Incidence and sequence of scoliosis and windswept hip deformity: which comes first in 4148 children with cerebral palsy? A longitudinal cohort study.

BACKGROUND: The aim was to analyse whether scoliosis or windswept hip deformity (WSH) occurs first for children with cerebral palsy (CP). METHODS: This longitudinal cohort study using data from 1994 - 2020 (26 years) involved 41,600 measurements of 4148 children (2419 [58.3%] boys) with CP born 1990 - 2018 and registered into the Swedish CP follow-up program. Children were followed from a mean age of 2.8 [SD 1.4] years, until they developed either scoliosis or WSH or were removed at surgery. RESULTS: WSH developed first in 16.6% of the children (mean age 8.1 [SD 5.0] years), and scoliosis in 8.1% (mean age 8.1 [SD 4.9] years). The incidence of WSH was higher than scoliosis across all levels I-V of the Gross Motor Function Classification System (GMFCS), both sexes, and for those with dyskinetic (20.0%) or spastic (17.0%) CP. The incidence of scoliosis was highest (19.8%) and developed earliest in children with GMFCS level V (mean age 5.5 [SD 3.5] years), and in children with dyskinetic (17.9%) CP (mean age 7.0 [SD 4.7] years). CONCLUSIONS: WSH presents earlier than scoliosis in most children with CP. Children with higher GMFCS level or dyskinetic CP are more likely to develop these deformities at a younger age.

Sequence of flexion contracture development in the lower limb: a longitudinal analysis of 1,071 children with cerebral palsy.

BACKGROUND: To prevent severe contractures and their impact on adjacent joints in children with cerebral palsy (CP), it is crucial to treat the reduced range of motion early and to understand the order by which contractures appear. The aim of this study was to determine how a hip-knee or ankle contracture are associated with the time to and sequence of contracture development in adjacent joints. METHODS: This was a longitudinal cohort study of 1,071 children (636 boys, 435 girls) with CP born 1990 to 2018 who were registered before 5 years of age in the Swedish surveillance program for CP and had a hip, knee or ankle flexion contracture of ≥ 10°. The results were based on 1,636 legs followed for an average of 4.6 years (range 0-17 years). The Cox proportional-hazards model adjusted for Gross Motor Function Classification System (GMFCS) levels I-V was used to compare the percentage of legs with and without more than one contracture. RESULTS: A second contracture developed in 44% of the legs. The frequency of multiple contractures increased with higher GMFCS level. Children with a primary hip or foot contracture were more likely to develop a second knee contracture. Children with a primary knee contracture developed either a hip or ankle contracture as a second contracture. CONCLUSIONS: Multiple contractures were associated with higher GMFCS level. Lower limb contractures appeared in specific patterns where the location of the primary contracture and GMFCS level were associated with contracture development in adjacent joints.

Postoperative growth rate affects time to growth arrest after percutaneous physiodesis: A radiostereometric analysis.

Purpose: The aim of this study was to determine the time at which physeal arrest is achieved after percutaneous physiodesis, and whether immediate postoperative growth rate affects the time to reach physeal arrest. Methods: Radiostereometric analysis, with implantation of tantalum balls as radiographic markers on each side of the physes, was used to measure residual longitudinal growth in 21 children (10 boys and 11 girls) after percutaneous physiodesis for leg length discrepancy or extreme tall stature. In total, 25 femoral and 20 tibial physes were operated on. Median age at surgery was 13.9 years (range = 11.4-16.1). Radiostereometric analysis was performed postoperatively and after 3, 6, 9, 12, 26, and 52 weeks. Longitudinal growth rate <50 µm per week was defined as physeal arrest. Descriptive statistics were used for evaluation. Results: Physeal arrest was obtained in 19 of the 21 children (40 physes) within 12 weeks postoperatively. One child was reoperated on in three out of four physes because of continued growth, and in one child, delayed physeal arrest was present at 26 weeks postoperatively. Time to physeal arrest was longer in physes with a higher immediate postoperative growth rate. Conclusion: Postoperative follow-up with radiostereometric analysis at 12 and 15 weeks can determine whether physeal arrest has been achieved. The immediate postoperative growth rate after physiodesis seems to affect the time to physeal arrest. This implies that the risk for complications is greater for children during an accelerated growth period, for example, in boys, younger children and in distal femoral physes. Level of evidence: level III.

