BESS Abstracts

THE ROLE OF LONG HEAD OF BICEPS INCARCERATION IN SHOULDER ADHESIVE CAPSULITIS

J Allport, A Kingman, J Dickinson and J Candal-Couto

Northumbria Healthcare NHS Foundation Trust, Northumbria

Purposes: In adhesive capsulitis histopathological changes typically begin in the rotator interval, in close proximity to the long head of biceps tendon (LHB). We investigated the incidence of LHB incarceration and its affects on patient outcomes in adhesive capsulitis.

Methods: 60 consecutive patients undergoing arthroscopic capsular release for adhesive capsulitis by a single surgeon were retrospectively identified from a locally held database. Cases were classified as “normal LHB” or “incarcerated LHB” based on findings recorded at time of surgery. No patients had an LHB tenodesis performed. Oxford Shoulder Score and EQ-5D scores were extracted from the database.

Results: Six cases had to be excluded (three operation note unavailable, two LHB findings not recorded and one absent LHB). 54 cases were included in the pre-operative analysis, 40 cases had post-operative scores available with a mean follow up of 300 days (range 89–623). LHB incarceration was found in 13 cases (24%). There was no statistical difference between the groups’ baseline demographics. Pre-operative OSS scores were significantly lower in the group where LHB incarceration was noted (15.3) than the normal group (21.0, P = 0.030). Post-operatively there was no significant difference between groups (36.6 vs 37.4, P = 0.795), this finding was also true when analysis was limited to cases with follow up of greater than 11 months (38.8 vs 40.6, P = 0.649).

Conclusion: The LHB tendon is closely related to the rotator interval, an area where to adhesive capsulitis histopathological changes begin. LHB incarceration is associated with more severe cases of adhesive capsulitis, however, as outcomes are comparable between the two groups it is not an indication for LHB tenodesis.

SURGICAL INTERVENTION IN THE SHOULDER: WHAT PROCEDURES ARE GAME CHANGERS AND WHAT ARE NOT?

R Paraparan, PH Lam and GAC Murrell

Orthopaedic Research Institute, St George Hospital Campus, University of New South Wales, Sydney, Australia

Background: We aimed to determine, from a patient’s perspective, which were the most effective commonly performed surgical procedures for disorders of the shoulder, and which were not.

Methods: The study was a retrospective analysis of prospectively collected data from patients who underwent shoulder surgery from a single surgeon from 2004–2015. To be included, at least 20 patients needed to have undergone that procedure and complete a questionnaire evaluating their shoulders function pre-operatively and six months post-operatively. The primary outcome was change in response to the question “How is your shoulder overall?”.

Results: 2206 surgical procedures in 13 categories met the inclusion criteria. All procedures were associated with improvements in patient-ranked overall shoulder status at 6 months (p < 0.01 – p < 0.0001). Reverse total shoulder arthroplasty (RTSA) provided the greatest effect size (improvement) in overall shoulder status (d = 3.14, 95% CI: 2.49–3.79) followed by total shoulder arthroplasty (TSA) (d = 2.60, 95% CI: 2.10–3.10) and capsular release (d = 1.41, 95% CI: 1.08–1.75). Reverse total shoulder arthroplasty (RSTA) provided the greatest effect size in patient reported shoulder pain, whilst capsular release provided the greatest effect size in patient reported shoulder function. Acromioclavicular joint resection (d = 1.22, 95% CI: 0.56–1.88) and acromioplasty (d = 1.29, 95% CI: 0.96–1.61) provided the least effect size overall.

Conclusion: All shoulder surgical procedures within this study provided a significant therapeutic effect size in a relatively short period of time (6 months). However, some procedures provided a greater improvement in patient reported outcomes than others. Reverse total shoulder arthroplasty provided the greatest improvement in patient reported shoulder status. Capsular release provided the greatest effect size in patient rated overhead function. Conversely, acromioplasty and acromioclavicular joint resections were associated with the least benefit in patient-assessed shoulder pain and function.

VALIDATION OF HES CODING FOR SHOULDER ARTHROSCOPY

BA Marson, BW Gooding, MD Wijeratna, P Manning and JM Geoghegan

Nottingham University Hospitals NHS Trust, Nottingham

Background: The volume of subacromial decompressions being performed in U.K. hospitals is under scrutiny. Subacromial decompressions are being identified from the Hospital Episode Statistics (HES) submitted by each unit. Numbers of procedures where O291 (subacromial decompression) is recorded as the main procedure are being analysed. The validity of these returns is crucial to prevent misrepresentation of a unit’s activity.

The aim of this study is to evaluate the accuracy of the coding of procedures coded with the subacromial decompression (O291) code.

Methods: 611 procedures were coded including the O291 code between 21/3/2016–29/3/2018. 438 (71%) of procedures were coded with O291 (subacromial decompression) as primary procedure. Operation notes were reviewed and compared against submitted code strings

Results: 201 (46%) of procedures coded with O291 as primary procedure were not subacromial decompressions. These included 59 (29.4%) rotator cuff repairs and 73 (36.3%) acromioclavicular joint excisions.

247 ‘true’ subacromial decompressions were identified. 237 (96.0%) were coded with O291 as the primary procedure. 57 different code strings were used to describe subacromial decompressions.

Discussion: Following NHS Digital guidelines, our coders are instructed to identify the ‘primary procedure’ for each case. Coders are identifying the right strings, but not the correct primary procedure.

The results from our unit are representative of the U.K. activity. In the 2016–2017 HES data 30,056 out of 39,850 (75.4%) of shoulder arthroscopies were recorded as a subacromial decompression as the main procedure. Yet in our series, almost half the recorded subacromial decompressions were different procedures and this is likely to be a national problem.

As surgeons we are evaluated based on the hospital level records in the HES dataset. Clearer guidance for coders is required from national bodies who understand shoulder surgery to improve the transparency of our work and ensure patients are being provided with evidence-based interventions.

ULTRASONOGRAPHIC AXILLARY NERVE MAPPING: AN ALTERNATIVE COST-EFFECTIVE TOOL WITH IMPLICATIONS IN SURGERY

AJ Thayaparan, S Shalaby, A Sahi and T Zaman

London North West University Healthcare NHS Trust, London

Aim: To identify the course of the axillary nerve using ultrasonography. We also wanted to ensure this techniques validity and reproducibility as a method of pre-operatively marking and avoiding iatrogenic injury.

Method: Patients recruited into this study were referred from their primary care physician with shoulder pain for an ultrasound assessment of their shoulder. Prospectively 123 patients were recruited in this study, of which 31 were partitioned due to the presence of a rotator cuff tear or age over 60 years.

Ultrasound scans were performed with identification of the axillary nerve and its course from the quadrangular space with measurements taken from the most proximal visualised part of the humeral head. Results were tabulated and the process was repeated.

Results: All patients recruited in the study underwent scans where the axillary nerve was identified in its course with the posterior circumflex humeral artery approximately 4 cm from the humeral head. Statistical analysis revealed no variability in the position of the axillary nerve comparing those with rotator cuff tears and without (p value 0.30). Furthermore, there was no statistically significant difference when observing for variability in distance of the axillary nerve with patient age.

Conclusion: We propose that ultrasonographic mapping of the axillary nerve is a valid, cost-effective method that can be performed with relative ease. This has major implications in surgery whereby the nerve is often at risk. Injury to this can cause devastating consequences to a patient's function. This study offers instructions and guidance in mapping the axillary nerve with the growing use of ultrasonography as an alternative to costly, time-consuming alternatives such as magnetic resonance imaging. Further studies will be conducted to confirm the findings as above and indeed expand to potential usage in handheld ultrasonography due to the validity of the method proposed above.

LOCAL ANAESTHETIC ADMINISTRATION FOR DAY CASE SHOULDER SURGERY: A COMPARISON OF 4 METHODS

A Akhtar, E Ibrahim, D Shaerf, M Smith, R Gupta, K Singh and D Ahearne

The Hillingdon Hospitals NHS Trust, London

The purpose of this study was to compare the efficacy of four methods of local anaesthetic administration for day case shoulder surgery - intralesional infiltration (IL); ultrasound (US)-guided interscalene brachial plexus block (ISB); US-guided combined peripheral nerve block (PNB) of the suprascapular nerve (SSN), axillary nerve (AXN) and sensory branch of the lateral pectoral nerve (SLPN) (US-PNB), and landmark-guided block of the same nerves (L-PNB) plus incision site infiltration.

Four groups of 20 consecutive patients had general anaesthetic plus bupivacaine administered by one of the aforementioned techniques. Open and arthroscopic day case procedures were included. Outcome data were collected in person on the day of surgery and by telephone the following day.

Patients in the US-PNB and L-PNB groups were most satisfied. Interscalene and both PNB techniques led to significantly better day-of-surgery pain scores and lower morphine requirements than IL; ISB resulted in the lowest morphine requirement of all but there was no significant improvement in 1-hour pain score compared to the PNB groups. However, pain score at 24 hours was significantly higher in the ISB group than the other groups. Most patients receiving ISB were unable to engage with therapy exercises on day 1 due to persistent paralysis or severe pain. Patients in the IL group experienced significantly more nausea and were more likely to experience disturbed sleep on the first night. Ultrasound guided-PNB was significantly slower to perform than L-PNB.

Combined PNB of the AXN, SSN and SLPN, guided by anatomical landmarks or US, provides the best balance of analgesia and patient satisfaction in the first 24 hours after day case shoulder surgery compared to ISB or IL alone.

DISAPPOINTING RESULTS WITH THE INSPACE® BALLOON SPACER FOR FAILED-REPAIRED/IRREPARABLE MASSIVE ROTATOR CUFF TEARS

SK Khan, M Sohatee, J Candal-Couto and D Cloke

Northumbria Healthcare NHS Foundation Trust, Northumbria

Purpose: We studied a carefully selected cohort of patients treated with the InSpace balloon for irreparable or failed-repair rotator cuff defects, with the aim of evaluating short and medium term results (Oxford shoulder score, and further surgery).

Methods: After institutional approval (new devices committee), two fellowship-trained shoulder surgeons recruited patients with massive irreparable or failed-repaired rotator cuff tears (confirmed on imaging or previous arthroscopic attempted repairs) between February 2016 and May 2017. Their notes were retrospectively reviewed as part of a service evaluation, and the following data were recorded: 1) demographics, 2) pre- and post-operative (6 weeks, 3,6,12 and 18 months) symptoms, ranges of motion (ROM), and Oxford Shoulder Scores (OSS), 3) previous/subsequent procedures, 4) features on standardised shoulder AP radiographs.

Results: Seven patients [M:F 7:1, mean age 66.3 (54.7–80.8)] had eight shoulders implanted with the balloon spacer for failed repairs (n = 6) and irreparable tears (n = 2). Pain improved in five while ROM improved in four out of these eight shoulders. OSS (where available) improved gradually up to 6 months and then declined to scores similar to pre-operative values. There were five returns to theatre in the subsequent 18 months: 1) one retrieval for early balloon rupture, 2) one margin-convergence repair with allograft, 3) three reverse shoulder replacements.

Conclusions: Our results with the balloon spacer are disappointing, and in stark contrast to those reported nationally (two studies) and internationally (ten studies). This is despite our series having similar patient inclusion criteria, age ranges and grades of cuff tear arthropathy. We heed caution in the use of this implant. Other options need to be explored for this problem.

WHAT IS THE OPTIMUM INSERTION-POINT FOR LATISSIMUS DORSI TENDON TRANSFER: A COMPUTATIONAL MODELLING AND CADAVERIC STUDY

S Thompson, S Roche, S Sivarasu and DJ Henderson

Institution University of Cape Town Department of Biomedical Engineering, South Africa

Aim: To determine the optimal insertion-point for a transferred Latissimus Dorsi tendon (LDT) in order to best restore both active external rotation (ER) and forward-flexion (FF) in the treatment of irreparable postero-superior rotator cuff tears, using combined computer modelling and cadaveric simulation.

Background: LDT transfer is an accepted treatment for irreparable postero-superior rotator cuff tears, however few studies have investigated how changing the insertion-point of a transferred LDT affects the resulting restoration of motion.

Methods: Virtual surgery was performed, simulating LDT transfer for irreparable postero-superior rotator cuff tear, on a modified Dutch Shoulder and Elbow computer model. Rotation and flexion moments were calculated for insertion-points across the entire proximal humerus, in multiple shoulder positions, to determine the effect on resulting muscle force in flexion and rotation. A cadaveric validation of these models was then carried out, using six fresh-frozen cadaveric shoulders, with transfer of the LDT to seven attachment-points across the proximal humerus. Muscle contraction was simulated using weights and resulting ER force in the humerus was recorded.

Results: Proximal humeral “efficiency maps” generated by the computer model demonstrated that a more anterior insertion-point generated improved external rotation throughout shoulder flexion and rotation, while a more superior insertion point improved flexion. A “best compromise zone”, combining both, was identified at the antero-superior greater tuberosity, balancing optimal flexion and external rotation.

These findings were supported by the cadaveric external rotation model, where insertion-points towards the anterior greater tuberosity generated greatest external rotation torque, though this became less significant with increased forward flexion.

Conclusions: This combined study, using cadaveric modelling to corroborate findings of a computational model, has identified an optimal insertion zone on the antero-superior greater tuberosity, which best allows a transferred LDT to restore both ER, and flexion, deficits resulting from an irreparable postero-superior rotator cuff tear.

ASSESSING THE RELIABILITY AND ACCURACY OF PRE-OPERATIVE DIAGNOSTIC ULTRASOUND TO EVALUATE THE ROTATOR CUFF IN ARTHRITIC PATIENTS

R Gogna, S Dhillon, R Dias, T McBride, S Thompson, S Roche, S Sivarasu and DJ Henderson

The Royal Wolverhampton NHS Trust, Wolverhampton Institution University of Cape Town Department of Biomedical Engineering

Objective: MRI remains the gold standard investigation to assess rotator cuff integrity prior to shoulder arthroplasty, however this is not always possible due to contraindications and patient claustrophobia. The aim of this study is to establish the sensitivity and specificity of USS to assess the rotator cuff in patients with underlying arthritis.

Materials and methods: A retrospective analysis of consecutive patients who underwent USS prior to shoulder arthroplasty surgery between February 2014 and December 2018 was performed. Demographic data was collected from electronic records. USS findings were correlated with intra-operative assessment of the rotator cuff performed by three consultant orthopaedic surgeons. The sensitivity, specificity, positive and negative predictive values were then calculated.

Results: Twenty-eight patients were identified who underwent pre-operative USS prior to shoulder arthroplasty surgery during the study period, with mean age 70 years (range 49 to 86). Final operative outcomes included reverse polarity shoulder replacement (n = 16), anatomic total shoulder replacement (n = 5) and resurfacing hemiarthroplasty (n = 7).

In six patients the pre-operative USS was unable to determine whether a full thickness cuff tear was present; of which four had intra-operative cuff tears. The overall sensitivity and specificity of US in detecting a full-thickness cuff tear was 0.7 (CI 0.35–0.92) and 0.5 (CI 0.22–0.78) respectively. The lowest sensitivity was seen for subscapularis tears 0.33 (CI 0.02–0.87). The positive predictive value was 0.54 (CI 0.26–0.8) with an overall diagnostic accuracy of 46%.

Conclusion: The use of ultrasound to assess cuff integrity has a very low sensitivity and accuracy for patients with underlying arthritis. This is likely due to restricted range of movement and abnormal anatomy. The poor correlation with intra-operative findings suggests that scan results should be interpreted with caution in this group and surgeons should be prepared for all eventualities.

PREVALENCE OF METAL DEBRIS WITH A TRABECULAR METAL- BACKED GLENOID PROSTHESIS

A Jaiswal, P Kankanalu, M Papanna, M Morgan, T Creswell, M Espag, A Tambe and D Clark

Royal Derby Hospital, Derby

Aim:

1. To evaluate prevalence of metal debris in patients who underwent anatomical total shoulder replacement using Zimmer Trabecular metal(TM) glenoid prosthesis.

2. To analyse clinical and radiological results of these patients.

Methods: 63 consecutive shoulders were included in the study with minimum follow up of 2 years. Mean age was 62 years (range 26–82). Thirty-four shoulders had long stem and 29 had short stem humeral prosthesis. Oxford shoulder score (OSS) and Constant score (CS) were used to assess functional outcome. Radiographs were reviewed for loosening and to grade presence of metal debris. Outcomes between groups with and without metal debris were compared. Statistical analysis was performed to assess the association between most recent debris grade and outcome scores, and association between age, gender, humeral stem type and preoperative Walch classification with presence of metal debris.

Results: At mean follow up of 3 years (range 2–7 years) 25 shoulders (39%) showed evidence of metal debris. Mean OSS in the group with metal debris was 25.5 and in the group without was 17.7, p = 0.02. Mean CS in the group with metal debris was 54.8 and in one without was 64.5, p = 0.06. There was no significant association of age, gender, type of humeral prosthesis and preoperative Walch classification with presence of metal debris. OSS showed significant poorer outcome in patients with any grade of metal debris, however this was not replicated in analysis of CS. Four patients demonstrated progressive radiolucent lines. Two of these had revision for septic loosening and cuff failure respectively.

Conclusion: In line with published literature, our study also demonstrated concern with metal debris with TM glenoids. We have stopped using this implant and all the patients are now being monitored at one-year interval. This study shows importance of surveillance of patients with new prostheses.

SPHERICAL PYROCARBON INTERPOSITIONAL SHOULDER ARTHROPLASTY: A PROSPECTIVE, SINGLE SURGEON, 4 - 7 YEAR FOLLOW-UP STUDY

MJC Stanislas

South Warwickshire Foundation Trust & Warwickshire Nuffield, Warwickshire

Aim: To assess the medium term outcome of a spherical Pyrocarbon Interpositional Shoulder Arthroplasty, especially the effect on the native Glenoid

The use of pyrocarbon implants for Interpositional Arthroplasty has been deemed successful in small joints of the human body. The excellent tribologic properties of pyrocarbon has enabled the use of this material in the arthroplasty of the shoulder joint. This particular implant is spherical and has a smaller radius of curvature than the native humeral head.

Method: A single centre prospective study of 21 patients aged between 29 and 72 years (average 49 years) who underwent pyrocarbon interpositional shoulder arthroplasty were assessed at a minimum of 4 year follow up. A total of 10 men and 11 women, 16 with a primary diagnosis of osteoarthritis, 2 with avascular necrosis, and 3 with post stabilisation arthritis were included.

Results: Follow-up ranged from 48 - 87 months (mean 64 months). Both Constant score and Oxford shoulder score improved. No cases have been revised, although assessment of 1 patient is incomplete after 4 years. There have been no fractures or dislocations. Serial radiographs were evaluated which revealed glenoid erosion in two patients, tuberosity thinning in one, and corticalisation of the bed of the implant in all. Postero-superior migration of the implant in relationship to the glenoid was seen in 3 patients.

Conclusion: Medium term results and indications are still developing for the use of pyrocarbon interpositional arthroplasty. Long term follow-up is essential to determine: the impact on the native glenoid; erosion of the proximal humerus; the effects on the rotator cuff; and the functional outcome of this material in shoulder arthroplasty.

