMP-2 group and the controls contained significantly less calcium than the OP-1 group. In conclusion, the three CDMPs appeared similar as regards improvement of tendon repair and osteogenicity in this setting.

Tendon healing in vitro: Promotion of collagen gene expression by bFGF with NF-κB gene activation

Purpose: Matrix synthesis of intrasynovial tendon cells and activation of nuclear transcription factors are pivotal to promotion of flexor tendon healing. We investigated the effects of basic fibroblast growth factor (bFGF) on synthesis of type I collagen and activation of nuclear transcription factor κB (NF-κB) in an in vitro culture model of intrasynovial tenocytes. Method: Tenocytes obtained from explant culture of rabbit intrasynovial tendon segments were treated with bFGF at concentrations of 0, 2, and 10 ng/mL. Expression of type I collagen and NF-κB genes was determined by quantitative analysis of products of reverse-transcription polymerase chain reactions. Proliferation of the cells was assessed by incorporation of bromodeoxyuridine into the DNA of the cells. Results: Expression levels of type I collagen and NF-κB genes of tenocytes were increased significantly by bFGF. Cell proliferation as indicated by DNA labeling was promoted significantly by bFGF. Expression of the NF-κB gene increased proportionately to the amounts of bFGF stimulating the cells and was correlated with increases in proliferation rate of tenocytes. Conclusions: Results of this study show that expression of type I collagen and NF-κB genes is promoted manifestly by bFGF. The effects were proportionate to in vitro proliferation rates of tenocytes. The study indicated that matrix synthesis of flexor tendons can be promoted by bFGF and that NF-κB may play a pivotal role in initiating proliferation and type I collagen synthesis of tenocytes. (J Hand Surg 2003;28A:215-220. Copyright © 2003 by the American Society for Surgery of the Hand.)

Estudo morfológico de tendões flexores de eqüinos

The objective of the present study was to describe the normal appearance of the superficial (SDFT) and deep (DDFT) digital flexor tendons of horses and to focus on the differences between their histological constituents. Morphological study was performed on the four tendon segments collected from the thoracic and pelvic limbs of seven fresh cadavers of adult horses without previously diagnosed clinical lesions. The segments analyzed were the muscle-tendon junction, middle third and third distal of the third metacarpus/metatarsus and distal extremity of the proximal phalanx for the SDFT, and muscle-tendon junction, middle third of the third metacarpus/metatarsus, region of the check ligament and region of the middle phalanx for the DDFT. The results showed some histological differences in the distribution and localization of the components between segments of the same tendon and between thoracic and pelvic tendons, especially in terms of the quantity of fibrocartilage cells; that is revealed higher in the middle third region of the third metacarpus/metatarsus both for the thoracic and the pelvic limbs for the SDFT and highest counts in the region of the middle third of the third metacarpus for the thoracic limbs and in the region of the middle phalanx for the pelvic limbs for the DDFT.

Cadaveric evaluation of canine arthroscopic bicipital tenotomy

SummaryThe purposes of this study were to determine the optimal portal location, limb position and instrumentation for arthroscopic assisted biceps tenotomy as an alternative to open tendon transection and humeral tenodesis and to evaluate anatomical location and support of the tendon following transection.Eight canine cadaver shoulder joints underwent arthroscopic visualization of bicipital tendon length via cranio-lateral and caudo-lateral camera portals in a variety of thoracic limb positions to determine maximal tendon length visualization by anatomical marking. Comparison of tenotomy time and ease was compared between radio frequency microscalpel, blade and arthroscopic shaver. Gross anatomical dissection was performed post-tenotomy to record tendon lengths, locations and supporting structures.The cranio-lateral camera port in conjunction with combined moderate shoulder and elbow flexion optimized tendon visualization, accessible length, and instrumentation ease. Visualized tendon length varied from 39-76% of total tendon length. Tenotomy times were lowest via blade and were unable to be performed with the shaver. After tenotomy the distal tendon segment remained loosely tethered within the in- tertubercular groove at the level of the intertubercular ligament by tendon sheath and capsular attachments.Biceps tenotomy is readily performed with standard arthroscopic equipment. Appropriate limb positioning and modification of previously described portals allows maximal access. Immediately posttenotomy the distal tendon is loosely maintained within the bicipital groove by tendon sheath and capsular attachments.

Laser photoirradiation in digital flexor tendon repair.

