Aim

Bone and joint infections (BJI) are associated with a heavy morbidity and high health costs. Comorbidities, device associated infections and complicated journeys are associated with increased mortality, treatment failures and costs. For this reason, 24 referral centers (RC) have been created in 2009 in order to advise about management of “complex” BJI in weekly multidisciplinary meetings (MM). Since end of 2012, data from these meetings are gathered in a national database. We aimed to describe the data from this French registry of BJI and determine factors associated with the definition of “complex” BJI.

Introduction

For preoperative planning of Total Hip Arthroplasty (THA) it is paramount to choose the correct implant size to avoid subsidence with too small a component or fracture with too large a component. This planning can be done either in 2D or 3D. 2D templating from X-rays frontal images remains the gold standard technique in THA preoperative planning despite the lower accuracy with uncemented components. 3D planning techniques require a CT-Scan examination overexposing patients to radiation. Biplanar EOS^® radiographs are an alternative to obtain bone 3D reconstructions with a very low dose of radiation. The objective of this study was to evaluate the accuracy and reproducibility a novel 3D technique for THA preoperative planning based on biplanar low-dose radiographs.

Purpose of the study: Nonunion, which is a biological failure, requires revision, usually an aggressive operation. Haematopoietic bone marrow contains colony forming unit fibroblasts (CFU-F) which could favour bone healing. The purpose of this work was to determine whether a minimally invasive procedure, injection of CFU-F into the nonunion space, could favour bone healing without further procedure.

Material and methods: Our series included 43 patients: 36 male and 7 female, mean age 39.9 years. Forty-seven atrophic and aseptic nonunions of long bones were treated with percutaneous injection of concentrated autologous bone marrow: 27 tibias, 17 femurs, 3 humeri. Bone marrow was harvested from the posterior iliac crests (346 ml) then centrifuged to keep the leuko-platelet fraction (78 ml). This concentrate was injection into the nonunion space under radioscopic guidance. Efficacy was assessed on the basis of clinical criteria (complete pain-free weight-bearing, absence of contention, absence of mobility) and on radiographic criteria (healing of 3/4 corticals).

Results: Thirty nonunions healed: 19 tibias (70%, 11 femurs (65%) and 0 humerus. Mean time to healing was 5.9 months (2.4–15.6). Factors of poor prognosis were: smoking, alcohol, diabetes, corticosteroids, radiotherapy, history of sepsis (p=0.01). Early grafting increased the chances of success (p=0.04). Age, initial skin opening, type of fixation did not have a significant impact on healing. The number of CFU-F had an effect on the rate of healing.

Discussion: This technique is effective for the treatment of nonunion of the lower limb, allowing bone healing in two thirds of the cases with a minimally aggressive procedure. The method is easy to perform but requires a rigorous technique for the different phases of puncture, concentration and reinjection. Nonunions unresponsive to conventional methods, and thus corresponding to multifactorial problems, probably constitute the limitation of this method. Cell expansion or differentiation techniques could be helpful in improving the success rate but at the present time the osteogenic potential of these cells remains to be elucidated as a function of their stage of maturation.

Conclusion: Percutaneous grafts of concentrated autologous bone marrow can be a useful contribution to the therapeutic armamentarium for nonunion. Morbidity is low and the method does not compromise future options. It can be proposed as a first-intention solution for the treatment of long bone nonunion.

Purpose of the study: Monolayer cultures of chondrocytes multiply and rapidly lose their chondrocyte phenotype, limiting their potential for tissue engineering. Mesenchymatous stem cells can preserve their phenotypic characteristics after several monolayer passages, offering a promising alternative for cartilage repair. The purpose of this work was to study the influence of transforming growth factor beta-1 (TGF-beta1) and bone morphogenic protein-2 (BMP2) and/or culture supplements (hyaluronic acid) on matrix synthesis and chondrocyte differentiation of human mesenchymatous stem cells (MSC) cultured on collagen sponges.

Material and methods: MSC were isolated from bone marrow harvested during hip arthroplsty. At the third passage in monolayer culture, the MSC were reseeded on collagen sponges and cultured in vitro for 28 days under seven differ conditions: insulin transferrin selenium (ITS), foetal calf serum (FCS), ITS+TGFbeta1, ITS+ hyaluronate, ITS+TGFbeta1+hyaluronate, ITS+TGFbeta1+BMP2, ITS +TGFbeta1+BMP2+hyaluronate. The phenotypic evolution was followed using the expression of different genes of interest with PCRq (collagen2, collagen1, collagen3, collagen10, agrecanne, versicanne, COMP, Sox9). Synthesis of matrix material was assessed histologically and immunohistochemically.