Knee and foot contracture occur earliest in children with cerebral palsy: a longitudinal analysis of 2,693 children.

Background and purpose - Joint contracture is a common problem among children with cerebral palsy (CP). To prevent severe contracture and its effects on adjacent joints, it is crucial to identify children with a reduced range of motion (ROM) early. We examined whether significant hip, knee, or foot contracture occurs earliest in children with CP.Patients and methods - This was a longitudinal study involving 27,230 measurements obtained for 2,693 children (59% boys, 41% girls) with CP born 1990 to 2018 and registered before 5 years of age in the Swedish surveillance program for CP. The analysis was based on 4,751 legs followed up for an average of 5.0 years. Separate Kaplan-Meier (KM) curves were drawn for each ROM to illustrate the proportions of contracture-free legs at a given time during the follow-up. Using a clustered bootstrap method and considering the child as the unit of clustering, 95% pointwise confidence intervals were generated for equally spaced time points every 2.5 years for each KM curve.Results - Contracture developed in 34% of all legs, and the median time to the first contracture was 10 years from the first examination. Contracture was most common in children with a higher Gross Motor Function Classification System (GMFCS) level. The first contracture was a flexion contracture preventing dorsiflexion in children with GMFCS level I or II and preventing knee extension in children with GMFCS level III to V.Interpretation - Early interventions to prevent knee and foot contractures in children with CP should be considered.

Windswept hip deformity in children with cerebral palsy: a population-based prospective follow-up.

PURPOSE: To analyze the development of windswept hip deformity (WS) in a total population of children with cerebral palsy (CP) up to 20 years of age, the association between WS and hip dislocation, and femoral varus osteotomy and scoliosis, and the impact of a hip surveillance program on the subsequent incidence of WS. METHODS: This is a prospective study on children with CP in southern Sweden included in the Swedish follow-up programme and registry for CP (CPUP). All children born between 1990 and 1995 with CP were included; those born between 1990 and 1991 did not partake in the hip surveillance program until they were older (3-5 years of age) and served as a historic control group. Children born between 1992 and 1995 were included in the hip surveillance program from about 2 years of age and constituted the study group. RESULTS: In the control group, 12 of 68 children (18 %) developed WS. In the study group of 139 children, 13 (9 %) developed WS (p = 0.071). Of all 25 children with WS, 21 also developed scoliosis and 5 developed a hip dislocation. The number of children with WS starting in the lower extremities was significantly lower in the study group (p = 0.028). No difference between the two groups was seen regarding WS that started in combination with scoliosis. CONCLUSION: With early inclusion in a hip surveillance program and early treatment of contractures, it appears possible to reduce the frequency of WS starting in the lower extremities.

Eight plate should not be used for treating leg length discrepancy.

PURPOSE: The Blount staple has been widely used for treating angular deformities and leg length discrepancy (LLD) of the lower limbs. In recent years, the Eight plate has replaced the Blount staple for treating angular deformities in many centres. Although not described in the literature, it has been proposed that the Eight plate could also be used for treating LLD. We studied the effectiveness of the device for this indication. METHODS: Ten patients with LLD were included prospectively. Medial and lateral plates were inserted for symmetrical growth reduction and the patients were followed by radiostereometric analysis (RSA) 0, 3, 6, 9, 12, 24, 52 and 80 weeks postoperatively. The error of measurement with this technique is less than 0.05 mm. RESULTS: Case 1 continued to grow an average of 0.08 mm per week (6.7 mm longitudinal growth in 1.5 years) postoperatively, while Case 2 showed 0.07 mm longitudinal growth per week (5.6 mm longitudinal growth in 1.5 years). The longitudinal growth, in both children, showed only slight growth retardation throughout the follow-up period. The study was interrupted and no further subjects were included after the results indicated that only minor growth reduction had occurred in both patients. CONCLUSION: The Eight plate does not significantly reduce growth when applied both medially and laterally in a symmetrical way at the proximal tibial physis.