This study evaluates tendon coaptation using Nd:YAG laser photoirradiation in an in vivo cockerel model. Using the intervinculum segments of the flexor profundus tendons, experimental transactions were performed. Tendon coaptation was then attempted using laser photoirradiation. Tendons were immediately examined for evidence of stable coaptation. After this assessment, specimens were excised and processed for electron microscopic examination and exposure to trypsin digestion. Despite varying multiple laser parameters, tissue welding was not observed. The subsequent functional and ultrastructural observations of irradiated tendon suggest that these changes are those of simple thermal denaturation. The results of this study suggest that when successful tissue welding has been observed in other tissue types, the mechanism is unlikely to be because of formation of intermolecular collagen bonds as hypothesized. An alternative hypothesis is that laser welding reflects photothermal coagulation of cytoplasmic peptides or nucleic acids liberated at the coaptation interface. This may explain the successful welding of cell-rich tissues such as bowel, vas deferens, and arteries and the observed failure of laser welding in collagen-rich but relatively hypocellular tendon.

The effect of variations in applied rehabilitation force on collagen concentration and maturation at the intrasynovial flexor tendon repair site

The biochemical means by which accelerated rehabilitation alters intrasynovial flexor tendon repair site collagen synthesis and extracellular matrix maturation are not fully understood. We hypothesized that an increased level of applied rehabilitative force in a clinically relevant animal model would hasten the maturation of the repair site extracellular matrix as demonstrated by total collagen and collagen cross-link assessment. Twenty-eight flexor digitorum profundus tendons from 14 adult dogs were transected and repaired. The animals received either low- or high-force rehabilitation and were killed 10, 21, and 42 days after surgery. A 10-mm segment of tendon surrounding the repair site was obtained. Biochemical analysis showed that total collagen concentration was significantly reduced at each time point, that the reducible cross-link ratio of dihydroxylysinonorleucine to hydroxylysinonorleucine was significantly increased at each time point, and that the nonreducible pyridinoline cross-link content was significantly decreased at 10 days in both rehabilitative groups. Total collagen content did not vary to a statistically significant degree with either time or as a function of rehabilitation type. Based on these findings several clinically relevant observations can be made. Increasing collagen concentration and repair site maturation do not explain the previously demonstrated increased tensile properties of tendon that occur between 3 and 6 weeks after repair. Higher force rehabilitation does not alter the biochemical composition of the healing tendon through 6 weeks. Coupled with other recent data these findings suggest that high-force rehabilitation does not stimulate accelerated healing after intrasynovial flexor tendon repair.

Tendon healing stimulated by injected CDMP-2.

Tendon healing stimulated by injected CDMP-2. Med. Sci. Sports Exerc., Vol. 33, No. 5, 2001, pp. 685-687. CDMP-2 is a member of the TFG-beta super-family. It is known to induce bone and cartilage formation but has also been shown under certain conditions to induce a tendon- and ligament-like tissue. The purpose of this study was to find possibilities for improvement of Achilles tendon repair during nonoperative treatment, by local injections of CDMP-2. Fifty rats had a 3-mm segment of the Achilles tendon removed. Six hours postoperatively, CDMP-2 was injected locally into the defect at a dose of 0, 2, 10, or 50 microg. Eight days after the operation, the rats were killed and the tensile strength of the repairing tendons was measured with a materials testing machine. After 8 d, the CDMP-2-treated tendons were 39% stronger than the controls (P = 0.0008). One single injection CDMP-2 can augment tendon repair. Mechanical stimulation is of great importance for tissue differentiation and tendon repair. The tendons in our model were mechanically loaded during healing and this might explain why CDMP-2 injections induced a strong tendon-like tissue instead of bone or cartilage in this model.

The effects of reduced oxygen tension on cell proliferation and matrix synthesis in synovium and tendon explants from the rabbit carpal tunnel: an experimental study in vitro

Local ischemia may play an important role in the development of tendon degeneration as well as entrapment neuropathies. In order to investigate the cellular effects of hypoxia on tendon and synovial tissue from the carpal canal, dose response effects of oxygen on cell proliferation and synthesis of matrix components were examined in segments of synovial and flexor digitorum profundus tendon from the carpal tunnel of rabbits during short term culture. Explants were incubated in airtight containers flushed with either 0%,1%,3%,20% O 2 plus 2% CO 2 and N 2 to balance and labeled with either 3 H -thymidine or 3 H -proline and 35 S -sulfate. Cell proliferation was significantly inhibited by hypoxia in synovium but not in tendon ( P=0.03). In parallel, the synthesis of non-collagenous proteins was significantly reduced in synovium but not in tendon ( P=0.006). In both tissues hypoxia significantly inhibited collagen synthesis. On the other hand, hypoxia had no significant effect on the synthesis of new proteoglycans as determined by 35 S -sulfate incorporation. Hypoxia can inhibit cell proliferation and alter synthesis of matrix components in synovial tissue, but may only have minor effects on non-collagen protein synthesis in tendon explants from the carpal canal of rabbit forepaws.