Results: Used alone, hyaluronic acid did not trigger chondrocyte differentiation of MSC. For the additives FCS, ITS, or hyaluronate, the synthesis of matrix material in the sponge was weak and poor in major constituents of cartilage. Conversely, the other conditions in presence of TGFbeta1±BMP2 induced important expression of collagen2, agrecanne and COMP as well as increased matrix synthesis with a strong content in proteoglycans and collagen.

Discussion: The usefulness of MSC is growing due to their pluripotent characteristics. The conditions leading to their differentiation into the chondrocyte phenotype remains a subject of discussion. Our results show the particular importance of TGFbeta1 in the process of differentiation.

Conclusion: Chondrogenic differentiation of MSC cultured in collagen sponges as well as the synthesis of the cartilaginous matrix requires the presence of TGFbeta1 in the culture medium and to a lesser extent BMP2. These results suggest the perspective of using MSC for guided cell therapy targeting cartilage.

Purpose of the study: The right position of total hip arthroplasty (THA) implants affects short-, mid- and long-term outcome and complications. Navigation can improve implant position relative to a reference plane, in particular during mini-invasive implantation. The purpose of this work was to compare the position of the prosthetic cup in two series, one implanted with a navigation system and one with the conventional technique.

Material and method: The same surgeon performed the operations in each group of 42 patients (matched for age, gender, BMI, side). In the historic non-navigated series, a press fit femoral implant was used (Excia). The cups were either press fit (Ovalock or Plasmacup), or cemented with polyethylene inserts. The Hardigne incision (15 cm) was used for the conventional implantations. In the prospective navigated series, a press fit femoral element (Excia) and a press fit cup (Plasmacup) were implanted. The Orhtopilot navigation system was used (reference plane: Lewinnek anterior pelvic plane). The adapted Hardinge incision (5 cm) was used for the mini-invasive implantations. Inclination was measured on the weight-bearing pelvis relative to the radiological U line; the Pradhan method was used for anteversion. The objective was to achieve 45° horizontal inclination and 15° anteversion.

Results: In the non-navigated series, the inclination was 53±8 and in the navigated series 44±5.6. On average, inclination decreased significantly (8). Anteversion in the non-navigated series was 7±4, and 12±5.3 in the navigated series. On average, anteversion increased by 6 (statistically significant). The number of cups in the Lewinnek safety zone was 21 of 42 (50%) in the non-navigated series and 38 of 42 (90%) in the navigated series (statistically significant). The increase in leg length was 6.2 mm in the non-navigated group and 4.4 in the navigated group.

Discussion: Positioning did not take into account the preoperative analysis of the hip, but could adapt to scanner or EOS data. Navigation should also integrate offset, femoral position, and leg length.

Conclusion: Navigation of the prosthetic cup improves precision positioning in relation to the reference objectives, in particular, for mini-invasive surgery.

Purpose of the study: Basimetatarsal osteotomy to correct hallux valgus deformity by subtraction of a lateral wedge does not take into account the distal angle of the first metatarsal (DMMA). The purpose of this study was to demonstrate that the preoperative DMMA has an effect on the correction of the metatarsophalangeal angle and the duration of the result.

Material and methods: This retrospective study included 76 patients, mean age 58 years (84 feet). The M1P1 angle of the first ray was 34 on average, the DMMA 10 (two-thirds of the patients had a DMAA > 10. The same operator used the same technique for all procedures: lateral wedge osteotomy of the base of the first metatarsal with metatarsophalangeal release. Basal osteotomy of the first phalanx was performed for severe deformity. Radiographic measures were made on the dorsoplantar anteroposterior images in the weight-bearing condition preoperatively, at four weeks and at last follow-up. All images were read by an independent observer. Mean follow-up was 11 months. The DMMA measurement was the angle between the distal joint surface of the first metatarsal and the alignment of its diaphysis.

Results: Mean postoperative correction of the hallux valgus was 25 with a mean M1P1 of 9. This result remained stable without loss of correction at last follow-up. The mean postoperative DMAA was 10 and remained unchanged.

Discussion: Determination of the DMAA can be difficult because of preoperative pronation of the forefoot, compromising the reliability of the measurement. The literature also reports intraobserver and interobserver variability of this angle. Mean follow-up was not greater than one year, but the loss of correction generally occurred during the first six postoperative months. The operative technique enabled sufficient and stable correction over time. A pathological value for DMAA, even if large and uncorrected, does not prevent a good correction of the M1P1 angle and to maintain that correction. The clinical result is also the same irrespective of the preoperative DMMA.

