BMJ Open Is Committed to Open Peer Review. As Part of This Commitment We Make the Peer Review History of Every Article We Publish Publicly Available

BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from

BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available.

When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to.

The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript.

BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay-per-view fees (http://bmjopen.bmj.com).

If you have any questions on BMJ Open’s open peer review process please email [email protected] http://bmjopen.bmj.com/ on September 24, 2021 by guest. Protected copyright. BMJ Open BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from

A quantitative analysis of medical students’ and physicians’ knowledge of degenerative cervical myelopathy

Journal: BMJ Open ManuscriptFor ID peerbmjopen-2018-028455 review only Article Type: Research

Date Submitted by the 08-Dec-2018 Author:

Complete List of Authors: Waqar, Mueez; University of Manchester, Academic Neurosurgery Wilcock, Jane; University of Liverpool, Department of Undergraduate Medical Education Garner, Jayne; University of Liverpool, Department of Undergraduate Medical Education Davies, Benjamin; University of Cambridge, Department of Clinical Neurosurgery Kotter, Mark; University of Cambridge, Department of Clinical Neurosurgery; Ann McLaren Laboratory of Regenerative Medicine, Department of Clinical Neurosciences

Cervical myelopathy, spondylosis, degenerative spine, physician Keywords: knowledge, gap analysis http://bmjopen.bmj.com/

on September 24, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 A quantitative analysis of medical students’ and 6 7 8 physicians’ knowledge of degenerative cervical 9 10 11 myelopathy 12 13 14 15 16 Authors and affiliations 17 18 1. Mueez WaqarFor1, [email protected] peer review only 19 20 2. Jane Wilcock2, [email protected] 21 22 3. Jayne Garner2, [email protected] 23 24 4. Benjamin M Davies3,4, [email protected] 25 26 5. Mark RN Kotter3,4, [email protected] 27 28 29 30 1Department of Academic Neurosurgery, Manchester Academic Health Sciences 31 32 Centre, The University of Manchester, Manchester, UK 33 34 35 36 2Department of Undergraduate Medical Education, The University of Liverpool,

37 http://bmjopen.bmj.com/ 38 Liverpool, UK 39 40 41 42 3Department of Academic Neurosurgery, Cambridge University Hospital, University 43 44 of Cambridge, Cambridge, UK

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 4WT MRC Cambridge Stem Cell Institute, Anne McLaren Laboratory, University of 49 50 Cambridge, UK 51 52 53 54 Corresponding author 55 56 Dr. Mark R.N. Kotter 57 58 WT MRC Cambridge Stem Cell Institute and Department of Clinical Neurosciences 59 60

Page 1 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 University of Cambridge 5 6 7 8 Acknowledgements 9 10 The authors would like to thank Passmedicine for their assistance with this work. 11 12 13 14 Declarations and conflicts of interest 15 16 Research in the senior author’s laboratory is supported by a core support grant from 17 18 the Wellcome TrustFor and peer MRC to thereview Wellcome Trust-Medicalonly Research Council 19 20 Cambridge Stem Cell Institute. MRNK is supported by a NIHR Clinician Scientist 21 22 Award. 23 24 25 26 Disclaimer: This report is independent research arising from a Clinician Scientist 27 28 Award, CS-2015-15-023, supported by the National Institute for Health Research. The 29 30 views expressed in this publication are those of the authors and not necessarily those 31 32 of the NHS, the National Institute for Health Research or the Department of Health. 33 34 35 36

37 Word count http://bmjopen.bmj.com/ 38 39 2425 40 41 42 43 Data 44

45 Additional data is provided in supplementary table 1. on September 24, 2021 by guest. Protected copyright. 46 47 48 49 Author statement 50 51 Each author met ICMJE criteria for authorship: 52 53  MW: coordinated creation of questions, performed statistical analysis, wrote up 54 55 all manuscript drafts. 56 57  JW: created first draft of questions for question-bank, aided in manuscript 58 analysis and write-up, helped to check of final draft. 59 60

Page 2 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 68 BMJ Open

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 3 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Abstract 5 6 7 8 Objectives. We have previously identified a delay in general practitioner (GP) referrals 9 10 for patients with degenerative cervical myelopathy (DCM). The aim of this study was 11 12 to evaluate whether an education gap existed for DCM along the GP training pathway 13 14 by quantitatively assessing training in, and knowledge of, this condition. 15 16 17 18 Design. Gap analysis:For comparison peer of reviewDCM to other conditions. only Comparators selected 19 20 on the basis of similar presentation/epidemiology (multiple sclerosis), an important 21 22 spinal emergency (cauda equina syndrome) and a common disease (diabetes 23 24 mellitus). 25 26 27 28 Subjects. Medical students, foundation doctors and GP trainees. 29 30 31 32 Primary and secondary outcome measures. (1) Assessment of training: quantitative 33 34 comparison of references to DCM in curricula (undergraduate/postgraduate) and 35 36 commonly used textbooks (Oxford Handbook Series), to other conditions. (2)

37 http://bmjopen.bmj.com/ 38 Assessment of knowledge: using standardised questions placed in an online question- 39 40 bank (PassMedicine). 41 42 43 44 Results. DCM had the joint lowest modal rank of references to the condition in curricula

45 on September 24, 2021 by guest. Protected copyright. 46 analysis and 2nd lowest modal rank in textbook analysis. In knowledge analysis 47 48 questions were attempted 127,457 times. The mean user performance decreased 49 50 sequentially for DCM and other neuroscience subjects with advancing question-bank. 51 52 However, performance for DCM questions in themes of presentation, workup and 53 54 management were all greater than the question-bank mean and within one standard 55 56 deviation. For students and junior trainees, there was a serial decrease in performance 57 58 from presentation to workup (-0.7 to +10.4 relative to question-bank mean) and 59 60 management (-0.6% to -3.9% relative to question-bank mean).

Page 4 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 Conclusions. Although infrequently cited in curricula and learning resources, 7 8 knowledge relating to DCM was above average. However, knowledge relating to its 9 10 management was relatively poor. 11 12 13 14 Keywords 15 16 Cervical myelopathy; spondylosis; degenerative spine; physician knowledge; gap 17 18 analysis For peer review only 19 20 21 22 Strengths and limitations of this study 23 24  Search terms relating to DCM were queried from three UK specific medical 25 26 school curricula and relevant postgraduate curricula 27 28  Only a handful of learning resources were searched to assess references to 29 30 DCM 31 32  A large number of responses were obtained by placing questions in an online 33 34 question-bank, relating to DCM 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 5 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Introduction 5 6 7 Degenerative cervical myelopathy (DCM) is a common and insidious condition that 8 9 can lead to severe disability.[1] It arises when degenerative changes of the spine 10 11 12 compress the spinal cord, causing a progressive spinal cord injury. Currently, 13 14 15 treatment is limited to surgical decompression, which is able to stop further injury but 16 17 due to the limited capacity of the spinal cord to repair, recovery is limited. The timing 18 For peer review only 19 20 of surgery is therefore crucial to recovery (‘time is spine’), and a recent meta-analysis 21 22 has demonstrated treatment within 6 months offers a greater chance of making a full 23 24 25 recovery.[2] Unfortunately, few patients are diagnosed promptly, with the majority 26 27 28 waiting more than 2 years for a diagnosis.[3] Consequently, most patients retain life- 29 30 long disabilities, contributing to quality of life scores lower than cancer, heart and lung 31 32 33 diseases.[4, 5] 34 35 36

37 http://bmjopen.bmj.com/ 38 The diagnostic pathway for DCM almost exclusively starts with assessment and triage 39 40 41 by a community physician, termed in the UK a General Practitioner (GP).[3] If 42 43 appropriate, the patient is then referred on for further investigation and management. 44

45 on September 24, 2021 by guest. Protected copyright. 46 Our analysis of this pathway has identified time to initial referral by GP (6.4±7.7) as 47 48 49 representing 51% of diagnostic delay.[6] 50 51 52 53 54 This period of the diagnostic pathway is difficult to examine in detail, and whilst 55 56 delayed patient presentation is likely to contribute, delayed detection measured by 57 58 59 multiple consultations and patient perspective, is certainly a significant component.[3, 60

Page 6 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 7] The cause for delayed detection is also likely to be multi-faceted, including subtle, 5 6 non-specific symptoms and incomplete clinical examination.[1] However, it is 7 8 9 conceivable that a lack of understanding of the workup and management of DCM may 10 11 12 also contribute. Indeed, DCM would be included within the spectrum of “neurophobia” 13 14 – an aversion to the neurosciences due to perceived difficulty, which has been 15 16 17 demonstrated in GPs and GP trainees.[8, 9] 18 For peer review only 19 20 21 22 Our objective therefore was to evaluate whether an education gap exists for DCM 23 24 25 amongst GPs, by quantitatively assessing their training in, and knowledge of, DCM. 26 27 28 29 30 Methods 31 32 33 34 35 36 Ethics

37 http://bmjopen.bmj.com/ 38 The present study design did not require ethical approval. 39 40 41 42 43 Education Gap Analysis 44

45 on September 24, 2021 by guest. Protected copyright. 46 A gap analysis is a process to identify gaps in existing systems such as education 47 48 49 curricula. It involves an assessment of existing knowledge or standards against pre- 50 51 determined standards that define core competency requirements.[10, 11] This differs, 52 53 54 for example, from simple cross sectional knowledge assessments whose primary 55 56 outcome is usually assessment of knowledge beyond core requirements. 57 58 59 60

Page 7 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 The study objective was therefore approached in two separate gap analyses (Table 5 6 1): 7 8 9 1. Assessment of Training: quantification of DCM in curricula and commonly used 10 11 12 learning resources, including assessment of relative importance through 13 14 comparison to other conditions. 15 16 17 2. Assessment of Knowledge: formal assessment of trainees’ knowledge using 18 For peer review only 19 questions placed in an online question-bank. 20 21 22 23 24 25 The GP Training Pathway 26 27 UK medical graduates complete the UK Foundation Programme to enter higher 28 29 30 specialty training as formal GP trainees. There are several assessments along this 31 32 route including: medical school final examinations to gain registration to practice in the 33 34 35 UK, or the equivalent Professional and Linguistic Assessments Board test (PLAB) for 36

37 http://bmjopen.bmj.com/ 38 international medical graduates; the Specialty Recruitment Assessment (SRA) – a 39 40 written assessment taken towards the start of the second foundation year prior to 41 42 43 higher specialty applications; and the Membership of The Royal College of General 44

45 on September 24, 2021 by guest. Protected copyright. 46 Practitioner’s (MRCGP), an exit exam for GP trainees. Whilst alternative entry routes 47 48 to General Practice exist, this pathway represents the most common training route for 49 50 51 GPs today. 52 53 54 55 56 Definition of DCM and comparators 57 58 59 60

