Introduction. Patient-reported outcome measures (PROMs) has gained momentum in the orthopaedic literature. The GOAL-LD (Gait Outcomes Assessment List for Limb Deformity) incorporates the framework of The WHO International Classification of Functioning, Disability and Health and correlates highly with the Paediatric Outcomes Data Collection Instrument (PODCI) when applied in a limb reconstruction setting. The Royal Children's Hospital Melbourne, Australia, supported by The University of Melbourne is designing and implementing a Limb Reconstruction Registry, prospectively enrolling patients with a number of key conditions and provide a mechanism for capturing surgical data and PROMs at regular follow-up intervals. LimbDREAM (The Limb Reconstruction Registry of Experience, Aspiration and Measures) has begun recruiting participants, and this paper serves as a narrative review of our design and implementation process. Materials & Methods. After provision of a research grant, audit was conducted to examine local case mix and estimate recruitment based on conditions selected. Literature review was focussed on (i) registries in the paediatric limb reconstruction domain and (ii) use of PROMs across conditions selected. A high-level process-flow chart was constructed in order to inform governance and staffing requirements in addition to REDCap Database structure. Local as well as scalable deployment was considered. Alpha and beta testing was performed by principal and associate investigators prior to participant recruitment. Results. Audit identified 6 condition streams for inclusion: Congenital limb deficiencies, congenital pseudarthrosis of the tibia, bone dysplasias, bone tumours, amputation, and complex deformity of other etiology. Literature review identified 20 PROMs used across these conditions, with limited validation studies and significant implementation heterogeneity. REDCap database design took longer than anticipated (planned 10 months, actual 18 months) before being recruitment ready, due to the complexity of ensuring data collection logic would work across highly diverse patient journeys. Initial recruitment across all streams has been promising. Email and text message reminders have proved helpful in increasing survey return rates. Conclusions. Prospective collection of PROMs as well as surgical data via a standardized dataset will provide the basis for future condition-specific research, with the ability to support nested trials. Implementation requires forethought with regard to time frames and staffing for non-automated steps as well as data integrity review. The next phase of the LimbDREAM project will be to integrate use of the Registry into our weekly Limb Reconstruction Meeting to ensure that operative plans and data are captured close to the point of care

Introduction. Osteoarthritis (OA), a painful, debilitating joint disease, often caused by excessive joint stress, is a leading cause of disability (World Health Organisation, 2003) and increases with age and obesity. A 5° varus malalignment increases loading in the medial knee compartment from 70% to 90% (Tetsworth and Paley, 1994). Internal unloading implants, placed subcutaneously upon the medial aspect of the knee joint, are designed to offload the medial compartment of the knee without violating natural joint tissues. The aim of this study is to investigate the effect of an unloading implant, such as the Atlas™ knee system, on stress within the tibiofemoral joint with different grades of cartilage defects. Methods. To simulate surgical treatment of medial knee OA, a three-dimensional computer-aided design of an Atlas™ knee system was virtually fixed to the medial aspect of a validated finite element knee model (Mootanah, 2014), using CATIA v5 software (Dassault Systèmes, Velizy Villacoublay, France). The construct was meshed and assigned material properties and boundary conditions, using Abaqus finite element software (Dassault Systèmes, Velizy Villacoublay, France). A cartilage defect was simulated by removing elements corresponding to 4.7 mm. 2. The international cartilage repair society (ICRS) Grade II and III damage were simulated by normalized defect depth of 33% and 67%, respectively. The femur was mechanically grounded and the tibia was subjected to loading conditions corresponding to the stance phase of walking of a healthy 50-year-old 68-Kg male with anthropometrics that matched those of the cadaver. Finite element analyses were run for peak shear and von Mises stress in the medial and lateral tibiofemoral compartments. Results. Von Mises stress distribution in the tibial cartilage, with ICRS Grade II and III defects, without the unloading implant, at the end of weight acceptance (15% of the gait cycle) were analysed. The internal unloading implant reduces peak von Mises stress by 40% and 43% for Grade II and Grade III cartilage defects, respectively. The corresponding reductions in shear stress are 36% and 40%. Consistent reduction in peak von Mises stress values in the medial cartilage-cartilage and cartilage-meniscus contact areas were predicted throughout the stance phase of the gait cycle for ICRS Grade II defect. Similar results were obtained for Grade III defect and for peak shear stress values. There were no overall increases in peak von Mises stress values in the lateral tibial cartilage. Discussion and Conclusions. The internal unloading implant is capable of reducing von Mises and shear stress values in the medial tibial cartilage with ICRS Grade II and III defects at the cartilage-cartilage and cartilage-meniscus interfaces throughout the stance phase of the gait cycle. This did not result in increased stress values in the lateral tibial cartilage. Our model did not account for the viscoelastic effects of the cartilage and meniscus. Results of this study are based on only one knee specimen. The internal unloading implant may protect the cartilage in individuals with medial knee osteoarthritis, thereby delaying the need for knee replacements