Scoliosis in a total population of children with cerebral palsy.

STUDY DESIGN: Epidemiological total population study based on a prospective follow-up cerebral palsy (CP) registry. OBJECTIVE: To describe the prevalence of scoliosis in a total population of children with CP, to analyze the relation between scoliosis, gross motor function, and CP subtype, and to describe the age at diagnosis of scoliosis. SUMMARY OF BACKGROUND DATA: Children with CP have an increased risk of developing scoliosis. The reported incidence varies, partly due to different definitions and study groups. Knowledge of the prevalence and characteristics of scoliosis in an unselected group of children with different CP types and levels of function is important for health care planning and for analyzing the risk in an individual child. METHODS: A total population of 666 children with CP, aged 4 to 18 years on January 1, 2008, followed with annual examinations in a health care program was analyzed. Gross Motor Function Classification System (GMFCS) level, CP subtype, age at clinical diagnosis of scoliosis, and the Cobb angle at the first radiographical examination were registered. RESULTS: Of the 666 children, 116 (17%) had mild and another 76 (11%) had moderate or severe scoliosis based on clinical examination. Radiographical examination showed a Cobb angle of more than 10° in 54 (8%) children and a Cobb angle of more than 20° in 45 (7%) children. The risk of developing scoliosis increased with GMFCS level and age. In most children, the scoliosis was diagnosed after 8 years of age. Children in GMFCS level IV or V had a 50% risk of having moderate or severe scoliosis by 18 years of age, whereas children in GMFCS level I or II had almost no risk. CONCLUSION: The incidence of scoliosis increased with GMFCS level and age. Observed variations related to CP subtype were confounded by the GMFCS, reflecting the different distribution of GMFCS levels in the subtypes. Follow-up programs for early detection of scoliosis should be based on the child's GMFCS level and age.

Outcomes of a standardized surgical and rehabilitation program in transtibial amputation for peripheral vascular disease: a prospective cohort study.

OBJECTIVE: To study the outcomes of a new surgical and rehabilitation program for initial unilateral transtibial amputation in patients with peripheral vascular disease. The program consists of sagittal incision, rigid dressing, compression therapy using silicone liner, and direct manufacturing prosthetic technique. DESIGN: A prospective cohort study with 1-yr follow-up. RESULTS: Of the 217 consecutive patients with peripheral vascular disease who underwent transtibial amputation (mean age, 77 yrs; 51% diabetic; 116 could walk before amputation), 119 (55%) were fitted with a prosthesis at a median time of 41 (range, 12-147) days after amputation. Of the prosthetic recipients, 76 (64%) obtained good function with the prosthesis within 6 mos. Within 1 yr, reamputation was performed on 8.2%, and contralateral amputation was performed on 5.5%. The 90-day mortality was 24% (53 patients). The total 1-yr mortality was 40% (86 patients): 17% among patients who received a prosthesis and 67% among those who did not receive a prosthesis or had undergone reamputation. CONCLUSIONS: Following this standardized surgical and rehabilitation program, prosthetic fitting was achieved in more than half of transtibial amputees, almost two-thirds of prosthetic recipients obtained good function, and the reamputation rate was low. Comparison with outcomes of alternative strategies is needed.

Development of lower limb range of motion from early childhood to adolescence in cerebral palsy: a population-based study.

BACKGROUND: The decreasing range of joint motion caused by insufficient muscle length is a common problem in children with cerebral palsy (CP), often worsening with age. In 1994 a CP register and health care programme for children with CP was initiated in southern Sweden. The aim of this study was to analyse the development of the passive range of motion (ROM) in the lower limbs during all the growth periods in relation to gross motor function and CP subtype in the total population of children with CP. METHODS: In total, 359 children with CP born during 1990-1999, living in the southernmost part of Sweden in the year during which they reached their third birthday and still living in the area in the year of their seventh birthday were analysed. The programme includes a continuous standardized follow-up with goniometric measurements of ROM in the lower limbs. The assessments are made by each child's local physiotherapist twice a year until 6 years of age, then once a year. In total, 5075 assessments from the CPUP database from 1994 to 1 January 2007 were analysed. RESULTS: The study showed a decreasing mean range of motion over the period 2-14 years of age in all joints or muscles measured. The development of ROM varied according to GMFCS level and CP subtype. CONCLUSION: We found a decreasing ROM in children with CP from 2-14 years of age. This information is important for both the treatment and follow-up planning of the individual child as well as for the planning of health care programmes for all children with CP.

