Tibiotalocalcaneal arthrodesis with an intramedullary nail: The functional and clinical outcome of a challenging patient group and its comparison to a below knee amputation.

BACKGROUND: Tibiotalocalcaneal arthrodesis (TTCA) is a salvage procedure to fuse the ankle and subtalar joints to treat severe ankle and hindfoot disease. Patients often have multiple operations before a TTCA. A below knee amputation (BKA) has major physical and psychological impacts. Our aim is to investigate the outcomes of these treatments in salvage situations, with a focus on objective functional measurements and patient reported outcome measures (PROMs). METHODS: 52 patients underwent TTCA with a retrograde intramedullary nail and contacted for clinical and functional assessments and compared to patients who underwent traumatic BKA. PROMS such as AOFAS score, SF-36 and foot function index (FFI), and objective functional outcome measures were used. RESULTS: Of the 52 TTCA patients, 28 patients were recruited for follow-up. 35.7% of patients had postoperative complications. Mean postoperative AOFAS score was 63.9 ± 8.4 (range, 47-81), FFI 48.8 ± 15.8 (range, 22.2-75.2). 11 BKA patients (mean age 46.4 years) were included as control group. BKA patients scored higher than the TTCA patients on SF-36 physical functioning (p < 0.01) and SF-36 mental health (p < 0.05) subscales. The flat-surface functional tests (timed up and go test, 2-minute walk test, 10-meter walk test) showed significantly (p < 0.05) better outcomes for the BKA compared to TTCA. CONCLUSIONS: A TTCA is a salvage procedure with high complication rates. Functional and psychometric results are reduced compared to the normal population. Patients after a BKA had significantly better scores on SF-36 functional and mental health subscales and better functional outcomes for flat ground activities compared to TTCA. Heterogeneity of the TTCA and BKA patient cohorts is a limitation of this study. With these results in mind, the outlook of a BKA is not necessarily a grim one. They may be used by surgeons to counsel patients preoperatively when managing complex ankle and hindfoot disease. LEVEL OF EVIDENCE: Level III, retrospective cohort study.

Anatomy of the tibial incisura as a risk factor for syndesmotic injury.

BACKGROUND: The study aims at comparing the bony anatomy of the syndesmosis in patients who sustained a high fibular fracture with syndesmosis disruption and that of the non-injured population. We hypothesised that there are certain anatomical features making the syndesmosis susceptible to injury. METHODS: The CT examinations of 75 patients who sustained a high fibular fracture with syndesmosis disruption and control group of 75 patients with unrelated foot problems were compared. The depth, fibular engagement and rotational orientation of the tibial incisura were analyzed. RESULTS: With the median values of the control group as cutoff there were 71% shallow, 71% disengaged and 77% retroverted syndesmoses in the injury group. The differences between the groups were statistically significant for every measure (P<.002 to P>.0001). CONCLUSIONS: Patients with a shallow, disengaged and retroverted bony configuration of the syndesmosis are overrepresented among patients with syndesmosis disruption.

["Calcaneogenesis" with secondary Achilles tendon-bone allograft for repair of the loss of hindfoot function : A 12-year case report]. / "Kalkanogenese" mit sekundärer Achillessehnen-Knochenblock-Alloplastik zur Wiedergewinnung der Rückfußfunktion : Eine 12-Jahres-Kasuistik.

BACKGROUND: Calcanectomy and Achilles tendon resection are very hard to repair. OBJECTIVE: Ilizarov's "calcaneogenesis" is possible with ankle joint preservation. Even after 3.5 years of functio laesa of the triceps surae muscle it can be rebuilt. MATERIAL AND METHODS: A 25-year-old motorcyclist suffered a 3rd degree open calcaneal dislocation fracture (type 5). Osteitis and necrosis required calcanectomy, resection of the Achilles tendon and a latissimus dorsi muscle transfer. A talus corpus osteotomy with Ilizarov distraction created in the 1st step a "neo-calcaneus". In a delayed 2nd step a fresh-frozen Achilles tendon-bone block allograft was transplanted to regain active plantar flexion. RESULTS: The initial AOFAS score of 35 points was significantly improved to 70 points 12 years after step 1. After both operations the patient could walk without an orthosis and regained 88% of normal plantar flexion strength. Quantitative measure of health outcome according to EQ-5D-5L was marked by the patient with 80 out of 100 points. DISCUSSION: "Calcaneogenesis" with preservation of the ankle joint is possible and innovative. Despite 3.5 years of disconnection of the triceps surae muscle, an Achilles tendon-bone block allograft could restore 88% of the push-off force even attached to a neo-calcaneus that is 1/3 smaller than normal, which is also new.