Page 8 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 DCM was chosen as an inclusive term for a variety of diseases resulting in 5 6 compressive myelopathy.[1] Comparator diseases were also selected to compare 7 8 9 findings within and outside the field of DCM: 10 11 12 13 14 1. Within Myelopathy: a differential for DCM with equivalent or greater incidence. 15 16 17 2. Within Degenerative Spine: an alternative degenerative spine disease that is 18 For peer review only 19 widely taught. 20 21 22 3. Outside Neurology: a common disease that all clinicians would have some 23 24 25 knowledge about and interaction with. 26 27 28 29 30 The a priori hypothesis was, if an education gap existed within DCM, we would expect 31 32 metrics of training and knowledge of DCM to rank inferior to controls. Controls were 33 34 35 identified through a consensus author meeting and finalised after confirmation of 36

37 http://bmjopen.bmj.com/ 38 epidemiological profiles through literature review. 39 40 41 42 43 Assessment of Training: Learning Resource Analysis 44

45 on September 24, 2021 by guest. Protected copyright. 46 We selected learning resources that were in common usage by trainees, available in 47 48 electronic format (to be amenable for searching) and able to be stratified by training 49 50 51 stage. Specifically, training curricula, the Oxford Handbook series [12] and online 52 53 question-banks were selected. Alongside the United Kingdom Foundation Programme 54 55 56 and MRCGP, an example from each of the three main, UK medical school teaching 57 58 59 models [13] were included: problem-based learning (The University of Manchester, 60

Page 9 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 UK); traditional lecture based (The University of Cambridge, UK); and an integrated 5 6 system including aspects of both (Imperial College London, UK). The choice of 7 8 9 medical school curricula was pragmatically selected, based on access. The 10 11 12 handbooks most pertinent to the GP training pathway were included in our analysis: 13 14 The Oxford Handbook of Clinical Medicine (OHCM), The Oxford Handbook for the 15 16 17 Foundation Programme (OHFP) and The Oxford Handbook of General Practice 18 For peer review only 19 (OHGP). 20 21 22 23 24 25 Search terms for analysing learning resources were created for DCM and each 26 27 comparator using relevant terms from the search syntax of Cochrane Reviews, or if 28 29 30 absent, recent systematic reviews. Curricula were searched for the number of 31 32 references per disease, text books for the number of words per disease section and 33 34 35 question banks for the number of questions. 36

37 http://bmjopen.bmj.com/ 38 39 40 Assessment of Knowledge 41 42 43 The author panel includes GPs with an interest in education (JW), university appointed 44

45 on September 24, 2021 by guest. Protected copyright. 46 educationalists (JG) and neurosurgeons (MW, BMD, MRK). The authors devised a set 47 48 of questions to be included in an online question bank, composed of both multiple 49 50 51 choice questions [MCQs] and extended matching questions [EMQs], designed to 52 53 cover the different components of medical assessment; presentation, investigation 54 55 56 and management. The final set consisted of 19 questions (13 multiple choice 57 58 59 60

Page 10 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 questions [MCQs] and 6 extended matching questions [EMQs]). The questions 5 6 together with answers are included in Supplement 1. 7 8 9 10 11 12 Online question-banks including subsections devoted to each stage of the GP training 13 14 pathway were contacted with details of the study. Only one question-bank - 15 16 17 Passmedicine [14], responded and was therefore selected for this arm of the study. 18 For peer review only 19 20 21 22 Analysis 23 24 25 Statistical analysis was performed using SPSS version 22 (SPSS Inc., Chicago, 26 27 Illinois, USA). For assessment of training, frequencies of search terms were described 28 29 30 and the mode and modal ranks determined. For assessment of knowledge, 31 32 histograms were constructed with either a condition or question-bank along the X-axis 33 34 35 and user performance relative to question-bank mean along the Y-axis. The user 36

37 http://bmjopen.bmj.com/ 38 performance relative to question-bank mean corresponded to the raw difference 39 40 between the mean value for each theme and the mean value for a particular question- 41 42 43 bank. 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 11 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Results 5 6 7 8 9 Selection of Comparators and Resources 10 11 12 13 14 15 Multiple sclerosis (MS), cauda equina syndrome (CES) and diabetes mellitus were 16 17 selected as the three disease comparators and list of search terms compiled 18 For peer review only 19 20 accordingly (Table 2).[15-18] MS is a common differential for DCM,[19] with 21 22 overlapping signs and symptoms. Whilst DCM is likely to be more prevalent in 23 24 25 reality,[1] currently their characterised epidemiology is comparable.[20, 21] CES is 26 27 28 uncommon, but its missed or delayed diagnosis carries significant consequences. 29 30 Diabetes is relevant to all medical fields. 31 32 33 34 35 36 Assessment of training

37 http://bmjopen.bmj.com/ 38 39 40 41 Curricula analysis 42 43 DCM and cauda equina syndrome had the lowest modal rank in curricula search 44

45 on September 24, 2021 by guest. Protected copyright. 46 analysis (Table 3). This was true for early stage undergraduate curricula and late stage 47 48 49 curricula for the foundation programme and MRCGP. Multiple sclerosis was 50 51 mentioned less frequently in early versus late stage curricula. The opposite trend was 52 53 54 observed for diabetes mellitus. 55 56 57 58 59 Textbook analysis 60

Page 12 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Overall, DCM had the second lowest modal rank but above that of cauda equina 5 6 syndrome (Table 4). The relative word count attributed to neurological conditions such 7 8 9 as DCM, multiple sclerosis and cauda equina syndrome decreased with advancing 10 11 12 stage of textbook. The opposite trend was observed for diabetes mellitus, for whom 13 14 the total word count seemed to increase. 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 13 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Question Bank Analysis 5 6 7 8 9 Assessment of Knowledge 10 11 12 Passmedicine was approached, on the basis it did not have any DCM questions 13 14 previously which could influence analysis. Questions were introduced between the 15 16 17 period June 2017 – October 2017. Section editors for each question-bank ultimately 18 For peer review only 19 selected which of the 19 questions were relevant, such that finals/PLAB contained 14, 20 21 22 SRA – 19 and MRCGP – 19. 23 24 25 26 27 Overall, questions were attempted 127,457 times; finals/PLAB - 36706; SRA - 47530; 28 29 30 MRCGP – 43221 (Supplementary Table 1). Data on the number of unique attempts or 31 32 first-time viewers was not extractable. 33 34 35 36

37 http://bmjopen.bmj.com/ 38 There were differences in user performance in the three DCM question themes – 39 40 presentation, workup and management, though average scores were all within 1 41 42 43 standard deviation of the mean (Fig. 1). Performance sequentially decreased across 44

45 on September 24, 2021 by guest. Protected copyright. 46 these themes for the finals/PLAB group, SRA group and when all groups were 47 48 considered together. In the MRCGP group however, performance clustered around 49 50 51 the mean for the question bank with little variability. 52 53 54 55 56 User performance by disease showed large variation between question-banks (Fig. 57 58 59 2). The mean user performance decreased sequentially for DCM and other 60

Page 14 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 neuroscience subjects with advancing question-bank. However, the mean user 5 6 performance for DCM was always greater than the question-bank mean. 7 8 9 10 11 12 Discussion 13 14 15 In this study, we found that DCM is relatively infrequently cited in curricula and 16 17 commonly used textbooks, even compared to diseases such as multiple sclerosis, with 18 For peer review only 19 20 similar epidemiology. Despite this however, user performance in DCM questions 21 22 remained consistently above question-bank averages, unlike other neuroscience 23 24 25 themes such as multiple sclerosis and cauda equina syndrome. There was a 26 27 28 sequential decrease in user performance across the themes of DCM presentation, 29 30 workup and management for early years’ trainees, whereas for senior trainees, 31 32 33 performance did not vary by theme of question. The user performance was either 34 35 36 consistently below average or decreased sequentially for DCM and other

37 http://bmjopen.bmj.com/ 38 neuroscience themes such as multiple sclerosis and cauda equina syndrome. This 39 40 41 was not observed for diabetes mellitus. 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 Although the present study design is novel, many of our findings correlate to previous 47 48 49 reports. Firstly, there is the concept of “neurophobia”, a term given to the relatively 50 51 poor performance of neurosciences subjects relative to others. The present study 52 53 54 adds to this literature surrounding neurophobia, which is well described for GPs and 55 56 GP trainees and students.[8, 9, 22] In one questionnaire study for example, GPs rated 57 58 59 neurology as the most difficulty medical specialty and the one for which they had the 60

Page 15 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 least confidence compared to cardiology, endocrinology, gastroenterology, geriatrics, 5 6 respiratory medicine and rheumatology.[9] DCM was unusual in that performance in 7 8 9 this particular theme remained consistently above average, with a sequential decrease 10 11 12 with increasing training stage. This correlated with decreasing emphasis in more 13 14 advanced curricula. 15 16 17 18 For peer review only 19 Secondly, our study supports the observation of poorer performance in management 20 21 22 themes by clinicians at an earlier training stage. Prior studies on medical students that 23 24 25 have evaluated knowledge formally have shown similar results in specialties such as 26 27 ophthalmology but not DCM.[23] This also appears to translate into patient encounters 28 29 30 during clinical assessments in neurology.[24, 25] In one study evaluating student 31 32 performance in neurology outpatients, students were more likely to make errors 33 34 35 regarding diagnostic tests or planning treatment than lesion localisation.[24] Another 36

37 http://bmjopen.bmj.com/ 38 study evaluating student performance in neurology themed objective structured 39 40 clinical examinations (OSCEs) also reported similar results, finding poorer student 41 42 43 performance in supportive management of neurological conditions versus 44

45 on September 24, 2021 by guest. Protected copyright. 46 diagnosis.[25] 47 48 49 50 51 Our original hypothesis attempting to correlate delays in diagnosis for DCM with a 52 53 deficiency in training and knowledge, was not conclusively demonstrated in the 54 55 56 present study. Earlier stage trainee doctors demonstrated comparably higher levels of 57 58 59 knowledge with regards to the presentation and work up of DCM and a potential gap 60

Page 16 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 with regards to management-related questions. These differences disappeared in 5 6 individuals studying for the GP exit exams. Nevertheless, the clinical argument for an 7 8 9 education gap is strong - our previous analysis of the referral pathway for DCM 10 11 12 identified a time to initial referral by the GP as representing 51% of the overall 13 14 diagnostic delay. 15 16 17 18 For peer review only 19 This study demonstrated DCM received relatively less training than other conditions. 20 21 22 In the West, the prevalence of DCM is underestimated but at least 60 per 100,000, 23 24 25 compared to around 100 per 100,000 for multiple sclerosis.[20, 21] Despite this 26 27 however, DCM was under-represented by eight-fold compared to multiple sclerosis in 28 29 30 curricula analysis. In late stage curricula (foundation programme, MRCGP), DCM was 31 32 not referred to at all. A similar trend was observed in textbook analysis, though neither 33 34 35 of these analyses were exhaustive. These results were in contrast to our knowledge 36