THE IMPORTANCE OF PRE-OPERATIVE OUTCOME SCORES FOR PREDICTION AND CASE-MIX ADJUSTMENT MODELS IN SHOULDER ARTHROPLASTY

RS Craig, GS Collins, A Rangan, A Sayers and JL Rees

Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford

Purpose: To evaluate the importance of pre-operative PROMS in predicting outcomes and ensuring appropriate case-mix adjustment in analysis of primary shoulder arthroplasty outcomes.

Methods: Data from 8871 elective shoulder replacement procedures recorded in the National Joint Registry between June 2012 and March 2015 were used to develop a linear model to predict post-operative Oxford Shoulder Scores (Q2 OSS). Missing data were handled with multiple imputation. The following candidate variables were specified a priori: age, indication, condition of rotator cuff, ASA grade, sex, previous surgery. This was compared to a model with the addition of the pre-operative (Q1) OSS. Model performance was compared using goodness of fit tests and calibration plots.

Results: A large proportion of scores were missing. Q1 OSS was 48% complete and Q2 OSS was 69% complete. Q1 OSS was an important predictor of Q2 OSS (coefficient = 0.39, p < 0.0001) and greatly improved the fit of the linear model: optimism corrected R2 0.131 vs 0.061, AIC 67195 vs 67870, LR Test Chisq 669 on 1df). Visual inspection of calibration plots confirmed this improved fit.

Conclusion: Pre-operative patient reported outcome measures (PROMs) scores are an important predictor of outcome and risk-adjustment models are significantly improved by their inclusion. Shoulder registries need to reliably collect pre-operative PROMs. These scores remain incomplete in the NJR but their collection is necessary if reliable models scrutinising both implants and providers are to be achieved. The collection of Q1 scores should be routine and nationally mandated in order to account for baseline imbalance as well as allowing change of score analysis.

SERIOUS ADVERSE EVENTS AND LONG-TERM REVISION RISK FOLLOWING SHOULDER REPLACEMENT FOR ACUTE TRAUMA: A POPULATION COHORT STUDY OF 14 052 PATIENTS FROM HOSPITAL EPISODE STATISTICS DATA

RS Craig, JCE Lane, D Furniss, SE Gwilym, GS Collins and JL Rees

Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford

Purpose: To use a large observational dataset from Hospital Episode Statistics (HES) to provide more accurate estimates of peri-operative risks, complications and revision rates in patients over 50 years having shoulder replacement surgery for acute proximal humeral fractures.

Methods: Shoulder replacements performed for acute fractures were identified within the HES database, together with all linked prior and follow-up episodes. Patients aged ≥ 50 years were included. The incidence of serious adverse events was calculated at 30 and 90 days post-surgery. Revision risk was estimated using the Kaplan-Meier method.

Results: 14 052 procedures were included (10 843 in women). The median age was 74 (IQR 67 to 80). Male patients were younger than women. Overall, 9.7% of patients experienced a serious adverse event within 30 days. By type these were: death 1.2%, pulmonary embolism 0.5%, myocardial infarction 0.6%, pneumonia 4.0%, acute kidney injury 1.8%, urinary tract infection 3.7%, stroke 0.5%. Within 90 days, the risk of any serious adverse event was 11.6%. Adverse events were strongly associated with increasing age, multimorbidity and male sex, but not implant type.

A total of 639 patients underwent revision surgery. More than 200 others needed additional shoulder surgery (without implant revision), and half of these were to address early post-operative stiffness, particularly following hemiarthroplasty. Revision rates across implant types at 5, 10 and 15 years were 4.9%, 5.8% and 6.3% by the Kaplan Meier method. Data comparing revision rates between reverse total shoulder arthroplasty (RTSA) and hemiarthroplasty is only available at two years.

Conclusion: Approximately 10% of patients undergoing shoulder arthroplasty for trauma experience a serious adverse event within 30 days. The risk of revision surgery appears low, but long-term data on RTSA is lacking. Strategies to reduce serious adverse events in the first 30 days should be a priority for upper limb trauma service providers.

METAPHYSEAL REVERSE TOTAL SHOULDER ARTHROPLASTY WITHOUT A STEM - LONG-TERM RESULTS WITH 5 - 11 YEARS FOLLOW-UP

O Levy, A Leonidou, G Panagopoulos, P Consigliere, O Tsvieli, C Witney-Lagen, L Natura, R Abraham, G Sforza, J Bruguera and E Atoun

Reading Shoulder Unit, Royal Berkshire Hospital, Reading

Introduction:

rTSA are increasingly used in recent years. Metaphyseal humeral components without a stem were developed to minimise bone resection and preserve bone.

The aim is to report the long-term clinical and radiologic results with a stemless rTSA.

Methods: 172 consecutive shoulders underwent rTSA between 2005 to December 2011, 149 with short metaphyseal stemless implant and 23 with stemmed implant.

The average follow up was 89 months (6.25 years) (range 60 – 138 Months). 41 males and 131 females; mean age at surgery 74.3 y (range 38–93 y).

111 patients for severe rotator cuff deficiency: 86 cuff tear arthropathy, 19 fracture sequelae, 24 rheumatoid arthritis, 16 patients following failed RC repair or massive irreparable cuff tear, 4 osteoarthritis with cuff deficiency or eroded glenoid, 18 failed anatomical prosthesis with cuff deficiency, and 5 for acute trauma. 13 patients (26 shoulders) underwent bilateral (staged) rTSA at that period.

50 patients were operated as revision arthroplasty (21 from stemmed implants to stemmed rTSA, 29 to short metaphyseal stemless rTSA (3 of them from stemmed implant to stemless implant).

Results: SSV improved from 1.1/10 to 9.3/10. Mean Constant Score (for all diagnoses) improved from 15.9 ± 8.6 preop to 59.7 ± 20.4 at last follow-up. Age/sex adjusted Constant score improved from 22.6 ± 12.3 preoperatively to 89.2 ± 30.3 at the last follow-up (p < 0.0001).

The mean active range of movement improved from 53.8° to 131.9° active elevation, 20.7° to 34.6° active external rotation and 32.3° to 68.8° active internal rotation. Radiographic analysis showed no lucencies, subsidence or stress shielding around the humeral or glenoid components.

Glenoid notching was found in 40 shoulders (23.2%) (34 grade 1–2 (19.8%) and 6 cases grade 3 (3.5%)).

Conclusion: The short metaphyseal stemless rTSA shows good long-term results with restoration of good active ROM and high patients’ satisfaction. There is no implant loosening, subsidence or stress shielding.

SURVIVORSHIP OF AUTOLOGOUS BONE GRAFT AT A MINIMUM OF 2 YEARS FOR GLENOID BONE LOSS IN PRIMARY AND REVISION SHOULDER ARTHROPLASTY: A CT AND CLINICAL REVIEW

J Singh, K Neelakandan, M Walton, P Monga and S Bale; Ian Trail

Wrightington Hospital, Wigan

Background: The purpose of this study was to assess the integrity of the autologous glenoid bone graft (humeral head or iliac crest) at two years in our series of primary and revision shoulder arthroplasty.

Methods: Ethical approval was sought and the study has a portfolio study status by the NIHR. We contacted patients who had primary and revision shoulder arthroplasty with Lima Axioma TT metal back glenoid with autologous bone graft and were more than two years since their operation. All eligible patients underwent CT evaluation, clinical review and functional scoring.

Results: 40 patients (42 shoulders) were reviewed. Mean age 65 yrs. (range 35–83 yrs.), 22 females and 18 males. The average follow up period was 40 months (range 24 - 59 months).

Primary arthroplasty was performed in 22 shoulders whereas other 20 had revision arthroplasty. 25 had reverse shoulder arthroplasties (RSR) and 17 had anatomical shoulder replacements (ASR). Two of the ASR sustained a cuff failure and were revised to a RSR. 23 shoulders had graft taken from the humeral head and 19 had iliac crest bone graft, reflecting the numbers of revisions.

Satisfactory bone graft incorporation (>50 %) was seen in 39 shoulders and only 3 patients had <50 % graft incorporation. The scans at 2 years showed no significant deterioration in the bone graft from the early post op scans.

Average forward elevation improved from 50 degrees (pre op) to 100 degrees (range 35–150). The mean improvement in average Oxford Shoulder Score was 15 (Pre op- 15, Post op 30) and the mean improvement in Constant Score improvement was 35 (Pre op 12, post op 47). Mean postoperative ASES was 62 (range 30–85).

Conclusion: The use of trabecular metal in conjunction with autologous bone graft provides a reliable method of addressing glenoid bone defects in shoulder arthroplasty.

PRE-OPERATIVE PARTIAL THICKNESS CUFF TEARS DO NOT COMPROMISE RESULTS OF ANATOMICAL TOTAL SHOULDER REPLACEMENT- 5 YEAR FOLLOW UP

P Raval and R Pandey

University Hospitals of Leicester, Leicester

Reverse shoulder arthroplasty (RSA) in elderly patients with primary osteoarthritis (OA) and cuff pathology is increasing, the purpose of our study was to determine the medium term results of anatomic total shoulder athroplasty (TSR) for OA in patients with pre-operative partial thickness (P/T) cuff tears on MRI scans.

Methods: We retrospectively reviewed patients who had TSR for OA with a preoperative MRI diagnosis of P/T cuff tear. Patients were assessed with pre and post-operative Oxford Shoulder Score(OSS), range of movements(ROM) and rotator cuff was clinically assessed. AP and Axillary radiographs were taken for proximal humeral migration (using Torchia classification) and evidence of loosening. Lazarus score was used to grade glenoid radiolucencies.

Results: Thirty-six patients (M14:F22), who underwent TSR had a P/T tear on MRI preoperatively, all demonstrated mild to moderate fatty infiltration. The mean age was 79.2(71–88) years; Mean follow-up was 5.1(4.8–7.0) years. Significant improvements in pain, ROM, was reported in all cases. At the final follow-up the OSS was 42(32–46) with minimum 14 points improvement(p = 0.001). External rotation (15°vs30°; P = 0.001), forward flexion (80°vs130°; P = 0.015) abduction (40°vs70°; P = 0.015) and internal rotation also improved. Lucencies has been observed in 8 glenoids. Grade1-6, Grade2-2 & none with Grade3. There were no cases of implant loosening. Clinically four patients had cuff weakness but only 2 showed evidence of proximal migration and one of anterior migration of humerus of which one patient with Torchia moderate grade proximal migration was revised for cuff failure; one patient had washout for infection.

Conclusion: There is paucity of literature on whether preoperative partial thickness rotator cuff tear has an adverse affect on the outcome of TSR. Our results show that the presence of P/T cuff tear on preoperative MRI does not significantly effect function after TSR in the medium term. The use of reverse shoulder arthroplasty in these cohorts of patients, therefore, may not be justified.

COMPARISON OF COMPLICATION TYPES AND RATES ASSOCIATED WITH ANATOMIC AND REVERSE TOTAL SHOULDER ARTHROPLASTY

SA Parada, RJ Friedman, JD Zuckerman, TW Wright, PH Flurin and CP Roche

Medical College of Georgia, Augusta University, USA

Introduction: Complications after anatomic (aTSA) and reverse (rTSA) total shoulder arthroplasty can be devastating to a patient’s quality of life and require revisions which are costly to both the patient and the hospital. The purpose of this study is to better quantify the type of failure modes for a single platform total shoulder system.

Methods: 2224 aTSA (1090M/1134F) and 4158 rTSA (1478M/2680F) patients were enrolled in an international database of primary shoulder arthroplasty performed by 40 different surgeons in the US/Europe. Adverse events and revisions reported for these 6382 patients were analyzed to identify the most common failure modes associated for both aTSA and rTSA.

Results: Of 2224 aTSA patients, 239 adverse events were reported for a complication rate of 10.7% and 124 revisions for a revision rate of 5.6%. The top three complications for aTSA were rotator cuff tear/subscapularis failure (n = 69, complication rate = 3.1%, revision rate = 1.9%), aseptic glenoid loosening (n = 55, complication rate = 2.5%, revision rate = 1.9%) and infection (n = 28, complication rate = 1.3%, revision rate = 0.8%)

Of 4158 rTSA patients, 372 adverse events were reported for a complication rate of 8.9% and 104 revisions for a revision rate of 2.5%. The top three complications for rTSA were acromial/scapular fracture/pain (n = 102, complication rate = 2.5%, revision rate = 0.0%), instability (n = 60, complication rate = 1.4%, revision rate = 1.0%) and pain (n = 49, complication rate = 1.2%, revision rate = 0.2%).

Conclusions: This large database analysis quantified complication and revision rates for aTSA and rTSA. We found aTSA and rTSA complication rates of 10.7% and 8.9%, respectively; with revision surgery rates of 5.6% and 2.5%, respectively. The two most common complications for each prosthesis type (aTSA: subscapularis/rotator cuff tears; rTSA: acromial/scapular fractures) were unique to each device. The rate of infection was similar for both. Future prosthesis and technique development should work to mitigate these common complication types in order to reduce their rate of occurrence.

BENFITS OF USING METALIC LATERALISED BASEPLATE SUPPORTED BY A LONG METALLIC POST, TO ADDRESS GLENOID BONE LOSS IN REVERSE SHOULDER ARTHROPLASTY

M Cartaya, JD Werthel, J Kany, Ph Valenti and D Rossouw

Paris Shoulder Unit, Clinic Bizet, Paris, France

Severe glenoid bone loss remains a challenge in shoulder arthroplasty surgery. This condition makes it challenging to implant a stable component.

Aim: The objective of this study was to evaluate clinical and radiological outcomes of a lateralised metal-backed 15.2 mm keeled baseplate, prolonged by a thin 24.8 mm metallic post transfixed directly in the subscapularis fossa. This implant was used in primary and revision cases of reverse shoulder arthroplasty (RSA) with severe glenoid bone loss. The medial fixation point is designed to provide greater primary stability to this construct.

Methods: Between January 2011 and December 2014, 51 shoulders (50 patients) underwent primary or revision RSA using this baseplate. Forty-five shoulders in 44 patients were followed for a minimum of two years (mean, 33 months; range, 24–60 months). The mean age of the patients was 76 years (range, 55–93 years). Outcome measures recorded were the Constant Score, X Rays and complications.

Results: The overall complication rates were 12% in primary cases and 25% in revisions(infection, pain, stiffness). One glenoid implant (4%) failed in a primary case and one glenoid implant (5%) failed in a revision case, both from early displacement in massive defect situations. All remaining cases showed stable implant position up to latest follow up. Pain and range of motion were significantly improved in both groups. The mean Constant Score improved from 24 (±7) to 62 (± 9) in primary cases and from 24 (± 10) to 58 (± 12) in revision cases.

Conclusion: A lateralised metal-backed 15.2-mm keeled baseplate prolonged by a thin 24.8-mm metallic post fixed directly in the subscapularis fossa provides good primary stability satisfactory early and mid-term outcomes in patients with large glenoid bone defects where initial press-fit of a regular baseplate is impossible to implant.

IN ANATOMIC SHOULDER ARTHROPLASTY THERE ARE LESS GLENOID RADIOLUCENT LINES AND HUMERAL OSTEOLYSIS WITH A CERAMIC THAN A METAL HUMERAL HEAD

S Bell, M Christmas and J Coghlan

Melbourne Shoulder and Elbow Centre, Monash University, Melbourne, Australia

Purpose: To compare in an anatomic total shoulder replacement the clinical outcome, the incidence of glenoid radiolucent lines, and the extent of proximal humerus osteolysis, utilizing a ceramic or a chrome-cobalt humeral head.

Background: Radiolucent lines leading to glenoid loosening and proximal humeral osteolysis, in total shoulder replacement, may be influenced by changes on the humeral side of the replacement.

Methods: An analysis was made of prospectively collected data comparing 43 patients having a total shoulder replacement utilizing a long humeral stem with a chrome-cobalt head, and 54 with a stemless humeral component with a ceramic head. The same two pegged polyethylene cemented glenoid prosthesis was used in all cases.

At 2 years and 5 years post arthroplasty radiological assessment of glenoid component radiolucent lines was made in both groups utilizing the Lazarus classification, and clinical results recorded. Also, the extent of proximal humeral osteolysis was assessed in both groups at 5 years.

Results: The demographics of the two groups were very similar, including the pre-operative Walch scores, and the utilized head and glenoid size. The function, satisfaction, and ASES scores were also very similar.

On analysis of the glenoid radiolucent lines, at both 2 and 5 years more lucent lines were present with the metal head than with the ceramic head (p = <0.001). At 5 years no >1 Lazarus lines were seen with ceramic heads but there were 16% with metal heads. There was a significant correlation (p.004) between the extent of proximal humeral osteolysis and the Lazarus scores, with considerable humeral osteolysis in the metal head group, and minimal with the ceramic heads.

Conclusion: The use of a ceramic head resulted in far less glenoid component radiolucent lines and proximal humeral osteolysis than a metal head.

IN-VIVO RANDOMISED CONTROL STUDY OF THE BONE RESPONSE OF ALL-SUTURE ANCHORS TO BIOCOMPOSITE ANCHORS IN THE GLENOID

C Stewart, H Raja, E Torrance and L Funk

HCA The Wilmslow Hospital & University of Manchester, Manchester

Purpose: We aimed to compare the bone response to two common suture anchors used for glenohumeral instability surgery: a traditional biocomposite push fit anchor (Osteoraptor, Smith & Nephew) and an all-suture anchor (Suturefix, Smith & Nephew).

Methods: Seventeen patients were included in this study with a total of 37 unique suture anchors. MRI scans were performed at 3 weeks and 6-months post-operatively. Thirty eight senior radiologists and shoulder surgeons evaluated the images, using a previously validated system for grading the bone response around suture anchors. The mean difference in grading at 3 weeks and 6 months was calculated using unpaired t-tests, and the inter-rater reliability was evaluated with an intraclass correlation coefficient. Ethical approval was obtained on initiation of the study.

Results: At 3 weeks, there was no statistically significant difference in the degree of osteolysis surrounding each suture anchor type (p = 0.258), with little bone response. However on the 6-month scans there was a significantly lower level of osteolysis seen in the all-suture anchors compared to the biocomposite anchors (p = 0.040) and inter rater reliability was excellent with an ICC value of 0.975 (95% CI [0.962–0.985]).

Conclusion: All suture anchors cause significantly less osteolysis in glenoid bone at 6 months, compared with biocomposite anchors.

FAILED LATARJET - WHY, HOW AND WHAT NEXT?

U Khan, E Torrance and L Funk

HCA The Wilmslow Hospital, University of Manchester, Manchester

Purpose: The aim of this study was to review the causes and management of recurrent instability after previous Latarjet stabilisation surgery.

Methods: A retrospective analysis of prospective data was conducted on patients undergoing revision surgery after failed Latarjet stabilization. Data was collected over a five-year period. This included patient demographics (age, gender), clinical presentation, cause of recurrent instability, indications for revision surgery, intra-operative analysis, outcome of revision surgery and return to sport.