Stability of locking and non-locking plates in an osteoporotic calcaneal fracture model.

BACKGROUND: The aim of this biomechanical cadaver study of calcaneal fractures was to investigate whether a locking calcaneal plate provides more stiffness in osteoporotic bone compared to a non-locking plate. MATERIALS AND METHODS: Sixteen fresh frozen bone mineral density (BMD)-matched cadaver feet were tested in a four-part model of a Sanders Type IIB calcaneal fracture. The fractures were fixed either with a non-locking AO (Sanders) plate or an interlocking AO plate (Synthes, Paoli, PA) to the lateral calcaneal wall with six screws. Specimens were subjected to cyclic loading which was increased stepwise to full body weight. Displacement of the posterior facet fragment was measured with an optical tracking system in the sagittal and transverse planes. RESULTS: No statistically significant differences were observed between the non-locking and the locking plates with respect to number of cycles to failure or 1-mm displacement of the posterior facet. The initial stiffness was significantly higher for non-locking plates. CONCLUSION: In osteoporotic bone, the greater stiffness of the screw-locking-plate construct was offset by the smaller diameter of the screw threads and the lower friction between the plate and bone when a locking plate was used. In clinical practice, the plate should first be compressed to osteoporotic bone with cancellous screws and at least two screws should be placed in the anterior process and in the tuberosity of the calcaneus.

Injuries to the distal tibiofibular syndesmosis: an evidence-based approach to acute and chronic lesions.

Injuries to the distal tibiofibular syndesmosis are frequent in collision sports. Most of these injuries are not associated with latent or frank diastasis between the distal tibia and fibula and are treated as high ankle sprains, with an extended protocol of physical therapy. Relevant instability of the syndesmosis results from rupture of two or more ligaments leading to a diastasis of more than 2 mm and requiring surgical fixation. Most of these syndesmosis ruptures are associated with bony avulsions or malleolar fractures. Treatment consists of anatomic reduction of the fibula and fixation with one or two tibiofibular syndesmosis screws. Proper reduction and positioning of the screws are more predictive of a good clinical result than the material, size, and number of cortices purchased. Chronic injuries without instability are treated by arthroscopic or open debridement and arthrolysis. Chronic syndesmotic instability can be treated with a three-strand peroneus longus ligamentoplasty in the absence of symptomatic arthritis or bony defects.

Transsection of the rectus abdominis muscle in the treatment of acetabular fractures: operative technique and outcome in 21 patients.

BACKGROUND: A standard ilioinguinal approach is often insufficient for reduction and stabilization of the medial acetabular wall and the dorsal column in acetabular fractures. To avoid extended approaches, we have used a medial extension of the approach by transverse splitting of the rectus abdominis muscle. We have thus been able to reduce and stabilize transverse and oblique fractures of the dorsal column and the medial acetabular wall and to fix plates in a mechanically better position below the pelvic brim. To evaluate the procedure, especially the risk of abdominal hernia, we started a prospective study. PATIENTS AND METHODS: Over 2 years, we treated 21 consecutive patients using a transverse splitting of the rectus abdominis muscle-either as an extension of the standard ilioinguinal approach or in combination with parts of this approach or a Kocher-Langenbeck approach. The patients were evaluated clinically and radiographically after 1 year. RESULTS: The clinical and radiographic results were excellent or good in 18 patients. Complications occurred in 5 patients. No hernias were observed. CONCLUSIONS: Our small study indicates that the procedure described is a useful and safe complement to the intrapelvic approaches. The procedure does not provide better reduction than extended approaches, but may help to avoid them in some cases.

Correlation of Incisura Anatomy With Syndesmotic Malreduction.