Conclusion: Wedge osteotomy of the base of the first metatarsal is a reliable procedure for the treatment of hallux valgus. The value of the DMAA has no effect, in our experience, on the quality of the correction, or on the duration of the result.

Navigation technology is a new tool which can help surgeons to a more accurate hip component implantation and a better reproducibility of the procedure.

The purpose of this study was to compare conventional and navigated technique and a new developed straight hip stem for uncemented primary total hip replacement.

The results of two consecutive implantation series of 42 patients (non navigated) and 42 patients (navigated) were analysed for implant positioning and short term complications. Non navigated components were implanted through conventional incision (15 cm), navigated component by minimal invasive surgery (5 cm). All surgeries were performed through Hardinge approach and by a single senior surgeon.

Radiographic analysis of cup position showed a significant improvement with reduced radiological inclination (53° non navigated/44° navigated, p< 0.001) and higher anteversion (7° non navigated/12° navigated, p< 0.001). The mean postoperative limb length difference was 6.2 mm (SD 9.0, non navigated) and 4.4 mm (SD 6.4, navigated). Intraoperative and early postoperative complications were not different. No dislocation occurred in both groups. There was one intraoperative trochanter fracture which was not revised (non navigated) and one revision because of a periprosthetic fracture caused by fall down during rehabilitation (navigated).

We conclude that acetabular implant positioning can be significantly improved by the use of navigated surgery technique even in minimal invasive surgery condition. The data for postoperative limb length difference was still similar but within the expected range in both groups. Navigation technology seems essential for minimal invasive surgical procedure yielding help and security to the surgeon. The effect of improved cup positioning on mid and long term results for both groups have to be further investigated.

Objectives. To evaluate the contribution of leptin, an adipose-derived hormone, to the pathophysiology of osteoarthritis (OA), by determining leptin in both synovial fluid and cartilage specimens from human joints, and by investigating the effect on cartilage of intra-articular injection of leptin in rat.

Methods. Leptin levels were measured by enzyme-linked immunosorbent assay (ELISA) in synovial fluids sampled from OA patients undergoing either knee replacement surgery or knee arthroscopy. Besides, histological sections of articular cartilage and osteophytes obtained during surgery for total knee replacement, were graded using the Mankin score, and were immunostained using antibodies to leptin, TGF_ and IGF-1. For experimental studies, various doses of leptin (10, 30, 100 and 300μg) were injected into the rat knee joint. Tibial plateaus were collected and further processed for proteoglycan synthesis by radiolabeled sulfate incorporation, and for expression of leptin, its receptor (Ob-Rb), and growth factors by RT-PCR and immunohistochemistry.

Results. Leptin was found in synovial fluids from human OA-affected joints, and concentrations were correlated to Body Mass Index. A marked expression of the protein was seen in OA cartilage and in osteophytes, while few chondrocytes produced leptin in normal cartilage. Furthermore, the pattern and level of leptin expression were related to the grade of cartilage destruction, and paralleled those of growth factors (IGF-1 and TGFb-1). Animal studies showed that leptin strongly stimulated anabolic functions of chondrocytes, and induced the synthesis of IGF-1 and TGFb-1 in cartilage at both mRNA and protein levels.

Conclusion. These findings provide a new peripheral function to leptin as a key regulator of chondrocytes metabolism, and indicate that leptin may play an important role in the pathophysiology of OA.

Introduction: A good cup positioning requires reliable anatomical landmarks expecially for navigation. The anterior pelvic plane (APP) seems to be a good reference for navigation because it is in relation with pelvic tilt which do affect the position of the cotyle and consequently the position of the cup. The value of this plane is not well known according to gender, age, weight… The aim of the study is to assess radiologically the APP in standing and supine position before and after total hip arthroplasty.

MATERIALS AND Methods: 92 Patients (32 males, 60 females, mean age 65 years) underwent strict lateral X-rays in standing and supine standardized position. Uninterpretable or unsatisfying X rays were withdrawn. 45 patients underwent a standing X-ray, 24 a supine X-ray, 21 a supine and standing X ray. Statistical analysis used a Student t-test.

Results: Non matched values showed a retroversion of the pelvis of 6.4° (+/− 6.9) in supine position, 0.3° (+/− 7.4) in standing position. Matches values showed an retroversion of the pelvic of 6.9° (+/− 5.3) in supine position, 0.3° (+/− 5.03) in standing position (significant difference). Extreme values varied from −15° to + 18° (3 patients showed no variation, 2 patients a retroversion from supine to standing position). There was no statistical difference between male and female but a statistical differences in females.