Results: Sixteen patients were identified undergoing revision surgery for recurrent instability after Latarjet stabilization. Twelve were male and four were female. Eleven were athletes: nine were professionals and two were amateurs. The mean age at revision was 29.9 years ± 8.9 (range, 15–50). Indications for revision were anterior instability in eleven patients and posterior instability in four, with one both anterior and posterior instability. 54.5% of anterior instabilities were due to coracoid non-union and 36.4% were due to capsular failure (retear). All of the posterior instabilities had posterior capsulo-labral injuries and the mean Beighton score in this group was 6 and above. One patient had a failed Latarjet with coracoid non-union and posterior labral tear.

Conclusion: This study found coracoid non-union to be the most common cause of recurrence following Latarjet stabilization, requiring an Eden-Hybinette procedure. Those patients returning with posterior instability had a high incidence of hypermobility and could be treated successfully by arthroscopic techniques.

CLINICAL EVALUATION OF THE ARTHROSCOPIC CONJOINT TENDON TRANSFER FOR THE MANAGEMENT OF FAILED BANKART REPAIR IN THE SHOULDER

V Patel, E Pearse, M Arnander and D Tennent

St Georges Hospital London, London

Background: Arthroscopic labral repair is the gold standard for the primary treatment of the majority of anterior instability cases. Difficulty arises when a labral repair fails in patients with no critical anterior glenoid bone loss and an on track Hill Sachs lesion. There are limited options of either a revision repair with reported higher failure rates or a Latarjet procedure. The latter has the potential for significant complications and bone reabsorption has been observed in up to 60% of cases.

Aim: To evaluate the clinical outcomes of the arthroscopic conjoint tendon transfer procedure as previously described us in active patients with a failed Labral repair and sub critical glenoid bone loss.

Methods: Inclusion criteria were active patients with recurrent shoulder instability due to failed labrum repair with < 10% anterior glenoid bone loss and an on track Hill Sachs lesion. Patients were counselled appropriately and offered revision Bankart repair, Latarjet or arthroscopic conjoint tendon transfer. Pre and post-operative Western Ontario Shoulder Instability Index (WOSI) and Oxford Instability Score (OIS) were collected.

Results: 8 patients met the inclusion criteria and consented for the procedure. Mean age was 35 with Male: Female of 7:1. No patients had hyper laxity clinically. Mean operative time was 95 minutes. There were no complications to report. At a mean follow up period of 26 months[14–35] there was a significant improvement in the WOSI and OIS from a mean of 50.2[SD18.5] to 17.5[SD8.4] (p = 0.0003) and 26[SD9.9] to 42.8[SD4.5] (p = 0.0017), respectively. No patient had a further dislocation and all had resumed contact/ non- contacts sports.

Conclusion: Early results demonstrate the conjoint tendon transfer technique provides stability in patients with a failed arthroscopic labral repair who actively engage in sports.

ACCELERATED REHABILITATION FOLLOWING DISTAL BICEPS TENDON REPAIR

V Jones, S Booker, D Potter, A Ali and D Thyagarajan

Sheffield Shoulder and Elbow Unit, Northern General Hospital, Sheffield

To evaluate the outcome of early mobilisation in patients undergoing distal biceps repair, following the introduction of an accelerated rehabilitation guideline.

A consecutive series of patients were identified, who were managed before and after the introduction of an accelerated rehabilitation guideline, following distal biceps tendon repair. The new guideline recommends immediate active elbow mobilisation within a defined safe range of motion. The previous guideline delayed active motion for a minimum for 2 weeks, and marked activity reduction thereafter.

The new guideline recommended early physiotherapy and gave clearly defined clinical milestones. The primary outcome measure was clinical integrity of the distal biceps tendon. Secondary outcomes were incidence of neurovascular or wound complications, and range of motion. Patients were followed up for a minimum of 6 months.

47 patients were identified between January 2016 and October 2018. 17 were treated 18 months prior to the guideline introduction, 30 afterwards. All patients were male and the dominant arm was affected in 57% of patients. The mean age at time of surgery was 42 (range 24 - 62).

At 8 weeks, 20 patients in the early mobilisation group (81.4%) had achieved full range of flexion/ extension, compared with 4 patients in the delayed mobilisation group (23.5%; p < 0.05). Mean time to recover full active motion was 7.73 weeks in the early mobilisation group, and patients required an average of 5.2 post-operative physiotherapy sessions, compared with 10.47 weeks in the delayed group, and 6.7 treatment sessions.

In both groups there were no wound complications, but there were 2 transient neuropraxias seen in both groups. One patient in the accelerated rehabilitation group had a re-rupture of his tendon repair 6 days after surgery, after ignoring post-operative instructions.

Early active mobilisation is safe and effective following distal biceps repair. Patients recovered faster without increasing the complication rate.

HAVE THE BESS SUBACROMIAL SHOULDER PAIN GUIDELINES CHANGED PRACTICE IN PRIMARY CARE IN THE UK?

E Whitby, K Walstow and M Bateman

University Hospitals of Derby & Burton NHS Foundation Trust; Derby Shoulder Unit, Derby

Background: The British Elbow & Shoulder Society published guidelines in April 2015 to guide practitioners working in Primary Care in the UK in how best to manage patients presenting with shoulder pain. The aim of this study was to assess the impact of this publication on clinical practice.

Methods: An audit was conducted in a large NHS Foundation Trust providing Primary Care physiotherapy services retrospectively between January and June 2017. Electronic coding was used to identify all patients with shoulder pain and notes were individually screened for accuracy. All patients with subacromial shoulder pain were included. Notes were independently reviewed and data extracted based upon pre-defined criteria.

Results: 84 patients attended during the analysed time with a mean age of 55 and 73% female. Median symptom duration was 8 months. 62% of patients had imaging arranged by their General Practitioner prior to referral: X-ray 14%, ultrasound 21%, both X-ray and ultrasound 26%. Patients were treated for a mean of 3.8 sessions over 3.1 months. 82% of patients received education and activity modification advice. 99% were taught exercises and only 6% manual therapy. 15% received steroid injections. Overall 71% improved or symptoms resolved. 11% were unchanged and in 4% symptoms deteriorated. 12% were lost to follow up. Only 2% of patients were subsequently seen in Secondary Care and underwent surgery.

Conclusion: Based upon this single centre study, the majority of patients presenting with subacromial shoulder pain can be successfully managed in Primary Care with physiotherapy based upon advice, education and exercise. The high number of patients undergoing unnecessary imaging in the form of X-ray and particularly diagnostic ultrasound is cause for concern and a potential target for efficiency savings. Further education of GPs is required to address this.

THE USE OF GAMIFIED VIRTUAL PHYSIOTHERAPY AS AN EFFECTIVE TREATMENT FOR PATIENTS WITH SHOULDER PROBLEMS

A Barratt, MH Granat, S Buttress and B Roy

University of Salford; Manchester Foundation Trust, Manchester

Statement of purpose: A study to investigate the use of exergames technology on the outcomes of physiotherapy treatment following shoulder surgery

Methods: Physiotherapy exergames were created using a combination of commercially available hardware. The Microsoft Kinect, and bespoke software incorporating games were mapped to physiotherapy goals applying principles of gamification to them.

This study was a randomised prospective controlled trial. The intervention group [n = 10] received physiotherapy aided by automated sensor-based technology which helped them perform exergames and track rehabilitation progress. The control group [n = 10] was treated by standard physiotherapy protocols. The two groups were compared using patient reported outcome measures and assessment of shoulder range of movement pre and post operatively.

Results: The study showed that there was an improvement in the range of movement in both groups. There was no difference in the between the groups in percentage changes 6 weeks post-operative and 12 weeks post-operative, compared to pre-operative, for external rotation, forward flexion and abduction. The Oxford Shoulder Score results showed a significant improvement for the control group at 12 weeks to pre – operative (p = 0.02), there was no significant change for the intervention group (p = 0.193). The results for the DASH scoring tool showed that there was no significant change for the control group (p = 0.01) compared to the intervention group (p = 0.088). The results using the T-test for the EQ5D score show that there is no difference in the intervention group p = 0.135 compared to the control group 0.171.

Conclusion: Exergames may be used as an alternative for standard physiotherapy regimes, however further analysis is required focusing on patient engagement.

MEDIUM-TERM RATES OF RADIOLUCENCY AFTER PRIMARY TOTAL SHOULDER ARTHROPLASTY USING A CEMENTLESS METAL-BACKED PEGGED POLYETHELENE GLENOID

NJ Murray, KA Al-Hourani, MAA Crowther, IN Packham and PA McCann

North Bristol NHS Trust, Bristol

Purpose: The aim of this study was to investigate periprosthetic radiolucency rates associated with an uncemented, metal-backed polyethylene glenoid component with medium term results as a biologically fixed implant may offer better survivorship than traditional cemented polyethylene glenoid components.

Methods: A single centre retrospective study examining radiological outcomes of the Epoca metal-backed glenoid component. Radiographs were analysed for postoperative adequacy of glenoid seating and most recent radiographs at follow-up assessed for periprosthetic lucencies and any revision procedures were recorded. Intra- and inter-observer reliability was assessed.

Results: 41 patients were followed up with a mean follow-up time of 5.5 years(3–8 years). Primary indication for total shoulder arthroplasty was osteoarthritis(80%). Mean age was 69 years(53–86 years). Ninety-five percent of glenoid components were completely seated post operatively. At follow-up six patients had undergone revision(14.6%), three for cuff failure and three for polyethylene wear and metalosis. Thirty of the remaining thirty-five patients(86%) did not demonstrate any evidence of radiolucency on follow-up radiographs. Complete glenoid seating post operatively was associated with lower rate of subsequent radiolucency and revision(P < 0.01).

Conclusion: Low rates of radiolucency at medium term follow-up with an uncemented, metal-backed, pegged glenoid however significant rates of revision. Complete seating of the glenoid component was associated with lower rates of radiolucency and revision.

RESULTS OF REVERSE SHOULDER ARTHROPLASTY (RTSA) IN ACUTE FRACTURES VERSUS FRACTURE SEQUELAE

A Leonidou, G Panagopoulos, F Butt, P Megaloikonomos and O Levy

Reading Shoulder Unit, Royal Berkshire Hospital, Reading

Purpose: Recent studies suggest conservative treatment for displaced proximal humerus fractures (PHF). This may lead to fracture sequelae (FSQ). rTSA has been suggested for treatment of PHF in elderly patients. We compared the results of rTSA in acute PHF, with rTSA in patients with FSQ.

Methods: Demographics, radiographs, and surgery data were prospectively collected and analysed. At the final follow up (FU), range of motion (ROM), subjective shoulder value (SSV), Constant Score (CS) and radiographic analysis were recorded.

Results: Between 2005–2017, 36 patients (28 women; 8 men) underwent rTSA for acute fractures. Mean age was 79.3 years (range, 40–93). The average time from injury was 4 weeks (range, 0–13). Mean FU was 24 months (range, 13–95). Mean SSV at final FU was 8.48/10 and mean adjusted CS 89.4. Final mean active ROM was: 122o forward flexion, 120o abduction, 33o external rotation and 57o internal rotation.

At the same period, 55 patients (42 women; 13 men) underwent rTSA for FSQ. 25 had type 1, 14 type 3, 6 type 2 and 3 type 4 sequelae (Boileau classification). Mean age was 72 years (range, 55–88). 15 patients had previous surgery. Average time from injury was 32 weeks (range, 1–111). Mean FU was 40 months (range, 3–111). At final FU, mean SSV was 7.48/10 and mean adjusted CS 78.5. Final mean active ROM was: forward flexion 113o, abduction 109o, external rotation 19o and internal rotation 53o.

Radiographic analysis of both groups no lucencies around the humeral or glenoid components and no evidence of loosening, subsidence or stress shielding. 1 case of grade I notching in the acute group, 2 cases of grade 2 notching in the sequelae group.

Conclusion: Patients treated with rTSA for acute proximal humerus fractures without previous surgery had better functional results than patients who had rTSA for fracture sequelae.

	Acute Fracture Group	Fracture Sequelae Group	p value
Final SSV	8.48/10	7.48/10	N/A
Final CS	59.1	55.9	p = 0.390
Final A/s Adj CS	89.4	78.5	p = 0.045
Forward Flexion	122	113	p = 0.02
Abduction	120	109	p = 0.655
Active External  Rotation	33	19	p = 0.478
Active Internal  Rotation	57	53	p = 0.730

SHOULDER ARTHROPLASTY ENHANCED RECOVERY – REDUCING LENGTH OF INPATIENT STAY

G Davies-Jones, M Morgan, E Ibrahim, S Booker, M Bateman, AA Tambe, M Espag, T Cresswell and DI Clark

Derby Shoulder Unit, Derby

Enhanced Recovery (ER) programmes are well established in hip and knee arthroplasty, they have not however been described in shoulder arthroplasty. This study analyses the effect of implementing an ER shoulder arthroplasty programme on length of stay, functional outcome and patient satisfaction.

73 consecutive patients who underwent total shoulder arthroplasty and were enrolled in an ER programme were compared to a control group of 79 consecutive patients who underwent total shoulder arthroplasty in the year before the programme started.

An MDT steering group was involved in service design and following this, a number of new interventions were implemented through the pre, peri and post-operative process.

A dedicated preoperative education class, run by physiotherapy, occupational therapy and nursing staff includes assessment of home circumstances, an informative DVD and opportunity for questions. A standardised peri-operative anaesthetic regime is based on regional anaesthetic techniques with pre-operative analgesic and nutritional loading. Post-operative rehabilitation was also standardised with slings for comfort only and early safe-zone mobilisation. Patients were encouraged to be out of bed in the afternoon or evening immediately after surgery.

For all patients, length of stay as well as mean Oxford and Constant scores were quantified and patient satisfaction qualified through an 8-item, 32-point survey.

Mean length of stay improved from 2.4 to 1.9 nights in hospital. Of those patients enrolled in ER, 46% went home the day after surgery compared to 39% in the control group.

Mean 3-month Oxford and Constant scores did not significantly change.

Satisfaction was already high before ER but scores stayed the same or improved across all areas surveyed.

Our Enhanced Recovery programme benefited patients and the Trust by reducing time in hospital and improving patient satisfaction without an adverse effect on complication rate.

ANTIBIOTIC-LOADED BIODEGRADABLE BEADS WITH CEMENTLESS TSA FOR REVISION OF INFECTED SHOULDER ARTHROPLASTY

G Panagopoulos, A Leonidou and O Levy

Reading Shoulder Unit, Royal Berkshire Hospital, Reading

Background: Treatment of infected shoulder arthroplasty is controversial: Single versus two-stage revision; use of antibiotic-loaded cement requires the use of cemented implants; use of antibiotic-loaded cement beads requires another procedure for bead removal. We introduced a novel technique, using antibiotic-loaded calcium sulfate biodegradable beads (ALCSB) in single stage or two-stage revision TSAs with cementless implants.

Design & Methods: Seven patients underwent revision TSA for infection or suspected infection between 2010–2017, using ALCSB. Following removal of infected implants and meticulous debridement/washout, ALCSB were inserted into the bones before re-implantation with cementless new implants and were also spread between the soft tissues. Demographics, radiographs, and surgery data were prospectively collected. At follow up, Constant Score (CS), Subjective Shoulder Value (SSV), range of motion and patient satisfaction were recorded. Radiographic analysis was performed in every follow-up.

Results: Mean age was 63.8 years (range, 51–78). Six underwent single-stage revision (1 TSA to rTSA, 2 resurfacing to rTSA, 1 infected plating to rTSA and 2 resection arthroplasty (post infection) to rTSA). One case, initial attempt of single-stage exchange of modular components for immediate post-op infection failed, and a formal 2-stage revision was successfully performed. Perioperative cultures grew Propionibacterium acnes in 3 cases; Staphylococcus aureus in 3 cases, in one case cultures failed to grow an organism. Mean follow up was 30.1 months (range, 14 to 63). Infection successfully eradicated in all, with no recurrence at final follow up. Good function was achieved. Mean CS improved from 31 to 66. Active movements improved significantly with elevation increased from 76o to 136o. On radiographs, all beads resorbed 3–6 weeks post op. No lucencies around the implants, loosening, subsidence or stress shielding were seen.

Conclusion: The use of biodegradable antibiotic-loaded calcium sulfate beads is an effective and practical option in revision TSA for periprosthetic shoulder infections.

THE RELEVANCE OF CRITICAL SHOULDER ANGLE IN THE SECONDARY CUFF FAILURE AFTER SHOULDER REPLACEMENT

SK Gill, S Haines, P McCann and P Sarangi

Bristol Royal Infirmary, Bristol

The relevance of Critical Shoulder Angle in the secondary cuff failure after shoulder replacement

Introduction Previous studies have shown that patients who have develop osteoarthritis have been associated with a decreased critical shoulder angle (<28°) compared to normal subjects. An increased critical shoulder angle (>35°–38°) has been associated with patients with a rotator cuff tear compared to normal subjects. Secondary cuff failure after shoulder replacement is disabling and often requires additional surgery. The aim of this study was to look at the critical shoulder angle (CSA) in patients undergoing revision for cuff failure following shoulder replacement, to see if there was an association between an increased CSA and failure rate.

Materials and methods All the patients from a single centre were reviewed from 2011–2017. Patients who were revised to a reverse total shoulder replacement following cuff failure after a hemiarthroplasty or anatomic shoulder replacement were compared to an age and sex match control group. The CSA was calculated on X-rays taken before the initial surgery.

Results: Eight patients with symptomatic cuff failure after anatomic shoulder replacement and 9 patients with symptomatic cuff failure after a hemiarthroplasty (mean time to revision 4.4 years) were compared to a control group of 17 age and sex matched patients showing no signs of symptomatic cuff failure after a minimum of 2 years from the time of surgery. The mean CSA in the study group was 32.9 +/- 3.0. The mean CSA in the control group was 29.3 +/- 2.8. (p = 0.05).

Conclusion: The mean CSA was higher in patients who had undergone shoulder replacement and experienced secondary cuff failure compared to those in whom the same procedure had been successful, although this was not statistically significant. Further studies are required to look into this further.

PREDICTORS OF LENGTH OF STAY FOLLOWING SHOULDER ARTHROPLASTY IN A HIGH VOLUME UK CENTRE

P Dacombe, L Harries, P McCann, M Crowther, I Packham, P Sarangi and MR Whitehouse

Avon Orthopaedic Centre, Southmead Hospital; Musculoskeletal Research Unit, Bristol Medical School, Bristol

Objective: To describe the length of inpatient stay following shoulder arthroplasty in a high volume UK centre and identify predictors of prolonged inpatient stay.

Methods: All shoulder arthroplasty cases performed between 2012–2018 were identified. A review of case notes, electronic patient record and pathology systems was completed to identify demographic data, Charlson co-morbidity score, length of inpatient stay and factors associated with length of stay.