Characteristics of children with hip displacement in cerebral palsy.

BACKGROUND: Hip dislocation in children with cerebral palsy (CP) is a common and severe problem. The dislocation can be avoided, by screening and preventive treatment of children with hips at risk. The aim of this study was to analyse the characteristics of children with CP who develop hip displacement, in order to optimise a hip surveillance programme. METHODS: In a total population of children with CP a standardised clinical and radiological follow-up of the hips was carried out as a part of a hip prevention programme. The present study is based on 212 children followed until 9-16 years of age. RESULTS: Of the 212 children, 38 (18%) developed displacement with Migration Percentage (MP) >40% and further 19 (9%) MP between 33 and 39%. Mean age at first registration of hip displacement was 4 years, but some hips showed MP > 40% already at two years of age. The passive range of hip motion at the time of first registration of hip displacement did not differ significantly from the findings in hips without displacement. The risk of hip displacement varied according to CP-subtype, from 0% in children with pure ataxia to 79% in children with spastic tetraplegia. The risk of displacement (MP > 40%) was directly related to the level of gross motor function, classified according to the gross motor function classification system, GMFCS, from 0% in children in GMFCS level I to 64% in GMFCS level V. CONCLUSION: Hip displacement in CP often occurs already at 2-3 years of age. Range of motion is a poor indicator of hips at risk. Thus early identification and early radiographic examination of children at risk is of great importance. The risk of hip displacement varies according to both CP-subtype and GMFCS. It is sometimes not possible to determine subtype before 4 years of age, and at present several definitions and classification systems are used. GMFCS is valid and reliable from 2 years of age, and it is internationally accepted. We recommend a hip surveillance programme for children with CP with radiographic examinations based on the child's age and GMFCS level.

Radiographic threshold values for hip screening in cerebral palsy.

PURPOSE: To analyse the consequences of using different radiographic measurements and different threshold values for hip screening in children with cerebral palsy (CP). METHODS: In a total sample of children with CP a standardised radiological follow-up of the hips was carried out as a part of a hip prevention programme. Acetabular index (AI) and migration percentage (MP) were measured on all radiographs. In this study, 1,067 radiographs of 272 children born 1992-1998 were analysed. RESULTS: Lateral displacement of the femoral head was common without acetabular dysplasia, and acetabular dysplasia occurred at a later stage than femoral head lateralisation. Hip dysplasia without lateral displacement of the femoral head was rare. In 16 of the 56 hips (29%) with AI >/= 27 degrees and in 23 of the 71 (32%) hips with MP >/= 33% the values decreased below the threshold value without operative treatment. In hips with AI >/= 30 degrees only 2 of 31 hips (6%) and in hips with MP >/= 40% only 5 of 44 hips (11%) decreased below the threshold values without operative treatment. CONCLUSIONS: Radiographic follow-up with only measurement of the MP seems sufficient in screening for dislocation in children with CP. MP >/= 33% is recommended as threshold for reaction or intensified observation. In children with MP >/= 40%, the lateral displacement increased over time in most hips, thus indicating the need for operative intervention. In children with MP 33-40%, treatment should be based on other clinical signs and the progression of MP over time.

Windswept hip deformity in children with cerebral palsy.

Windswept hip deformity describes an abduction and external rotation position of one hip with the opposite hip in adduction and internal rotation. Windswept hip deformity may occur in association with hip dislocation and scoliosis. We analysed the prevalence of this deformity in a total population of children with cerebral palsy, and the impact of hip prevention and early treatment of contractures on the prevalence and severity of windswept hip deformity. The frequency of windswept hip deformity was 12% in the control group and 7% in the study group, comprising children in the hip prevention programme. The children with this deformity in the study group had a lower frequency of scoliosis and none had hip dislocation. It thus seems that the hip prevention programme results in a decrease in the number of children with windswept hip deformity, and a decrease in the severity of the deformity.