37 http://bmjopen.bmj.com/ 38 analysis, which showed consistently above average performance for DCM unlike 39 40 multiple sclerosis. 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 One potential way of reconciling the absence of a knowledge gap with the deficiencies 47 48 in the diagnosis and management of DCM, is that factual knowledge may not translate 49 50 51 into recognition of the disease in the clinical context. Causal association between 52 53 training and clinical practice have not been clearly demonstrated, albeit prior studies 54 55 56 have found good correlation between exam performance and clinical performance in 57 58 59 a variety of medical and surgical specialties.[26-28] Most of this data is based on the 60

Page 17 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 United States Medical Licensing Examination (USMLE), which appears to correlate 5 6 well with residency performance.[26, 27] Beyond medical school exams, in another 7 8 9 study, internal medicine clinicians failing their maintenance of certification exams had 10 11 12 a more than double chance of disciplinary action versus those who passed.[28] 13 14 15 16 17 There are several limitations to be considered in the context of these findings. Firstly, 18 For peer review only 19 only select learning resources were used in this study and questions were inserted 20 21 22 into a single online question bank. Secondly, question bank data was not extractable 23 24 25 per user, nor per first time answer. Whilst this was not a comprehensive strategy, this 26 27 is unlikely to have limited our results for at least two reasons: questions were 28 29 30 attempted more than 100,000 times and analysis was relative within learning 31 32 resources, and therefore limitations were applicable to each comparator. 33 34 35 36

37 http://bmjopen.bmj.com/ 38 The methodology employed in the present study may not have been sufficiently 39 40 sensitive to detect a knowledge gap for DCM, for several reasons. Firstly, our 41 42 43 questions may not have been challenging enough given that they tested core 44

45 on September 24, 2021 by guest. Protected copyright. 46 principles all clinicians would be expected to achieve. Secondly, there was no first- 47 48 time answer data available for analysis. Thirdly, our controls may not have been 49 50 51 appropriate and certainly, educational comparisons between specialties is not always 52 53 the best method of assessing knowledge adequacy.[29] Finally, our methodology 54 55 56 assumed a direct correlation between “on paper” knowledge and clinical care provided 57 58 59 to patients – though this was not proven in the present study and could still differ vastly. 60

Page 18 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 7 Conclusion 8 9 In this study, we set out to evaluate whether there is a deficiency in training and 10 11 12 knowledge regarding DCM in the GP training pathway. Although DCM was 13 14 15 infrequently referenced in learning resources, trainees performed above average for 16 17 DCM questions in an assessment of knowledge using an online question-bank. The 18 For peer review only 19 20 clinical suggestion of an education gap is strong, and these contrasting findings do not 21 22 conclusively allay this concern. Future studies are required to better understand these 23 24 25 observations. 26 27 28 29 30 31 References 32 33 34 35 36 1. Davies BM, Mowforth OD, Smith EK, et al. Degenerative cervical myelopathy. BMJ

37 http://bmjopen.bmj.com/ 38 39 2018;360:k186. 40 41 2. Fehlings MG, Tetreault LA, Riew KD, et al. A Clinical Practice Guideline for the 42 43 44 Management of Patients With Degenerative Cervical Myelopathy: Recommendations

45 on September 24, 2021 by guest. Protected copyright. 46 for Patients With Mild, Moderate, and Severe Disease and Nonmyelopathic Patients 47 48 49 With Evidence of Cord Compression. Global spine journal 2017;7(3 Suppl):70S-83S. 50 51 52 3. Behrbalk E, Salame K, Regev GJ, et al. Delayed diagnosis of cervical spondylotic 53 54 myelopathy by primary care physicians. Neurosurgical focus 2013;35(1):E1. 55 56 57 4. Fehlings MG, Ibrahim A, Tetreault L, et al. A global perspective on the outcomes of 58 59 surgical decompression in patients with cervical spondylotic myelopathy: results from 60

Page 19 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 the prospective multicenter AOSpine international study on 479 patients. Spine 5 6 2015;40(17):1322-8. 7 8 9 5. Oh T, Lafage R, Lafage V, et al. Comparing Quality of Life in Cervical Spondylotic 10 11 12 Myelopathy with Other Chronic Debilitating Diseases Using the Short Form Survey 36- 13 14 Health Survey. World Neurosurg 2017;106:699-706. 15 16 17 6. Hilton B T-MJ, Davies BM, Kotter MRN. Route to diagnosis of degenerative cervical 18 For peer review only 19 myelopathy in a UK healthcare system: a retrospective cohort study. BMJ Open (in- 20 21 22 press) 2018 23 24 25 7. Elemental Ideas – Cervical Spondylotic Myelopathy Part One [cited 2018 October 26 27 12th]. Available from: http://archive.cambridge-tv.co.uk/elemental-ideas-cervical- 28 29 30 spondylotic-myelopathy-part-one/ accessed October 12th 2018. 31 32 8. McCarron MO, Stevenson M, Loftus AM, et al. Neurophobia among general practice 33 34 35 trainees: the evidence, perceived causes and solutions. Clinical neurology and 36

37 http://bmjopen.bmj.com/ 38 neurosurgery 2014;122:124-8. 39 40 9. Loftus AM, Wade C, McCarron MO. Primary care perceptions of neurology and 41 42 43 neurology services. Postgraduate medical journal 2016;92(1088):318-21. 44

45 on September 24, 2021 by guest. Protected copyright. 46 10. Fater KH. Gap analysis: a method to assess core competency development in the 47 48 curriculum. Nurs Educ Perspect 2013;34(2):101-5. 49 50 51 11. O'Brien CE, Stafford R, Franks AM. Establishment of a patient-centered 52 53 communication course to address curricular gaps. Curr Pharm Teach Learn 54 55 56 2018;10(7):933-39. 57 58 59 60

Page 20 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 12. Warriner D. The Oxford Handbook of Clinical Medicine. BMJ 2008;336(7640):393- 5 6 93. 7 8 9 13. NHS careers. Teaching styles at medical school [27/05/2018]. Available from: 10 11 12 https://www.healthcareers.nhs.uk/sites/default/files/documents/Teaching%20styles% 13 14 20at%20medical%20school.pdf accessed 27/05/2018. 15 16 17 14. Passmedicine 2018 [cited 2018 12th October]. Available from: 18 For peer review only 19 https://www.passmedicine.com accessed 12 October 2018. 20 21 22 15. Kiely PD, Quinn JC, Du JY, et al. Posterior surgical treatment of cervical 23 24 25 spondylotic myelopathy: review article. HSS J 2015;11(1):36-42. 26 27 16. Khan F, Amatya B. Rehabilitation in Multiple Sclerosis: A Systematic Review of 28 29 30 Systematic Reviews. Arch Phys Med Rehabil 2017;98(2):353-67. 31 32 17. Chau AM, Xu LL, Pelzer NR, et al. Timing of surgical intervention in cauda equina 33 34 35 syndrome: a systematic critical review. World Neurosurg 2014;81(3-4):640-50. 36

37 http://bmjopen.bmj.com/ 38 18. Richter B, Bandeira-Echtler E, Bergerhoff K, et al. Pioglitazone for type 2 diabetes 39 40 mellitus. Cochrane Database Syst Rev 2006(4):CD006060. 41 42 43 19. Davies BM, Munro CF, Kotter M. A Novel Insight into the Challenges of Diagnosing 44

45 on September 24, 2021 by guest. Protected copyright. 46 Degenerative Cervical Myelopathy using Online Symptom Checkers. J Med Internet 47 48 Res 2018 49 50 51 20. Nouri A, Tetreault L, Singh A, et al. Degenerative Cervical Myelopathy: 52 53 Epidemiology, Genetics, and Pathogenesis. Spine 2015;40(12):E675-93. 54 55 56 21. Dilokthornsakul P, Valuck RJ, Nair KV, et al. Multiple sclerosis prevalence in the 57 58 59 United States commercially insured population. Neurology 2016;86(11):1014-21. 60

Page 21 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 22. Flanagan E, Walsh C, Tubridy N. 'Neurophobia'--attitudes of medical students and 5 6 doctors in Ireland to neurological teaching. Eur J Neurol 2007;14(10):1109-12. 7 8 9 23. Esparaz ES, Binder SB, Borges NJ. How prepared are medical students to 10 11 12 diagnose and manage common ocular conditions. J Educ Eval Health Prof 13 14 2014;11:29. 15 16 17 24. Davis LE, King MK. Assessment of medical student clinical competencies in the 18 For peer review only 19 neurology clinic. Neurology 2007;68(8):597-9. 20 21 22 25. Lukas RV, Blood A, Park YS, et al. Assessment of neurological clinical 23 24 25 management reasoning in medical students. Journal of clinical neuroscience : official 26 27 journal of the Neurosurgical Society of Australasia 2014;21(6):919-22. 28 29 30 26. Thundiyil JG, Modica RF, Silvestri S, et al. Do United States Medical Licensing 31 32 Examination (USMLE) scores predict in-training test performance for emergency 33 34 35 medicine residents? J Emerg Med 2010;38(1):65-9. 36

37 http://bmjopen.bmj.com/ 38 27. Sutton E, Richardson JD, Ziegler C, et al. Is USMLE Step 1 score a valid predictor 39 40 of success in surgical residency? Am J Surg 2014;208(6):1029-34; discussion 34. 41 42 43 28. McDonald FS, Duhigg LM, Arnold GK, et al. The American Board of Internal 44

45 on September 24, 2021 by guest. Protected copyright. 46 Medicine Maintenance of Certification Examination and State Medical Board 47 48 Disciplinary Actions: a Population Cohort Study. J Gen Intern Med 2018 49 50 51 29. Di Paola M, Bennet GC, Shearer JR. Orthopaedic teaching in United Kingdom 52 53 medical schools. Med Teach 1986;8(2):155-8. 54 55 56 57 58 59 60

Page 22 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Figure captions 5 6 7 8 9 Figure 1. Trainee knowledge analysis: user performance by DCM question theme. 10 11 12 DCM questions assessed three themes - presentation, workup and management. 13 14 15 Early stage trainees, those sitting finals/PLAB and the SRA, performed worse in 16 17 questions relating to management versus more senior trainees taking the MRCGP. 18 For peer review only 19 20 Performance for MRCGP trainees was more consistent. The average performance for 21 22 each question theme was within 1 standard deviation (SD) of the mean. 23 24 25 26 27 28 Figure 2. Trainee knowledge analysis: user performance by disease. The mean user 29 30 performance for DCM, neurology and cauda equina syndrome decreased with 31 32 33 advancing question-bank. This trend was observed in most neurological pathologies. 34 35 36 User performance of diabetes mellitus was more variable. * = there were no questions

