Session 13: Eksperimental/non-clinical

Thomas Colding-Rasmussen^1,2, Nanett Kvist Nikolaisen³, Peter Horstmann⁴, Michael Mørk Petersen¹, Daniel John Hutchinson⁵, Michael Malkoch⁵, Stine Jacobsen ³, Christian Nai En Tierp-Wong¹
1. Department of Orthopaedic Surgery, Rigshospitalet, University Hospital. 2. Department of Orthopaedic Surgery, Hvidovre University Hospital. 3. Department of Veterinary Clinical Sciences, Faculty of Health and Medical Sciences, University of Copenhagen. 4. Department of Orthopaedic Surgery, Gentofte University Hospital 5. Department of Fibre and Polymer Technology, KTH Royal Institute of Technology, Stockholm, Sweden.

Background: Certain fractures, especially in osteoporotic bone, may benefit from patient specific implants to enhance fragment reposition and healing. A novel in situ customizable fracture fixation platform (Bonevolent™ AdhFix) has shown promising biomechanical properties in ex vivo ovine phalanx fracture models. Accordingly, AdhFix might be an adjuvant in the surgical management of complex fractures.
Aim: To evaluate stability and clinical applicability of AdhFix compared to conventional metal plates in a complex fracture model in vivo.
Materials and Methods: Seven skeletally mature Texel sheep underwent a midline osteotomy with a 4.5 mm circular unicortical defect on the lateral proximal phalanx of both front limbs. Fractures were treated with either AdhFix or a locking plate (DePuy Synthes 1.5) and bandaged with a custom-made wooden block under the medial phalanx and either a cast or elastic bandage. Free movement was allowed post-surgery. Bone healing and implant stability were assessed via weekly x- ray evaluations and post-euthanasia dual- energy x-ray absorptiometry (DXA) with measurement of bone mineral density (BMD). Descriptive statistics and Wilcoxon signed-rank test were applied to assess differences in BMD between the two groups using RStudio.
Results: All initial intraoperative x-rays confirmed satisfactory reduction and hardware placement. All AdhFix patches had failed at the 1-week mark; complete fracture (n=6) or screw dislodgement (n=2). All metal plates were intact at the 1-week mark, except for one case of screw loosening. Subsequently, a total of four screw failures were observed in the plate side, but without loss of osteotomy reduction. Callus formation was observed in all samples starting at 6 weeks. Mean BMD in the fracture gap was 0.45 g/cm2 and 0.60 g/cm2 for AdhFix and plates, respectively (p=0.078) and in the distal tuft 0.36 g/cm2 for AdhFix and 0.41 g/cm2 for plates (p=0.016).
Interpretation / Conclusion: AdhFix demonstrated inadequate biomechanical properties for this fracture model, suggesting limited clinical application in its current form to treat weight-bearing fractures. Additional in vivo studies are warranted to further characterize the clinical applicability of this osteosynthesis technique.

Anna Bertoli Borgognoni¹, Michael Melchior Bendtsen¹, Jørgen Baas¹, Jeppe Skinnerup Byskov², Thomas Baad-Hansen¹
1. Department of Orthopaedics, Aarhus University Hospital; 2. Danish Technological Institute, Aarhus

Background: Total tumour removal is the primary factor of consideration in surgical resection. Several techniques exist for tumour reconstructive surgery in musculoskeletal oncology: allografts, autografts, recycled bone and endoprosthesis. Recently, 3D printing has undergone tremendous development and now has important applications in various fields of medicine. A major improvement is the possibility to print prosthesis, which are custom made for the single patient. Currently, when a customized 3D printed prosthesis is needed, an outside order must be placed, and the procedure is usually time consuming, making almost impossible to fall within the time interval of the law A collaboration between a 3DP centre in Aarhus University Hospital (AUH) and the Danish Technological Institute (DTI), allow us to manufacture custom-made 3D printed metal implant in-house.
Aim: Aim of this study was to assess the osseointegration of 3D-printed titanium implants through a validated randomized animal study.
Materials and Methods: 20 stable, non-weight-loaded, 6*10 mm cylindrical implants were 3D printed by DTI: 10 with a rough and 10 with a smooth surface. Implants were randomized and implanted into the left humerus of 20 skeletally mature sheep. After 4 weeks of observation all sheep were euthanized. The specimens were collected and cut into blokes, each containing an implant and surrounding tissue. Each block was then cut into a 3 mm block for mechanical test, closest to the surgical entry site, and a 6 mm block for future histomorphometrical evaluation. Biomechanical testing was performed as failure by push-out test on an Instron Universal Test Machine.
Results: Implants with a smooth surface demonstrate complete absence of osseointegration, as they fall out of the bone during sample preparation. Testing this group was therefore not possible. Porous implants showed macroscopic integration and breaking point at implant´s surface. We measured a median Ultimate Shear Strength of 0,06 MPa (IQR:1,14), a median Apparent Shear Stiffness of 0,16 MPa/mm (IQR:0,48) and a median Energy Absorption of 19,98 J/m2 (IQR:25,80).
Interpretation / Conclusion: Our study shows superior osseointegration in 3DP implants with a porous surface.