Descriptive statistical analysis was performed on all variables. Multiple regression analysis was conducted to determine which factors were associated with the length of inpatient stay. Independent variables included in the multiple regression model were age, gender, ASA, Charlson score, trauma or elective, primary or revision, indication for surgery, procedure, side, hand dominance, operating surgeon grade, previous surgery, fixation of humeral component, intraoperative complication and transfusion. Multicollinearity was assessed and addressed where it was identified (R2 > 0.75).

Results: 640 shoulder arthroplasty cases were performed in 566 patients. 451 cases (70%) were performed in female patients; the mean age was 72 years (24–95). The median length of stay were 2 days, the mean was 4 days (0–71). 430 patients (67%) had an inpatient stay of 3 days or less, whilst 83 (13%) required an inpatient stay of greater than 7 days.

The multiple regression model significantly predicted the length of stay (p < 0.0001). The length of stay was predicted by the following variables: age (p = 0.033), gender (p = 0.021), chronic kidney disease (p = 0.0008), congestive cardiac failure (p = 0.035), previous myocardial infarction (p = 0.030), intraoperative complication (p = 0.014) and postoperative transfusion (p = 0.0006). None of the other variables were associated with length of stay.

Conclusion: Median length of stay following shoulder arthroplasty was 2 days, the mean was 4 days. Increasing age, female sex, chronic kidney disease, congestive cardiac failure, previous myocardial infarction, intra-operative complication and transfusion were independent predictors of increased length of stay.

A TWO ARM RANDOMISED CONTROL TRIAL USING RADIOSTEREOMETRIC ANALYSIS(RSA) TO COMPARE IMPLANT MIGRATION OF HA COATED AND NON HA COATED METAL BACKED GLENOID COMPONENTS IN ANATOMIC PRIMARY TOTAL SHOULDER REPLACEMENT

L Cunningham, IA Trail, AC Watts and M Moffatt

Upper Limb Research Unit, Wrightington Hospital, Wigan

Background: NJR statistics show 1,884 PA-TSR were undertaken in the UK in 2017 with the frequency steadily increasing since 2012.

Metal-backed glenoid components are an alternative to cemented polyethylene glenoid components aiming to achieve more stable fixation and corresponding increased implant survival. The LIMA SMR metal-backed glenoid component is available in standard porous titanium, and hydroxyapatite (HA) coated versions. Both of these are available in the UK and are designed to be used un-cemented with additional screw fixation.

The purpose of the current study is to compare the amount of implant migration in both types of the glenoid component using RSA during a 2 year follow-up period.

Methods: Ethics Committee approval granted. A 2 arm RCT recruited and randomised 33 participants having PA-TSR. No formal sample size calculation was performed, however small sample sizes have been shown to allow for robust statistical analysis when using RSA.

RSA images were taken immediately post-operatively and at 3, 6, 12 and 24 months using standard uniplanar RSA technique with the UmRSA system and computed radiography (CR). Image analysis was performed using Model Based RSA software from RSAcore, using Computer Aided Design models supplied by LIMA.

Results: Two participants from the non-HA arm were excluded from the analysis due to imaging quality factors. There were 16 patients remaining in the non HA group and 12 in the HA coated group. The measured translation (in mm) in the X, Y & Z axes were compared between the two groups. Statistical comparison shows no significant difference in migration between the two groups.

Conclusion: Within the limits of RSA analysis, that is accurate to 1 mm, little movement of the prosthesis is seen in either group and there appears to be no additional benefit of using HA coating.

COST-EFFECTIVENESS OF THE REVERSE TOTAL SHOULDER ARTHROPLASTY. DOES INDICATION AFFECT OUTCOME?

R Jones, JA Nicholson, D MacDonald, IDM Brown and J McBirnie

University of Edinburgh, Edinburgh

Background: Few prospective studies explore the cost-effectiveness and patient satisfaction of the Reverse Total Shoulder Arthroplasty (RTSA) over a range of indications.

Objectives: To investigate patient satisfaction and cost-effectiveness following RTSA for cuff arthropathy, osteoarthritis with rotator cuff deficiency, and revision from previous arthroplasty.

Study design and methods: Patients who underwent RTSA over a two year period were prospectively monitored over the first post-operative year. All patients completed an Oxford questionnaire, DASH score and EQ-5D. Satisfaction was determined by the Net Promoter Score (NPS). Complications and use of health care resources were recorded to estimate the cost-effectiveness of the intervention. The incremental cost effectiveness ratio (ICER) was used to express the cost per quality-adjusted life year (QALY) gained.

Results: 67 patients underwent RTSA (44.8% OA with incompetent cuff, 23.9% cuff arthropathy, 31.3% revision arthroplasty). Median age was 72.4 years (IQR 46.2) and the majority were female (73.1%). Those who had cuff arthropathy were significantly older (p = 0.003). There was a significant improvement in the EQ-5D score at one-year post-operatively (pre-op 0.23, 1-year 0.78, p = 0.002). Shoulder scores also showed significant improvement (DASH pre-op 33.0 vs. 1-year 63.6, p < 0.001; Oxford pre-op 19.0 vs. 1-year 38.5, p < 0.001). The revision arthroplasty group had the lowest of the functional and health index gains of the cohort. There was one post-operative superficial infection and one scapula stress fracture requiring revision fixation. The total cost of the procedure and surrounding care was estimated as £6,699/patient. The median change in EQ-5D for the whole cohort was 0.55, giving an ICER of £12,180/QALY at 1-year post-operatively. The NPS for the cohort was 69 (73% promoters, 22.4% passive, 4.5% detractors).

Conclusions: RTSA provides a cost-effective intervention with excellent patient outcomes and satisfaction. The greatest patient benefit was derived from primary procedures with rotator cuff deficiency.

REVISION TOTAL ELBOW ARTHROPLASTY; IS IT SAFE TO PERFORM A SINGLE-STAGE REVISION FOR PRESUMED ASEPTIC LOOSENING BASED ON CLINICAL ASSESSMENT, NORMAL INFLAMMATORY MARKERS AND A NEGATIVE ASPIRATION?

AI Eyre-Brook, MJ Gandhi, P Gopinath, V Jones, E Williams, R Townsend, DS Thyagarajan, D Stanley and AA Ali

Northern General Hospital, Sheffield

Background: Revision Total Elbow Arthroplasty (TEA) is a challenging procedure that is becoming increasingly common. In our unit the cause of TEA loosening is based on history, examination, radiographs, inflammatory markers and joint aspiration. Open biopsies are not routinely performed. Our aim was to ascertain whether it is safe to perform a single-stage revision for presumed aseptic loosening on this basis. We also investigated whether there was any difference between single- and dual-component revisions in this group.

Methods: We conducted a retrospective review of consecutive Revision TEAs performed in our unit over a 10-year period. Case notes, radiographs, blood results, aspiration results and microbiology of intraoperative tissue samples were reviewed. Single-stage Revisions performed for presumed aseptic loosening were identified.

Results: A total of 123 Revision Elbow Arthroplasty cases were performed in this period. Fifty-eight cases revised for preoperatively proven infection, instability or implant failure were excluded from this study.

In 65 cases aseptic loosening was diagnosed based on history, clinical examination, blood markers and aspiration. There were 22 dual-component and 43 single-component revisions. In the dual-component revision group, two cases (9%) had positive results on 14-day culture from tissue samples taken at time of revision surgery. In the single-component revision group, positive culture samples were present in three cases (7%). Chi-squared analysis showed no significant difference between single- and dual-component revisions (p = 0.76).

No cases with positive culture samples from either group have required subsequent revision surgery.

Conclusion: Given the results of this study we conclude that it is safe to perform single-stage revision arthroplasty for implant loosening based on history, examination, normal inflammatory markers and negative aspiration results without the need for open biopsies. There is no difference if dealing with single- or dual-component loosening.

LONG-TERM REVISION RISK FOLLOWING PRIMARY TOTAL ELBOW REPLACEMENT: A NATIONWIDE POPULATION BASED COHORT STUDY OF 8682 CASES FROM HOSPITAL EPISODE STATISTICS DATA

RS Craig, JCE Lane, D Furniss, CP Little, GS Collins and JL Rees

Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford

Purpose: Because the NJR has only been recording total elbow replacements (TER) since 2012, we used a much larger data set from Hospital Episode Statistics (HES) to establish real-world estimates of revision risk following TER surgery.

Methods: We analysed data for all TERs recorded in HES between 1998 and 2017, including only patients aged 18 and older. The primary outcome was implant revision risk (Kaplan-Meier method) at 5, 10 and 15 years. The secondary outcome was standardised annual incidence of TER procedures. Results were stratified by indication.

Results: 8682 TERs with a median follow-up of 6.4 years (IQR 3.1 to 10.5) were included. The median age was 68 (59 to 76) and 6299 procedures were performed in women. Indications for surgery were: inflammatory 3921, osteoarthritis (OA) 1605, acute trauma 1588, trauma sequalae 509, other 1059. The overall annual incidence of TER increased marginally from 1.1 per 100,000 to 1.2 per 100,000 over the study period but halved for inflammatory indications. 914 TERs were revised and a further 739 required other elbow operations. At 5, 10 and 15 years, overall revision rates were 8.1%, 13.7% and 20.0%. Revision rates by indication at 10 years were: acute trauma 8.8%, inflammatory 12.3%, OA 16.8%, other 18.2%, trauma sequelae 20%. Infection was the indication for 21% of these first-time revisions. Frequent non-revision operations included fracture fixations, washouts/debridements and ulnar nerve procedures.

Conclusion: Overall rates of primary TER surgery have remained static over a 19-year period, but acute trauma is now a more common indication than inflammatory arthropathy. Revision rates are higher than those from designer series, but similar to those from arthroplasty registries. More research on adverse events is needed in order to reliably inform patients on expected outcomes, while prospectively collected NJR data will be important in assessing outcome variations between different implants.

DISTAL BICEPS TENDON RUPTURE: OUTCOMES FOLLOWING DELAYED REPAIR WITH AUTOLOGOUS HAMSTRING GRAFT

RJM Morrison, S Venkatachalam, SD Muller and J Candal-Couto

Department of Trauma and Orthopaedics, Northumbria Healthcare NHS Foundation Trust, Northumberland

There are few studies reporting the clinical and patient-reported outcomes following reconstruction of chronic distal biceps tendon ruptures using an autologous hamstring graft; we present our series of patients using this technique.

All consecutive patients with a delayed presentation to our unit following distal biceps tendon rupture were included in this study. Patients had surgery under general anaesthesia, and a semitendinosus-gracilis tendon graft was harvested by a knee surgeon. The graft was attached to the radial tuberosity through drill holes with Fibrewire, using Morrey’s two-incision approach, then sutured into the residual biceps tendon utilising a weave-through technique. Patients were placed in an above-elbow backslab for two-weeks, then changed to an above-elbow split cast for six weeks to allow physiotherapy-guided rehabilitation involving passive then active movements. Post-operative patient-reported outcomes were measured using the Quick-DASH and Mayo Elbow Performance Score (MEPS) to assess arm and elbow function, and the Lysholm score to assess knee function.

Eleven patients underwent delayed reconstruction in our unit between 2013 and 2018. All were male with an average age of 41 years (range 28–53) and underwent surgery a median of 17 weeks (range 6–261) following injury. Six patients completed post-operative outcome scores at an average of 30 months following surgery, with a median quickDASH score of 1.15 (0 = no disability, 100 = severe disability), ‘excellent’ elbow function according to the MEPS, and ‘excellent’ function in the knee using the Lysholm score. One patient reported ongoing elbow pain following surgery.

We suggest delayed reconstruction of distal biceps tendon ruptures using an autologous hamstring graft has good clinical and patient-reported outcomes and should be considered as a surgical option for the treatment of this injury.

A SYSTEMATIC REVIEW OF THE COMPLICATIONS OF CONTEMPORARY TOTAL ELBOW ARTHROPLASTY

P Parker, ND Furness, TJ Batten, WJ White and CD Smith

Royal Devon and Exeter Hospital, Exeter

Background: Total Elbow Arthroplasty (TEA) is a low volume procedure. Indications and implants have evolved in recent times. We aimed to evaluate complication rates associated with the most frequently used contemporary implants and for the most common indications.

Methods: A systematic literature search was undertaken of all studies reporting complications following TEA. Inclusion criteria were studies in English with a 12-month minimum follow up. Quality of the included studies was assessed with the Methodological Index for Non-Randomised studies criteria. British NJR data was used to identify the commonly implanted prostheses in the UK and the most common indications. The complication rates for all undesirable events contributing to patient outcome were extracted and cumulative rates calculated.

Results: 117 studies were identified and screened with 25 studies included, totalling 1185 procedures. Mean follow up time was 3.8 years. The overall complication rate was 41.7%, significantly higher than that seen following arthroplasty in other joints and included a 2.3% deep infection rate. Nerve injury was comparable between implants at around 4.4%. Aseptic loosening has a cumulative rate of 16.8%. Bushing wear in the Coonrad-Morrey group was 6%. Loosening of the radial component in the Latitude significantly contributed to an overall rate of aseptic loosening of 46%. Periprosthetic fracture was notably higher with the Discovery implant at 8.4%.

Conclusion: This is the largest systematic review of the complications of contemporary total elbow arthroplasties. Surgeons should be aware of differing complications related to their implant of choice with each having its own specific complication. Trauma and inflammatory arthropathy appear to have similar complication rates. However, there is a paucity of literature on the complication rates associated with the indication of osteoarthritis.

COMPLICATIONS FOLLOWING DISTAL BICEPS TENDON REPAIR

M Amarasooriya, I Karim, T Roper, K Ruxton, GI Bain and J Phadnis

Brighton and Sussex University Hospitals NHS Trust, Brighton; Flinders Medical Centre, South Australia

Aim: The aim of this study was to report the rate of complications following repair of complete distal biceps ruptures, and to analyse risk factors and outcomes of the individual complications.

Methods: Studies published in English on primary repair of the distal biceps between January 1998 and January 2019 were identified and a systematic review was performed according to PRISMA guidelines. Complications were classified as major and minor for analysis.

Results: 72 articles including 3091 primary distal biceps repairs were identified. The overall complication rate was 24% (n = 754).

The major complication rate was 4% (n = 126) and included a 1.7% (n = 51) rate of posterior interosseous nerve injury; 0.4% (n = 10) of median nerve injury; 1.4% rate (n = 43) of re-rupture and a 0.1% (n = 4) rate of synostosis. Brachial artery injury, ulnar nerve injury, compartment syndrome, proximal radius fracture and CRPS occurred at a rate of < 0.1% each. The vast majority of nerve injuries resolved with an expectant approach.

The minor complication rate was 20% (n = 621). The most common complication was Lateral cutaneous nerve (LCN) injury (rate 9.2% n = 283).

An extensile single incision approach was associated with higher rate of superficial radial nerve injury (6% vs 2.1%, p = 0.002) and a limited anterior single incision technique had a higher rate of LCN injury (9.7% vs 5.2%, P = 0.03). Synostosis only occurred with a double incision approach.

Fixation technique had no significant effect on the rate of complications, although there was an increased trend of nerve injuries with cortical button fixation and re-rupture with suture anchor fixation.

Conclusion: This is the largest analysis of complications following distal biceps repair indicating a major complication rate of 4%. This study provides valuable data with regard to choice of technique, surgical approach and rate of complications which is essential for surgical planning and patient consent.

PRIMARY OPEN ARTHROLYSIS IN POSTTRAUMATIC ELBOW STIFFNESS

A Murray, DLJ Morris, T Cresswell, M Espag, AA Tambe and DI Clark

Derby Shoulder Unit, Derby Teaching Hospitals, Derby

Purpose: Stiffness is a sequelae of elbow trauma. Arthrolysis may be considered to increase range of movement (ROM). Little is published on elbow arthrolysis outcomes. We present our series of primary open arthrolysis in posttraumatic elbow stiffness.

Methods: Consecutive patients that underwent primary open arthrolysis for posttraumatic elbow stiffness in our unit between 2011–2018 were analysed. Procedures were performed by an elbow surgeon with metalwork removed concurrently as required. Postoperative rehabilitation was as per standard protocol with early motion; continuous passive motion (CPM) was utilised if requested. Data collected included patient demographics, traumatic injury type, arthrolysis technique, preoperative and postoperative elbow ROM, complications and postoperative Oxford Elbow Score (OES).

Results: 41 patients were included. 59% were male. Mean age at time of arthrolysis was 43 years (range 12–79 years). Mean duration of follow-up was 53 months (range 8–100 months). Median duration from time of injury to arthrolysis was 11 months (range 2–553 months). Initial injury necessitated surgery in 78%.

Mean preoperative flexion-extension arc (FEA) was 70°, improving to 104° postoperatively (p < 0.001). Mean preoperative pronosupination arc (PSA) was 125°, improving to 165° postoperatively (p < 0.001).

Mean postoperative OES was 37 (n = 28). Complication rate was 24% with 7 recurrence requiring surgery, 2 nerve injuries and 1 infection.

CPM was utilised in 10 patients. In this group, mean FEA improvement was 56°. Complication rate for these patients was 40%.

Severe traumatic injury (fracture-dislocation, intraarticular humerus fracture) was associated with increased preoperative stiffness (FEA 61° vs 84°, PSA 111° vs 149°) but larger improvements in ROM (postoperative FEA 98° [p < 0.001], PSA 165° [p < 0.001]).

Conclusion: This series demonstrates improvement in elbow ROM following open arthrolysis; with moderate results in a patient-reported outcome measure. Patients considering arthrolysis should be counselled regarding expectations and complications.

MIDTERM OUTCOME OF REVERSE V OSTEOTOMY IN THE TREATMENT OF CUBITUS VARUS IN CHILDREN AND ADOLESCENT

S Hussain, Khalid, K Chin and S Lambert

Lady Reading Hospital Peshawar Pakistan, District Headquarter Hospital Timergara KPK Pakistan, Akbar Medical Centre Peshawar, Pakistan

Aim: The aim of the study is to report the outcome of corrective osteotomy for cubitus varus in children and adolescent.

Method: Twenty-three patients with cubitus varus deformity underwent corrective reverse V osteotomy as described by Yun et al through posterior triceps splitting approach. All osteotomies were stabilised using 2 crossed K-Wires. The average time since injury was 4.7 years and the mean age at operation is 11.9 (5–15). The mean follow-up time was 5 years (5–12 years). The outcomes were evaluated using the humeral-elbow-wrist (HEW) angles, lateral condyle prominence index (LPI) and Oppenheim’s criteria.

Result: The mean Hew angles improved from −20 degrees (range: −15 to −28) pre-operatively to +6 degrees (range: 5 to7) post-operatively (mean difference: 27 degrees; p < 0.01). The mean post-operative LPI was −0.80 (- 0.95% to + 6.58%). There were no early, nor late complications. There is no significant difference in the pre and post-operative range of motion (mean: 0–130). Sixteen patients were graded as excellent and 7 as good on the Oppenheim’s criteria.

Conclusion: Correction of elbow cubitus varus with reverse V osteotomy and two cross K-wires is a mechanically stable and reproducible technique. Significant correction with optimal residual lateral condylar prominence can be achieved.