The influence of flip angle on the magic angle effect.

To assess the impact of flip angle with gradient sequences on the "magic angle effect". We characterized the magic angle effect in various gradient echo sequences and compared the signal-to-noise ratios present on these sequences with the signal-to-noise ratios of spin echo sequences. Ten normal healthy volunteers were positioned such that the flexor hallucis longus tendon remained at approximately at 55 degrees to the main magnetic field (the magic angle). The tendon was imaged by a conventional spin echo T1- and T2-weighted techniques and by a series of gradient techniques. Gradient sequences were altered by both TE and flip angle. Signal-to-noise measurements were obtained at segments of the flexor hallucis longus tendon demonstrating the magic angle effect to quantify the artifact. Signal-to-noise measurements were compared and statistical analysis performed. Similar measurements were taken of the anterior tibialis tendon as an internal control. We demonstrated the magic angle effect on all the gradient sequences. The intensity of the artifact was affected by both the TE and flip angle. Low TE values and a high flip angle demonstrated the greatest magic angle effect. At TE values less than 30 ms, a high flip angle will markedly increase the magic angle effect.

Reconstruction of the Achilles tendon and overlying soft tissue using antero-lateral thigh free flap

Reconstruction of combined loss of the Achilles tendon and overlying soft tissue was performed using an antero-lateral thigh free flap in three patients. The cutaneous portion is used to cover the open wound, and a piece of fascia lata is utilised to replace the missing segment of the Achilles tendon. The skin defect ranged from 5 x 2.5 to 7 x 5 cm, and the tendon loss measured from 3.5 to 5.5 cm in length. All of the patients showed satisfactory functional results with a follow-up period from 3 to 9 months. The advantages of the procedure are that: it is a single-staged operation; it promotes rapid healing of the tendo Achilles since the tendon substitute is well vascularised; it is adaptable to a wide range of defect sizes and shapes; it can be performed in the supine position without the need for postural change; and it can restore good contour and causes minimal morbidity at the donor site.

Tenascin-C in tendon regions subjected to compression.

Tendon regions subjected almost exclusively to tension differ from regions subjected to high levels of compression as well as tension. Regions not exposed to compression consist primarily of spindle-shaped fibroblasts surrounded by densely packed longitudinally oriented collagen fibrils formed principally from type-I collagen. In contrast, regions subjected to compression have a fibrocartilagenous structure and composition: they consist of spherical cells surrounded by a matrix containing hyaline cartilage proteoglycans (aggrecan) and type-II collagen as well as type-I collagen. Reducing their adhesion to the matrix may help cells in the latter regions establish and maintain a spherical shape and minimize their deformation when the tissue is subjected to mechanical stress. We hypothesized that expression of tenascin-C, an anti-adhesive protein, is part of the adaptation of tendon cells to compression that helps establish and maintain fibrocartilagenous regions. To test this hypothesis, we compared segments of bovine flexor tendons subjected to repetitive compression (distal) with segments that are not subjected to compression (proximal) to determine whether they differed in tenascin-C content and expression. RNA and protein analyses showed that tenascin-C expression was elevated in the distal tendon. Tendon cells from the distal segment expressed more tenascin-C mRNA than did cells from the proximal segments for as long as 4 days in cell culture, indicating that increased tenascin-C expression is a relatively stable feature of the distal cells. Moreover, purified tenascin-C inhibited the attachment of cultured tendon cells to fibronectin. These observations support the hypothesis that tenascin-C expression is a cellular adaptation to compression that helps establish and maintain fibrocartilagenous regions of tendons by decreasing cell-matrix adhesion.