Thomas Colding-Rasmussen^1,2, Peter Schwarzenberg ³, Peter Horstmann ⁴, Casper Bent Smedegaard Ottesen ¹, Jorge Garcia ⁵, Daniel John Hutchinson ⁵, Michael Malkoch⁵, Peter Varga ³, Christian Nai En Tierp-Wong ¹
1. Department of Orthopaedic Surgery, Rigshospitalet, Copenhagen 2. Department of Orthopedic Surgery, Hvidovre University Hospital, Hvidovre 3. AO Research Institute Davos, Davos, Switzerland. 4. Department of Orthopedic Surgery, Gentofte Hospital, Gentofte. 5. Department of Fibre and Polymer Technology, KTH Royal Institute of Technology, Sweden.

Background: A novel osteosynthesis platform (Bonevolent™ AdhFix) has shown promising biomechanical properties in preliminary ex vivo studies. As this technique relies on an unconventional surgical approach that involves the light curing of a viscous compound adhered to bone via metal screws, an evaluation of clinical usability is warranted. Furthermore, as AdhFix is customizable in situ, potential variability in biomechanical stability is anticipated.
Aim: To assess inter- and intra-surgeon biomechanical variability and usability of the AdhFix osteosynthesis platform.
Materials and Methods: Six surgeons, ranging from interns to consultants, reviewed a written instruction manual and were subsequently supervised in one trial osteosynthesis before each conducting ten consecutive procedures on a synthetic cylindrical (12 x 30 mm) bone model (Synbone, Switzerland), with a reduced midline osteotomy. Subsequently, all constructs underwent 4-point bending at a quasi-static loading rate. Maximum bending moment (BM) and bending stiffness (BS), as well as AdhFix cross-sectional area (CSA: mm²), were recorded for each sample. A linear mixed effect model were used to calculate and describe differences in BM and BS between and within surgeons (SPSS v. 27).
Results: All constructs exhibited a consistent appearance and suitability for biomechanical testing. The mean BM was 2.64 ± 0.57 Nm, and the mean BS was 4.35 ± 0.44 Nm/mm. Statistically significant differences were observed among surgeons in both BM (p < 0.001) and BS (p = 0.004). Across ten trials, only one surgeon demonstrated an increase in BM (p < 0.025), and one surgeon showed increased BS (p < 0.01). A larger CSA correlated with higher BM (p < 0.001), but not BS (p = 0.594)
Interpretation / Conclusion: Conclusion: Consistent biomechanical stability was observed within surgeons, suggesting that AdhFix can be learned and applied with minimal training, regardless of prior surgical experience. Although statistically significant variability was noted in maximum bending moment (BM) and bending stiffness (BS) among surgeons, we anticipate that these differences are unlikely to have a clinically significant impact. Further investigations are warranted to validate these findings.

Bagge Jasmin¹, Mahmood Haider¹, Nehlin Jan², Hölmich Lisbet³, Blønd Lars⁴, Barfod Kristoffer^1,5, Hölmich Per¹
1. Sports Orthopedic Research Center - Copenhagen (SORC-C), Department of Orthopedic Surgery, Copenhagen University Hospital - Hvidovre 2. Department of Clinical Research, Copenhagen University Hospital - Hvidovre 3. Department of Plastic Surgery, Copenhagen University Hospital – Herlev and Gentofte 4. Department of Orthopedic Surgery, Zealand University Hospital – Køge 5. Unit of Sports Traumatology, Copenhagen University Hospital - Bispebjerg