37 http://bmjopen.bmj.com/ 38 on cauda equina syndrome present in the SRA question-bank. 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 23 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Table 1. Summary of gap analysis methods. 5 6 7 Medical Foundation GP Training Metric 8 9 School 10 11 Curricula UoM Foundation MRCGP References to 12 13 UoC programme curriculum search terms 14 Imperial curriculum 15 16 Text Book OHCM OHFP OHGP References to 17 18 For peer review only search terms 19 20 21 Online PLAB SRA AKT Performance 22 23 Question Bank Medical finals in questions 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Table 2. Selected diseases for textbook and curricula searches. The reasoning for 5 each of the comparators is explained in this table. The search terms employed for 6 7 the curricula searches are shown below this. 8 9 10 11 12 Degenerative cervical Cauda equina Multiple sclerosis Diabetes mellitus 13 myelopathy syndrome 14 Disease of interest A disease with a similar Widely taught spinal Non-spinal control 15 16 incidence and morbidity emergency, selected 17 to degenerative cervical as a spinal control. 18 For peermyelopathy. review Selected only Reasoning 19 as a clinico- 20 epidemiological control 21 [19] 22 23 24 • Cervical myelopathy • Multiple sclerosis • Cauda equina • Diabetes mellitus 25 • Cervical myelopathy • Demyelinating • Saddle anaesthesia • Insulin dependent 26 • Cervical disease diabetes mellitus 27 myeloradiculopathy • Demyelination • Type 1 diabetes 28 • Cervical stenosis mellitus 29 30 • Cervical • Non insulin 31 compression dependent diabetes 32 • Cervical herniation mellitus 33 • Cervical • Type 2 diabetes 34 degeneration mellitus 35 Search terms • Ossification of • Maturity onset 36

37 for curricula posterior longitudinal diabetes of young http://bmjopen.bmj.com/ 38 searches ligament • Gestational diabetes 39 • Spinal osteophytosis mellitus 40 • Spinal cord • Late onset diabetes 41 compression • Maturity onset 42 • Spondylosis diabetes 43 44 • Insulin resistance

45 • Diabetic ketoacidosis on September 24, 2021 by guest. Protected copyright. 46 • Hyperosmolar 47 hyperglycaemic state 48 • Hyperosmolar non 49 ketotic coma 50 51 52 53 54 55 56 57 58 59 60

Page 25 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 68

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Table 3. Curricula analysis. Electronic copies of curricula were queried with relevant 5 search terms. 6 7 8 9 Number of references to term 10 Degenerative 11 Curriculum Multiple Cauda equina Diabetes 12 cervical sclerosis syndrome mellitus 13 myelopathy 14 15 Undergraduate - PBL 0 3 1 43 16 Undergraduate – traditional 1 13 1 66 17 18 Undergraduate – integratedFor peer3 review15 only 0 30 19 Foundation Programme 0 0 0 0 20 21 MRCGP 0 4 2 113 22 Total 4 32 4 252 23 24 Rank 3 2 3 1 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 68 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Table 4. Learning resource analysis. The number of words devoted to DCM and 5 other diseases were determined. Importantly, only the words contained within the 6 7 main section for the particular disease were considered. 8 9 10 11 Word count devoted to 12 Degenerative 13 Resource Multiple Cauda equina Diabetes 14 cervical 15 sclerosis syndrome mellitus myelopathy 16 17 OHCM 870 1104 567 5165 18 For peer review only 19 OHFP 0 112 58 3599 20 OHGP 252 679 120 5736 21 22 Cumulative 1122 1895 745 14500 23 24 Modal rank 3 2 4 1 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 27 of 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 68 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 Figure 1 29 30 364x239mm (300 x 300 DPI) 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 on September 24, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 68 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 Figure 2 28 381x236mm (300 x 300 DPI) 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 on September 24, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from

BMJ Open Page 30 of 68

1 2 3 4 Supplementary Table 1. Trainee knowledge analysis. A = number of questions used in question-bank; B = number of correct 5 responses / number of attempts; C = correct response rate (%). Note that each individual EMQ had 3 distinct clinical scenarios. 6 7 8 9 Question-bank 10 Cumulative results by theme Theme Type Finals/PLAB (beginner) SRA (intermediate) MRCGP (advanced) 11 12 A ForB peerC A reviewB C A onlyB C A B C 13

14 MCQ 1 658/1718 38.0% 3 407/1280 31.8% 3 1303/4667 27.9% 7 2367/7665 30.9% 15 16 Presentation EMQ 2 5665/7914 73.2% 2 12589/17208 73.2% 2 6747/9246 73.0% 6 29286/40056 73.1% http://bmjopen.bmj.com/ 17 18 Both 3 6323/9632 65.6% 5 12996/18488 70.2% 5 8050/13913 57.9% 13 31653/47721 66.3% 19 20 MCQ 4 3801/5378 70.7% 5 630/1198 52.6% 5 3960/5837 67.8% 14 8391/12413 67.6% 21 22 Assessment EMQ 1 1618/2952 54.8% 2 8360/14019 59.6% 2 4784/8988 53.2% 5 14762/25959 56.9% 23 Both 5 5419/8330 65.1% 7 8990/15217 59.1% 7 8744/14825 59.0% 19 23153/38372 60.3% 24 on September 24, 2021 by guest. Protected copyright.

25 MCQ 5 4653/6750 68.9% 5 792/1324 59.8% 5 3850/5612 68.6% 15 9295/13686 67.9% 26 27 Management EMQ 1 3224/6306 51.1% 2 6939/12501 55.6% 2 4806/8871 54.2% 5 14969/27678 54.1% 28 29 Both 6 7877/13056 60.3% 7 7731/13825 55.9% 7 8656/14483 59.8% 20 24264/41364 58.7% 30 31 MCQ 10 9112/13846 65.8% 13 1828/3802 48.1% 13 9113/16116 56.5% 36 20053/33764 59.4% 32 Cumulative results by EMQ 4 14792/22860 64.7% 6 27888/43728 63.8% 6 16337/27105 60.3% 16 59017/93693 63.0% 33 question-bank 34 Both 14 23904/36706 65.1% 19 29716/47530 62.5% 19 25450/43221 58.9% 52 79070/127457 62.0% 35 36 37 38 39 40 41 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 31 of 68 BMJ Open

1 2

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Passmedicine questions 5 6 Authors 7 1. Mueez Waqar 8 2. Jane Wilcock 9 10 3. Jayne Garner 11 4. Benjamin M Davies 12 5. Mark RN Kotter 13 14 Introduction 15 The following is a set of questions prepared for the degenerative cervical myelopathy education 16 initiative. We would request a link to our website - www.myelopathy.org, our website to promote 17 18 this initiative, below theseFor questions. peer review only 19 20 21 22 23 24 25 26 Methods 27 28 Questions have been prepared as follows: 29 • Clinical presentation - focussing on differential diagnosis: 5 MCQs, 2 EMQs 30 • Assessment - focussing on examination and workup: 5 MCQs, 2 EMQs 31 • Management - focussing on referral pathways and follow-up: 5 MCQs, 2 EMQs 32 33 These questions are targeted to the following question banks: 34 35 • MRCGP 36 • MSRA

37 • Medical finals http://bmjopen.bmj.com/ 38 39 Output 40 The aim of this initiative is to provide data on clinician knowledge on degenerative cervical 41 myelopathy. We would like to analyse the following metrics based on these questions: 42 43 • Distribution of answers on 1st attempt 44 • Distribution of answers on 2nd attempt

45 on September 24, 2021 by guest. Protected copyright. 46 In addition, we would like to compare the correct response rate of this sample to existing questions 47 in the question-bank along the following themes: 48 • Multiple sclerosis 49 Cauda equina syndrome 50 • 51 • Diabetes mellitus 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 32 of 68

1 2 Theme 1: Clinical presentation

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 A 65-year-old female reports progressive unsteadiness, mild weakness, pain and 6 paraesthesias in her legs over the last 2-3 months. She reports no symptoms in her upper 7 limbs and there is no history of trauma. Examination reveals increased lower limb tone with 8 brisk reflexes. Which of the following is the most likely explanation for her symptoms? 9 10 11 A. Motor neuron disease 12 B. Degenerative cervical myelopathy 13 C. Multiple sclerosis 14 15 D. Syringomyelia 16 E. Cauda equina syndrome 17 18 Answer: B For peer review only 19 20 21 Degenerative cervical myelopathy (DCM) refers to cervical spinal cord compression 22 23 due to cervical spondylosis. The term replaces CSM (Cervical Spondylotic 24 25 Myelopathy). Patients with DCM can present with predominantly lower limb 26 27 symptoms. Examination findings, especially early on, are often subtle. DCM is the 28 29 most common cause of non-traumatic paraparesis in developed countries and 30 31 becomes more common with age [2]; it is estimated to have a prevalence of 5% in 32 33 over 50-year-olds [3, 4]. In a series of patients presenting to Neurology outpatients, 34 it was twice as common as Multiple Sclerosis [2]. Early diagnosis is key to ensuring 35 36 good clinical outcomes. Currently most patients wait over 2 years for a diagnosis. 37 http://bmjopen.bmj.com/ 38 39 40 Incorrect options: 41 42 • Motor neuron disease is a disease of motor neurons, with mixed upper and lower 43 44 motor features. Patients are classically described to have brisk reflexes in wasted

45 on September 24, 2021 by guest. Protected copyright. 46 and fasciculating limbs. As a rule, motor symptoms of weakness predominate and 47 48 patients do not have sensory findings. 49 50 • Multiple sclerosis can also cause paraparesis. As a disease of the central nervous 51 system, the paraparesis is usually associated with upper motor neuron signs. A 52 53 first presentation of multiple sclerosis is most common between the ages of 20 54 55 and 40, younger than this patient. Degenerative Cervical Myelopathy is also more 56 57 common with age and more common than MS overall, making that diagnosis 58 59 more likely. 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 33 of 68 BMJ Open

1 2 • Syringomyelia refers to the development of a syrinx in the spinal cord. It presents

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 with a central cord syndrome, with predominantly upper limb signs. It is a 5 6 relatively uncommon condition. 7 8 • Cauda equina syndrome results from compression of the cauda equina and 9 10 classically includes leg weakness, saddle anaesthesia and sphincter disturbance. 11 12 It is usually an acute syndrome with progressive signs. It is also a relatively 13 14 uncommon condition. 15 16 17 18 References: For peer review only 19 20 1. Kjaer P, Leboeuf-Yde C, Korsholm L, Sorensen JS, Bendix T. Magnetic 21 resonance imaging and low back pain in adults: a diagnostic imaging study of 22 23 40-year-old men and women. Spine. 2005 May 15;30(10):1173–80. 24 25 2. Moore AP, Blumhardt LD. A prospective survey of the causes of non-traumatic 26 27 spastic paraparesis and tetraparesis in 585 patients. Spinal Cord. 1997 28 29 Jun;35(6):361-7. 30 31 3. Kovalova I, Kerkovsky M, Kadanka Z, Kadanka Z, Nemec M, Jurova B, et al. 32 33 Prevalence and Imaging Characteristics of Non-Myelopathic and Myelopathic 34 35 Spondylotic Cervical Cord Compression. Spine. 2016 Aug 9. 36