BACKGROUND: The anatomy of the syndesmosis is variable, yet little is known on the correlation between differences in anatomy and syndesmosis reduction results. The aim of this study was to analyze the correlation between syndesmotic anatomy and the modes of syndesmotic malreduction. METHODS: Bilateral postreduction ankle computed tomography (CT) scans of 72 patients treated for fractures with syndesmotic disruption were analyzed. Incisura depth, fibular engagement into the incisura, and incisura rotation were correlated with degree of syndesmotic malreduction in coronal and sagittal planes as well as rotational malreduction. RESULTS: Clinically relevant malreduction in the coronal plane, sagittal plane, and rotation affected 8.3%, 27.8%, and 19.4% of syndesmoses, respectively. The syndesmoses with a deep incisura and the fibula not engaged into the tibial incisura were at risk of overcompression, anteverted incisuras at risk of anterior fibular translation, and retroverted incisuras at risk of posterior fibular translation. CONCLUSIONS: Certain morphologic configurations of the tibial incisura increased the risk of specific syndesmotic malreduction patterns. LEVEL OF EVIDENCE: Level III, comparative study.

Effect of modification of hydroxyapatite/collagen composites with sodium citrate, phosphoserine, phosphoserine/RGD-peptide and calcium carbonate on bone remodelling.

This study describes the early interface reaction of cancellous bone to a nanocrystalline hydroxyapatite cement containing type I collagen (HA/Coll) and its modifications with sodium citrate (CI), calcium carbonate (CA), phosphoserine (P) and phosphoserine plus RGD-peptide (RGD). Cylindrical implants of HA/Coll and its modifications were inserted into the tibia of Wistar rats. We analysed 6 specimens per group at days 2, 4, 7, 14 and 28. CI, P and RGD modifications showed improved material properties (finer microstructure and higher compressive strength) compared to CA and HA/Coll implants. The powder X-ray diffraction (XRD) showed that the addition of P and CI led to an increase of alpha-TCP peaks while the diffraction patterns of the non-modified cement (HA/Coll) were quite similar with that of natural bone. All of the implants healed without adverse reactions. A significantly higher number of TRAP-positive osteoclasts were observed around CI, RGD and P on day 7 compared to CA and HA/Coll. Around CI, P and RGD a significantly delayed increase of ED1-positive mononuclear cells was detected. The amount of direct bone contact after 28 days was significantly higher around CI, P and RGD compared to CA and HA/Coll implants. The addition of CI, P and RGD appears to enhance bone remodelling at the early stages of bone healing, leading to increased bone formation around HA/Coll composite cements.

Biomechanical Analysis of Stability of Posterior Antiglide Plating in Osteoporotic Pronation Abduction Ankle Fracture Model With Posterior Tibial Fragment.

BACKGROUND: We performed a biomechanical comparison of 2 methods for operative stabilization of pronation-abduction stage III ankle fractures; group 1: Anterior-posterior lag screws fixing the posterior tibial fragment and lateral fibula plating (LSLFP) versus group 2: locked plate fixation of the posterior tibial fragment and posterior antiglide plate fixation of the fibula (LPFP). METHODS: Seven pairs of fresh-frozen osteoligamentous lower leg specimens (2 male, and 5 female donors) were used for the biomechanical testing. Bone mineral density (BMD) of each specimen was assessed by means of dual-energy x-ray absorptiometry. After open transection of the deltoid ligament, an osteotomy model of pronation abduction stage III ankle fracture was created. Specimens were systematically assigned to LSLFP (group 1, left ankles) or LPPFP (group 2, right ankles). After surgery, all specimens were evaluated via CT to verify reduction and fixation. Axial load was then applied onto each specimen using a servohydraulic testing machine starting from 0 N (Zwick/Roell, Ulm, Germany) at a speed of 10 N/s with the foot fixed in a 10 degrees pronation and 15 degrees dorsiflexion position. Construct stiffness, yield, and ultimate strength were measured and dislocation patterns were documented with a high-speed camera. The normal distribution of all data was analyzed using Shapiro-Wilk test. The group comparison was performed using paired Student t test. Statistical significance was assumed at a P value of .05. RESULTS: All specimens had BMD values consistent with osteoporosis. BMD values did not differ between the left and right ankles of the same pair ( P = .762). The mean BMD values between feet of men (0.603 g/cm2) and women (0.329 g/cm2) were statistically different ( P = .005). The ultimate strength for LSLFP (group 1) with 1139 ± 669 N and LPPFP (group 2) with 2008 ± 943 N was statistically different ( P = .036) as well as the yield in LSLFP (group 1) 812 ± 452 N and LPPFD (group 2) 1292 ± 625 N ( P = .016). Construct stiffness trended to be higher in group 2 (179 ± 100 kNn) compared to group 1 (127 ± 73 kN/m) but this difference was not statistically significant ( P = .120). BMD correlated with bone-construct failure. CONCLUSION: Fixation of the posterior tibial edge with a posterolateral locking plate resulted in higher biomechanical stability than anterior-posterior lag screw fixation in an osteoporotic pronation-abduction fracture model. CLINICAL RELEVANCE: The clinical implication of this biomechanical study is that the posterior antiglide plating might be advantageous in patients with osteoporotic pronation abduction stage III ankle fracture.