Background: Treatment of knee osteoarthritis (OA) with autologous stem cells from microfragmented adipose tissue (MFAT) has shown promising, but varying results. Multiple stem cell types have been identified in MFAT, such as CD34+ (adventitial stem cells), CD146+ (pericytes), and CD271+ stem cells. These subtypes have shown varying differentiation potential when derived from bone marrow. The patient- dependent heterogeneity of the stem cell population and content of highly potent cells may be determining factors for a successful treatment outcome.
Aim: To identify the most promising stem cell type from MFAT for the treatment of OA.
Materials and Methods: CD34+, CD146+, and CD271+ stem cells from MFAT of 8 knee OA patients were separated by magnetic activated cell sorting (MACS) and analyzed as subtypes. Efficiency of sorting was measured by flow cytometry. Unsorted cells were used as a control. The chondrogenic and osteogenic in vitro differentiation performance were assessed using quantitative Safranin-O staining, pellet size, and qPCR for chondrogenesis, and Alizarin Red S staining and qPCR for osteogenesis.
Results: CD34+, CD146+, and CD271+ stem cells can be doubled using MACS with a mean sorting efficiency on 84% (CD34), 60% (CD146), and 33% (CD271). All subtypes were able to undergo osteogenic differentiation with a significant difference between induced and non-induced controls when using Alizarin Red S staining for calcium deposits (p=0.001). CD146+ stem cells showed significantly higher osteogenic Alizarin Red S performance compared to CD34+, CD271+ and unsorted stem cells (p=0.03). All subtypes were able to make 3D chondrogenic pellets. CD271+ stem cells showed significantly higher chondrogenesis when measuring proteoglycans with Safranin-O staining compared to CD34+ and CD146+ stem cells (p=0.02), but no difference compared to unsorted. No significant difference was seen in chondrogenic pellet size between cell types. qPCR is under analysis.
Interpretation / Conclusion: CD146+ stem cells showed highest osteogenic differentiation performance. CD271+ stem cells showed highest chondrogenic differentiation performance. Composition of stem cell types might be important to their clinical effectiveness when treating OA.

Ali Yalcinkaya^1,2, Hans-Christen Husum^1,2, Jens Svendsson ², Michael Skovdal Rathleff ^3,4, Søren Kold^1,2, Ole Rahbek^1,2, Shima Gholinezhad ^1,2
1Interdisciplinary Orthopaedics, Aalborg University Hospital, Aalborg, Denmark 2Department of Orthopaedics, Aalborg University Hospital, Aalborg, Denmark 3Department of Health Science and Technology, Aalborg University, Aalborg, Denmark 4Center for General Practice at Aalborg University, Aalborg University, Aalborg, Denmark

Background: Patellofemoral Pain Syndrome (PFP) and Osgood- Schlatter Disease (OSD) are two of the most common causes of anterior knee pain. The role of inflammation in PFP is not well-established, however, there are indications of inflammation in OSD. Infrared thermography, a non- invasive imaging method, has the potential to visualize areas of pain by detecting temperature variations resulting from inflammation.
Aim: Our objective was to examine potential connections between the subjective reporting of pain location and measurements obtained through thermography in adolescents experiencing pain in one knee.
Materials and Methods: Adolescent patients with unilateral knee pain (32 PFP and 24 OSD) were recruited. They marked the location of their pain on a body chart that depicted both the left and right knee, creating a two-knee pain map. Infrared images were captured using a FLIR C3 thermal camera. To assess temperature differences between normal and painful knees, a symmetric comparison was performed by comparing average temperatures of the pain area in the painful knee with the same regions in the pain- free contralateral knee.
Results: In OSD, painful knees were consistently warmer (31.41 ± 1.16) than pain-free contralateral knees (30.83 ± 0.94) with a statistically significant difference (p=0.042). In PFP, painful knees (30.32 ± 1.29) didn't show a clear temperature pattern while pain-free knees were slightly warmer (30.92 ± 1.43) with no statistical significance (p=0.43).
Interpretation / Conclusion: Significant temperature differences in OSD were observed, but this difference could not be demonstrated in the PFP group. Results supporting the role of inflammation and the accompanying temperature increase in OSD were observed. A clear pattern did not emerge in PFP patients. The idea that many factors are at play in the pathophysiology was supported.

Mads K. D. Mikkelsen^1,2, Maya Bang^1,2, Hans Christian Rasmussen^1,2, Johanne G. Lilleøre^1,2, Josephine O. Kipp^1,2, Lars Lindgren^1,3, Janni K. Thillemann^1,2, Annemarie Brüel⁴, Mats Bue^1,2, Maiken Stilling^1,2
1. Orthopaedic Research Laboratory, Aarhus University Hospital; 2. Department of Orthopaedic Surgery, Aarhus University Hospital; 3. Department of Radiology, Aarhus University Hospital; 4. Department of Biomedicine, Aarhus University