37 4. Bednarik J, Kadanka Z, Dusek L, Kerkovsky M, Vohanka S, Novotny O, et al. http://bmjopen.bmj.com/ 38 Presymptomatic spondylotic cervical myelopathy: an updated predictive model. 39 40 Eur Spine J. 2008 Mar;17(3):421–31. 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 34 of 68

1 2 A 73-year-old male presents with progressively worsening gait and urinary urgency. He is

3 diagnosed with degenerative cervical myelopathy. Which ONE of the following is true BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 regarding this condition? 6 7 A. Smoking is not a risk factor in isolation 8 B. Asians can have a different underlying aetiology than caucasians 9 10 C. Bowel and bladders symptoms are rare and should prompt consideration of cauda equina 11 syndrome 12 D. Most patients present with a classic triad of neck pain, finger paraesthesias and weak legs 13 E. Family history is of limited value 14 15 16 Answer: B 17 18 For peer review only 19 Asian populations have a higher rate of ossification of the posterior longitudinal 20 21 ligament (OPLL), which can result in myelopathy. 22 23 24 Degenerative cervical myelopathy (DCM) has a number of risk factors, which 25 26 include smoking due to its effects on the intervertebral discs (A, false), genetics 27 28 (option E, false) and occupation - those exposing patients to high axial loading [1]. 29 30 31 32 The presentation of DCM is very variable (option D, false). Early symptoms are 33 34 often subtle and can vary in severity day to day, making the disease difficult to 35 36 detect initially. However as a progressive condition, worsening, deteriorating or new

37 http://bmjopen.bmj.com/ 38 symptoms should be a warning sign. 39 40 41 DCM symptoms can include any combination of [1]: 42 43 • Pain (affecting the neck, upper or lower limbs) 44 45 • Loss of motor function (loss of digital dexterity, preventing simple tasks such as on September 24, 2021 by guest. Protected copyright. 46 47 holding a fork or doing up their shirt buttons, arm or leg weakness/stiffness 48 49 leading to impaired gait and imbalance 50 51 • Loss of sensory function causing numbness 52 53 • Loss of autonomic function (urinary or faecal incontinence and/or impotence) - 54 55 these can occur and do not necessarily suggest cauda equina syndrome in the 56 57 absence of other hallmarks of that condition 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 35 of 68 BMJ Open

1 2 The most common symptoms at presentation of DCM are unknown, but in one

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 series 50% of patients were initially incorrectly diagnosed and sometimes treated 5 6 for carpal tunnel syndrome [2]. 7 8 9 10 References 11 12 1. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 13 14 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 15 16 Supp1 1):S35-41. 17 18 2. Behrbalk E, SalameFor K, peer Regev GJ,review Keynan O, only Boszczyk B, Lidar Z. Delayed 19 20 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 21 Focus. 2013 Jul;35(1):E1. 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 36 of 68

1 2 Stem 1: A 60-year-old gentleman with a background of hypertension and high cholesterol

3 presents with a long history of neck pain, lower limb stiffness and urinary hesitancy. Which BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 of the following is most likely? 6 7 Stem 2: A 56-year-old gentleman presents with lower limb stiffness and imbalance. His 8 only past medical history of note is carpal tunnel syndrome that was diagnosed a year ago 9 10 on clinical grounds and has been refractory to treatment with splints and steroid injections. 11 Which of the following is most likely? 12 13 A. Cauda equina syndrome 14 15 B. Subacute combined degeneration of the cord 16 C. Degenerative cervical myelopathy 17 D. Parkinson’s disease 18 E. Multiple sclerosisFor peer review only 19 20 21 Answer: C 22 23 24 The presentation of degenerative cervical myelopathy [DCM] is variable. Early 25 26 symptoms are often subtle and can vary in severity day to day, making the disease 27 28 difficult to detect initially. However as a progressive condition, worsening, 29 deteriorating or new symptoms should be a warning sign. 30 31 32 33 DCM symptoms can include any combination of [1]: 34 35 • Pain (affecting the neck, upper or lower limbs) 36

37 • Loss of motor function (loss of digital dexterity, preventing simple tasks such as http://bmjopen.bmj.com/ 38 39 holding a fork or doing up their shirt buttons, arm or leg weakness/stiffness 40 41 leading to impaired gait, imbalance and 42 43 • Loss of sensory function causing numbness 44 Loss of autonomic function (urinary or faecal incontinence and/or impotence). 45 • on September 24, 2021 by guest. Protected copyright. 46 47 48 The most common symptoms at presentation of DCM are unknown, but in one 49 50 series 50% of patients were initially incorrectly diagnosed and sometimes treated 51 52 for carpal tunnel syndrome [2]. 53 54 55 56 Other answers: 57 58 • Cauda equina syndrome results from compression of the cauda equina and 59 60 classically includes leg weakness, saddle anaesthesia and sphincter disturbance.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 37 of 68 BMJ Open

1 2 It is usually an acute syndrome with progressive signs. It does not cause leg

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 stiffness. 5 6 • Subacute combined degeneration of the cord results from long-standing vitamin 7 8 B12 deficiency, classically presenting as a posterior cord syndrome – with 9 10 impaired proprioception. It can feature both upper and lower motor neuron signs. 11 12 B12 deficiency can be associated with several neurological features. These 13 14 include a myelopathy (classically the subacute combined degeneration of the 15 16 cord), neuropathy and paraesthesias without neurological signs [3]. Subacute 17 18 combined degenerationFor peer is extremely review rare in developedonly countries, though in 19 20 tropical countries it is frequently the commonest cause of non-traumatic 21 myelopathy [4]. 22 23 • Idiopathic Parkinson’s disease is a tetrad of Tremor, Rigidity, Akinesia and 24 25 Postural Instability (this can be remembered using the TRAP mneumonic). In the 26 27 early stages pain is not a typical feature and it does not cause numbness. 28 29 • Multiple Sclerosis [MS] can have a variable presentation, with both sensory and 30 31 motor symptoms and signs. Inflammatory changes are often present at multiple 32 33 sites, which can cause symptoms at more than one site; a ‘dissociated sensory 34 35 loss’, that is numbness at different and unlinked sites, is a hallmark of MS. Often 36

37 patients will recall previous episodes of odd neurological deficits, which resolved. http://bmjopen.bmj.com/ 38 MS predominantly affects woman (3-4 times common) and usually presents 39 40 before the age of 45. 41 42 43 44 References:

45 on September 24, 2021 by guest. Protected copyright. 46 1. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 47 48 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 49 50 Supp1 1):S35-41. 51 52 2. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 53 54 diagnosis of cervical spondylotic myelopathy by primary care physicians. 55 Neurosurg Focus. 2013 Jul;35(1):E1. 56 57 3. Kumar N1. Neurologic aspects of cobalamin (B12) deficiency. Handb Clin 58 59 Neurol. 2014;120:915-26. 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 38 of 68

1 2 4. Pinto WB, de Souza PV, de Albuquerque MV, Dutra LA, Pedroso JL, Barsottini

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 OG. Clinical and epidemiological profiles of non-traumatic myelopathies. Arq 5 6 Neuropsiquiatr. 2016 Feb;74(2):161-5. 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 39 of 68 BMJ Open

1 2 Select the most likely diagnosis for the following clinical scenarios. Each option can be

37 http://bmjopen.bmj.com/ 38 within the carpal tunnel, and results in lower motor neuron signs, with thenar 39 muscle wasting and weakness of the LOAF muscles (lateral lumbricals, opponens 40 41 pollicis, abductor pollicis brevis and flexor policis brevis). Patients can have 42 43 paraesthesias in the median nerve distribution, classically at night. Tinel’s test and 44

45 Phalen’s test can be positive. on September 24, 2021 by guest. Protected copyright. 46 47 • 3E - Degenerative cervical myelopathy leads to loss of fine motor function in both 48 49 upper limbs. There is a delay in diagnosis of degenerative cervical myelopathy, 50 51 which is estimated to be >2 years in some studies [1]. It is most commonly 52 53 misdiagnosed as carpal tunnel syndrome and in one study, 43% of patients who 54 55 underwent surgery for degenerative cervical myelopathy, had been initially 56 diagnosed with carpal tunnel syndrome [1]. 57 58 59 60 References:

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 40 of 68

1 2 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 5 6 Focus. 2013 Jul;35(1):E1. 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 41 of 68 BMJ Open

1 2 Select the most likely diagnosis for the following clinical scenarios. Each option can be

3 chosen once or not at all. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 A. Multiple sclerosis 7 B. Aortic aneurysm 8 C. Fractured clavicle 9 10 D. Peripheral neuropathy 11 E. Degenerative cervical myelopathy 12 F. Median nerve entrapment 13 G. Ulnar nerve entrapment 14 15 H. Lumbar canal stenosis 16 I. Saturday night palsy 17 J. Adhesive Capsulitis 18 For peer review only 19 20 1. A 60 year-old male presents with clumsy hands. He has been dropping cups around the 21 house. His wife complains he doesn’t answer his mobile as he struggles to use it. His 22 symptoms have been gradually deteriorating over the preceding months. 23 2. A 32 year-old female presents with a 3 day history of altered sensation on her left foot 24 25 and right forearm. On examination she has clonus in both legs and has hyperreflexia in 26 all limbs. 27 3. A 45 year-old female presents with stiffness and pain in her left shoulder, which started 28 around a month ago. She had a similar episode that resolved by itself. Examination 29 30 reveals limited external rotation. 31 32 Answers 33 34 • 1E - Degenerative cervical myelopathy leads to loss of fine motor function in both 35 36 upper limbs. There is a delay in diagnosis of degenerative cervical myelopathy, 37 which is estimated to be >2 years in some studies [1]. It is most commonly http://bmjopen.bmj.com/ 38 39 misdiagnosed as carpal tunnel syndrome and in one study, 43% of patients who 40 41 underwent surgery for degenerative cervical myelopathy, had been initially 42 43 diagnosed with carpal tunnel syndrome [1]. 44 45 • 2A - Multiple sclerosis (MS) can have a variable presentation, affecting both the on September 24, 2021 by guest. Protected copyright. 46 47 sensory and/or motor systems. Inflammatory changes are often present at 48 49 multiple sites, which can cause symptoms at more than one site; a ‘dissociated 50 51 sensory loss’, that is numbness at different and unlinked sites, is a hallmark of 52 53 MS. Often patients will recall previous episodes of odd neurological deficits, which 54 resolved. MS predominantly affects woman (3-4 times common) and usually 55 56 presents before the age of 45. 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 42 of 68

1 2 • 3J - Adhesive capsulitis or ‘frozen shoulder’ is most common in the fifth or sixth