Aims. Tuberculosis (TB) infection of bones and joints accounts for 6.7% of TB cases in England, and is associated with significant morbidity and disability. Public Health England reports that patients with TB experience delays in diagnosis and treatment. Our aims were to determine the demographics, presentation and investigation of patients with a TB infection of bones and joints, to help doctors assessing potential cases and to identify avoidable delays. Patients and Methods. This was a retrospective observational study of all adults with positive TB cultures on specimens taken at a tertiary orthopaedic centre between June 2012 and May 2014. A laboratory information system search identified the patients. The demographics, clinical presentation, radiology, histopathology and key clinical dates were obtained from medical records. Results. A total of 31 adult patients were identified. Their median age was 37 years (interquartile range (IQR): 29 to 53); 21 (68%) were male; 89% were migrants. The main sites affected were joints (10, 32%), the spine (8, 26%) and long bones (6, 19%); 8 (26%) had multifocal disease. The most common presenting symptoms were pain (29/31, 94%) and swelling (26/28, 93%). ‘Typical’ symptoms of TB, such as fever, sweats and weight loss, were uncommon. Patients waited a median of seven months (IQR 3 to 13.5) between the onset of symptoms and referral to the tertiary centre and 2.3 months (IQR 1.6 to 3.4.)) between referral and starting treatment. Radiology suggested TB in 26 (84%), but in seven patients (23%) the initial biopsy specimens were not sent for mycobacterial culture, necessitating a second biopsy. Rapid Polymerase Chain Reaction-based testing for TB using Xpert MTB/RIF was performed in five patients; 4 (80%) tested positive for TB. These patients had a reduced time between the diagnostic biopsy and starting treatment than those whose samples were not tested (median eight days versus 36 days, p = 0.016). Conclusion. Patients with bone and joint TB experience delays in diagnosis and treatment, some of which are avoidable. Maintaining a high index of clinical suspicion and sending specimens for mycobacterial culture are crucial to avoid missing cases. Rapid diagnostic tests reduce delays and should be performed on patients with radiological features of TB. Cite this article: Bone Joint J 2018;100-B:119–24

There is evidence that preoperative physical fitness impacts surgical outcomes, specifically preceding abdominal, cardiovascular and spine surgery. To our knowledge, there are no papers on self-reported exercise frequency as a predictor of cervical spine surgery outcomes. Our objectives were to quantify self-report of exercise frequency in cervical spine surgery patients, and to elucidate if self-reported exercise prior to surgery confers less pain, improved health state and/or less disability post-surgery. We performed a retrospective review of prospectively collected data from the Canadian Spine Outcomes and Research Network (CSORN) Database from the time of its inception. Inclusion criteria specified all elective cervical surgery patients over 18 years old with degenerative pathology who proceeded to surgery and completed the pre- and post-operative outcomes measures up to 24 months post surgery (n=460). Outcome measures were visual analog scales (VAS) for neck and arm pain, Neck Disability Index (NDI), and EuroQOL score at baseline and 3, 12 and 24 months post-op. Exercise frequency was self-reported as “none” (n=212) versus “some” (n=248). These groups were further categorized into “none due to physical limitations”, “none” (not due to physical limitations), those to exercised “once or less per week” and those that exercised “twice or more per week”. Student's t-tests were used to compare the mean scores of the outcome measures, and analysis of variance for subgroup comparisons, with results considered significant at p < 0 .05. At baseline, 56% of total patients reported exercise prior to surgery, of which 73% reported doing so twice or more per week. Of the 44% reporting no exercise, 74% could not exercise due to physical limitations. Those who reported “some” exercise had more favorable VAS neck and arm pain scores pre-operatively (neck: 5.55 vs 6.11, p < 0 .001) (arm: 5.69 vs 6.04, p=0.011), but no difference at 3 and 24 months post-operatively. Significantly lower NDI scores and higher EuroQOL Index scores were seen in the exercise group compared to the no exercise group pre-operatively (NDI: 39 vs 48, p < 0 .001) (EuroQOL: 0.60 vs 0.50 p < 0 .001) as well as at 3, 12, and 24 months post-op (NDI: 24 vs 31, p=0.007) (EuroQOL: 0.75 vs 0.68, p=0.001). Further subgroup analysis demonstrated that compared to the “no exercise due to physical limitation” group, the “twice or more” exercise group showed favorable NDI and EuroQOL scores up to 24 months post-op (NDI: 24.32 vs 32.33, p=0.001) (EuroQOL: 0.76 vs 0.66, p=0.001), whereas the “once or less times per week” group no longer demonstrated any significant difference at 24 months (NDI: 28.79 vs 32.33, p=1) (EuroQOL: 0.73 vs 0.66, p=0.269). Self-reported exercise prior to cervical spine surgery does not predict improved long-term neck and arm pain at 2 years post-op. However, self-reported exercise does demonstrate less disability and improved health state at baseline and up to 2 years post-op and this relationship is dose dependent