Background: In Denmark, it is recommended to treat stable scaphoid fractures in a short arm cast without including the thumb. However, this recommendation is supported by sparse evidence. Thus, the treating clinician’s preference and individual patient needs may favour other cast types.
Aim: To evaluate the immobilizing properties of three different cast types commonly used for wrist immobilisation in the treatment of scaphoid fractures.
Materials and Methods: Ten fresh frozen human arm-and-elbow specimens without radiological or clinical signs of pathology were included. Three different cast types (X-lite) were applied to the specimen’s wrists: 1) a dorsal half cast (DHC) not including the thumb, 2) a radial half cast (RHC) including the thumb, and 3) a circular cast (CC) including the thumb. Anterior-posterior and lateral radiographs were taken in neutral position. Next, the immobilised wrists were fixed in a radiolucent fixture and loaded in extension, flexion, radial, and ulnar deviation by applying an orthogonal load of 42 N. The mean angle (CI95%) of deviation was defined as the difference between the radiologically measured loaded wrist angle and the neutral position.
Results: The mean angle of deviation allowed with CC was 11° (7-14) flexion, 7° (4-11) extension, 12° (8-15) radial deviation, and 11° (8-14) ulnar deviation. Compared to CC, the DHC allowed for an additional 26° (23- 30) flexion, 7° (3-10) extension, 16° (12-20) radial deviation, and 13° (9-17) ulnar deviation. Compared to CC, the RHC allowed for an additional 23° (21-28) flexion, 24° (21-28) extension, 16° (12-20) radial deviation, and 17° (13-21) ulnar deviation. CC allowed for significantly less wrist movement in all directions. RHC and DHC stabilised the wrist equally in radial deviation.
Interpretation / Conclusion: Casting of nondisplaced scaphoid fractures using a dorsal half cast without thumb immobilisation or a radial half cast including the thumb may be used for initial immobilisation. However, significantly better wrist stabilisation was observed using a circular cast including the thumb. This supports exchanging the cast throughout the remaining healing period when the fracture swelling is reduced and a circular cast can be tolerated.

Christian Hoegsbjerg^1,2, Peter Schjerling^1,2, Michael R. Krogsgaard³, Michael Kjaer^1,2, Arvind von Keudell^4,5, Abigail L. Mackey^1,2
1 Institute of Sports Medicine Copenhagen, Department of Orthopaedic Surgery, Copenhagen University Hospital - Bispebjerg and Frederiksberg, Copenhagen, Denmark 2 Department of Clinical Medicine, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark. 3 Section for Sports Traumatology M51, Department of Orthopaedic Surgery, Copenhagen University Hospital - Bispebjerg and Frederiksberg, Copenhagen, Denmark 4 Orthopaedic Trauma Section, Department of Orthopaedic Surgery, Copenhagen University Hospital - Bispebjerg and Frederiksberg, Copenhagen, Denmark 5 Brigham and Women’s Hospital, Department of Orthopaedic Surgery, Havard Medical School, Boston

Background: The interface between the muscle and tendon, i.e. the myotendinous junction (MTJ), presents a crucial anatomical region for the force transmission between these two distinct tissues. Despite the optimization for an environment of high stress and strain, the MTJ remains the predominant site of injury within the musculoskeletal unit. It can be assumed that cells on the tendon and muscle sides of the MTJ contribute to tissue repair and yet the number of these cells is unknown.
Aim: To perform a fibre type-specific and 3D analysis of both the muscle (myonuclear) and tendon cell populations at the MTJ in single human myofibre.
Materials and Methods: We isolated 10 type I and 10 type II single myofibres with an intact MTJ from waste gracilis graft tissue from 6 patients undergoing anterior cruciate ligament reconstruction surgery (age: 30.5 ± 7.6 SD). We then employed confocal imaging to generate a model of the single human myofibres with known 3D coordinates of all nuclei. Data were analysed by a Wilcoxon signed rank and Friedman signed rank statistical test.
Results: We observed a surprisingly large aggregation of cells at the MTJ, with fibre type differences in numbers (p<0.05); median of 29 (range: 16-63) cells for type I and 16 (range: 9-23) cells for type 2 myofibres. Density of myonuclei at the MTJ was similar for fibre types. However, we found a gradual increase (p<0.05) in myonuclear number and density from the main portion of the myofibre to the tip for both fibre types, with both type fibre types displaying an excess of 10 myonuclei at the MTJ (range: 6-15).
Interpretation / Conclusion: These results highlight fibre type differences in MTJ nuclear organization both on the muscle and tendon sides of the MTJ. The high concentration of cells at the MTJ is unexpected given the current thinking on maximising the amount of contact area between muscle and tendon for resisting damage during force transmission. Future work could explore the capacity of these cells to rebuild the MTJ after rupture.