Discussion: The APP is easily assessable by X rays in standing as in supine position. Bony landmarks of the plane are also assessable by navigation tools and to can be a good plane as reference. Several authors showed the repercussion of the pelvic tilt on the cotyle position. The difference between standing and supine position is about 6°. But for some patients the difference is may be of 20°and that could explain some impigment and instability. A cup well positioned in supine position may be not so good in standing position because of the pelvic tilt.

Conclusion: The value of the APP is important to know before THA and seems to be a good plane as reference for navigation.

Purpose of the study: Correct positioning of the prosthetic cup requires reliable anatomic landmarks, particularly for navigation systems. Referring uniquely to the three dimensions fails to recognize interindividual differences in pelvic position. The anterior plane of the pelvis is a good indicator of the pelvic position which can be determined from radiographic measurements. Standard values are poorly known (age, gender, weight). The purpose of this study was to measure the APP radiographically in the upright and reclining positions before and after total hip arthroplasty and to correlate the observed values with those obtained with navigation and ultrasound.

Material and methods: Strictly standardized x-rays of the pelvis in the upright and standing position were obtained in 110 patients (40 men, 70 women, mean age 65 years). Films which did not meet strict standard criteria were removed from the analysis which thus included upright views in 57 patients, reclining view in 36, and upright and reclining views in 28. Navigation measurements were made in 20 patients and ultrasound measurements in 10.

Results: Before arthroplasty, anteversion was 6.42±6.9° in the reclining position, 0.29±7.39° in the upright position (significant difference). After arthroplasty, anteversion was 6.9±5.3° in the upright position and 0.28±5.03 in the reclining position (significant difference). The values ranged from −15° to +18° (three patients without change, four with anteversion). There was no significant difference by gender. There was no clear correlation between the navigation values and those measured on the standard x-rays. The navigation and ultrasound values appeared to be correlated.

Discussion: The anterior pelvic plane can be easily measured on standard x-rays (upright and reclining position). Its landmarks can be easily accessed by navigation enabling the constitution of a reference plane. Several authors have demonstrated the influence of pelvic tilt on the position of the prosthetic cup. Posterior tile produces acetabular anteversion and inversely. The difference between the reclining and upright position is to the order of 6°. There are however variants up to 20° observed in certain patients and which might explain malpositions or instabilities. A cup with correct anteversion in the reclining position may be malpositioned on the upright film because of pelvic tilt.

Conclusion: Pelvic tilt should be taken into consideration when positioning the cup. The anterior pelvic plane can be correctly measured on standard x-rays and used to evaluate this tilt then serve as a reference for navigation. It should be proposed in all patients to search for extreme values.

Purpose of the study: Percutaneous compression plating (PCCP) is a new method for minimally invasive fixation of intratrochanteric fractures. Fixation is achieved with two neck screws and a 3-hole plate. This prospective study of a non-randomized series was designed to compare results in a monocentric cohort of patients treated by PCCP or dynamic hip screw (DHS).

Material and methods: From September 2003 to December 2004, all patients presenting an A1 (75.8%) or A1 (24.2%) (AO classification) intratrochanteric fracture were treated with PCCP (n=37) or DHS (n=20). Female gender predominated (86.5%) in this elderly population, mean age 83.2 years. The following variables were studied: operative time, radiation time, blood loss, hemoglobin level, blood transfusion, bone healing, complications, quality of the reduction.

Results: Mean follow-up was 8.3 months. The two groups were similar regarding bone healing, functional outcome and mortality. Intraoperative blood loss was less with PCCP (63 ml) than with DHS (120 ml). Mean fall in hemoglobin level was 2 after PCCP and 3 after DHS. The transfusion rate was 28% for PCCP and 40% for DHS. Mean operative time was 50 for PCCP and 30 minutes for DHS. Men radiation exposure was 4 minutes for PCCP and 1 minute for DHS. The positions of the screw (DHS) and the two PCCP screws were considered good for 68% of the PCCP and 75% of the DHS, acceptable for 29% PCCP and 20% DHS, and poor for 3% PCCP and 5% DHS. Complications were similar (one disassembly in each group).

Discussion: Although this was a preliminary study, PCCP was found to provide an attractive alternative for the treatment of intratrochanteric fractures. Results are similar to those obtained with the DHS but with a less aggressive method (limited approach, less blood loss). A learning curve (at least 10 implantations) appears indispensable to achieve maximum skill. The main drawback is the duration of the radiation. This implant would not be acceptable for subtrochanteric fractures which would require another type of implant.

Purpose of the study: Static metatarsalgia is a common complaint in podology surgery. Most cases are related to the great toe, but in certain cases, isolated metatarsal disharmony, without hallux vlgus, can be observed. We report 45 such cases.