HYALURONIC ACID MODULATES CELL PROLIFERATION UNEQUALLY IN INTRASYNOVIAL AND EXTRASYNOVIAL RABBIT TENDONS IN VITRO

As tendons differ in biochemical composition and cellular capacities, we have compared dose response effects of hyaluronic acid on cell proliferation and synthesis of matrix components in intermediate and proximal segments of intrasynovial deep flexor tendons and extrasynovial peroneus rabbit tendons in vitro. Compared with matched control tendons, hyaluronic acid inhibited cell proliferation in intermediate and proximal intrasynovial flexor tendon segments at the concentrations of 0.1–2.0 mg/ml and 0.5–2.0 mg/ml respectively, but in extrasynovial tendon segments only at the concentration of 0.5 mg/ml. Hyaluronic acid did not affect synthesis of proteoglycan, collagen and non-collagen protein in either type of tendon. These results show that hyaluronic acid modulates cell proliferation unequally in intra- and extrasynovial tendons without affecting the synthesis of matrix components in the two types of tendons, indicating differential hyaluronic acid sensitivity and a possible mechanism of action.

Tear of the peroneus longus tendon: MR imaging features in nine patients.

To determine the magnetic resonance (MR) imaging features that characterize tear of the peroneus longus tendon at the midfoot. Medical records and MR images in nine patients with a tear of the middle segment of the peroneus longus tendon were retrospectively reviewed. All nine patients had undergone routine ankle MR imaging; three had undergone additional oblique coronal MR imaging. Surgical proof of a tear was available for three patients. Partial tear was present in four patients, and complete tear was present in five. Partial tears were characterized by heterogeneous signal intensity and thickening of the tendon. Complete tears were characterized by discontinuity of the tendon. Additional findings included fluid in the tendon sheath (n = 6), marrow edema of the lateral calcaneal wall (n = 3), enlarged peroneal tubercle (n = 3), and tear of the peroneus brevis tendon (n = 2). The extent of the tear was better assessed with oblique coronal MR images. The characteristic MR imaging appearance of complete or partial tear of the middle portion of the peroneus longus tendon includes foci of increased signal intensity in the distal tendon, morphologic alterations, and/or discontinuity of tendon. Bone marrow edema along the lateral calcaneal wall may be suggestive of the diagnosis. Additional oblique coronal midfoot MR images may help in assessment of the extent of the tear.

Distribution of sonographically detected tendon abnormalities in patients with a clinical diagnosis of chronic achilles tendinosis.

We conducted a retrospective study of the distribution of sonographically detected abnormalities in the heels of patients who had a clinical diagnosis of Achilles tendinosis. One hundred eighteen symptomatic heels in 73 patients who had a clinical diagnosis of chronic Achilles tendinosis were examined over a 12-month period by the same experienced sonologist. The distribution of altered tendon architecture and features suggesting retrocalcaneal bursitis or Achilles paratendinosis were evaluated. Sonograms of 118 symptomatic heels demonstrated that 96 (81%) had abnormalities confined to the proximal two thirds of the Achilles tendon, 9 (8%) had abnormalities in the distal third alone, and 13 (11%) had abnormalities at both sites. Of the 109 heels with proximal two-third Achilles tendon disease, 99 (91%) had medial tendon involvement; 22 of the 99 showed diffuse tendon changes. Lateral tendon segment changes were seen in 22 (19%) of the 118 symptomatic heels. No lateral tendon segment was involved in isolation. Of the 22 heels with distal third abnormalities, 14 (64%) had sonographic evidence of Achilles paratendinitis, and 13 (59%) had sonographic evidence of Achilles tendinosis. Eighteen of the 22 had sonographic evidence of retrocalcaneal bursitis. In all cases of distal third tendinosis, the deep surface of the tendon was primarily involved. In the heels with both proximal and distal changes, superficial segment involvement of the mid-Achilles tendon was present. Sonography provides information that helps to accurately diagnose clinical Achilles tendinopathy and may help to determine the biomechanical processes involved in the injury.