Reconstruction after talar fractures.

Malunited and nonunited talar fractures cause significant disability. Distinction between partial and total avascular necrosis (AVN) determines the choice of treatment. Patients who have minimal or no AVN and well-preserved joint cartilage may be amenable to corrective osteotomy through the malunited fracture or removal of the pseudoarthrosis. Secondary reconstruction with joint preservation leads to considerable functional improvement in painful talar malunions and nonunions in reliable patients. If symptomatic arthritis is present, arthrodeses and correction of deformity through the fusion or with additional osteotomies provide predictable results, although they do not restore normal foot function. Fusions should be limited to the affected joint. If the subtalar joint shows severe arthritic changes, every attempt should be made to salvage the ankle and talonavicular joints.

[Tibiotalocalcaneal arthrodesis using a distally introduced femur nail (DFN)].

OBJECTIVE: Simultaneous arthrodesis of ankle and subtalar joints and, at the same sitting, correction of axial malalignment of hindfoot, treatment of bony defects and of sequelae of circulatory disturbances after multiple previous interventions. Internal stabilization with a short distal femur nail. Restitution of a pain-free weight bearing. INDICATIONS: Failure of arthrodesis of ankle and subtalar joint causing pain in patients with severely altered bone structures particularly at the level of the talar dome. Malalignment of hind- and/or forefoot after previous arthrodesis of ankle and subtalar joint. Failure of conservative therapy in both above-enumerated conditions. CONTRAINDICATIONS: Poor skin or soft-tissue conditions. Reflex sympathetic dystrophy. Acute osteitis/osteomyelitis. SURGICAL TECHNIQUE: Posterolateral approach. Resection of the articular cartilage and the areas of sclerosis of the ankle and of the posterior facet of the subtalar joint. Interposition of bone grafts harvested from the iliac crest. Correction of malalignment of hind- and forefoot. Locked nailing with a short distal femur nail. POSTOPERATIVE MANAGEMENT: Fitting of a flexible custom-made arthrodesis boot; weight bearing in boot not exceeding half of body weight until the 8th week. Gait training. After 12 weeks wearing of normal shoes. Radiographs after 6 and 12 weeks. RESULTS: Between February 1, 2002 and September 1, 2003 this technique was performed on 18 feet in 17 patients (three women, 14 men, average age 53 years [38.9-73.7 years]). Average duration of follow-up: 1.2 years (0.6-2.1 years). The goal of surgery was reached in all patients. Subjective assessment: 14 times good, three times satisfactory. Four complications: one loss of nail purchase, one dislocation of locking screw, one breakage of locking bolt, one prolonged bone healing.

[Ankle arthrodesis after failed joint replacement].

UNLABELLED: Report of four patients. OBJECTIVE: Restoration of a painless gait through ankle arthrodesis after failed total ankle replacement. INDICATIONS: Loosened or infected total ankle replacement. CONTRAINDICATIONS: Poor general health. Absent patient compliance. SURGICAL TECHNIQUE: Removal of total joint components. Filling of the defect with tricortical bone grafts harvested from the ipsilateral iliac crest and internal fixation. In instances of suspected infection a two-stage procedure is recommended, the first stage consisting of a removal of the components, a meticulous debridement, and filling of the defect with gentamycin-laden PMMA beads. Second stage: arthrodesis. RESULTS: The goal of surgery was reached without complications in two out of four patients. An absence of bony bridging was noted in the fourth patient after 9 months. A revision adding cancellous bone grafts and resorting to an intramedullary fixation led to a success.

[Ankle arthrodesis with four cancellous lag screws].