Material and methods: This was a retrospective analysis of isolated metatarsal disharmony observed in patients who underwent metatarsal surgery between 1986 and 2003. There were 36 women and 9 men, mean age 49 years. Three subgroups were distinguished: posttraumatic disharmony, isolated disharmony of the second ray, iatrogenic disharmony. Conditions related to rheumatoid disease, aseptic osteonecrosis of the metatarsal heads, and rear foot disorders were excluded. Surgical treatment was osteotomy of the base of the metatarsal for 24 patients, and Weil’s osteotomy for 21. Clinical and radiological assessment used the Kitaoko and Maestro criteria.

Results: Mean preoperative score was 38 (range 21–58). Mean gain one year after osteotomy was 35 points. The score was 76 after osteotomy of the metatarsal base and 79 after Weil osteotomy. The less favorable results were observed in the group of posttraumatic metatarsalgias. Outcom was less satisfactory in the male population where residual metatarsalgia was noted in 75%. Reflex dystrophy occurred in 15% of the patients who had multiple osteotomies. Radiographically, The SM4 line was centered with progressive geometry in 50%.

Discussion: While the short-term results obtained with these two surgical methods were similar, osteotomy of the metatarsal base offers better long-term outcome. The osteotomy improved the functional score, even without improvement of radiological criteria. Initial treatment of metatarsal fractures should attempt to restore correct alignment of the metatarsals because of the poor results obtained with corrective osteotomy for posttraumatic misalignment.

Conclusion: When metatarsal disharmony is symptomatic, we propose osteotomy of the base of the metatarsals for the median rays in order to avoid transfer metatarsalgia. Complementary osteotomy of the fifth metatarsal is not always necessary.

Purpose of the study: This study was designed to assess the role of leptin in the development of osteoarthritis (OA) by searching for its presence in the synovial fluid (SF), tissues, and cartilage of osteoarthritic joints in humans and by observing the effect of intra-articular injections of leptin in the rat.

Material and methods: The leptin level in SF was measured (ELISA) in twenty patients (ten female, ten male, mean age 70 years). Presence of leptin, TGF beta and IGF1 in cartilage (and osteophytes) was detected by immunohistochemistry after histological evaluation (Mankin). In the rat, leptin was injected into the knee joint at the dose of 30 and 100 μg. After the immunohistological study, proteoglycan synthesis was assessed (S35 integration) as was the expression of leptin, TGF beta1 and IGF1 using RT-PCR.

Results: This study demonstrated for the first time the presence of leptin in synovial fluid (0.6–17.4 and 5.3–28.4 μg in male and female specimens respectively). There was a significant correlation with body mass index. Leptin was over expressed in chondrocytes of osteoarthritic cartilage and was correlated with the histological score (leptin not detecable in normal cartilage). IGF1 and TGF beta1 were expressed in osteoarthritic chondrocytes. The topographic distribution and the intensity of labeling varied with the histological score. There was a strong expression of TGF beta 1 only in osteophytes. In the rat, leptin stimulated anabolic functions of the chondrocyte: maximal effect at 30 μg (medial tibial plateau) and 100 μg (lateral tibial plateau). Leptin over expressed transcripts IGF 1 and TGF beta 1. This effect was confirmed at the protein level.

Discussion: Leptin is an adipocytokin which regulates food intake and energy expenditure at the hypothalamic level. A mechanical mechanism is the primary explanation of osteoarthritis in weight-bearing joints in obese patients. But leptin is also present specifically in non-weight-bearing joints in obese subjects. A biological factor is thus incriminated which might be leptin produced by adipose tissue. Leptin is overexpresssed in the cartilage of the osteoarthritic knee. This is in favor of a role for leptin in the pathogenesis of OA via synthesis of TGF beta 1 and IGF 1. This effect of leptin could explain the relationship between body mass index and the risk factor for osteoarthritis.

Purpose: Two questions arise after metatarsophalangeal hallux arthrodesis: what are the 3D changes observed when walking on flat ground ? and is the arthrodesis compensation essentially at the talocrural or interphalangeal level ?

Material and methods: Our series included twelve patients (ten women and two men) mean age 60.7 years. Nine patients had unilateral arthrodesis and three bilateral arthrodesis. The optoelectronic exploration was conducted barefoot. The patients wore underclothes and 27 reflectors. A miniaturised reflector was placed on the distal end of each hallux. Three valid recordings were made.

Results: General gait parameters and kinematic and kinetic values were unchanged (excepting nonsignificant maximal ankle dorsiflexion). On the arthrodesis side we observed: significant decline in propulsion force in the anteroposterior and vertical planes; significantly later heel lift-off; systematic anterior displacement of the ground reaction force of the metatarsophalangeal joint (not seen on the healthy side).