CLINICAL AND RADIOGRAPHIC OUTCOMES OF SALVAGE RADIAL HEAD ARTHROPLASTY

J Singh, Z Hamoodi and AC Watts

Wrightington Hospital, Wigan

Introduction: There is limited evidence on the outcome of radial head arthroplasty (RHA) as salvage. The aim of this study is to examine the medium term outcomes of salvage RHA.

Method: A retrospective review was conducted of fourteen consecutive patients (9F/5M) with an average age of 48 years having revision RHA from 2011 to 2018. The mean follow-up was 27 months (6–84).

The outcome measures were implant removal, range of movement (ROM), Mayo Elbow Performance Score (MEPS) and radiographic loosening.

Results: Twelve patients had revision of a primary radial head replacement. Two patients had radial head arthroplasty as salvage for excisional arthroplasty complicated by stump impingement or valgus instability. Painful loosening of a radial head component was the most common indication for revision (6 patients). Four patient had chronic longitudinal instability and all had interosseous membrane reconstruction in addition to RHA.

Eight patients had revision surgery using a cemented long stem bipolar implant, four with a short stem uncemented implant and 2 with a long stem uncemented implant.

Only 1 of 14 patients had a revision failure and underwent further revision surgery from a short uncemented to a long cemented implant that resulted in an excellent outcome (MEPS score 100). On average patients had a good functional outcome with an average MEPS of 77 (40–100) and average ROM from 24 to 121 degrees of flexion.

Radiographic assessment showed evidence of loosening in 3 patients with short stem uncemented implants. One patient required revision and 2 have non-progressive asymptomatic loosening. Three patients had capitellum erosion. Seven patients have ulnahumeral osteoarthritis and one is currently awaiting total elbow replacement.

Conclusion: The overall midterm outcomes with revision radial head surgery are satisfactory. Cemented implants may produce a more favourable outcome in revision surgery.

REVISION OF INFECTED ELBOW ARTHROPLASTY –ECONOMIC BURDEN IN RELATION TO THE NATIONAL TARIFF

R Shenoy, D Butt, W Rudge, D Higgs, M Falworth and A Majed

Royal National Orthopaedic Hospital, Stanmore

Aim: The aim of this study was to evaluate the hospital cost of revision for infected elbow arthroplasty in relation to the Health Resource Group (HRG) national tariff and compare this with primary elbow arthroplasty, highlighting the ‘Getting It Right First Time’ (GIRFT) principle in elbow arthroplasty.

Methods: Within a six year study period 81 primary elbow arthroplasties and 27 revisions for infected elbow arthroplasties were performed. Twenty patients had a first stage or a single stage revision and fifteen patients had a second stage revision. Data collected included length of hospital stay and number of outpatient visits. HRG tariff for these episodes were recorded and the total including market forces factor (MFF) were calculated.

Results: Mean hospital stay of patients who had primary elbow arthroplasty, first stage and second revision of infected elbow arthroplasty were 4, 15 and 7 days respectively. These episodes attracted a mean tariff of £6307, £9558 and £13114 respectively with a mean tariff of £24444 for patients who had a two stage revision. Mean outpatient visits per patient for primary elbow arthroplasty were 5 while this was 21 for patients with infection related revision arthroplasty. Excluding cost of implants and high cost antibiotics, the mean total tariff including inpatient stay and outpatient visit tariff for a patient having primary elbow arthroplasty was £6602. The corresponding tariff for patients undergoing a two stage revision of infected elbow arthroplasty was £29183 including a six week course of standard antibiotics.

Conclusion: Revision of infected elbow arthroplasty results in additional hospital cost of over 442%, with the eventual financial burden to the healthcare system being higher when costs of implants and antibiotics are included; this study supports need to maintain high standards during primary elbow arthroplasty.

EARLY OUTCOMES OF RADIAL HEAD REPLACEMENT FOR TRAUMA IN YOUNG AND MIDDLE-AGED PATIENTS

K Kulkarni, G Johal, R Anakwe, P Reilly and D Griffiths

Imperial College Healthcare NHS Trust, London

Purpose: Evaluation of early outcomes in patients aged under 60 years following radial head replacement for trauma.

Methods: Retrospective review of 31 consecutive young and middle-aged patients with radial head fractures who underwent Anatomic Radial Head Replacement (Acumed, US) at a single institution between 2014 and 2017. Patients assessed post-operatively for range of motion (ROM), function (Oxford Elbow Score, OES), and radiographic change by an independent assessor.

Results: 9 patients were excluded due to incomplete records (29%). The remaining 22 patients had a mean age of 42 years (range 24 to 58) and mean follow-up of 27 months (range 8 to 55). The radial head fractures were associated with dislocation in 8 and polytrauma in 3 (1 patient with both). All 8 dislocations underwent acute lateral ligament repair and 7 patients underwent additional fixation of the elbow or forearm. 2 cases were revisions to a radial head replacement following failed fixation.

Mean ROM was 12 to 126 degrees (range 0–135). 20 regained full rotation. Mean OES was 43.2 (range 33 to 48), with no significant difference in clinical outcomes seen in those with associated dislocation or associated other trauma. Radiographic evidence of heterotopic ossification was observed in 5, and osteolysis around the implant was observed in 5 (2 patients with both). Subset analysis did not show any correlation with functional outcomes. No capitellar wear was observed.

3 patients underwent further elbow surgery (1 stem cerclage wire removal, 1 prominent suture removal, 1 ulna nerve decompression). 1 excluded patient (with open polytrauma and neurovascular injury) required implant removal for aseptic loosening.

Conclusion: Early outcomes of anatomic radial head replacement are encouraging. Longer-term data is required to assess progression of osteolysis around the prosthesis and the overall prosthesis longevity in this active population.

THE ACCURACY OF COMPUTERISED TOMOGRAPHY FOR CLAVICLE NONUNION EVALUATION

JA Nicholson, BJ Fox, D Rishi and CM Robinson

University of Edinburgh, Edinburgh

The use of computed tomography (CT) is increasingly employed to determine bone union, however the accuracy with regards to clavicle nonunion has not been evaluated to date.

Methods: A retrospective review of 10 years (2007–2017) of CT scans performed for suspected nonunions of midshaft displaced clavicle fractures were reviewed. The influence of scan timing, callus finding and patient characteristics on diagnosis accuracy was explored. Blinded review of CT and radiographs were performed to determine if CT had better agreement over radiographs for nonunion diagnosis.

Results: 184 clavicle CT scans were evaluated (85 unions and 99 suspected nonunions). No patient was incorrectly diagnosed with union. Of the 99 suspected nonunions, 19 patients were incorrectly diagnosed, or represented delayed unions, with union confirmed at operation (n = 7) or on repeat imaging (n = 12). Time from injury (or operation date) to CT scan, comminution, post-fixation or smoking did not affect the accuracy of diagnosis.

The majority of the nonunions were atrophic (n = 57, 71.3%), however all of the delayed unions were hypertrophic (p < 0.001), resulting in a 54.8% positive predictive value for correct diagnosis of nonunion with a hypertrophic callus. CT evaluation compared to radiographs showed greater inter-observer agreement for the evaluation of bridging callus (Weighted Kappa 0.75 vs 0.50 respectively). Overall the accuracy of CT is 100% sensitive to determine union and 81.7% specific for the diagnosis of nonunion.

Discussion: CT has excellent accuracy at determining clavicle union but misdiagnosed approximately one in five nonunions. Hypertrophic nonunions carried the greatest risk of incorrect nonunion diagnosis.

RADIAL HEAD REPLACEMENT FOR ACUTE FRACTURES: A STUDY OF LONG-TERM OUTCOMES

CD Cristofaro, TH Carter, NR Wickramasinghe, MM McQueen, TO White and AD Duckworth

Edinburgh Orthopaedic Trauma

Purpose: This study reports the long-term complications and patient reported outcomes following radial head replacement (RHR) for acute complex fractures of the radial head.

Methods: We retrospectively identified 119 patients over a 16-year period who underwent primary RHR for an acute complex radial head fracture. We reviewed electronic records to document post-operative complications, including prosthesis revision or removal. Patients were contacted to confirm complications and long-term patient reported outcomes. The primary outcome measure was the QuickDash.

Results: The mean age was 50yrs (16–94) and 63 (53%) were female. There were 102 fractures (85.7%) classified as Mason III injuries, with 28 injuries associated with an elbow dislocation (19 terrible triad) and 32 associated with a proximal ulna fracture. Most implants were uncemented ‘loose-fit’ monopolar prostheses; 86% (n = 102) metallic and 14% (n = 17) silastic. Thirty patients (25%) required revision surgery (n = 3) or prosthesis removal (n = 27) at a median of 7 months (0–125), with 70% (21/30) of these occurring within one year post-operatively. At a mean of 12yrs (7.5–23.5), 80% (80/100; 19 deceased) of patients were contacted. The median QuickDASH was 6.8 (range, 0–63), the median OES was 46 (range, 21–48), and the median EQ-5D was 0.8 (range, -0.3–1.0). Overall satisfaction was high with a median of 10 (range, 2–10). Median return to sport was 20 weeks (range, 4–104) and return to work was 6 weeks (range, 0–100). There was no significant difference in outcome for those patients requiring revision/removal surgery (all p > 0.05).

Conclusion: This is the largest series in the literature documenting the long-term patient reported outcome after acute RHR. Despite a quarter of patients requiring further surgery, RHR is supported by positive long-term results for the treatment of complex radial head fractures. The peak incidence of prosthesis revision or removal occurs within the first year following implantation.

OPEN REDUCTION AND INTERNAL FIXATION OF CLAVICLE FRACTURES AFTER 3 MONTHS IS ASSOCIATED WITH AN INCREASED RISK OF COMPLICATIONS AND REVISION SURGERY

JA Nicholson, H Gribbin, ND Clement and CM Robinson

University of Edinburgh, Edinburgh

Background: The primary aim of this study was to determine if delayed clavicle fixation results in a greater risk of operative complications and revision surgery.

Methods: A retrospective case series was undertaken of all displaced clavicle fractures that underwent plate fixation over a 10 year period (2007–2017). Patient demographics, time to surgery, complications and mode of failure were collected. Logistic regression was used to identify independent risk factors contributing towards operative complications. Receiver operating curve (ROC) analysis was used to determine if a potential ‘safe window’ exists from injury to delayed surgery. Propensity-score-matching was used to construct as case control study for comparison of risk.

Results: 259 patients were included in the analysis. Post-operative infection occurred in 3.9% of all patients, the only variable associated was a greater time interval from injury to fixation (p = 0.001). Failed primary surgery requiring revision fixation was required in 7.7% of the cohort with smoking (p < 0.001), presence of a post-operative infection (<0.001), increasing age (p = 0.018), and greater time delay from injury to surgery (p = 0.015) identified as significant independent predictors on regression analysis. ROC analysis revealed that surgery beyond 96 days from injury increased the rate of major complications and revision surgery. Using a matched case cohort of cases before (n = 67) and after the ‘safe window’ (n = 77), the risk of post-operative infection increased (Odds ratio (OR) 7.7, p = 0.028), fixation failure (OR 3.8, p = 0.017) and revision surgery (OR 4.8 p = 0.004).

Conclusions: Delayed clavicle fixation is associated with a significantly higher risk of major complications. A delay to primary fixation of up to 3 months following injury may be acceptable, beyond which there is an increased risk of major operative complications and revision surgery.

Level of Evidence: Therapeutic Level III.

REDUCING UNNECESSARY FIXATION OF MIDSHAFT CLAVICLE FRACTURES

N Murray, T Johnson, I Packham, M Crowther and T Chesser

North Bristol NHS Trust, Bristol

Purpose: The aim of this study is to report on a protocol established for the management of middle third clavicle fractures, identifying those that may go on to non-union whilst reducing the rate of unnecessary fixation.

Method: A protocol was introduced, where all isolated closed midshaft clavicle fractures were initially managed non-operatively in a sling. At 2 weeks, all patients were reviewed and those who were struggling with their symptoms proceeded to surgery with the remainder mobilised as comfortable. All cases treated at one centre over a three-year period, with a minimum follow up of one-year underwent case note review.

Results: Between 2015 and 2017, 613 clavicle fractures were managed through fracture clinic. 347 were middle third (56%) and of these 75% were male, mean age 41(range 16–97) and 225(65%) were displaced. 41 middle third clavicle fracture patients underwent early fixation, with 8 of these managed out-with the protocol and were excluded. A total of 33(9.5%) were managed operatively through the protocol with the number of days to operation 4–57(median 15 days). 11 patients required late fixation for symptomatic non or mal union, of these 4 had not been managed through the protocol. Of the 7 cases included time to surgery was 193–494 days (mean 378 days). 6 were symptomatic non-unions(1.7%) and 1 was a symptomatic malunion(0.3%).

Using the protocol the early operative rate was 10%(33 of 339 patients or 15% of the displaced fractures), the later non-union or symptomatic malunion rate was 2%(3% displaced). This lead to a total operative rate of 12% (18% displaced).

Conclusion: A reproducible protocol devised and introduced for managing middle third clavicle fractures in this hospital has demonstrated an effective means of managing these injuries which is cost effective, reducing the rate of fixation(10%) whist maintain low rates of symptomatic non and malunion(2%).

FUNCTIONAL OUTCOMES OF ELBOW INJURIES MANAGED ACCORDING TO THE WRIGHTINGTON CLASSIFICATION OF ELBOW FRACTURE DISLOCATIONS

Z Hamoodi, J Singh, M Elvey and AC Watts

Wrightington Hospital, Wigan

Objectives: The aim of this study is to review the functional outcomes of patients managed by application of the Wrightington elbow fracture-dislocation classification system and its corresponding management algorithm.

Methods: This is a retrospective case-series of all adult patients who were managed according the Wrightington classification. Images were retrospectively reviewed and classified by all investigators. Any disagreement was resolved by discussion. The content validity of the classification was assessed by comparing each subgroup classification with the intraoperative findings. Primary outcome measures were the range of movement (ROM) and Mayo Elbow Performance Score (MEPS). Complications were collected as a secondary outcome.

Results: Sixty patients were managed using the Wrightington classification (32F/28M) with a mean age of 48(19–84). In order of observed frequency 20 cases (33%) were classified as type B+,18(30%) Type C,11(28%) Type B and 7(12%) Type A. There were 3 Type D+ and 1 Type D.

Fifty eight (96.7%) patients completed a minimum of 3 months follow up. Average follow up was 5.7 months (2–18). The average MEPS at final follow up was 93(55–100) and mean ROM of 15 to 131 degrees. Sub-group analysis showed average MEPS of 91(65–100) in group A, 93(70–100) in group B, 92(55–100) in group B+,94 (65–100) in group C,100 in group D and 90(70–100) in group D+. The average ROM is 10–136 degrees in group A, 22–128 in group B, 17–127 in group B+, 12–134 in group C, 0–130 in group D and 18–131 degrees in group D+. Four patients underwent secondary surgery. All 4 patients had improved outcomes with average MEPS score improvement from 65 to 94.

Conclusions: The Wrightington Classification system facilitates pattern recognition and provides an algorithm for management for these complex injuries. Our results suggest that predictably good outcomes can be achieved by application of surgical algorithms related to this classification.

RADIOGRAPHIC AND CT CHARACTERISTICS OF CORONOID FRACTURES AND INTER-OBSERVER RELIABILITY OF CORONOID FRACTURE CLASSIFICATION

PA Cameron, H Syed and J Phadnis

Brighton and Sussex University Hospitals NHS Trust, Brighton

Purpose: To compare X ray measurement of coronoid fracture size and morphology with CT scans and to determine the inter-observer reliability of the Regan-Morrey and O’Driscoll classification systems.

Methods: 72 fracture dislocations of the elbow with the presence of a coronoid fracture were included. The coronoid fragment size was measured using plain radiographs and as a percentage of the intact coronoid. Paired CT scans on the same patients were used to perform the same measurements. Fractures were classified on X ray using the Regan-Morrey system and on CT using the O’Driscoll classification. Measurements were taken independently by two researchers and compared using Cohen’s Kappa method.

Results: The mean x ray coronoid fragment height was 5.5 mm (0.7 - 12.6), compared to 6.6 mm (1.8 - 14.7) on CT.

According to the Regan-Morrey classification, there were 37 Type 1; 31 type 2 and 4 type 3 fractures. For the O’Driscoll classification, there were 11 Type 1–2; 5 type 2–1 (isolated anteromedial); 54 type 2–2 and 2 type 2–3 fractures.

91% of O’Driscoll type 1 fractures were <5 mm in size on X ray measurement. The majority of fractures assessed (85%), involved the anteromedial facet of the coronoid. Fracture size >5 mm on x ray had a sensitivity of 0.59 and specificity of 0.91 for involvement of the anteromedial coronoid.

X rays underestimated the fragment size by a mean of 17% when compared to CT. Inter-observer reliability was moderate (k = 0.42) for Regan Morrey; and fair (k = 0.27) for the O’Driscoll system.

Conclusion: Coronoid fracture size and morphology is underappreciated on plain x rays. Fractures >5 mm are likely to involve the anteromedial part of the coronoid. Based on these findings, CT scans are recommended for all coronoid fractures. A simplified classification system with improved inter-observer reliability is desirable.

ROUTINE FIXATION OF DISPLACED MIDSHAFT CLAVICLE FRACTURES IS NOT COST EFFECTIVE: A COST ECONOMIC ANALYSIS FROM A RANDOMISED CONTROL TRIAL

JA Nicholson, ND Clement, E Goudie and CM Robinson

University of Edinburgh, Edinburgh

Aims: The primary aim of this study was to establish the cost-effectiveness of acute fixation of displaced midshaft clavicle fractures.

Patients and Methods: A cost economic analysis was conducted within a randomised control trial comparing conservative management (n = 92) to acute plate fixation (n = 86) of displaced midshaft clavicle fractures over 12-months. The incremental cost effectiveness ratio (ICER) was used to express the cost per QALY. The short-form 6 dimensional (SF-6D) score was used as the preference based index to calculate the cost per QALY at 12-months.

Results: The 12-month SF-6D of acute fixation was 0.9607 (95% CI 0.9447–0.9767) versus conservative management 0.9522 (0.9355–0.9689) giving an advantage of 0.0085 (p = 0.46). The mean cost per patient was £5,418.91 for acute fixation and £1,322.69 for conservative management. This gave a ICER of £481,907.26/QALY. For a threshold of £20,000/QALY the benefit of acute fixation would need to be present for 24.1 years to be cost-effective over conservative treatment. Linear regression analysis identified nonunion as the only factor to influence the SF-6D at 12-months (p < 0.001).

A propensity-score-matched cohort of patients that underwent delayed fixation for nonunion following conservative management (14.1%, n = 13) were compared against patients who underwent acute fixation. Nonunion patients had a significantly worse 12-month SF-6D score compared to acute fixation by 0.0912 (p = 0.018) with similar healthcare costs (acute fixation £170.13 more expensive). Modelling of acute fixation versus delayed fixation at 6 months for nonunion following conservative management resulted in a ICER of £1,865.46/QALY at 12-months.