6 Weeks with the von Rosen splint is sufficient for treatment of neonatal hip instability.

BACKGROUND: There is no concensus on the optimal treatment time for unstable hips in the newborn. We analyzed the efficiency of a treatment program that has been used for 10 years at our hospital, in which all unstable hips (subluxatable, Barlow-positive and Ortolani-positive) are treated with the von Rosen splint for 6 weeks. PATIENTS AND METHODS: Between 1988 and 1997, 32,171 children were born alive at the hospital. During this period 247 children had a clinically unstable hip diagnosed. 223 of the 247 children underwent a radiographic follow-up after 5-15 years. RESULTS: 1 patient with bilateral instability and treated with a splint for 6 weeks showed a dislocated left hip at the radiographic examination at 8 months, which is part of the screening program, and needed operative treatment. 1 patient did not follow the treatment program and showed a dislocated hip at the age of 3. Another 4 patients required more treatment than the 6 weeks with the splint. We found no dysplastic hips at the radiographic follow-up. There was no late dysplasia and there were no late dislocations in children born in Lund between 1988 and 1997 who were diagnosed at other Swedish centers that treat developmental dysplasia of the hip (DDH). INTERPRETATION: We conclude that the present screening and 6-week treatment in a von Rosen splint prevent almost all cases of late dysplasia and late dislocation of the hip.

Prevention of severe contractures might replace multilevel surgery in cerebral palsy: results of a population-based health care programme and new techniques to reduce spasticity.

During the 1990s three new techniques to reduce spasticity and dystonia in children with cerebral palsy (CP) were introduced in southern Sweden: selective dorsal rhizotomy, continuous intrathecal baclofen infusion and botulinum toxin treatment. In 1994 a CP register and a health care programme, aimed to prevent hip dislocation and severe contractures, were initiated in the area. The total population of children with CP born 1990-1991, 1992-1993 and 1994-1995 was evaluated and compared at 8 years of age. In non-ambulant children the passive range of motion in hip, knee and ankle improved significantly from the first to the later age groups. Ambulant children had similar range of motion in the three age groups, with almost no severe contractures. The proportion of children treated with orthopaedic surgery for contracture or skeletal torsion deformity decreased from 40 to 15% (P = 0.0019). One-fifth of the children with spastic diplegia had been treated with selective dorsal rhizotomy. One-third of the children born 1994-1995 had been treated with botulinum toxin before 8 years of age. With early treatment of spasticity, early non-operative treatment of contracture and prevention of hip dislocation, the need for orthopaedic surgery for contracture or torsion deformity is reduced, and the need for multilevel procedures seems to be eliminated.

The effect of bone quality on the stability of ankle arthrodesis. A finite element study.

BACKGROUND: Despite uniform operating techniques, lack of fusion still occurs after ankle arthrodesis. Differences in the biological healing potential may be a factor but the mechanical performance of the arthrodesis construct because of varying bone quality also may be important. Internal compression techniques are preferred because of higher union rates, shorter fusion times, and fewer complications. A three-screw configuration has been shown to be more stable than a two-screw configuration, but it is not obvious when it should be used. METHODS: Three-dimensional finite element models of intact and flat-cut ankle arthrodeses were built, using two and three screws in different configurations. Poor bone quality was simulated by decreasing Young's modulus of the bone. The constructs were loaded in torsion and dorsiflexion, and micromotions at the fusion site were measured. RESULTS: Bone quality had a marked effect on the stability at the arthrodesis site. Inserting two screws at 30 degrees relative to the longitudinal axis of the tibia in an intact arthrodesis seemed the best option, especially as bone quality worsened. The addition of a third screw increased the stability at the arthrodesis site. CONCLUSIONS: Overall, intact joint surfaces and three-screw fixation, with the lateral and medial screws inserted produced the most stable arthrodesis constructs when bone quality was poor. CLINICAL RELEVANCE. Ankle arthrodeses are technically demanding because of the shape and small size of the talus. Preoperative planning is an absolute necessity to determine placement and number of screws. This study shows that poor bone quality decreases the stability of the arthrodesis constructs, suggesting that an attempt should be made to create the most stable three-screw configuration. Finite element models can be used as an effective preoperative tool for planning screw number and placement.