Purpose. The objective of this study was to compare items from musculoskeletal outcome questionnaires with items generated by pre- and post-operative ankle arthrodesis and arthroplasty patients (patient-selected portion of the Patient-Specific Index (PASI-P)) to determine if existing questionnaires address patients' concerns. Materials/Methods. Patients (n=142) completed the PASI-P. Items from 6 standardised questionnaires (AAOS, patient-reported portion of AOFAS, FFI, LEFS, SMFA, WOMAC) and PASI-P were matched by 3 reviewers to corresponding categories in the International Classification of Functioning, Disability and Health (ICF). The ICF classification hierarchy codes broad first categories and increasingly detailed second, third, and fourth categories. The standardised questionnaires were then compared to the ‘gold standard’ of PASI-P. Results. 690 concepts were identified in the patients' responses, which corresponded to 45 ICF categories. Most PASI-P concepts fell into Activities and Participation (60.3%) and Body Functions and Body Structures (35.2%), including the second level categories ‘walking’ (21.6%), ‘pain’ (18.7%), and ‘recreation and leisure’ (17.4%). A total of 237 concepts was identified in the 6 questionnaires and linked to 39 second level ICF categories. Overall, SMFA addressed the most second level categories and had the closest proportion of Activities and Participation (68.9%) and Body Functions (23.0%) concepts compared to the patient-generated responses. The patient-reported portion of AOFAS addressed the fewest categories. LEFS only contained items from Activities and Participation. AAOS was the only questionnaire to address the issue of ‘swelling’, though it represented 5.6% of all patient-generated responses. The concepts of ‘swelling’, ‘recreation and leisure’, and ‘sports’ were seldom included in the questionnaires. Conclusion. No single questionnaire captured all concerns identified by patients, and the standardised questionnaires differed largely in content. Clinicians should recognise patient concerns relevant to the ankle currently not included in available questionnaires. This analysis will guide the development of a more comprehensive instrument for evaluating ankle outcomes

Purpose. No knee-specific outcome measures have demonstrated reliability, validity or responsiveness in patients with multiligament knee injuries. Furthermore, the content validity of existing questionnaires has been challenged and remains unknown for patients with concomitant neurovascular injury. As a first step in developing a disease-specific outcome measure, the objective of this study was to identify items from existing knee questionnaires pertinent to patients with multiligament knee injuries using established a priori criteria. Method. Eighty-five consecutive patients from a level one trauma centre were mailed a questionnaire comprising 124 items from 11 knee-specific instruments. They rated the frequency and importance for each item on a five-point Likert scale. Criteria for item selection included a mean importance rating (MIR) ≥ 3.5 and frequency < 30% for the response never experienced. The World Health Organization International Classification of Functioning, Disability and Health (ICF) framework was used to support the content. Results. The average age of the 60/85 respondents (70.6%) was 34.7 years and most were male (n=45). Average time from injury was 845 days. Using the Schenck classification, 19 patients were KD I, two were KD II, 39 were KD IIIL or IIIM, and nine were KD IV. Nineteen had peroneal nerve injury and five had a vascular injury. Forty-four patients were treated surgically, six nonoperatively and 10 are pre-operative. Sixty-one of 124 questionnaire items met the a priori inclusion criteria. Based on the ICF, 22 of the retained items were physical impairments (PI), 15 items were emotional impairments (EI), 18 items were activity limitations (AL), and six items were participation restrictions (PR). The 10 items with the highest MIR included six EI, one PI and three PR. Twenty-four items (PI=6 EI=13, AL=2, PR=3) of the Anterior Cruciate Ligament Quality of Life Questionnaire (ACL-QoL) satisfied the a priori criteria and this instrument also endorsed the most items within the EI and PR constructs. The Knee Osteoarthritis and Injury Outcome Score (KOOS) had 21 items (PI=7, EI=2, AL=11, PR=1) which satisfied the a priori criteria and also had the most items within the PI and AL constructs. The International Knee Documentation Committee (IKDC) Subjective Knee Form had 12 items (EI=0, PI=3, AL=9, PR=0) meeting criteria while the remaining questionnaires contained 11 items with concepts redundant to above. Combining the ACL-QoL and KOOS resulted in representation of 41/61 of the retained items (PI=13, EI=13, AL=12, PR=3). No existing instrument includes content that may result from neurological or vascular injury. Conclusion. This work has identified over sixty pertinent items across the various ICF domains relevant to patients with multiligament knee injuries. The ACL-QoL and KOOS together address over two thirds of these items. Patient and clinician focus groups will be conducted to address missing content and to further refine the questionnaire items