Tenoplastia experimental do calcâneo em cães com peritônio bovino conservado em glicerina

No presente estudo, avaliou-se a eficácia do emprego do peritônio bovino, conservado em glicerina a 98%, no reparo de lesões induzidas no tendão calcâneo (TC) de cães, quando um fragmento de aproximadamente 1cm do TC foi excisado e o espaço resultante preenchido por um fragmento de peritônio. Foram utilizados 21 cães, pesando entre 10 e 15kg, divididos em 7 grupos de 3, sacrificados aos 02, 07, 15, 30, 60, 90 e 120 dias de pós-operatório. Analisaram-se os aspectos clínico-cirúrgicos referentes à recuperação funcional motora, bem como, a integração do peritônio com o tecido tendíneo mediante avaliação macroscópica, por microscopia óptica e por microscopia eletrônica de varredura. Clinicamente, verificou-se que, por volta do 55º dia de pós-operatório, os animais já apresentavam deambulação normal e que o "neotendão" apresentou resistência suficiente para suportar o estresse normalmente aplicado ao TC. Microscopicamente, o peritônio implantado esteve presente em todos os períodos de observação. Proliferação fibroblástica e neoformação vascular foram observadas de forma incipiente no segundo dia; entretanto, no sétimo dia de pós-operatório, esta condição foi exacerbada. Com a evolução, as fibras de peritônio tendiam a se dissociar, entrando em estreita associação com fibras conjuntivas, fibroblastos e colágeno. Aos 30, 60, 90 e 120 dias de pós-operatório, notava-se maior presença de colágeno que se tornava cada vez mais organizado. Concluiu-se que o peritônio estimulou uma rápida deposição de tecido conjuntivo com mínima reação inflamatória, sendo incorporado ao tecido cicatricial e servindo como alicerce para o desenvolvimento de um novo tecido, restabelecendo assim a estrutura do tendão.

In vitro effects of epidermal growth factor or insulin-like growth factor on tenoblast migration on absorbable suture material.

To determine the effects of epidermal growth factor (EGF) or insulin-like growth factor (IGF) on tenoblast migration on absorbable suture material using an in vitro model. An in vitro evaluation of tenoblast migration. Segments of the long digital flexor tendon were obtained from Cobb chickens (9-11 weeks old) immediately after the birds were euthanatized. Tissue culture explants of tendons containing absorbable suture material were treated with either EGF or IGF. Tenoblast migration was assessed daily using an inverted microscope equipped with bright field and phase optics. Tenoblast migration was assessed according to the following criteria: time of first cell appearance, percent of explant interfaces producing cells, migration distance, and terminal migration index at 120 and 168 hours. EGF had a stimulatory effect on tenoblast migration for cells originating from the endotenon interfaces. No significant effect was noted on migration distance for cells originating from epitenon interfaces. A stimulatory effect on the percentage of interfaces producing cells and a significant decrease in time of first cell appearance were also observed after EGF treatment. IGF-stimulated cell migration distance for epitenon interfaces but this stimulatory effect did not occur at a higher concentration. IGF was inhibitory to percent of epitenon and endotenon interfaces producing cells but decreased time of first cell appearance at low concentration. Using an in vitro model, EGF had a stimulatory effect on tenoblast migration. IGF was stimulatory at low concentration levels but inhibitory at a higher concentration. Increased migration distance was observed for endotenon interfaces after EGF treatment and for epitenon interfaces after IGF treatment. EGF or IGF might enhance tendon repair if they could be delivered to the repair site. Incorporation of EGF or IGF into suture material would allow slow release and prolonged exposure of migrating tenoblasts to growth factors.

In reply

There are many solutions to this problem, and we commend Dr Bunata's comments. Our thoughts are that a loop can be made too tight, as can a sutured tendon segment. Adhesions will probably prevent any significant excursion of tendon through the loop. The key to full tendon hood excursion includes having the repair graft perpendicular to the extensor tendon with the finger in midrange of motion. Following repair, the surgeon should pull on the extensor tendon and ensure full extension of all three joints. Additionally, he should check that full flexion occurs without lateral deviation of the extensor tendon. Suturing to or around the lateral band has the advantage of being less likely to restrict motion, but the disadvantage of a less secure purchase for maintaining a centralized extension tendon in full flexion. Proximal interphalangeal motion in our patients was normal.

Ideal concentration of growth factors in rabbit's flexor tendon culture.