OBJECTIVE: Arthrodesis of the ankle at 90 degrees and perfect axial alignment for restoration of a painless function. Early functional postoperative care. INDICATIONS: Painful posttraumatic or idiopathic osteoarthritis of the ankle either unresponsive to conservative measures or where these measures are not expected to be successful. Posttraumatic malalignment of the ankle, paralysis or instability, that cannot be improved or eliminated by joint-preserving measures. Joint destruction after infection. Failure of total joint replacement. CONTRAINDICATIONS: Acute osteitis. Poor skin or soft-tissue conditions. Severe peripheral arterial occlusive disease. SURGICAL TECHNIQUE: Anterior approach, judicious resection of the remaining articular cartilage. Freshening of the zones of sclerosis. Preservation of the anatomic shape of the mortise. Correction of malalignments in the sagittal and frontal planes and placement of the talus in line with the tibial axis. Tibiotalar stabilization with four 7.3-mm self-cutting cannulated cancellous lag screws or with four 6.5-mm cancellous lag screws. RESULTS: Between January 1, 1994 and December 31, 1998 this technique was performed in 50 ankles of 48 patients. 40 patients could be followed up for an average of 5.6 years (4.8-7.6 years). No serious complications. The average compensatory movement of the Chopart joint amounted to 26 degrees . Osteoarthritis of the subtalar joint was seen in 13%, and of the talonavicular joint in 12.5% of patients. Preexisting osteoarthritis of these joints remained in general unchanged. The AOFAS Score was assessed pre- and postoperatively. Preoperatively, 17.5% of patients showed a satisfactory and 82.5% a poor score. Postoperatively, 52.5% had an excellent, 30% a good, 10% a satisfactory, and 7.5% a poor outcome.

Peroneus longus ligamentoplasty for chronic instability of the distal tibiofibular syndesmosis.

The distal tibiofibular syndesmosmotic ligament complex is important for dynamic stability and congruency of the ankle joint. Syndesmotic lesions in the ankle fracture-dislocations are well recognized and classified systematically. Chronic insufficiency of the syndesmosis leads to a lateral shift of the talus and under eversion stress permits a pathological rotation of the talus. There is also retroversion of the distal fibula representing a painful deformity. Little experience exists with surgical reconstruction of the syndesmosis. This article describes a new ligamentoplasty with a split peroneus longus tendon graft that mimics the normal anatomic conditions of the syndesmotic complex in 16 patients with symptomatic chronic syndesmotic insufficiency after pronation-external rotation and pronation abduction injuries to the ankle joint. Postoperatively, no infections or hematomas were seen. One patient had asymptomatic breakage of the syndesmosis screw; one patient had a 10 degree decrease of dorsiflexion at the ankle because of a partial anterior tibiofibular synostosis. Fifteen of 16 patients had pain relief at a mean follow-up period of 16.4 months (range, 13-29 months); all patients had relief of the chronic swelling of the ankle and the giving way. The mean Karlsson score at follow-up was 88 (range, 70-100) points. It may be concluded that peroneus longus ligamentoplasty in a preliminary series resulted in reliable ankle stability and considerable pain relief in patients with chronic syndesmotic instability.

The role of external fixation in acute ankle trauma.

The use of external fixation methods has become increasingly popular throughout the past decade in the treatment of tibial pilon fractures to prevent the dreaded soft tissue complications after high velocity injuries. A variety of methods has been proposed, including ankle-spanning half-pin frames; circular (Ilizarov) frames with tensioned wires; or hybrid frames, either as the sole treatment or, more frequently, in conjunction with limited internal screw fixation. External fixation also has a role in staged protocols as a primary tool for reduction and preliminary fixation until soft tissue consolidation makes internal fixation feasible. Although good to excellent results are reported in a high percentage of cases in most studies and infection rates have dropped to less than 10%, even for high velocity injuries with considerable soft tissue compromise, no single form of treatment seems to be suitable for all types of pilon fractures. Major concerns after external fixation are the development of pin track infections, malunions or nonunions, and the danger of imperfect reduction of the articular surface. Staged protocols that are based on the severity of the fracture and soft tissue injury are likely to play a major role in the future treatment of pilon fractures. In the treatment of acute malleolar fractures, ankle-spanning external fixation is reserved for fractures with considerable soft tissue compromise, open fractures, or compartment syndrome as a temporary transfixation until internal fixation becomes feasible.

Ligamentous injuries about the ankle and subtalar joints.