Initial stability of ankle arthrodesis with three-screw fixation. A finite element analysis.

OBJECTIVE: Compare the initial stability at the fusion site of ankle arthrodesis fixed with two and three screws. DESIGN: Finite element models of ankle arthrodesis were developed from computed tomography images. Two-screw constructs were augmented with a third screw in different orientations and subjected to loads likely to affect the ankle postoperatively. BACKGROUND: More stable fixation seems to increase the chance of fusion, as it minimises the motion between the tibiotalar interfaces. METHODS: Non-linear elastic finite element analyses were performed in external torsion and dorsiflexion. The micromotions at the tibiotalar interface were computed to compare the two- and three-screw fixation in intact and flat-cut arthrodesis. RESULTS: Adding a third screw reduced the micromotions at the fusion site. Inserting the third screw anteriorly predicted lower peak micromotions than inserting the screw posteriorly, except for the intact arthrodesis tested in dorsiflexion. Three-screw intact arthrodesis predicted lower peak micromotions than flat-cut arthrodesis. CONCLUSIONS: Better stability was predicted for three-screw ankle arthrodesis. In flat-cut arthrodesis, a third screw inserted anteriorly performed better than a posterior screw. In intact arthrodesis, a posterior screw seemed a better option when flexion stability was the main concern. Even with three-screw fixation, the configuration of the first two-crossed screws may still be important to improve the stability at the fusion site.

Synovial fluid depletion: successful arthrodesis without operative cartilage removal.

Operative arthrodesis has been thought to require surgical removal of remaining joint cartilage, but we have found in rabbits that even a joint with intact cartilage can fuse if rigidly fixed. This may enable new percutaneous techniques for arthrodesis. Rigid adaptation of the joint surfaces deprives the cartilage of mechanical stimulation and depletes the cells of synovial fluid transport of oxygen and nutrition. To better understand the requirements for successful arthrodesis, we studied the histological consequences of the complete absence of mechanical stimulation alone or in combination with synovial depletion by placing a metal cap over part of the joint cartilage in rabbits. The cap was either closed or had an opening to permit synovial fluid to reach the cartilage. We also studied if penetration of the bone-cartilage junction by a drill hole would facilitate cartilage resorption. Synovial fluid depletion in combination with a drill hole through the bone-cartilage junction led to disappearance of all cartilage matrix after 7 weeks. Synovial fluid depletion with an intact bone-cartilage junction led to complete disappearance of the cartilage matrix in four of seven rabbits after 7 weeks. With a hole in the cap for synovial fluid, the cartilage matrix was still present to varying degrees after 7 weeks in all the rabbits. In conclusion, percutaneous arthrodesis by rigid adaptation may lead to cartilage disappearance due to synovial depletion rather than due to the absence of mechanical stimulation. A combination with perforation of the bone-cartilage junction appears to lead to reasonably quick removal of the cartilage matrix.

Finite element analysis of the initial stability of ankle arthrodesis with internal fixation: flat cut versus intact joint contours.