Growth factors have the ability to stimulate matrix synthesis and cell proliferation in rabbit flexor tendon. Maximal stimulation effects of growth factors have a wide variation. It depends upon the different anatomic sites of the tendon segment, the kinds of growth factor, the concentration of growth factors, and the time sequence. Since proliferation was an early component of intrinsic tendon healing, we investigated the short-term dose response to four different growth factors on in vitro rabbit's tendon culture. We evaluated the effects according to the various concentrations of recombinant human insulin-like growth factor 1 (IGF), recombinant human epidermal growth factor (EGF), fibroblast growth factor (FGF), and recombinant human platelet-derived growth factor-BB (PDGF). Fetal calf serum was the most potent stimulator of cell proliferation and protein synthesis in in vitro rabbit's tendon culture. Matrix synthesis and cell proliferation were stimulated dose-dependently by IGF between the doses of 50 and 150 ng/ml. The maximum mitogenic effect of EGF was observed at the concentration of 100 ng/ml (1.3 times more than the media-only control culture). The rabbit's tendon responded significantly dose-dependently to PDGF, whereas there was no significant response to FGF.

Interference screw position and hamstring graft location for anterior cruciate ligament reconstruction

Anterior cruciate ligament reconstruction with hamstring tendon graft and interference screw fixation has recently been considered. Concerns for the use of interference screws with soft tissue grafts include damage to the graft during screw insertion, decreased fixation strength, and a decrease in the bone-tendon contact area for healing within the tunnel when the screw is placed in an eccentric position. This last concern could be addressed by placing the interference screw centrally between the four limbs of the hamstring graft. The purpose of this study was to determine the mode of failure, the pullout force, and graft slippage before graft fixation failure of hamstring tendons fixed with an interference screw positioned eccentrically in relation to the hamstring tendons verses an interference screw positioned centrally between the four graft limbs. The semitendinosus and gracilis tendons were harvested from six, fresh cadaveric specimens. Each tendon was divided into two segments of equal length. Both the semitendinosus and gracilis tendon segments were looped to form four strands. The specimens were then fixed with a bioabsorbable interference screw in the two different positions and pulled from a standardized polyurethane foam. All tendons in both groups failed by pulling out from between the interference screw and tunnel, regardless of the screw position. No tendon was cut by the screw in either group. There was no significant difference between the forces required to produce specific amounts of graft slippage between the two fixation techniques tested. There was no significant difference between the average total slippage at maximum pullout, 11.8 mm for the screw placed in the eccentric position and 13.7 mm for the screw placed in the central position. The maximum pullout force averaged 265.3 N for the screw placed in the eccentric position, and 244.7 N for the screw placed in the central position; these values were not significantly different. Placement of the interference screw in the central position did not compromise strength and it improves graft contact within the bone tunnel. Interference screw fixation, when applied against a bone plug, has been shown to consistently have a pullout force of more than 400 N. Arthroscopy 1998 Jul-Aug;14(5):459-64

The interface between bone and tendon at an insertion site: a study of the quadriceps tendon insertion.

Traumatic avulsions of ligament or tendon insertions rarely occur at the actual interface with bone, which suggests that this attachment is strong or otherwise protected from injury by the structure of the insertion complex. In this study we describe the terminal extent of quadriceps tendon fibres where they insert into the patellae of adult rabbits, humans, dogs and sheep. Specimens were examined by scanning electron microscopy (SEM) and light microscopy (LM). To facilitate tracing of tendon fibres the specimens were decalcified for SEM, and polarised light microscopy (PLM) was used in the LM segment of the study. By SEM it was possible to identify mature bone by the presence of osteocytes and a lamellar organisation. PLM and SEM showed that, unlike tendon fibres elsewhere, those in the calcified fibrocartilage were not crimped. No specific cement line was identified by SEM. Tendon fibres interdigitated among separate bone lamellar systems, (osteons or marrow spaces), but did not merge with the collagen systems of individual lamellae. The interdigitation was more extensive and the margin between tendon and bone was less distinct in the anterior third of the insertion. The segment of calcified tendon which interdigitated with bone stained less intensely blue and was less cellular than the more proximal calcified fibrocartilage zone adjacent to the tidemark. Lamellar collagen fibres of the bony trabeculae in the anterior patella were unusually parallel and longitudinal in orientation, making distinction of interposed tendon fibres difficult on LM and PLM sections. LM, SEM and transmission electron microscopy of rabbit patellae at birth revealed that anterior quadriceps tendon fibres extended over the patella in a fibrous cellular layer. By 2 wk of age, this layer had acquired chondroid features (i.e. cell lacunae and metachromasia) and contained vessels extending from patellar marrow. At 6 wk of age, part of this fibrocartilaginous layer was replaced by mature bone and osteoid. In the young adult animal, the quadriceps tension interdigitates extensively with the patellar bone. This segment of the insertion is perhaps the remnant of calcified fibrocartilage which has been remodelled by bone formation.