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 decade of life. Women are more likely to be affected than men. It is also more 5 6 common in patients with diabetes mellitus. 7 8 9 References: 10 11 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 12 13 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 14 15 Focus. 2013 Jul;35(1):E1. 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 43 of 68 BMJ Open

1 2 Theme 2: Assessment

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 A 58 year old gentleman presents with left sided paraesthesias affecting his thumb and first 6 finger. He complains of grip weakness and dropping objects unintentionally. On 7 examination, there is wasting over the thenar eminence. Which of the following signs would 8 suggest a diagnosis other than carpal tunnel syndrome? 9 10 11 A. Positive Hoffman’s sign 12 B. Thenar muscle wasting 13 C. Unilateral weakness of pincer grip 14 15 D. Positive Phalen’s test 16 E. Positive Tinnel’s test 17 18 Answer: A For peer review only 19 20 21 22 A positive Hoffman’s sign is a sign of upper motor neuron dysfunction and points to 23 24 a disease of the central nervous system - in this case from the history degenerative 25 26 cervical myelopathy [DCM] affecting the cervical spinal cord is most likely. To elicit 27 28 it, the examiner should flick the patient’s distal phalanx (usually of the middle finger) 29 to cause momentary flexion. A positive sign is exaggerated flexion of the thumb. 30 31 32 33 DCM is often missed initially and there is a delay in the diagnosis of this condition 34 35 by >2 years in some studies [1]. This is a problem as delayed treatment limits 36

37 recovery. It is most commonly misdiagnosed as carpal tunnel syndrome and in one http://bmjopen.bmj.com/ 38 39 study, 43% of patients who underwent surgery for degenerative cervical 40 41 myelopathy, had been initially diagnosed with carpal tunnel syndrome [1]. DCM is 42 43 therefore an important differential in patients suspected to have Carpal Tunnel 44

45 Syndrome [CTS]. on September 24, 2021 by guest. Protected copyright. 46 47 48 CTS is a disease of the peripheral nervous system, resulting from median nerve 49 50 compression at the wrist inside the carpal tunnel. It therefore affects only the 51 52 aspects of the hand innervated by the median nerve: 53 54 • Sensation; Thumb / Index / Middle Finger. This typically manifests as intermittent 55 56 pain or parasthesiae. 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 44 of 68

1 2 • Motor; ‘LOAF Muscles’(lateral lumbricals, opponens pollicis, abductor pollicis

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 brevis and flexor policis brevis). Motor signs are less commonly seen with 5 6 presentations of CTS, but wasting of the thenar eminence may be present. 7 8 9 10 Tinel’s test and Phalen’s test can be positive, but not always. Both tests aim to 11 12 increase the pressure within the carpal tunnel, to try to exacerbate symptoms; 13 14 Tinel’s test via tapping on it and Phalen’s test by sustained full flexion of the wrist. 15 16 17 18 In focal central nervousFor systempeer disorders, review like DCM,only examination features are 19 20 known to have low sensitivity but high specificity [2]. As a disease of the cervical 21 spinal cord, DCM can affect the sensory, motor and autonomic nervous systems 22 23 from the neck downwards. Motor signs will be upper motor neuron signs such as 24 25 increased toned, hyper-reflexia and pyramidal weakness. Note that the neurological 26 27 signs of DCM are often subtle initially and easily missed, but as a progressive 28 29 condition they are likely to get worse [3]. Therefore detecting early DCM can be 30 31 challenging. A high index of suspicion, alongside a comprehensive neurological 32 33 examination and monitoring for progression is required. 34 35 36

37 References: http://bmjopen.bmj.com/ 38 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 39 40 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 41 42 Focus. 2013 Jul;35(1):E1. 43 44 2. Nicholl DJ, Appleton JP. Clinical neurology: why this still matters in the 21st

45 on September 24, 2021 by guest. Protected copyright. 46 century. Journal of Neurology, Neurosurgery & Psychiatry 2015;86:229-33. 47 48 3. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 49 50 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 51 52 Supp1 1):S35-41. 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 45 of 68 BMJ Open

1 2 Which of the following statements is most accurate regarding the usefulness of cervical

3 spine radiographs (X-rays) in the assessment of degenerative cervical myelopathy (DCM)? BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 A. Cervical spine radiographs should be obtained in all patients suspected of having DCM. 7 B. Where DCM is suspected, AP (anteroposterior), lateral and oblique cervical spine 8 radiographs should be requested 9 10 C. Cervical spine radiographs are a useful first line investigation where a diagnosis of 11 DCM is suspected 12 D. Cervical spine radiographs have a low sensitivity but high specificity for DCM 13 E. Cervical spine radiographs cannot diagnose DCM 14 15 16 Answer: E 17 18 For peer review only 19 Radiographs are of limited value where a diagnosis of degenerative cervical 20 21 myelopathy is suspected [1] as they cannot visualise the soft tissue, such as the 22 23 spinal cord. 24 25 26 Spine radiographs have a high sensitivity, but limited specificity to diagnose most 27 28 spinal conditions. Oblique spine radiographs are usually requested in the lumbar 29 30 spine region to pick up defects in the pars interarticularis. They have no value in 31 32 setting of DCM. 33 34 35 36 The finding of spondylosis is common in spinal x-rays of adults over 40 [2]. Its

37 http://bmjopen.bmj.com/ 38 absence does not exclude neural compression. 39 40 41 Degenerative Cervical Myelopathy [DCM] is spinal cord compression due to 42 43 degenerative changes of the surrounding spinal structures; e.g. from disc 44 45 herniation, ligament hypertrophy or calcification, or osteophytes. Therefore in order on September 24, 2021 by guest. Protected copyright. 46 47 to visualise these structures, a MRI is gold standard and first line. 48 49 50 51 Again the presence of such degenerative changes is common on MRI; in one 52 53 study, 57% of patients older than 64 years of age had disc bulging, though only 54 55 26% had spinal cord compression [3]. Therefore a diagnosis of DCM requires the 56 57 finding of MRI compression in concert with appropriate signs and symptoms. 58 59 60 References

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 46 of 68

1 2 1. Nouri A, Tetreault L, Singh A, Karadimas SK, Fehlings MG. Degenerative

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Cervical Myelopathy: Epidemiology, Genetics, and Pathogenesis. Spine (Phila 5 6 Pa 1976). 2015 Jun 15;40(12):E675-93. 7 8 2. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 9 10 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 11 12 Supp1 1):S35-41. 13 14 3. Teresi LM, Lufkin RB, Reicher MA, Moffit BJ, Vinuela FV, Wilson GM, Bentson 15 16 JR, Hanafee WN: Asymptomatic degenerative disk disease and spondylosis of 17 18 the cervical spine:For MR imaging.peer Radiology review 164:83–88, only 1987. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 47 of 68 BMJ Open

1 2 A 67-year-old male undergoes investigations for bilateral paraesthesia in the radial aspects

3 of both hands, over the thumbs and first fingers. He also has paraesthesia in the lateral BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 aspects of both forearms and lower limb spasticity. Blood tests reveal a HBA1c of 46. He 6 undergoes nerve conduction studies and EMG with evidence of denervation. Which ONE of 7 the following diagnoses is most likely? 8 9 10 A. Bilateral carpal tunnel syndrome 11 B. Degenerative cervical myelopathy 12 C. Multiple sclerosis 13 D. Syringomyelia 14 15 E. Diabetic neuropathy 16 17 Answer: B 18 For peer review only 19 20 21 This patient’s twitches are probably fibrillations, a sign of lower motor neuron 22 23 dysfunction. This is confirmed on the neurophysiology report, with evidence of 24 denervation. His symptoms are predominantly in the C6 dermatome distribution 25 26 bilaterally. Although median nerve compression at the elbow bilaterally could in 27 28 theory produce his symptoms, it would be less likely to explain his symptoms given 29 30 his age. He is likely to have degenerative cervical myelopathy. This condition is 31 32 associated with a delay in diagnosis, estimated to be >2 years in some studies [1]. 33 34 35 36 Patients with degenerative cervical myelopathy can present with a number of

37 http://bmjopen.bmj.com/ 38 problems [2]: 39 40 • Pain/stiffness: affecting the neck, upper and/or lower limbs. L’hermitte’s sign is a 41 sharp pain radiating down the spine on flexion of the neck, which is classically 42 43 associated with multiple sclerosis, though it can occur in cervical myelopathy. 44 45 • Loss of function: Clumsiness (e.g. can’t do shirt buttons, hold cup), leg weakness on September 24, 2021 by guest. Protected copyright. 46 47 leading to impaired gait, imbalance and falls. 48 49 • Sphincter disturbance: this can range from frequency and urgency to 50 51 incontinence. 52 53 54 55 Neurological examination can reveal lower motor neuron signs at the level of the 56 57 lesion and upper motor neuron signs below. Note that neurological signs can be 58 subtle and a high degree of suspicion is needed [2]. 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 48 of 68

1 2 The other answers in this question are unlikely for the following reasons:

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 A: bilateral carpal tunnel syndrome would not cause forearm symptoms. Carpal 5 • 6 tunnel syndrome results from median nerve compression at the wrist and results 7 8 in a lower motor neuron picture, with thenar muscle wasting and weakness of the 9 10 LOAF muscles (lateral lumbricals, opponens pollicis, abductor pollicis brevis and 11 12 flexor policis brevis). Tinel’s test and Phalen’s test can be positive. 13 14 • C: multiple sclerosis (MS) is rare in this age group. MS predominantly affects 15 16 woman (3-4 times common) and usually presents before the age of 45. It can 17 18 have a variable presentation,For peer affecting review both the only sensory and/or motor systems. 19 20 Inflammatory changes are often present at multiple sites, which can cause 21 symptoms at more than one site; a ‘dissociated sensory loss’, that is numbness at 22 23 different and unlinked sites, is a hallmark of MS. Often patients will recall previous 24 25 episodes of odd neurological deficits, which resolved. 26 27 • D: Syringomyelia refers to the development of a syrinx in the spinal cord. It 28 29 presents with a central cord syndrome, with predominantly upper limb signs. It is a 30 31 relatively uncommon condition. 32 33 • E: His HBA1c is not within the diagnostic range of diabetes mellitus. Diabetes 34 35 mellitus can cause a peripheral neuropathy presenting in a glove and stocking 36

37 distribution, as well as neuropathy of peripheral nerves - mononeuritis multiplex. http://bmjopen.bmj.com/ 38 39 40 References: 41 42 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 43 44 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg

45 on September 24, 2021 by guest. Protected copyright. 46 Focus. 2013 Jul;35(1):E1. 47 48 2. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 49 50 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 51 52 Supp1 1):S35-41. 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 49 of 68 BMJ Open

1 2

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 50 of 68

1 2 Which of the following investigations is the most important for diagnosing degenerative