OBJECTIVE: Qualitative comparison of the initial stability provided by two joint preparation techniques and various screw configurations in ankle arthrodesis, using the finite element method.Design. A three-dimensional model of a healthy ankle was developed from computed tomography images. Two groups of models were built, one with the joint contours resected to produce flat surfaces, and the second with the joint contours preserved. In each case, a variety of screw orientations were examined. BACKGROUND: Despite the improved results of ankle arthrodesis, failure rates due to non-union are still reported. The initial stability of the arthrodesis construct seems important in the final outcome of the fusion. METHODS: Non-linear contact finite element analyses were performed in the arthrodesis constructs subjected to internal/external torsion and dorsiflexion. Micromotions at the bone-to-bone interface were calculated for frictionless and Coulomb friction contact, and compared for the two joint preparation techniques and screw configurations. RESULTS: Overall lower peak micromotions were predicted when preserving the joint contours both in torsion and dorsiflexion. For both preparation techniques, the lowest micromotions tended to occur with the screws inserted at 30 degrees with respect to the long axis of the tibia, crossing above the fusion site. Inclusion of friction in the models caused a general decrease on the magnitude of the micromotions as compared to the frictionless case, but did not affect the ranking of the models. CONCLUSIONS: The finite element method can be used as a qualitative tool to study the initial stability of ankle arthrodesis, overcoming the difficulties of measuring bone-to-bone interface micromotions experimentally. Better initial stability was predicted for ankle arthrodesis when the joint contours were preserved rather than resected. Crossing the screws above the fusion site at a steeper angle also tended to increase the stability at the fusion site. RELEVANCE: Finite element analyses can help during the pre-operative planning of ankle arthrodesis. When bone density is not compromised, preserving the joint contour and inserting the screws at less than 45 degrees to the long axis of the tibia, crossing over the arthrodesis site, may offer better initial stability.

Percutaneous arthrodesis.

It has been generally accepted that residual cartilage and subchondral bone has to be removed in order to get bony fusion in arthrodeses. In 1998 we reported successful fusion of 11 rheumatoid ankles, all treated with percutaneous fixation only. In at least one of these ankle joint there was cartilage left. This was confirmed by arthrotomy in order to remove an osteophyte, which hindered dorsiflexion. More than 25 rheumatoid patients with functional alignment in the ankle joint have subsequently been operated on with the percutaneous technique, and so far we have had only one failure. Patients with rheumatoid arthritis are known to sometimes fuse at least their subtalar joints spontaneously, and the destructive effect of the synovitis on the cartilage could contribute to fusion when using the percutaneous technique. In a rabbit study we therefore tested the hypothesis that even a normal joint can fuse merely by percutaneous fixation. The patella was fixated to the femur with lag screw technique without removal of cartilage, and in 5 of 6 arthrodeses with stable fixation bony fusion followed. Depletion of synovial fluid seemed to be the mechanism behind cartilage disappearance. The stability of the fixation achieved at arthrodesis surgery is an important factor in determining success or failure. Dowel arthrodesis without additional fixation proved to be deleterious. A good fit of the bone surfaces appears necessary. In the ankle joint, it would be technically demanding to retain the arch-shaped geometry of the joint after resection of the cartilage. Normally the joint surfaces are resected to produce flat osteotomy surfaces that are thus easier to fit together, encouraging healing to occur. On the other hand it is considered an advantage to preserve as much subchondral bone as possible, as the strong subchondral bone plate can contribute to the stability of the arthrodesis. Ankle arthrodesis can be successfully performed in patients with rheumatoid arthritis by percutaneous screw fixation without resection of the joint surfaces. This procedure has two advantages: first, it is less surgically traumatic, second, both the arch-shaped geometry and the subchondral bone are preserved, and thus both could contribute to the postoperative stability of the construct. Intuitively, preservation of the arch-shape should increase rotational stability. The results of our experimental sawbone study indicate that the arch shape and the subchondral bone should be preserved when ankle arthrodesis is performed. The importance of this is likely to increase in weak rheumatoid bone. In a finite element study the initial stability provided by two different methods of joint preparation and different screw configurations in ankle arthrodesis, was compared. Better initial stability is predicted for ankle arthrodesis when joint contours are preserved rather than resected. Overall, inserting the two screws at a 30-degree angle with respect to the long axis of the tibia and crossing them above the fusion site improved stability for both joint preparation techniques. The question rose as to whether patients with osteoarthritis could also be operated on solely by percutaneous fixation technique. The first metatarsophalangeal joint in patients with hallux rigidus was chosen as an appropriate joint to test the percutaneous technique. In this small series we have shown that it is possible to achieve bony fusion with a percutaneous technique in an osteoarthrotic joint in humans, but failed to say anything about the fusion rate.