Discussion: We propose a coherent explanation of these observations. The kinetics of balance movement under the head of the first metatarsal head is changed. When the foot is flat on the ground, as the ankle balance movement occurs, the weight of the body is transferred earlier and massively to the forefoot. While in the healthy foot this occurs under the metatarsophalangeal joint of the great toe, in arthrodesis patients body weight is transferred under the interphalangeal joint of the great toe. The balance movement of the interphalangeal joint of the great toe occurs when the ankle balance movement is terminated. The centre of the balance movement is more distal and heel lift-off tends to occur later. During the propulsion phase, the greater lever arm limits the propulsion force, explaining the lesser peak force observed on the arthrodesis side. Use of reflectors on the distal end of the hallux demonstrated that the arthrodesis compensation occurs essentially at the interphalangeal level, exposing this joint to greater risk of degeneration.

Purpose: The purpose of this retrospective analysis over a seven-year period was to evaluate outcome of centromedullary nailing without reaming using the UTN for leg fractures.

Material and methods: A nailing procedure was performed in 106 patients (71 men and 35 women), mean age 38.2 years (16–76); 31.1% had multiple trauma injuries. Fractures were closed (77.4%) or open (22.6%): Gustilo I (n=19), Gustilo II (n=5). The fracture involved the shaft (77.4%), the lower quarter of the tibia (12.2%) or was bifocal (10.3%). There were five cases of vessel injury at diagnosis and two cases of neurological injury. A static assembly was used in all cases. Weight-bearing was resumed after a 6-week period of rest before unlocking.

Results: Mean follow-up was 13.1 months. Bone healing was achieved in 85.8% of the cases in 17 weeks on average. There were five cases of deformed callus (5–10° valgus or varus which did not require surgical revision). Late healing was noted in 7.5% and true nonunion in 6.6% which required either fibular osteotomy to achieve healing (n=6) or insertion of a new nail (Grosse and Kempf, n=1) after reaming. There were ten cases of locking screw fracture. Functional outcome was considered good or very good in 89.6% of patients. Twelve patients presented limited dorsal flexion of the foot and nine had pain at the upper end of the nail.

Discussion: Not reaming offers a clear advantage over reaming, particularly for preservation of endosteal vessels. Unreamed nailing is an interesting alternative to external fixation for open fractures. Despite the absence of infection, we do not have any argument in favour of a solid nail to decrease the risk of infection. The time to healing and the rate of true nonunion were not sufficient to clearly favour this method over reamed nailing.

Purpose: Damaged cartilage has very limited potential for self-repair. Tissue bioengineering offers an interesting alternative for repair of cartilage injury caused by joint trauma or osteochondritis dessicans. The purpose of this work was to use primary chondrocytes cultivated in vitro on collagen gel to produce a neocartilage which can be reimplanted.

Material and methods: Chondrocytes were extracted by enzymatic digestion from calf feet harvested from animals aged less than six months. Two million cells were seeded on collagen gels in multiple-well plates and covered with culture medium (1 ml). Type I collagen was acquired from ground calf skin used at a concentration of 1.25 mg/ml. The culture medium was a v/v mixture of RPMI 1640 and NCTC 109. This mixture was supplemented with 10% foetal calf serum, 100 U/ml penicillin, and 250 ng/ml amphotericin B. Cell proliferation was assess fluorometrically and synthesis of glycosaminoglycans (sGAG) by colorimetric assay. Histological study (safranine O) and immunohistochemistry tests (type I and II collagen) were performed to monitor synthesis of matrix components. Expression of genes coding for certain matrix proteins (collagen Ia 2 and 1, II, X, agrecan and MMP13) was studied using RT-PCR.

Results: The chondrocyte phenotype was preserved. Type II collagen as well as agrecan was expressed and expression of type I collagen did not increase during the culture. Progressive synthesis of sGAG was observed as was moderate cell proliferation. Cell distribution within the gel was apparently homogeneous. The chondrocytes retained their round shape throughout the study. Type II collagen deposits were visible on day 9 in peripheral cells in areas of high-cell density, then progressed with time.

Discussion: Our in vitro results show that three-dimensional cultures of chondrocytes using a collagen gel can produce construction of an extracellular matrix with preservation of chondrocyte phenotype during the culture period.

Conclusion: The collagen matrix offers an environment favouring the formation of a functional artificial cartilage by chondrocytes and opens promising perspectives for repairing damaged cartilage.