Ligamentous injuries at the ankle and subtalar joint range from simple sprains to severe talar dislocations. While lateral ankle sprains are among the most frequently encountered injuries and do not pose a greater diagnostic problem, the surgeon must be suspicious not to overlook associated ligamentous injuries at the subtalar and midtarsal level that may result in chronic painful conditions. Syndesmotic instabilities with or without ankle fractures must be assessed carefully and treated properly, since these are prearthrotic conditions. In the treatment of chronic ankle or subtalar instability tenodeses provide mechanical stability while reducing subtalar mobility. Anatomic reconstruction methods therefore should be considered for both conditions.

Arthrodesis of the ankle joint.

Arthrodesis of the ankle that uses lag screws for internal fixation is a safe and biomechanically stable method to obtain a solid ankle fusion; it generates good to excellent results in most patients. Prompt bone healing can be expected and allows a functional rehabilitation with full weight bearing. The surgical technique can be simplified further when using cannulated screws. Malalignment hazards while doing the ankle fusion are minimized by respecting the shape of the ankle mortise because no osteotomy of the lateral malleolus is performed. Generally, removal of the implanted material is not necessary. Major complications such as infection, stress fractures, or nonunion were not seen in our series. A meticulous resection of all cartilage and sclerotic bone and an atraumatic surgical technique are essential for preventing those major complications. The need for revision surgery is minimized by correction of talar malalignment, fusion with the ankle in a 90 degrees position, and preoperative evaluation of the subtalar joint. External fixation methods are used in cases of osteitis, osteonecrosis, osteoporosis, and poor soft tissue conditions. With severe loss of bone stock at the distal tibia, stability can be achieved by using an intramedullary nail for ankle fusion.

Joint-preserving osteotomy for malunited intra-articular calcaneal fractures.

Displaced intra-articular fractures of the calcaneus can lead to severe pain and disability if not treated appropriately. Failure to treat may require fusion of the subtalar joint, supplemented by additional osteotomies. Occasionally, these fractures are seen after the fracture has just healed, and the decision to treat can be a difficult one. Over the course of 10 years, 5 patients were treated with a corrective osteotomy along the primary fracture line, joint realignment, soft tissue balancing, and secondary internal fixation at a mean of 2.9 months after the injury. At a mean of 4.1 years (range, 2-10 years), all patients were satisfied with their result. Two patients underwent implant removal and subtalar arthrolysis 1 year after correction. No secondary fusions were required. The mean American Orthopaedic Foot & Ankle Society (AOFAS) hindfoot score improved significantly from 19.0 preoperatively to 81.2 at follow-up (P < 0.001). The radiographic parameters (the Böhler angle, talocalcaneal height, and heel width) were substantially corrected. A joint-preserving osteotomy with axial realignment can be a treatment option for malunited intra-articular calcaneal fractures encountered early on, before the development of subtalar arthrosis.

In vivo effects of modification of hydroxyapatite/collagen composites with and without chondroitin sulphate on bone remodeling in the sheep tibia.

The addition of chondroitin sulphate (CS) to bone cements with calcium phosphate has lead to an enhancement of bone remodeling and an increase in new bone formation in small animals. The goal of this study was to verify the effect of CS in bone cements in a large animal model simulating a clinically relevant situation of a segmental cortical defect of a critical size on bone-implant interaction and bone remodeling. The influence of adding CS to hydroxyapatite/collagen (HA/Col) composites on host response was assessed in a standard sheep tibia model. A midshaft defect of 3 cm was created in the tibiae of 14 adult female sheep. The defect was filled with a HA/Col cement cylinder in seven animals and with a CS-modified hydroxyapatite/collagen (HA/Col/CS) cement cylinder in seven animals. In all cases the tibia was stabilized with an interlocked universal tibial nail. The animals in each group were analyzed with X-rays, CT scans, histology, immunohistochemistry, and enzymehistochemistry, as well as histomorphometric measurements. The X-ray investigation showed a significantly earlier callus reaction around the HA/Col/CS implants compared to HA/Col alone. The amount of newly formed bone at the end point of the experiment was significantly larger around HA/Col/CS cylinders both in the CT scan and in the histomorphometric analysis. There were still TRAP-positive osteoclasts around the HA/Col implants after 3 months. The number of osteopontin-positive osteoblasts and the direct bone contact were significantly higher around HA/Col/CS implants. We conclude that addition of CS enhances bone remodeling and new bone formation around HA/Col composites.