Conclusion: Routine plate fixation of displaced midshaft clavicle fractures is not cost-effective. Nonunion following conservative management has increased morbidity with comparable expense to acute fixation. This may suggest a targeted approach of fixation in patients at high risk of nonunion would be more cost-effective.

WHEN IS EARLY RETURN TOO EARLY? CLAVICLE PLATE FIXATION IN AUSTRALIA RULES FOOTBALL ATHLETES

M Ricks, P Boras, M Perret, S Warby and G Hoy

Melbourne Orthopaedic Group, Melbourne, Australia

Purpose: To review the return to sport times and results for a series of professional Australian Rules football (AFL) players after clavicle fracture management using a clavicle plate and screws.

Intro: We allowed an early use and aimed to assess if this construct is strong enough to withstand collision sport activities before radiological union is confirmed.

Methods: We reviewed 13 consecutive senior AFL players who underwent plate over a 10 year period. The players were injured during competition, and were in a position to return to play during the same season. Outcome measures were taken at 36 months post operatively and included time to return to play, on field performance (number of kicks, marks, handballs, tackles, Fantasy Scores and Supercoach Scores, the Nottingham Clavicle Score (NCS), and a Subjective Shoulder Value (SSV).

Results: The median time to return to competition play was 3 weeks (range 2 to 16 weeks). All fractures eventually united. Three of the nine players (33%) that returned before 6 weeks post-operatively suffered significant impacts that caused a bend in the titanium plates. One of these required refixation surgically due to perceived weakness in the bent plate construct and cosmetic deformity. One had the plate removed electively at 1 with full union noted. The performance scores were the same or minimally reduced in the first 5 games after return.

Conclusion: The decision to operate early on AFL players who suffered a clavicle fracture in competition play resulted in excellent long term results. However, returning players before bony union resulted in the bending of plates in a significant number, and the risk of further injury must be weighed up in a collision sport. The early return to play time had no adverse affects on performance. Perceived functional status was high, as reflected by the NCS and SVV.

ALL SUTURE REPAIR FOR OLECRANON FRACTURE: RESULTS OF A CONSECUTIVE SERIES OF 30 PATIENTS

M Dunne, D Gill and N Kang

Addenbrookes, Cambridge

The purpose of this study was to assess the outcomes of a consecutive series of patients undergoing an all suture technique for Mayo Type II intra-articular fractures and osteotomies of the olecranon in our institution.

A retrospective analysis of 30 patients who were identified using the hospital electronic records system and surgeon’s records was carried out for a 2 year period. Successful outcomes were defined as a functional range of movement with radiographic union at or before 6 months post operatively. Re-operation for any reason was recorded.

A total of 4 osteotomies and 26 acute traumatic fractures were identified. Mean age was 52.1 years with 16 males and 14 females. Mean follow up was 4 months (Max 12 months) with all patients followed up to a minimum of three months. 2 patients required re-operation. One for failure due to a further fall causing failure of fixation and one for a granuloma secondary to the suture material used that was removed after bony union was confirmed. All remaining patients achieved radiographic bony union with a functional range of movement at final follow up.

The surgical management of olecranon fractures has remained a controversial topic with much debate as to the effectiveness of tension band constructs compared with plate fixation with high complication rates associated with metalwork prominence and wound infection associated with each. A novel technique for olecranon fracture fixation has been previously described and our results suggest that it is an effective technique for olecranon fracture fixation resulting in lower overall complication rates compared to reported studies in the literature.

PREDICTING HUMERAL SHAFT FRACTURE NON-UNION: THE RADIOGRAPHIC UNION SCORE FOR HUMERAL FRACTURES (RUSHU)

WM Oliver, T Smith, JA Nicholson, SG Molyneux, TO White, ND Clement and AD Duckworth

Royal Infirmary of Edinburgh, Edinburgh

Purpose: To develop a reliable and effective radiological score – the Radiographic Union Score for HUmeral fractures (RUSHU) – to identify patients at risk of non-union following a humeral shaft fracture.

Methods: Patients were retrospectively identified from a trauma database held at the study centre. Twenty randomly-selected patients with AP and lateral radiographs taken 6 weeks following their humeral shaft fracture were rated by 3 observers, with scoring based on the Radiographic Union Scale for Tibial fractures (RUST) system. After refinement of RUSHU scoring criteria, 60 sets of 6-week radiographs (40 patients who achieved fracture union, 20 who developed non-union) were rated by 2 observers blinded to patient outcome.

Results: The inter-observer intraclass correlation coefficient (ICC) was 0.68 (95% CI 0.46–0.85). This improved after refinement to 0.79 (95% CI 0.67–0.87), indicating substantial agreement. At 6 weeks post-injury, patients that went on to unite (n = 40, median RUSHU = 10) had a significantly (p < 0.001) higher score than those who developed non-union (n = 20, median RUSHU = 7). A receiver operating characteristic curve determined a RUSHU cut-off of <8 to predict non-union (area under the curve = 0.84, 95% CI 0.74–0.94). Sensitivity was 75%, specificity 80%, positive predictive value (PPV) 65% and negative predictive value 86%. Patients with a RUSHU < 8 (n = 23) were 12 times more likely to develop non-union than those with a RUSHU ≥ 8 (n = 37, odds ratio = 12.0, 95% CI 3.4–42.9). Based on a PPV of 65%, if all patients with a RUSHU < 8 underwent fixation, the number of procedures needed to avoid one non-union would be 1.5.

Conclusions: The RUSHU is reliable and effective in identifying patients at risk of humeral shaft fracture non-union at 6 weeks post-injury. The RUSHU requires validation in other centres, but may enhance decision-making in patients with a humeral shaft fracture and reduce the morbidity associated with the delayed treatment of an established non-union.

OUTCOME OF PROXIMAL HUMERAL FRACTURES TREATED WITH INTRAMEDULLARY NAIL FIXATION WITH MEDIAL STABILISATION USING A LOCKED CURVED BLADE: IS THIS THE FUTURE?

T Antonios, E Fawzy, D Karadaglis and S Orakwe

Institution Lewisham and Greenwich NHS Trust, Queen Elizabeth Hospital, London

Introduction: Proximal humeral fractures treatment remains challenging and controversial. Fixed-angle locking plate is currently the gold standard implant for surgical fixation. Although intramedullary nail fixation showed superior biomechanical properties and comparable clinical results to plate fixation; it remains less favorable due to fear of rotator cuff violation and failure to maintain medial buttress leading to varus collapse of the construct.

We report on a retrospective cohort of 72 patients treated using third generation antegrade straight nail with a locking blade. The blade provides medial buttressing and head triangular stability, reducing the risk of varus collapse. Compared to plate fixation, it is less invasive with no clinical complications related to the entry point being at the muscular part of the cuff medial to its insertion.

Methods: Seventy-two patients were reviewed. Mean age was 68.3 years (range 60–87). Minimally invasive technique was used. Based on Neer's classification 18%, 58%% and 24% were 2-, 3- and 4-part fractures, respectively. Clinical evaluation was carried out using Disability of Arm, Shoulder and Hand (DASH) score and Oxford Shoulder Score. Radiological assessment of time to union, metalwork complications and varus angulation were examined.

Results: Mean follow-up was 21 months (range 9–26). Except for one patient, all fractures went into union with an average healing time of 3.7 months (range 3–7). One patient had prominent metalwork needing nail removal. Axillary nerve neuropraxia was noted in two patients that resolved. One superficial wound infection was treated with antibiotics. DASH score was 28 and Oxford shoulder score was 42. There were no cases of varus malalignment or pins/blade back-outs. Ten patients suffered postoperative stiffness.

Conclusions: Intramedullary humeral nail is a reliable option for treating proximal humeral fractures. It is less invasive, has better biomechanical properties, and when combined with the blade, has the properties to resist varus collapse.

LONG-TERM FOLLOW-UP OF PLATELET-RICH PLASMA INJECTIONS FOR REFRACTORY LATERAL EPICONDYLITIS

M Brkljac, J Conville, U Sonar and S Kumar

Royal Lancaster Infirmary, Lancaster

Background: Lateral epicondylitis (LE) affects between 1 and 3% of the population. Normally treated non-operatively, more recently, platelet-rich plasma (PRP) has become popular.

Aim: We aim to assess the long-term outcomes of PRP injection for patients suffering from refractory LE.

Methods: We prospectively evaluated 34 patients. All patients had failed a trail of conservative management. PRP was prepared using the Arthrex ACP system. The Oxford Elbow Score (OES) evaluated functional outcomes.

Results: Thirty-one patients were included, the mean age was 46 years (range 33–61 years). The mean follow-up was 5.2 years (range 4.2–6.1 years). No complications occurred.

87.1% of patients exhibited minimum clinically important difference (MCID) in pain symptoms between pre-op and long-term scores. 90.3% displayed MCID in function and psycho-social domains.

Statistical significance was found using repeated one-way ANOVA analysis to compare means of OESs taken pre-op, post-op and at long-term follow up (p = <0.01).

Two patients had a repeat injection and six patients underwent open release. The 3.5 year survival free rate was 74.2% (95% CI = 4.1 to 5.6 years).

Discussion: A single injection of PRP is successful in treating refractory LE in most patients and avoiding surgery. PRP is safe and improves symptoms years are the initial injection.

THE ASSESSMENT OF GLENOID TRACK ON AN MR ARTHROGRAM IS PREDICTIVE OF FAILURE FOLLOWING ARTHROSCOPIC SHOULDER STABILISATION

E Crighton, G Butterworth, C Elias-Jones, AJ Brooksbank and PJ Jenkins

Glasgow Royal Infirmary, Glasgow

Aim: The aim of this study was to investigate whether assessment of the glenoid track on a pre-operative MR arthrogram could be used to predict the recurrence of instability in a typical UK population following arthroscopic stabilisation.

Methods: A retrospective study was undertaken of 175 primary arthroscopic stabilisation procedures of which 82% (n = 143) were men. The median age was 26 years (IQR 22 to 32, range 16 to 77). The median follow-up was 76 months (range 21 to 125). A pre-operative MR arthrogram was used to determine if the shoulder was on-track or off-track. The prevalence of “off-track” shoulders was 14.2% (n = 25). The endpoint of recurrent dislocation was examined.

Results: There were 6 (24%) dislocations in the off-track group compared with 5 (3.33%) dislocations in the on-track group (RR 7.2, 95% CI 2.45 to 20.5, p = 0.001). At 5 years, the cumulative (Kaplan-Meier) redislocation rate was 26.1% in the off-track group compared with 8.7% in on-track group. The rate of any recurrent instability was 60% (n = 15) v 18% (n = 27) (RR 3.33, 95% CI 2.02 to 5.20, p < 0.0001). Off-track shoulders were more common in older patients (mean age 34.2 v 27.6 yrs, p = 0.004), but there was no association with gender (p = 0.411).

Discussion: In a typical UK population assessment of the glenoid track on an MR arthrogram can be used to risk stratify patients with shoulder instability. This study has identified a subset of older patients with bone loss who appear to be at greater risk of subsequent instability following arthroscopic stabilisation.

RATIONAL USE OF SHOULDER MRI IN THE PRIVATE SETTING: HOW OFTEN DOES THE MRI ACTUALLY INFLUENCE CLINICAL MANAGEMENT?

R Devlin and RA Delaney

Institution Dublin Shoulder Institute, Dublin

Aim: The purpose of this study was to determine whether there is a significant difference in the proportion of shoulder MRIs that actually influence clinical management based on whether the MRI was ordered by a shoulder specialist orthopaedic surgeon or a primary care provider (a GP, a sports medicine physician or a physiotherapist).

Methods: Retrospective review of 153 MRIs performed on 151 patients in a private practice setting; 77 ordered by a shoulder surgeon, 76 by a primary care provider (PCP).

Results: Shoulder specialist-ordered MRIs influenced patient management significantly more often than primary care-ordered MRIs (82% vs. 22%, p < 0.01).

54% of referral letters from PCPs to a shoulder specialist did not have any documentation of a physical examination of the shoulder, yet an MRI had been ordered.

For primary care-ordered MRIs which did not have any influence on patient management, the most common diagnoses were subacromial bursitis (39%) and adhesive capsulitis (24%).

Conclusions: Primary care-ordered shoulder MRIs influenced clinical management significantly less often than consultant shoulder surgeon-ordered MRIs. Furthermore, the data suggests that over half the time in primary care, the MRI is taking the place of clinical examination of the shoulder, with 54% of referrals containing no documented examination prior to sending the patient for MRI.

This highlights an opportunity for education.

The fact that primary care-ordered MRIs changed clinical management only 22% of the time raises the question of indications for MRI and lack of controls on this. If access to private shoulder MRI was to be rationalised, perhaps then shoulder specialist-ordered X-rays and CT scans, which are currently not covered by insurance in our system, could be covered, as these are much more likely to influence clinical management than primary care-ordered MRIs, which are currently covered by insurance without restriction on indications.

ARTHROSCOPY OF THE SYMPTOMATIC SHOULDER ARTHROPLASTY

C Doherty, ND Furness, TJ Batten, WJ White, J Kitson and CD Smith

Royal Devon and Exeter Hospital, Exeter

Background: Assessment of the painful or stiff shoulder arthroplasty can be challenging. The cause of pain can sometimes be easily identified. However, sometimes patients have normal inflammatory markers, normal plain films and no clinical signs leading to a diagnosis. We report the utility of performing diagnostic arthroscopy in these patients for diagnostic advantage.

Methods: We retrospectively reviewed records of all patients who underwent diagnostic shoulder arthroscopy over a three-year period. Patients were included if they were aged 18 or over, had undergone previous arthroplasty surgery and had symptoms of shoulder pain or stiffness. Patients were excluded if they had any signs of infection or had raised serum white cell count (WCC) or C-reactive protein (CRP).

Results: 14 patients met the initial inclusion criteria. Mean age was 71 years. The mean interval between index surgery and arthroscopic evaluation was 65.4 months. Rotator cuff tears not identified on pre-operative imaging were noted in 57% of patients. Capsular contraction requiring release was performed in 14%. Glenoid wear was identified in 14% of patients. A fractured and dislocated poly liner was identified in one patient. Arthroscopic biopsies returned positive cultures for 21% of patients. Aspiration result correlated with biopsy findings in only 20%. In all patients the diagnostic arthroscopy directed the next stage of management.

Conclusions: Diagnostic arthroscopy allows a full assessment of implants, rotator cuff, native articular surfaces, scar tissue and biopsy for indolent infection in patients considering revision arthroplasty surgery. This allows a more informative consent process for the patient, directs surgical management and on occasions allows for therapeutic intervention in the painful or stiff shoulder arthroplasty.

A PROTOCOL FOR DAY-CASE TOTAL SHOULDER ARTHROPLASTY

S Fogerty and A Mayor

Calderdale and Huddersfield NHS Trust, Huddersfield

The purpose of this study was to develop a protocol to select suitable patients for day-case arthroplasty of the shoulder and then safely and successfully navigate patients through to discharge the same day.

Suitability criteria include primary surgery, a responsible adult at home, lives no more than 1 hour travel from hospital, no major medical problems and have reasonable exercise tolerance. In addition they should not taking regular opiates, is able to and happy to be contacted by telephone the following day and consent to a nerve block.

The patient is listed first on morning list. They receive a general anaesthetic (GA) and a supraclavicular nerve block (patient information leaflet is provided). Post operatively they have haemoglobin level check and a radiograph of the shoulder. At 1600 hours, a full blood count is taken. At 1800, 10 mg po oxycodone MR and 4 mg po/iv ondansetron administered At 2000, the patient reviewed by surgeon and anaesthetist and if both are satisfied, the patient is discharged home. Discharge medications include sevredol prescription (10 mg morphine (sevredol preparation), po, 4 hourly, PRN, maximum 40 mg in 24 hours, 4 tablets only to be dispensed). The patient is given a leaflet on “Pain relief advice for after your shoulder replacement”. They are advised to have a supply of paracetamol and ibuprofen at home. Patient must meet standard discharge criteria but additionally must have Hb > 10 and SaO2 > 93% on room air. They are contacted by telephone the following day for follow up.

We have had 6 patients through our protocol. There have been no readmissions. There has been a 100% satisfaction rate reported by patients. We have monitored for adverse events and there have been none and we continue to monitor.

We have developed a safe and effective protocol to follow for successful day case shoulder arthroplasty.

RELIABILITY ANALYSIS OF RADIOLOGICAL AND INTRA-OPERATIVE LOOSENING IN TOTAL ELBOW ARTHROPLASTY

MJ Gandhi, AI Eyre-Brook, P Gopinath, V Jones, S Booker, DS Thyagarajan and AA Ali

Northern General Hospital, Sheffield

Background: Revision of Total Elbow Arthroplasty (TEA) are commonly performed due to prosthetic loosening. According to NJR 2018, 76 TEA revisions were performed in 2017. TEA radiological assessment is based on subjective interpretation. We seek to standardise radiological assessment of TEA loosening by improving inter-observer assessment and subsequent management of the patient. We defined loosening for TEA by the presence of at least one of the following criteria:

1. Progressive widening of bone-cement, bone-prosthesis or cement-prosthesis interface

2. Fragmentation or fracture of cement

3. Prosthetic component migration

4. Bead shedding in porous coated prostheses.

Using this definition, we looked at inter-observer agreement of radiological loosening and compared this assessment with intra-operative findings.

Methods: Ninety-three sets of radiographs of Revision TEA cases performed in our unit were selected and anonymised. Each case had radiographs selected from 2 different time points pre-revision surgery. Radiological implant loosening was assessed by 5 independent orthopaedic surgeons. Inter-observer agreement (kappa coefficient) was calculated. The majority’s view of radiological loosening was then compared with the intra-operative findings.

Results: Ninety-three sets of radiographs were reviewed. Kappa coefficient between assessors for humeral loosening was 0.87 (almost perfect). Kappa coefficient for ulna loosening was 0.75 (substantial). Kappa coefficient for radiological and intra-operative findings for humeral loosening and ulna loosening were 0.67 and 0.71, respectively (substantial).

Conclusion: Our definition of loosening provides a reproducible inter-observer agreement of radiographic component loosening. In our centre’s experience, radiological findings may not translate to intra-operative findings and we would advise surgical strategies to be prepared for dual-implant exchange.

INTRA-OSSEUS FIXATION OF THREE AND FOUR PART PROXIMAL HUMERUS FRACTURES: A REVIEW OF PATIENT REPORTED OUTCOMES IN A SINGLE CENTRE CASE SERIES

N Jacob, S Sabharwal, I Sinha, D Griffiths and P Reilly

Department of Trauma and Orthopaedics, Imperial College Healthcare Trust, London

Introduction: The outcomes of fixation in a case series of 24 consecutive patients with Neer 3 and 4 part fractures who underwent surgical fixation using an intra-osseous suture technique in isolation or in combination with a 1/3 tubular plate is presented.

Methods: All operations were performed at a single centre by 3 surgeons, with a minimum follow-up of 1 year. Outcomes were determined by assessment of range of motion, an Oxford Shoulder Score (OSS) and radiographic assessment.