Purpose: Global metatarsus varus is a deformity of the forefoot characterised by medial deviation of all the metatarsals. The condition is often associated with hallux valgus and pes cavus (metatarsus varus sometimes being considered a clinical form of cavus), as well as metatarsal verticalisation predominating on the first ray.

Material and methods: Among twenty patients with metatasus varus in this study fifteen had undergone surgical correction. There were seven men and eight women, mean age 43 years. The deformity was bilateral in most patients (n=10, operated =9) but predominantly on one side in five of them. Metatarsalgia was the common complaint, considered to be global involving the middle rays (n=12 patients), or localised under the head of the first metatarsal (n=3). The calcanean tendon was short in all cases, aggravating subcapital anterior loading. Likewise, 12 of the 15 patients had hallux valgus; mean 45°. For one patient calcaneal osteotomy for valgisation was also performed.

Hallus valgus was corrected whenever present: in three cases an osteotomy was performed alone to raise the first metatarsal followed by a plantar prosthesis. A basal osteotomy of the three middle metatarsals was performed to achieve elevation, valgisation, and shortening: for three cases a Weil osteotomy was performed.

Results: Hallus valgus recurred in all patients who had had correction surgery, with recurrence of metatarsus varus. Basal metatarsal osteotomies healed but after a long period (about six months. Metatarsalgia was unchanged and was situated under the heads that were not perfectly aligned. The Weil osteotomies did not improve the evolution because, like basal osteotomies, they did not prevent recurrence of global metatarsus varus.

Discussion and conclusion: Metatarsus varus is a challenging deformity which responds very poorly to classical treatments of the forefoot. Osteotomy, irrespective of the type, does not prevent recurrence, both for the lateral rays but also for the hallux. Certain authors suggest that metatarsophalangeal arthrodesis could be useful to stabilise the hallux in the corrected position. We have no experience with this method but in light of our results after conservative treatment for joint motion, this would appear to be an interesting option. We have modified our indications in this direction, using a basi-metatarsal osteotomy instead of a Weil osteotomy for the small toes. Lengthening the extensor tendons may be needed.

Purpose: The purpose of this study was to analyse mechanical failures involving dismonted osteosynthesis materials implanted to fix pertrochanteric or subtrochanteric fractures with a dynamic hip screw (DHS), a Gamma nail, or a plate nail system (STACA).

Material and methods: Our cohort included 16 patients among a series of 350 patients who had been treated with 240 DHS, 80 Staca nailplates, and 30 Gamma nails between 1996 and 1999. We used the Ender classification for the x-ray analysis and the Cuny criteria which describe the most common causes of dismounted material.

Results: According to the defined criteria, 70/350 osteosynthesis assemblies (20%) were considered insufficient on the immediate postoperative x-rays and eventually dismounted in 16 patients. Two of these patients had major osteopaenia according to the Singh criteria and ten underwent revision because of poor clinical tolerance. These patients had six DHS (3 “swinging” cervical nails, two dismounted plates, and one screw protruding into the joint space). A protruding screw was the problem for the eight Staca nail-plates. Two Gamma nails had a “swinging” screw. These cases of dismounted material were predominantly observed in patients with Ender grade 5, 7 and 8 pertrochanteric fractures.

Discussion: Our analysis of these cases revealed several important factors: 1) the quality of the fracture reduction with restitution of the medial pillar of the per-trochanteric mass; 2) central anchoring in the femoral head essential for stable fixation; 3) superior stability of the DHS in grade 1 and 6 fractures due to the greater projected surface improving hold in cancellous bone. Inversely, for subtrochanteric fractures (grade 7 and 8), centromedulary shaft anchoring with a Gamma nail reduces mechanical stress in varus and thus the risk of “swinging” screws. Finally, the monoblock construct of the Staca nail-plate, which does not have the dynamic potential of the DHS and the Gamma nail, raises a risk of protrusion, particularly in case of “internal rotation” fractures with major metaphyseal comminution (grades 4 and 5). This latter type of fixation is however very effective for simple pertrochanteric fractures with minimal comminution (grades 1 to 3).

Conclusion: Material dismounting results from a series of factors related both to the material used and to the operative technique. We thus reserve the Staca nail-plate for grade 1 to 3 fractures in the Ender classification, the DHS for grades 1 to 6 and the nail-screw fixation for subtrochanteric (grade 7 and 8) fractures.

Purpose: Traffic accidents and high level falls are the principal causes of femur trauma. Fractures generally involve the shaft but the proximal or distal metaphyseal zones may also be involved. Skin opening, vascular injury or associated lesions in multiple injury patients are all reasons for emergency treatment with an external fixator.