3 cervical myelopathy ? BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 A. Nerve conduction studies and EMG 7 8 B. MRI Cervical spine 9 C. CT myelogram 10 D. CT C-spine 11 E. AP and lateral C-spine radiographs 12 13 14 Answer: B 15 16 17 18 An MRI of the cervicalFor spine peer is the gold review standard test only where cervical myelopathy is 19 suspected. It may reveal disc degeneration and ligament hypertrophy, with 20 21 accompanying cord signal change. 22 23 24 25 Other answers: 26 27 • CT imaging is reserved for patients with contraindications to magnetic resonance 28 29 imaging. A CT myelogram is the first line investigation in this case. 30 31 • Radiographs are not clinically useful in the workup of these patients, though 32 33 osteoarthritic changes (e.g. osteophytes) can be visible if they are performed. 34 35 • Other investigatons (e.g. nerve conduction studies, EMG) may be performed 36 when the clinical picture is unclear. These can help to exclude mononeuropathies 37 http://bmjopen.bmj.com/ 38 and other lower motor neuron disorders. However, where there is strong clinical 39 40 suspicion and the diagnosis is suspected, an MRI of the cervical spine should be 41 42 performed. 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 51 of 68 BMJ Open

1 2 A 60-year-old gentleman with a background of lumbar spondylosis and chronic back pain

3 presents with gradually worsening bilateral upper limb paraesthesias and leg stiffness. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 Which one of the investigations below is diagnostic for his likely condition? 6 7 A. Nerve conduction studies and EMG 8 B. MRI Cervical spine 9 10 C. MRI Lumbar Spine 11 D. CT C-spine 12 E. AP and lateral C-spine radiographs 13 14 15 Answer: B 16 17 18 The presence of upperFor limb peer neurological review symptoms onlyindicates that there is pathology 19 20 either within his cervical spinal cord or brain. Brain disease is more likely to cause 21 22 unilateral problems. 23 24 25 26 A MRI lumbar spine would therefore not provide a unifying diagnosis here. 27 28 29 In the context of known lumbar degenerative spine, degenerative cervical 30 31 myelopathy is the number one differential for this presentation. An MRI of the 32 33 cervical spine is the gold standard test where cervical myelopathy is suspected. It 34 35 may reveal disc degeneration and ligament hypertrophy, with accompanying cord 36

37 signal change. It is not uncommon for patients to suffer from tandem (cervical and http://bmjopen.bmj.com/ 38 39 lumbar) stenosis. 40 41 42 43 Other answers: 44

45 • CT imaging is reserved for patients with contraindications to magnetic resonance on September 24, 2021 by guest. Protected copyright. 46 imaging. A CT myelogram is the first line investigation in this case 47 48 • Radiographs are not clinically useful in the workup of these patients, though 49 50 osteoarthritic changes (e.g. osteophytes) can be visible if they are performed. 51 52 • Other investigatons (e.g. nerve conduction studies, EMG) may be performed 53 54 when the clinical picture is unclear. These can help to exclude mononeuropathies 55 56 and other lower motor neuron disorders. However, where there is strong clinical 57 58 suspicion and the diagnosis is suspected, an MRI of the cervical spine should be 59 60 performed.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 52 of 68

1 2

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 53 of 68 BMJ Open

1 2

3 Select the most likely positive examination finding for the following clinical scenarios. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Each option can be chosen once, more than once or not at all. 5 6 7 A. Kernig’s sign 8 B. Ankle Brachial Pressure Index 9 C. Tinel’s test 10 11 D. Straight leg raise 12 E. Tongue fasciculations 13 F. Hoffman’s sign 14 G. Limited external rotation of the shoulder 15 16 I. Hypothenar wasting 17 J. Limited internal rotation of the shoulder 18 For peer review only 19 1. A 70 year-old male with a background of diabetes and hypertension presents with pain 20 21 and weakness in both legs on walking. It settles with rest. 22 2. A 54 year-old female complains of right hand pain radiating into her thumb, index and 23 middle finger. It often wakes her up from sleep. 24 3. A 65 year-old female presents with a loss of dexterity in both hands. She has been 25 26 struggling to type at work and use her mobile phone. Her symptoms have been 27 deteriorating gradually over the preceding months.3 28 29 Answers 30 31 1 - B: this patient is likely to have peripheral vascular disease [PVD] given his 32 33 background risk factors for this condition. The ankle brachial pressure index [ABPI] 34 35 is a simple method of assessing the peripheral circulation. It is calculated by 36

37 dividing systolic blood pressure in the ankle by the the systolic blood pressure in http://bmjopen.bmj.com/ 38 39 the arm. These are equal in health (ABPI = 1). The ABPI is reduced in PVD. 40 2 - C: this patient is likely to have carpal tunnel syndrome. This occurs due to 41 42 median nerve entrapment beneath the flexor retinaculum. Clinical tests to raise 43 44 carpal tunnel pressure can exacerbate symptoms and support a diagnosis. One 45 on September 24, 2021 by guest. Protected copyright. 46 such example is Tinel’s test, includes tapping over the volar surface of the wrist 47 48 joint i.e. over the carpal tunnel, may reproduce paraesthesias. A normal Tinel’s test 49 50 does not exclude carpal tunnel syndrome. 51 52 3 - F: this patient is likely to have degenerative cervical myelopathy [DCM], which is 53 54 associated with upper motor neuron signs. Hoffman’s sign is elicited by flicking the 55 56 distal phalaynx of the middle finger to cause momentary flexion. A positive result is 57 exaggerated flexion of the terminal phalanyx of the thumb. Patients with DCM often 58 59 have subtle signs that are easily missed [1], but as a progressive condition, these 60 are likely to get worse [2]. Whilst the sensitivity of signs is low (i.e. their absence

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 54 of 68

1 2 does not rule out a problem), their specificity is high (i.e. there will be a problem).

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Therefore, in order to diagnose early DCM and improve patient outcomes, a high 5 6 index of suspicion, alongside a 4comprehensive neurological examination and 7 8 monitoring for progression is required. 9 10 11 12 Other signs mentioned: 13 14 • Kernig’s sign refers to painful knee extension, from a position of hip flexion and 15 16 knee flexion. It suggest meningeal irritation e.g. meningitis, subarachnoid 17 18 haemorrhage. For peer review only 19 20 • Straight leg raise: this is positively associated with radicular pathology such as 21 disc herniation. The patient feels pain in the back when the leg is raised between 22 23 30-60 degrees. 24 25 • Limited external rotation is classically found in adhesive capsulitis. Patients have 26 27 global restriction of shoulder movements, in at least two axes, though external 28 29 rotation is usually the most affected and painful. 30 31 32 33 References: 34 35 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 36

37 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg http://bmjopen.bmj.com/ 38 Focus. 2013 Jul;35(1):E1. 39 40 2. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 41 42 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 43 44 Supp1 1):S35-41.

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 55 of 68 BMJ Open

1 2 Select the most likely positive examination finding for the following clinical scenarios.

3 Each option can be chosen once, more than once or not at all. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 A. Kernig’s sign 7 B. Ankle Brachial Pressure Index 8 C. Tinel’s test 9 10 D. Straight leg raise 11 E. Tongue fasciculations 12 F. Hoffman’s sign 13 G. Limited external rotation of the shoulder 14 15 I. Hypothenar wasting 16 J. Limited internal rotation of the shoulder 17 18 1. A 60 year-old maleFor presents peer with clumsy review hands. He onlyhas been dropping cups around the 19 20 house. His wife complains he doesn’t answer his mobile as he struggles to use it. His 21 symptoms have been gradually deteriorating over the preceding months. 22 2. A 32 year-old female presents with a 3 day history of altered sensation of her left foot 23 and right forearm. She had an episode of visual loss a few months ago and says her 24 25 friends have noted her eyes be flickery and jerky. 26 3. A 45 year-old female presents with stiffness and pain in her left shoulder, which started 27 around a month ago. She had a similar episode that resolved by itself. 28 29 30 Answers 31 1 - F: this patient is likely to have degenerative cervical myelopathy [DCM], which is 32 33 associated with upper motor neuron signs. Hoffman’s sign is elicited by flicking the 34 35 distal phalaynx of the middle finger to cause momentary flexion. A positive result is 36

37 exaggerated flexion of the terminal phalanyx of the thumb. Patients with DCM often http://bmjopen.bmj.com/ 38 39 have subtle signs that are easily missed [1], but as a progressive condition, these 40 41 are likely to get worse [2]. Whilst the sensitivity of signs is low (i.e. their absence 42 43 does not rule out a problem), their specificity is high (i.e. there will be a problem). 44 Therefore, in order to diagnose early DCM and improve patient outcomes, a high 45 on September 24, 2021 by guest. Protected copyright. 46 index of suspicion, alongside a comprehensive neurological examination and 47 48 monitoring for progression is required. 49 50 2 - F: this patient is likely to have Multiple Sclerosis (MS). As a disease of the 51 52 central nervous system, MS is usually associated with only upper motor neuron 53 54 signs such as Hoffman’s sign (see above). The patient’s visual loss was probably 55 56 secondary to optic neuritis, a common presentation of MS. Cerebellar signs are 57 58 particularly common with MS and include nystagmus, which is likely to be the jerky 59 60 eye movements noted by her friends.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 56 of 68

1 2 3 - G: this patient is likely to have adhesive capsulitis. Patients have global

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 restriction of shoulder movements, in at least two axes, though external rotation is 5 6 classically described as the most affected and painful. 7 8 9 10 Other signs mentioned: 11 12 • Kernig’s sign refers to painful knee extension, from a position of hip flexion and 13 14 knee flexion. It suggest meningeal irritation e.g. meningitis, subarachnoid 15 16 haemorrhage. 17 18 • Straight leg raise:For this ispeer positively review associated with only radicular pathology such as 19 20 disc herniation. The patient feels pain in the back when the leg is raised between 21 30-60 degrees. 22 23 • The ankle brachial pressure index [ABPI] is a simple method of assessing the 24 25 peripheral circulation. It is calculated by dividing systolic blood pressure in the 26 27 ankle by the the systolic blood pressure in the arm. These are equal in health 28 29 (ABPI = 1). The ABPI is reduced in peripheral vascular disease. 30 31 • Tinel’s test includes tapping over the volar surface of the wrist joint i.e. over the 32 33 carpal tunnel. This can reproduce paraesthesias in patients with carpal tunnel 34 35 syndrome. 36

37 http://bmjopen.bmj.com/ 38 References: 39 40 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 41 42 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 43 44 Focus. 2013 Jul;35(1):E1.