An activity based costing analysis was also performed.

Results: Of the 24 patients, 3 were lost to follow-up, leaving 21 patients with follow-up ranging from 1 to 6 years (Mean 2.2 years). 14 were classified as 3-part fractures, 7 were classified as 4-part fractures. The 4-part fracture group had a high failure rate (4 of 7 fixations failed- 57%), where ‘failure’ was deemed as unacceptable re-displacement of the fragments. Three failures were revised to a Philos plate, hemiarthroplasty and inverse shoulder replacement, while one patient declined revision. There were no failures in the 3-part group and all fractures achieved radiographic union.

The mean OSS for the entire group was 39/48 excluding those deemed as ‘failures’ (42 for the 3 part fractures and 34 for 4 part fractures). The mean forward flexion was 137.5o, abduction 140o, external rotation 49o and internal rotation to the level 10th thoracic vertebra. These results include the 4 part fractures in which the range of motion was inferior.

Conclusion: Based on our experience, we would advocate the use of this method specifically for valgus impaction type displaced Neer 3-part proximal humeral fractures. It is easily revisable, cheaper than most other methods, with good clinical outcomes and low failure rates.

The results in 4 part fractures are inferior. Appropriate consideration of other management options is advised for these fractures.

IMPLANT SURVIVAL IN ADULT PATIENTS WITH UNILATERAL AND BILATERAL TOTAL ELBOW ARTHROPLASTY

V Jones, D Stanley, D Thyagarajan and A Ali

Sheffield Shoulder and Elbow Unit, Northern General Hospital, Sheffield

Total Elbow Arthroplasty (TEA) is indicated in the treatment of end stage arthropathy secondary to various aetiologies or following trauma in the elderly. The aim of this study was to evaluate the difference on implant survival in adult patients with unilateral and bilateral TEAs performed for a variety of indications.

A retrospective review was undertaken of all primary TEA patients within a single surgical unit from 1993 to 2016. Survivorship analysis was carried out using Kaplan-Meier method with revision surgery classed as the end-point for implant failure. Results were analysed to ascertain whether there was any statistical difference between survivorship of implants in patients with unilateral and bilateral TEA.

133 primary TEAs were performed in 111 patients including 22 patients who had staged bilateral procedures. Mean age at the time of index surgery was 62.64 years (23–93). In the unilateral group 70% of the patients were female compared with 86.3% of patients in the bilateral group.

In the unilateral group, 55.1% TEAs were performed for inflammatory arthritis and 40.5% for acute trauma or trauma sequelae. In the bilateral group, 86.3% were implanted for inflammatory arthritis and 9% for trauma.

Minimum follow-up was 2 years with a mean of 10 years (2 - 25 years).

A total of 45 revisions (33.8%) were performed. Sub-group analysis showed 5, 10, 15 and 20 year survival rates for unilateral TEA for all indications was 86.0%, 69.6%,54.3% and 49.3% respectively. In the bilateral group, survival rates were 88.6%, 71.7%, 64.8% and 52.3% respectively, with no significant difference between groups (p > 0.05).

This study shows that undergoing staged bilateral elbow replacements does not have an adverse effect on elbow arthroplasty survival rates. Staged bilateral TEA can be considered a good option in patients with bilateral elbow joint pathology especially those with inflammatory arthritis.

INCIDENCE OF SYMPTOMATIC ROTATOR CUFF TEAR IN PATIENT'S OVER 40 YEARS OF AGE WITH PRIMARY ANTERIOR GLENOHUMERAL DISLOCATION – ARE WE OVER INVESTIGATING?

C Manafa, A Barakat, F Auld and J Phadnis

Royal Sussex County Hospital, Brighton

Purpose: The incidence of symptomatic cuff tears following primary anterior shoulder dislocation in patients over 40 years is thought to be high. Current guidelines recommend all patients should have specialist clinic review and advanced imaging. The aim was to report the incidence of cuff tear in our population using a virtual clinic protocol.

Methods: 432 consecutive patients with a primary, traumatic anterior glenohumeral dislocation were identified. Patients below 40 years were excluded leaving a study group of 239 patients. Patients were grouped by age: Group 1 (40–70 years, n = 150) and Group 2 (>70 years, n = 89). Patients were reviewed in the virtual fracture clinic. 75 patients (31%) with pain free motion and no subjective weakness were discharged to physiotherapy. The remaining 164 patients were reviewed in a specialist shoulder clinic. Of these, 89 patients (54%) were discharged to physiotherapy following clinical examination. The remaining 75 patients underwent either Magnetic resonance imaging (MRI) Ultrasound scan (USS) (MRI = 35, USS = 40). Cost analysis was performed using national figures for clinic costs, virtual appointments and advanced imaging to compare management using this protocol versus the published guidelines by BESS.

Results: The incidence of symptomatic cuff tears in all patients was 31%. The incidence was significantly higher in Group 2 (>70 years) than in Group 1 (40–70 years) (47% vs 22%, p = <0.001). Only 11 patients (5%) required surgical treatment for a symptomatic rotator cuff tear sustained as a result of their dislocation.

The cost analysis demonstrated a 42% reduction (£35,542 vs. £61,662) using this selective management protocol compared to the recommendation for specialist clinic and advanced imaging in all cases.

Conclusion: The incidence of symptomatic cuff tear in this population was lower than reported in the national guidelines and limited available literature. Reconsideration of the current guidelines is recommended based on these results.

AN UPPER LIMB ORTHOPAEDIC NETWORK: SUPPORTING REGIONAL SHOULDER AND ELBOW ARTHROPLASTY

BA Marson, L Mends, MD Wijeratna, BWT Gooding, P Manning and JM Geoghegan

Nottingham University Hospitals NHS Trust, Nottingham

Background: The management of complex upper limb arthroplasty is a current important topic of discussion. Many hospitals are completing relatively few procedures and attempting difficult cases without the benefit of institutional experience. Centralising services may be attractive for healthcare providers, but when your ‘local’ teaching hospital is 82 miles away then there is a significant incentive for patients to have operations performed in their local hospital.

An upper limb Orthopaedic Network has been established to share experience and best practice for primary and revision shoulder and elbow replacement between 5 shoulder and elbow arthroplasty units using online videoconferencing. This study evaluates the early impact from the first year of delivering these meetings

Methods: Data was collected prospectively from the network meetings and complimented by review of electronic records and radiographs

Results: 59 patients have been discussed in 14 meetings. 73% of patients were referred from the hub centre. Network discussion resulted in changes to the management plans in 54% of cases and identified the need for loan kit in 44% of cases. Network discussion resulted in 69% of complex cases being retained by spoke hospitals for local treatment.

Discussion: The specialist network is an ideal method to ensure that every patient gains the advantage of shared experience while retaining the ability to have their surgery at a convenient, local hospital. The network meetings have not resulted in an increased surgical workload at the hub hospital with most referred patients retained by spoke hospitals. The specialist network is an ideal method to ensure that every patient gains the advantage of shared experience while retaining the ability to have their surgery at a convenient, local hospital.

POSTERIOR ACROMIAL MORPHOLOGY IS SIGNIFICANTLY ASSOCIATED WITH POSTERIOR SHOULDER INSTABILITY

DC Meyer, L Ernstbrunner, MA Imam, R El Nashar, G Boyce and C Gerber

Department of Orthopaedics, Balgrist University Hospital, University of Zurich, Switzerland

Background: To determine whether acromial morphology, in addition to glenoid version, influences antero-posterior shoulder stability. We hypothesized that posterior slope, orientation and position of the acromion in the sagittal plane relative to the glenoid are the most significant factors of the direction of glenohumeral instability.

Methods: In this retrospective study, patients with unidirectional posterior instability were age- and gender-matched to a cohort of patients with unidirectional anterior instability. Both cohorts were compared with a control group of patients with no instability and no degenerative glenohumeral (rotator cuff and/or joint surface) changes. Measurements on radiographs and CT scans included glenoid version, posterior acromial tilt, anterior and posterior acromial coverage (AAC; PAC), posterior acromial height (PAH) and the critical shoulder angle (CSA).

Results: Forty-one shoulders in each instability group were enrolled based on a priori power analysis. The control group consisted of 82 shoulders. Logistic regression analysis for posterior and anterior instability showed that PAH (odds ratio (OR) 1.8; p < 0.001) followed by glenoid retroversion (OR 1.7; p = 0.001) are the most significant factors influencing the direction of instability. The PAH and glenoid retroversion were significantly higher in the posterior instability group compared with the anterior instability group (30.9 vs. 19.5 mm; p < 0.001, and −6.2° vs. −0.8°; p = 0.006). With a cut-off value of PAH of 23 mm, the odds ratio for posterior instability was 39. Shoulders with posterior instability were also significantly different from normal with regard to PAH (p < 0.001), AAC (p < 0.001) and PAC (p < 0.001), as opposed to shoulders with anterior instability, in which these values were not different from those of normal shoulders.

Conclusions: In the group with posterior instability, the acromion is situated higher and is oriented more horizontally in the sagittal plane. Higher sagittal plane steepness essentially excludes the presence of recurrent posterior shoulder instability in an albeit relatively small cohort of patients.

THE DEVELOPMENT OF A STANDARDISED TREATMENT PROTOCOL FOR FUNCTIONAL POSTERIOR SHOULDER INSTABILITY: AN INTERNATIONAL DELPHI SURVEY

P Moroder, P Boileau, V Danziger, C Gerhardt, B Kibler, S Lambert, G Porcellini, M Scheibel, M Tauber and M Wellmann

Centrum fur Muskuloskeletale Chirurgie, Charite- Universitätsmedizin Berlin, Augustenburger Platz, Berlin, GDR

Aim: To develop a standardised treatment protocol for the treatment of Functional Posterior Shoulder Instability (Type III, uncontrollable).

Method: Currently, there exists no published consensus on a specific physiotherapeutic protocol for functional non-controllable posterior shoulder instability (FPSI, so-called Type III). Accurate and specific diagnosis of the syndrome is crucial. Hypoactivity in scapular retractors and external rotators is considered relevant and consistent.

An international Delphi consensus survey process involving nine experts and their senior physiotherapy colleagues familiar with the challenges of treating this syndrome was used to evaluate and modify a protocol developed at the study host institution. The protocol comprised three groups of exercises with increasing difficulty performed under therapeutic supervision and focussing on the scapular retraction and external rotation muscles. The level of agreement was set at 95%. After the first round there was 83% agreement for level 1, 90% agreement for level 2, and 96% agreement for level 3 exercises, with concern about the early introduction of open-chain exercises in level 1. After the second round there was a 100% agreement for the structure of the protocol and for all exercises involved.

Conclusion: A consensus was reached for a specific structured protocol to be used in research designed to evaluate treatment of this enigmatic condition. The protocol will be utilised in a randomised controlled trial of the use of a shoulder pacemaker device in the treatment of FPSI.

ALL-INSIDE ARTHROSCOPIC ROTATOR CUFF REPAIR: RESULTS IN THE FIRST 1000 CASES

GAC Murrell, A Elkins and PH Lam

Orthopaedic Research Institute, St George Hospital Campus, University of New South Wales, Sydney, Australia

Introduction: Arthroscopic rotator cuff repair can be technically difficult with reported mean operative times ranging from 32 to 113 minutes, and re-tear rates between 15 and 90%. We devised a technique where the repair is performed while visualised entirely from the glenohumeral joint using a single-row, knotless, tension-band, inverted mattress sutures. The arthroscope does not enter the subacromial space, but stays beneath the rotator cuff. The aim of this study was to evaluate the safety and efficacy of this technique in the first 1000 consecutive patients.

Methods: The first 1000 patients who underwent a primary arthroscopic rotator cuff repair by one surgeon using the aforementioned technique and attended 6-month follow-up with ultrasound evaluation to determine the integrity of the repair were evaluated.

Results: The mean operative time for this technique began around 30 minutes and then reduced to 15 minutes after the first 200 cases. There were no infections. The only complication was re-tear. The re-tear rate was initially around 17%, reduced to 10% at case 400, and then 5% at case 700. Smaller tears were repaired faster and had better healing rates.

Conclusion: The all-inside undersurface repair technique was fast with a mean operative time of 16 minutes. There were no adverse effects of not performing an acromioplasty. The technique was also safe, with a 2% re-operation rate, and a re-tear rate of 8%. Repair integrity was better in patients with partial thickness tears and small full thickness tears.

	Partial Thickness	Full thickness <1 cm2	Full thickness 1–4 cm2	Full thickness >4 cm2	All Tears
Number	478	21	367	134	1000
Number	478	21	367	134	1000
Mean ± SEM operative time (mins)	15 ± 0.4	14 ± 1.8	17 ± 0.5	22 ± 0.9	16 ± 0.3
Range operative time (mins)	5–60	5–33	4–75	5–60	4–75
% intact at 6 months	98%	100%	91%	70%	92%
%r e-tear at 6 months	2%	0%	9%	30%	%r e-tear at 6 months

CLINICAL PREDICTORS OF CLAVICLE UNION AT 6 WEEKS. HOW USEFUL IS CLINICAL ASSESSMENT ALONE?

JA Nicholson, A Clelland, D MacDonald, N Clement and CM Robinson

University of Edinburgh, Edinburgh

Aims: Clinical recovery following clavicle fracture may enable identification of patients at high risk of nonunion and therefore limit operative intervention and prolonged delay to diagnosis.

Methods: All patients who sustained a 100% displaced midshaft clavicle fracture were enrolled into a prospective study in our department and underwent clinical review at 6, 12 and 24 weeks. Clinical assessment consisted of shoulder functional scores, radiographic evidence of callus and a range of examination features. Nonunions were confirmed on CT at six months and offered delayed fixation. Regression modelling was used to establish the most accurate clinical predictors for nonunion based on the 6 week assessment.

Results: Over a two year period (2016–2018) 236 adult patients sustained a midshaft fracture. 10 opted for acute fixation and 26 patients were lost to follow up. 200 patients completed the study with a response rate of 88.5% for eligible patients (200/226). The nonunion rate was 14% and this was associated with smokers (Odds Ratio (OR) 4.97, p < 0.001), females (OR 3.96, p = 0.001) and age (p = 0.004).

Receiver operator curves showed a threshold cut-off on QuickDASH at 40, above which was associated with nonunion (Area under curve 0.768, OR 8.9, p < 0.001). Above the QuickDASH cut-off the positive predive value for nonunion was 32% (Number Needed to Treat 3.1), increasing to 53% (NNT 1.9) in smokers. 69% of the cohort had a QuickDASH score below the cut-off and this had a 95% predictive value for union.

Regression revealed QuickDASH > 40 (p = 0.001), no callus on radiograph (p = 0.011), fracture movement on exam (p = 0.017) and smoking (p = 0.038) were the strongest clinical predictors. Irrespective of smoking status, when all three clinical predictors were present this was 100% predictive of nonunion.

Conclusion: Clinical assessment at 6 weeks can be used to identify high risk groups of nonunion for consideration of early intervention.

RTSA CLINICAL OUTCOMES WITH STANDARD AND CONSTRAINED HUMERAL LINERS

SA Parada, PH Flurin, TW Wright, JD Zuckerman and CP Roche

Medical College of Georgia at Augusta University, USA

Purpose: rTSA is an increasingly popular treatment for cuff-tear arthropathy (CTA), however instability and recurrent dislocations are concerns for which there are few solutions. Options for instability include the use of a constrained humeral liner. This study compares demographic, implant, preoperative, and post-operative factors and outcomes associated with rTSA using both standard and constrained liners.

Methods: 882 patients (mean 71.6 ± 7.9 yrs; BMI = 29.0 ± 6.2) with 2 yr minimum follow-up were treated with rTSA for CTA by 5 surgeons using either a standard or constrained humeral liner. 844 rTSA patients had a standard liner and 38 rTSA patients had a constrained liner. Validated outcome scores and ROM were recorded. Average follow-up was 47.7 ± 23.7 months.

Results: All patients demonstrated significant improvements in pain and function with standard and constrained humeral liners. Constrained liners were more often used in older patients, shorter patients and patients of lower weight and more often used in combination with a cemented humeral stem and more liner + tray offset compared to standard liner patients. Constrained liner patients had a significantly greater rate of humeral radiolucent lines and scapular notching compared to standard liner patients. There was no significant difference in any post-op metric between humeral liner cohorts and no significant difference in any pre-to-post-op improvement between cohorts.

Conclusion: The use of constrained liners is a safe and effective solution to achieve joint stability, with similar outcomes at 2 yr follow-up as compared to standard liners. Differences in patient demographics, implant usage, and radiographic outcomes were observed between standard and constrained liner patients. No cases of instability were reported in the constrained liner cohort and no significance decrease in active ROM was observed with the use of constrained liners. Patients with constrained humeral liners had a significantly higher scapular notching grade relative to the standard liner patients.

SURGICAL MANAGEMENT OF UPPER LIMB TENDON RUPTURES USING SINGLE INCISION ENDOBUTTON TECHNIQUE: A PROSPECTIVE CASE SERIES USING VALIDATED OUTCOME MEASURES

R Pulavarti, D Harmer, C Corbin and S Hussain

Lincoln County Hospital, Lincoln

Background: Traumatic tendon ruptures of large muscles of upper limb are common in active adults involved in heavy-demand activities. Distal biceps and pectoralis major tendons insert into proximal metaphysio-diaphyseal portions of long bones and repair using Endobutton technique is well known. We report a series of cases operated by open single incision Endobutton technique between March 2016 and January 2019.

Methods: A total of twenty-two male patients, sixteen distal biceps ruptures and six pectoralis major tendon ruptures (mean age 31 years), were operated by single surgeon. All patients were healthy, heavy manual workers and all had ultrasound and MRI scan. The mean duration of time of injury to operation was 13 days (7 to 31). Patients were followed up prospectively using Quick DASH score and/or Oxford Shoulder Scores.

Results: All patients had full deformity correction and full range of movements at the latest follow up. All but two distal biceps tendon rupture patients reported full strength in their involved muscle when compared to uninjured limb. The mean Oxford Shoulder Score of 14 (out of 60) for pectoralis major repair cases. Mean Quick DASH score was 11.5 at the latest follow up (range 3–25 months, mean 13.1 months). Eighteen of twenty-two patients reported the outcome as excellent, three patients good and one patient with distal biceps tendon repair reported fair. Complications in this series were one late presentation of deep infection (pectoralis tendon repair), three cases of neuropraxias with return to full function at latest follow-up.

Conclusion: Traumatic tendon rupture involving large muscles of the upper limb can adversely affect the appearance and function when inadequately treated. Multidisciplinary approach and single incision Endobutton technique would improve patient reported outcome.

ARTHROSCOPIC STABILISATION OF ATRAUMATIC SHOULDER INSTABILITY: MINIMUM TWO YEAR OUTCOMES

S Raja, S Jassim, D Butt, W Rudge, M Falworth, D Higgs and A Majed

Royal National Orthopaedic Hospital, Stanmore

Statement of purpose: We report the safety and efficacy of arthroscopic inferior capsular shift in the context of atraumatic instability at minimum two-year follow up.