Material and methods: We report a retrospective series of 23 cases who were treated with this technique between 1996 and 2000. There were 15 men and seven women, mean age 36 years (17–92) who were traffic accident victims in 17 cases. Fourteen had multiple injuries. The mean Index Severity Score was 28 points. Fractures were located in shaft in 16 cases, the proximal metaphysis in three, and in the supra and intercondylar zone in nine. The Chauchoix and Duparc classification was grade 2 in eight cases, and grade 3 in three cases. The fixation was installed with two or three pins in the lateral position; The knee was bridged in cases with an associated injury to the proximal tibia (floating knee) (two cases) or severe injury to the knee ligaments. Patients were reviewed clinically and radiographically. Bone healing was considered to be achieved when full weight bearing was possible without osteosynthesis contention.

Results: Twenty patients were reviewed. Mean follow-up was 20 months (7–42). Bone healing was achieved in 100% of the cases with a mean delay of 9.4 months (4–32). In three patients 13%) a complementary procedure (cancellous graft or bone marrow graft) was needed to achieve healing, the delay in these patients was 22 months compared with 7.5 months without secondary procedures. One patient developed a callus with a > 10° deviation of the AP view and five permanent flexion greater than 10°. Supra and intercondylar fractures healed at a mean 6.6 months in five cases with a deformed callus. The force moment related to excessive spread of the fixator pins (greater than 20 cm) was not a factor of poor final radiological outcome. Mean knee flexion was greater than 100° in only three cases. These amplitudes were not obtained until the fixator pins were removed. No releasing procedures were needed. Three mobilisations under general anaesthesia were needed.

Discussion: External fixation allows stable and dynamic osteosynthesis of femoral shaft fractures. It is indicated when centromedulary nailling is impossible or for patients with an excessively high risk of infection. The insertion of the pins must be rigorously control (perpendicular to the diaphysis, bicortical insertion, far enough apart). Supra and intracondylar fractures remain difficult to stabilise with external fixations and internal plate fixation may be discussed in grade 2 open fractures. These can give poor functional results despite rapid consolidation. An immediate corticocancellous graft may be indicated when metaphyseal comminution requires stabilisation, mainly on the medial aspect to avoid varisation.

Conclusion: External fixation is an interesting solution for safe fixation of open shaft fractures (grade 2 and 3) or fractures associated with vascular and nerve injury. It would be more indicated for shaft fractures than for supracondylar fractures which are difficult to stabilise, particularly in case of major comminution.

Purpose: Local factors such as poor vascular supply, open fracture, or infection can affect the potential for bone formation after fracture, arthrodesis or distraction. The fundamental principal for the treatment of late healing or nonunion is to supplement the local supply of the elements necessary for bone maturation. Centrifuged bone marrow is known to have a osteogenic effect in the treatment of femoral head necrosis or as a complement to conventional grafts. We examined the effect of bone marrow grafts used with conventional grafts.

Material and methods: This retrospective analysis included 14 cases where centrifuged bone marrow graft was used as complementary treatment for post-traumatic nonunion (10 cases), distraction callus (three cases) or late healing after arthrodesis (one case). Bone marrow (300 ml) was harvested from the posterior iliac crest then centrifuged to isolate the maximum number of nucleated cells and stem cells. The centrifugate (60–80 ml) was injected into the fracture site with a trocar during the same operative time. Cell concentrations (total nucleated cells, stem cells (CFU-GM), fibroblastic colonies) were noted. Patients were followed at regular visits. Bone healing was considered to be acquired when weight-bearing was possible without fixation or immobilisation.

Results: Definitive bone healing was achieved rapidly in two cases. Two patients required a conventional graft of a nonunion to achieve consolidation. For six patients, consolidation could not be achieved (three nonunions and three distraction calluses). Final outcome was good or very good in 57% of the cases. Mean delay to bone healing was 6.5 months. The infectious context had no effect on the method. The mean number of nucleated cells injected was 3.9•109 cells in successful cases and 2.8•109 cells in unsuccessful cases. These concentrations affected outcome.

Discussion: This technique for stimulating bone maturation by supplying bone generating cells is indicated for late healing or recent nonunion. It is less effective for distraction calluses or for very old nonunions. Morbidity and iatrogenic effects are minimal. A rigorous harvesting method is required since the result is highly dependent on the cell concentrations and the number of injected cells. Bone marrow injections after centrifugation should be greater than 85 ml and have a cell concentration around 45•106 cells/ml. The method is less successful for old injuries and in patients with arteritis.

Conclusion: Bone marrow grafts are indicated for the treatment of late healing or recent nonunion. Morbidity is low but a rigorous harvesting method is required. The method should be implemented shortly after the fracture without waiting for potential signs of nonunion.