Patients and methods: From April 2013 to January 2017, 72 patients with atraumatic instability underwent arthroscopic inferior capsular shift. Patient-reported outcome scores were prospectively collected during routine outpatient follow up, as determined by the local research and innovation committee, and stored in physiotherapy and surgical databases. These included Oxford Shoulder (OSS), Oxford Shoulder Instability (OSI), pain using a numerical rating system (NRS) and Stanmore Percentage of Normal Shoulder Assessment (SPONSA).

Results: Mean age was 27.4 years with a large female preponderance (46 F:26M). Median Beighton score was 5, with 14 diagnosed pre-operatively with a connective tissue disorder e.g. Ehler-Danlos syndrome. Mean follow up was 2.7 years (range 2 to 5.8 years).

As of January 2019, 59 patients have attended two year follow up. Significant improvements in pain and functional outcomes were seen. Mean pain score improved from 7.3 (SE ± 0.34) to 2.7 (SE ± 0.35) and mean SPONSA increased from 43.8% (SE ± 3.31) to 71.6% (SE ± 2.92). Mean OSS increased from 29.6 (SE ± 1.81) to 39.1 (SE ± 1.34), while mean post-operative OSI was 34.6 (SE ± 1.47).

At two years, 23 patients (39%) reported instability symptoms, with 5 requiring further stabilisation surgery (8.5%). Using the FEDS (Frequency, Aetiology, Direction, Severity) classification the majority of recurrences were atraumatic anterior subluxations with more than 5 episodes. No predictors of failure were identified. There was one complication of adhesive capsulitis, which required capsular release.

Conclusions: Although 39% have symptoms of recurrent instability, interestingly patients report improved frequency/severity of instability symptoms, reduced pain and better function. Longer term studies are required to confirm whether these benefits persist. When non-surgical treatments fail, surgery may be considered but patients should be counselled of the risk of recurrence accordingly.

EARLY OUTCOME OF THE BIOLOGICAL ACROMIAL RESURFACING (BAR) PROCEDURE; A NOVEL SOLUTION FOR MASSIVE IRREPARABLE ROTATOR CUFF TEARS

M Ravenscroft and BW Morgan

Stepping Hill Hospital, Stockport

Background: Irreparable rotator cuff tears (RCT) can be a source of significant pain and functional loss. Of the arthroscopic options available to manage this condition, only the biodegradable subacromial balloon spacer and Superior Capsular Reconstruction (SCR) provide inter-position between the humero-acromial articulation & addresses re-centring of the humerus in the glenoid. As the balloon spacer bio-degrades over time, this effect is only temporary. The SCR using ArthroFLEX® (Arthrex, LifeNet) dermal allograft has promising early results, but is a technically demanding procedure and requires an intact infraspinatus tendon for attachment to the graft.

We present a novel technique as a solution to this issue. Resurfacing the under-surface of the acromion with the ArthroFLEX® human dermal allograft provides an inter-position graft which does not bio-degrade. The surgical technique is straightforward, minimising surgical time.

Method: In this single arm, prospective study, elderly patients with irreparable RCT, but without significant arthritis underwent minimal sub-acromial preparation, followed by implantation with the ArthroFLEX®, which was secured in place using suture-tape tied sub-dermally over the acromion. Follow-up visits were scheduled according to routine clinical practice, including MRI to confirm graft position and integration. Shoulder function was evaluated using Oxford Shoulder Score (OSS) and Visual Analogue Score (VAS). All procedures were performed by the authors.

Results: We have performed this procedure in 19 patients with a minimum 3 months follow-up (mean 4.4). OSS improved significantly from pre-op mean 23.9 to post-op 35.2 at 3 months (p < 0.05) and 44 at 6 months (p < 0.05). VAS improved from pre-op mean 8.58 to post-op mean 3.75 at 3 weeks, 2.6 at 3 months and 1.5 at 6 months (p < 0.05).

Conclusion: We conclude that in this initial series that the BAR procedure represents a viable and straightforward alternative to the existing arthroscopic options & has potential to maintain improvement in the long term.

FACTORS ASSOCIATED WITH NEED FOR BLOOD TRANSFUSION AND PROLONGED LENGTH OF STAY (LOS) FOLLOWING SHOULDER ARTHROPLASTY

M Rijab Agha, W Al-Azzani, T Matthews, R Williams, P Carter, M Sandby-Thomas and T Bailey

University Hospital Llandough, Cardiff

The aim of this study was to identify factors associated with the need for blood transfusion and prolonged length of stay (los) following shoulder arthroplasty.

100 consecutive shoulder arthroplasty cases, over a two year period (August 2015 – August 2017) were identified. Demographics, type of surgery, indication, LOS, pre and post-operative haemoglobin, haematocrit and Total blood loss (TBL) were recorded. Multivariate analysis was performed.

There were 49 Reverse Total Shoulder Replacements, 28 Anatomic Total Shoulder Replacements, 12 Hemi-arthroplasties and 11 revisions.

The mean age was 74 (range 25–90). The average LOS was 4 days (1–17). 28% were discharged by day 1 post-operatively, 50% by day 2 and 63% by day 3.

Factors associated with prolonged LOS were related to anticoagulation and social factors. Transfusion rate was 9% with pre-operative anaemia and fracture treated with Reverse Replacement as risk factors. The average LOS for transfused patients was 6 days [2–14]. The average haemoglobin drop was 21 [3–56] and the TBL (using Hb-balance method) average was 160mls [22–367]. This did not correlate with need for blood transfusion.

Patients with pre-existing anaemia and those with fractures treated by Reverse Total Shoulder Replacements have a higher risk for requiring blood transfusion and prolonged LOS. These patients may benefit from targeted pre-operative optimisation protocols.

PATIENTS’ AND HEALTH PROFESSIONALS’ EXPERIENCES IN THE UK FROZEN SHOULDER TRIAL

C Srikesavan, F Toye, S Brealey, L Goodchild, M Northgraves, S Lamb and A Rangan

Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal disorders, University of Oxford, Oxford

Purpose: To explore patients’ and health professionals’ experiences in a clinical trial evaluating the effectiveness of early structured physiotherapy (ESP), manipulation under anaesthesia (MUA), and arthroscopic capsular release (ACR) in people with frozen shoulder.

Methods: Semi-structured interviews were conducted with 44 patients randomised to the treatment groups (ESP: 14; MUA: 15; ACR: 15); and 16 health professionals who delivered the treatments (Surgeons: 8; Physiotherapists: 8). Interviews were analysed thematically using NVivo software.

Results: Patients described that frozen shoulder had a major impact on all aspects of their life. All patients described they had improvements in pain, movements, and function and were satisfied with the treatment received. Patients who had an ACR described quicker recovery of pain and shoulder movements than they expected. They also preferred ACR before and after the trial while other group patients had mixed treatment preferences.

Surgeons preferred ACR above MUA and felt physiotherapy alone would not be effective. Physiotherapists had mixed treatment preferences and felt that patient outcomes might be similar in all groups. Both surgeons and physiotherapists described that some patients declined to participate in the trial as they did not want to receive further physiotherapy.

Our key shared themes with patients and health professionals were: (i) Patients were happy to be involved in the trial (ii) Relief from pain and regaining shoulder movements and function are important outcomes to be achieved (iii) Steroids help in pain relief in early stages so patients can do their exercises (iv) A graduated shoulder physiotherapy programme would help improve movements and function after surgery and (v) Patients’ adherence to exercise would lead to better treatment outcomes.

Conclusions: In general, patients had positive experiences in terms of trial participation and treatments. It is important to understand how patients’ and health professionals’ treatment preferences may influence trial results.

MANAGEMENT OF PRIMARY OSTEOARTHRITIS IN PATIENTS 75 YEARS AND OLDER: REVERSE SHOULDER ARTHROPLASTY VERSUS TOTAL SHOULDER ARTHROPLASTY

P Valenti, J Kany, JD Werthel and D Rossouw

Paris Shoulder Unit, Clinic Bizet, Paris, France

Background: The indications of reverse shoulder arthroplasty continue to expand subsequent to its description for managing cuff tear arthropathy. Functional results of total shoulder arthroplasty are superior to those obtained after RSA and it is accepted that TSA should be used in preference to RSA, in younger patients with an intact rotator cuff. However, as elderly patients have a fragile cuff and poor bone stock, RSA for primary osteoarthritis with an intact cuff may be beneficial.

The objective of this study was to compare the results of TSA and RSA in the treatment of primary osteoarthritis in patients aged 75 years and older with an intact cuff.

Materials and methods: Patients aged 75 years and older undergoing primary shoulder arthroplasty (TSA or RSA) for primary osteoarthritis with an intact rotator cuff between 2003 and 2015 were assessed. Minimum follow-up was 2 years (mean 4 years). Pain (VAS), range of motion, SST, SSV and Constant score were recorded.

Results: Seventeen RSAs (15 women, 2 men; mean age 79.2 years) and 14 TSAs (12 women, 4 men; mean age: 80.4 years) were included. All analysed parameters improved between the pre- and the postoperative evaluations in both groups. At last follow-up (48 months; range: 24–98 months), forward elevation, external rotation with the elbow on the side and Constant score were significantly higher after a TSA than after a RSA (147° versus 128°, 40° versus 29°, 69.4 versus 60.2 respectively). Although Pain scores and external rotation with the arm abduction improved, the difference was not statistically significant. 2 RSAs (12%) were revised and 3 TSASs (18.75%).

Conclusions: TSAs showed significant improvement in terms of range of motion and Constant score compared to RSA. The difference in revision rates was not significant

ARE THE CLINICAL OUTCOMES OF REVERSE SHOULDER ARTHROPLASTY WITH AN EPISCOPO PROCEDURE MAINTAINED OVER THE MEDIUM AND LONG TERM?

P Valenti, L Zanjani, JD Werthel and D Rossouw

Paris Shoulder Unit, Clinic Bizet, Paris, France

Background: Reverse total shoulder arthroplasty combined with an Episcopo procedure allows a restoration of active external rotation in patients with a pseudoparalytic shoulder cause by a massive cuff tear.

The goal of this study was to determine if the good short term results (2 to 5 years) are maintained over the medium and long term follow up (5 to 12years).

Methods: This procedure was performed in 33 patients (18 women,15 men) with a mean age of 68.7 years (47–83), through a delto pectoral approach. All the patients had irreparable rotator cuff tears including teres minor. Latissimus dorsi and teres major were transferred to the lateral aspect of the humerus with tranosseous sutures. Postoperatively, the arm was immobilized in neutral rotation and physiotherapy was started at 4 weeks. Outcomes included range of motion, SSV, VAS, and Constant scores.

Results: For the cohort, the follow up was 73 months (25–141), all clinical outcomes were significantly (p < 0.05) improved. VAS score improved from 6 ± 2.5 to 1.6 ± 1.2; SSV improved from 34 ± 15 to 68 ± 17; active forward elevation increased from 70 ± 33 to 130 ± 37; active external rotation arm at the body −13 ± 22 to 19 ± 14 and in 90° of abduction from −9 ± 17 to 37 ± 24; active internal rotation from L3 to L1. The mean Constant score improved from 25 ± 9 to 75 ± 9.

Stratifying into ‘early’ and ‘late’ showed no deterioration comparing the 2 to 5 years group (19 patients) and 5 to 12 years group (14 patients). SSV was 69 ± 16 at 2 to 5 years FU and 67 ± 19 at the latest FU and the respective Constant score were 86% ± 16 and 84% ± 19%.

Discussion: The good clinical outcome at the 2 to 5 years follow up of reverse shoulder arthroplasty combined with an Episcopo procedure does not deteriorate in the longer term.

ADDRESSING THE CHALLENGE OF PROVIDING ARTHROSCOPIC TRAINING INTERNATIONALLY

W A Wallace, A Lawley and MJ Gandhi

University of Nottingham, Nottingham

Purpose of the Study: To explore if training in shoulder arthroscopic knowledge and skills can be successfully carried out collaboratively & internationally

Background: In March 2018 we set up the SETGY Programme (Shoulder & Elbow Training for Generation Y) aimed at developing a curriculum and training Programme covering shoulder & elbow surgery for young surgeons across the world. Our aim is to standardise shoulder & elbow diagnostic arthroscopy to ensure that the majority of pathological lesions are picked up and identified during the arthroscopy. We also set out to improve early arthroscopic skills by simulation training using Apps on mobile phones.

A smartphone App has been developed with both educational content and arthroscopic skills training using handles with markers attached that can be tracked as mock instruments. The App is downloaded onto a smartphone from the App Store for iPhones and from the Google Play Store for android phones and is available world-wide.

Methods: This study explores whether a structure of National Ambassadors can be developed to support the provision of training of both arthroscopic knowledge and skills training in different countries internationally using smartphones.

Results: We have been able to establish a team of National Ambassadors in the following countries – UK; South Korea; Brazil; India; Hong Kong; South Africa; Iraq; Greece; Japan and Hungary

We have been able to coordinate collaboration using Skype conferences. The Ambassadors provide us with valuable feed-back about the value of this training in their own countries and how modifications of the App will be needed in the future to improve the delivery of this training.

Conclusion: We have demonstrated that simulation training is a resource to enhance the trainer-trainee relationships and can be rolled out internationally. Our experience has encouraged the further development of the shoulder App.

IT'S NOT ALL ABOUT REDISLOCATION: A SYSTEMATIC REVIEW OF COMPLICATIONS AFTER ANTERIOR SHOULDER STABILISATION SURGERY

HLM Williams, JP Evans, ND Furness, TJ Batten, WJ White and CD Smith

Royal Devon and Exeter Hospital, Exeter

Background: Surgical treatment of recurrent shoulder instability has in recent years developed to include both soft tissue and bone block procedures, undertaken with open or arthroscopic approaches. Although the utilisation of such techniques has rapidly expanded, the associated risk of complications remains poorly established. This information is vital for clinical decision making and patient counselling. We aimed to quantify the complication rate associated with all types of surgery for anterior glenohumeral joint dislocation.

Methods: A systematic search was undertaken of studies reporting complications from primary anterior shoulder stabilisation surgery. Studies published in English between 2000 and 2017 with a minimum 2-year follow-up were included. Quality was assessed with the Methodological Index for Non-Randomized Studies criteria. Complication rates for any undesirable event contributing to the patient outcome were extracted and cumulative rates calculated.

Results: Out of 1942 references, 56 studies were included, totalling 4362 procedures. Arthroscopic soft tissue repair had a complication rate of 1.6%. When repair was combined with arthroscopic remplissage, the rate was 0.5% with no difference in stiffness between the two groups. Open soft tissue repair had a complication rate of 6.2% and open soft tissue repair with remplissage, a rate of 2.3%. An open bone block procedure had a complication rate of 7.2% and an arthroscopic bone block procedure, a rate of 13.6%. Arthroscopic bone block had a higher rate of hardware problems, temporary nerve injury and graft osteolysis than any other technique.

Conclusion: This systematic review reports the overall complication rates (not purely re-dislocation rates) in modern shoulder stabilisation surgery. With the increased use of bone block stabilisation procedures, including arthroscopically performed surgery, surgeons should familiarise themselves with the 10-fold increase in complications for these procedures compared with soft tissue arthroscopic surgery and should counsel their patients accordingly.

REDUCING ERRORS IN REVERSE SHOULDER ARTHROPLASTY WITH FAILURE MODES AND EFFECTS ANALYSIS

C Witney-Lagen, M Wood and L Morgan

Surgical Sciences and Practice, University of Oxford, Oxford

Introduction: Failure Modes and Effects Analysis (FMEA) is a prospective technique to analyse potential causes of error before they occur. It is recommended by the UK National Patient Safety Agency and has proven effective in reducing errors in healthcare.

Purpose: To produce a FMEA for reverse total shoulder arthroplasty (rTSA) and identify methods to minimise errors.

Methods: A process map of steps required to perform rTSA was constructed. For each step, potential errors were considered by 4 surgeons, 3 nurses, 1 ODP, 1 director and 1 implant designer. The likelihood of occurrence, severity of impact and detectability of errors were considered and used to calculate Risk Priority Numbers (RPN).

Results: The greatest risks (RPN > 100) were the increased likelihood of infection with inadequate wound washing (RPN = 180) or inadequate wound dressings (RPN = 270), the risk of dislocation with incorrect sling application (RPN = 112), the risk of nerve injury with a poor surgical approach (RPN = 162) and the risk of reduced implant longevity with incorrect implant orientation (RPN = 160).

Conclusion: Raising awareness of the most likely causes of error can help the team prioritise areas for special attention to minimise risk. Many risk factors can be easily mitigated without the need for extra financial outlay. For example, simple measures such as special attention to wash the wound thoroughly, applying dressings securely and correct sling application can significantly reduce three of the highest risks. FMEA can also help to identify areas where extra training could be particularly beneficial. This study suggests extra time training on the surgical approach and practising implant orientation would produce the greatest risk reduction for surgeons. Such targeted training could help inexperienced surgeons to gain operative competencies more rapidly.

GLOBAL TRENDS IN PATIENT REPORTED OUTCOMES MEASURES FOLLOWING ARTHROSCOPIC SUBACROMIAL DECOMPRESSION: EVALUATION OF ONLINE REGISTRY DATA

A Yousaf, A Townson, S Webb, M Dunne and N Kang

Cambridge University Hospitals NHS Foundation Trust, Cambridge

The purpose of this study was to evaluate whether the treatment of subacromial impingement syndrome (SAIS) with arthroscopic subacromial decompression (ASD) results in clinically significant improvements in a global dataset, using a web-based platform that collects patient reported outcomes.

Using the Surgical Outcome System™, an online global registry for collecting PROMs, 1585 patients were identified with a diagnosis of SAIS that had undergone ASD. Various PROMs (ASES, VAS, VR-12, SANE and OSS) were analysed at 1, 2 and 5 years. Using minimal clinically important difference (MCID) values (17.0, 1.4, 4.94, 27.3, 6.0 respectively) and comparing these against each PROM pre- and post-treatment, any clinically significant improvement in patients was evaluated.

For all PROMS, a clinically significant improvement following treatment was observed at the 1 and 2 year marks with improvement also shown at 5 years where data was available. These differences were observed to increase between data collection points suggesting patients continued to derive benefit up to a minimum of 24 months post intervention. Values for the ASES, showed pre-treatment and post-treatment differences of 32.85, 37.32 and 41.27 at 1, 2 and 5 years respectively; for VAS, differences of 3.57, 3.97 and 4.30 respectively; VR-12, 8.88, 11.90 and 14.44 respectively; for SANE, differences of 29.91, 31.99 and 27.97 respectively; and OSS, differences of 11.95 and 13.59 at 1 and 2 years respectively. Overall patient compliance rates were similar pre and post treatment for all PROMs.

This study demonstrates the value of online PROMs collection and that continued clinically significant improvement is derived in patients undergoing ASD up to 5 years post intervention. Further investigation to determine if non-surgical intervention yields similar results would help establish the best approach when managing this controversial condition and is of particular importance in a climate where health resource utilisation is under scrutiny.