45 on September 24, 2021 by guest. Protected copyright. 46 2. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 47 48 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 49 50 Supp1 1):S35-41. 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 57 of 68 BMJ Open

1 2 Theme 3: Management and follow-up

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 A 75-year old gentleman presents with a short history of neck pain, paraesthesia in his 6 finger tips and progressive leg weakness. Following a MRI scan of his spine, he is 7 diagnosed with degenerative cervical myelopathy due to a C4/5 disc prolapse. Which of the 8 following is the most appropriate management? 9 10 11 A. Cervical decompressive surgery 12 B. Cervical nerve root injection 13 C. Analgesia and referral to physiotherapy 14 15 D. Analgesia and review in 4 weeks time 16 E. Analgesia, a hard cervical collar and review in 4 weeks 17 18 Answer: A For peer review only 19 20 21 22 All patients with degenerative cervical myelopathy should be urgently referred for 23 24 assessment by specialist spinal services (neurosurgery or orthopaedic spinal 25 26 surgery). This is due to the importance of early treatment. The timing of surgery is 27 28 important, as any existing spinal cord damage can be permanent. Early treatment 29 (within 6 months of diagnosis) offers the best chance of a full recovery but at 30 31 present, most patients are presenting too late. In one study, patients averaged over 32 33 5 appointments before diagnosis, representing >2 years [1]. 34 35 36

37 Currently, decompressive surgery is the only effective treatment. It has been shown http://bmjopen.bmj.com/ 38 39 to prevent disease progression. Close observation is an option for mild stable 40 41 disease, but anything progressive or more severe requires surgery to prevent 42 43 further deterioration. Physiotherapy should only be initiated by specialist services, 44

45 as manipulation can cause more spinal cord damage. on September 24, 2021 by guest. Protected copyright. 46 47 48 Prompt diagnosis and onward referral are therefore key to ensuring good outcome 49 50 for your patients. There are national initiatives to raise awareness of the condition 51 52 to try and improve referral times (www.myelopathy.org). All of the other listed 53 54 options in this question do not control the patient’s primary pathology. 55 56 57 58 References: 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 58 of 68

1 2 1. Behrbalk E, Salame K, Regev GJ, et al. Delayed diagnosis of cervical

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 spondylotic myelopathy by primary care physicians. Neurosurg Focus 2013;35:E1. 5 6 doi:10.3171/2013.3.FOCUS1374 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 59 of 68 BMJ Open

1 2 A 65-year old gentleman with a background of osteoarthritis and previous cervical

3 laminectomy for degenerative cervical myelopathy presents with a 2-month history of BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 worsening gait instability and urinary urgency. Which of the following is the most likely 6 explanation for his symptoms? 7 8 A. Transverse myelitis 9 10 B. Recurrent degenerative cervical myelopathy 11 C. Multiple sclerosis 12 D. Cauda equina syndrome 13 E. Spinal metastases 14 15 16 Answer: B 17 18 For peer review only 19 • Postoperatively, patients with cervical myelopathy require ongoing follow-up as 20 21 pathology can “recur” at adjacent spinal levels, which were not treated by the 22 23 initial decompressive surgery. This is called adjacent segment disease. 24 Furthermore, surgery can change spinal dynamics increasing the likelihood of 25 26 other levels being affected. Patients sometimes develop mal-alignment of the 27 28 spine, including kyphosis and spondylolisthesis, and this can also affect the spinal 29 30 cord. All patients with recurrent symptoms should be evaluated urgently by 31 32 specialist spinal services. 33 34 • Transverse myelitis usually presents more acutely than in this case, with a 35 36 sensory level and upper motor neuron signs below the level affected. It can occur

37 http://bmjopen.bmj.com/ 38 in patients with multiple sclerosis or Devic’s disease (neuromyelitis optica). These 39 40 patients tend to also have features such as optic neuritis. 41 • Cauda equina syndrome results from compression of the cauda equina and 42 43 classically includes leg weakness, saddle anaesthesia and sphincter disturbance. 44 45 This gentleman’s history is much more likely to be in keeping with recurrent on September 24, 2021 by guest. Protected copyright. 46 47 cervical myelopathy, given his background and given the subacute presentation 48 49 • Spinal metastases are uncommon, especially in a patient without a known 50 51 primary. Given previous DCM, recurrence is more likely. 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 60 of 68

1 2 A 70 year old man has decompressive surgery for degenerative cervical myelopathy. Three

3 years later he presents with neck pain and hand paraesthesias. Which one of the following BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 management strategies is recommended? 6 7 A. Trial of neuropathic analgesia and cervical nerve root injections 8 B. Investigate with nerve conduction studies and EMG in the first instance 9 10 C. Urgent AP/lateral cervical spine radiographs as an MRI scan is contraindicated 11 D. Urgent referral to spinal surgery or neurosurgery 12 E. Refer to physiotherapy services 13 14 15 Answer: D 16 17 Postoperatively, patients with cervical myelopathy require ongoing follow-up as 18 For peer review only 19 pathology can “recur” at adjacent spinal levels, which were not treated by the initial 20 21 decompressive surgery. 22 23 24 25 Recurrent symptoms should be treated with a high degree of suspicion. Although 26 27 peripheral neuropathy can occur in any patient, this should not be the diagnosis 28 29 that is the most strongly suspected as delays in diagnosis and treatment of DCM 30 31 affect outcomes. Therefore, B is false. 32 33 34 35 All patients with recurrent symptoms should be evaluated urgently by specialist 36 spinal services (A and E, false). Axial spine imaging is necessary and a MRI scan 37 http://bmjopen.bmj.com/ 38 is first line. In patients unable to to have a MRI, CT or CT myelogram may be 39 40 considered. AP and lateral radiographs are of limited use when myelopathy is 41 42 suspected (C, false). 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 References 47 48 1. Kong L, Cao J, Wang L, Shen Y. Prevalence of adjacent segment disease 49 50 following cervical spine surgery: A PRISMA-compliant systematic review and meta- 51 analysis. Medicine (Baltimore). 2016 Jul;95(27):e4171. 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 61 of 68 BMJ Open

1 2 A 65-year-old gentleman is referred to neurology outpatients with arm pain, stiffness and

3 imbalance. Following investigations he is diagnosed with degenerative cervical myelopathy. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 Unfortunately, he misses his next outpatient clinic due to admission with acute coronary 6 syndrome. He attends his GP 2 months later and mentions his ongoing neurological 7 symptoms. Which of the following is the most important next step in his care? 8 9 10 A. Refer to spinal surgery or neurosurgery 11 B. Refer for cervical nerve root injections 12 C. Commence neuropathic analgesia 13 D. Reassure the patient of his diagnosis 14 15 E. Refer for physiotherapy 16 17 Answer: A 18 For peer review only 19 20 Management of patients with cervical myelopathy should be by specialist spinal 21 22 services (neurosurgery or orthopaedic spinal surgery). Decompressive surgery is 23 24 the mainstay of treatment and has been shown to stop disease progression (B, 25 26 false). Close observation is an option for mild stable disease, but anything 27 28 progressive or more severe requires surgery to prevent further deterioration. Pre- 29 operative physiotherapy should only be initiated by specialist services, as 30 31 manipulation can cause more spinal cord damage. 32 33 34 35 The timing of surgery is important, as any existing spinal cord damage can be 36

37 permanent. Treatment within 6 months offers the best chance of making a full http://bmjopen.bmj.com/ 38 39 recovery. At present most patients wait more than 2 years for a diagnosis [1]. 40 41 42 43 Other incorrect options: 44

45 • Neuropathic analgesia is important for symptomatic relief but will not prevent on September 24, 2021 by guest. Protected copyright. 46 further cord damage. 47 48 • Physiotherapy does not replace surgical opinion, it can in fact cause more spinal 49 50 cord damage in patients yet to receive surgical treatment. It should therefore only 51 52 be initiated by specialist services. 53 54 55 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 56 57 diagnosis of cervical spondylotic myelopathy by primary care physicians. 58 59 Neurosurg Focus. 2013 Jul;35(1):E1. 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 62 of 68

1 2

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 63 of 68 BMJ Open

1 2 A 67-year old male recently attended A&E, with a 3 month history of bilateral paraesthesias

3 and twitching affecting the thumb, first finger and lateral forearm. He denied any trauma. A BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 MRI scan of his spine was performed and revealed cervical canal stenosis with mild cord 6 compression. He was discharged and advised to see his GP for follow-up. Which of the 7 following is the most appropriate initial step in management? 8 9 10 A. Refer to spinal surgery services 11 B. Refer for locally commissioned cervical root injections and review after 6 weeks 12 C. Enlist on the weekly minor ops clinic for carpal tunnel decompression 13 D. Commence neuropathic analgesia in the first instance and consider surgical evaluation if 14 15 this does not work 16 E. Refer to physiology services and review in 6 weeks 17 18 Answer: A For peer review only 19 20 21 Bilateral median nerve dysfunction is very suggestive of a diagnosis of 22 23 degenerative cervical myelopathy (DCM) rather than bilateral carpal tunnel 24 25 syndrome (option C). DCM should be suspected in elderly patients presenting with 26 27 limb neurology. His twitches are probably fibrillations, a sign of lower motor neuron 28 29 dysfunction. 30 31 32 33 Degenerative cervical myelopathy is associated with a delay in diagnosis, 34 estimated to be >2 years in some studies [1]. It is most commonly misdiagnosed as 35 36 carpal tunnel syndrome and in one study, 43% of patients who underwent surgery 37 http://bmjopen.bmj.com/ 38 for degenerative cervical myelopathy, had been initially diagnosed with carpal 39 40 tunnel syndrome [1]. Management of these patients should be by specialist spinal 41 42 services (neurosurgery or orthopaedic spinal surgery). Decompressive surgery is 43 44 the mainstay of treatment and has been shown to stop disease progression.

45 on September 24, 2021 by guest. Protected copyright. 46 Physiotherapy and analgesia does not replace surgical opinion, though they may 47 48 be used alongside (options D and E). Nerve root injections do not have a role in 49 50 management (option B). 51 52 53 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 54 55 diagnosis of cervical spondylotic myelopathy by primary care physicians. 56 57 Neurosurg Focus. 2013 Jul;35(1):E1. 58 59 60 5

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 64 of 68

1 2

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 65 of 68 BMJ Open

1 2 Select the best management option for the following clinical scenarios. Each option can be

3 chosen once, more than once or not at all. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 A) Arrange X-rays of the cervical spine 7 B) Refer to pain clinic 8 C) Trial of analgesia and re-review 9 10 D) Refer to spinal surgery 11 E) Refer for physiotherapy 12 F) Lifestyle modification and co-morbidity optimisation 13 G) Trial of a cervical collar 14 15 H) Refer for a minor ops procedure, division of flexor retinaculum 16 I) Refer to neurology 17 J) Refer to elderly care medicine 18 K) Refer to vascularFor surgery peer review only 19 20 21 1. A 70 year-old male with a background of diabetes and hypertension presents with pain 22 and weakness in both legs on walking. It settles with rest. His most recent HBA1c is 56 23 and an ankle-brachial pressure index is calculated as 0.7. 24 25 2. A 35 year-old female who is 30 weeks pregnant complains of right hand pain radiating 26 into her thumb, index and middle finger. It often wakes her up from sleep. 27 3. A 65 year-old female presents with neck pain and loss of dexterity in both hands. She 28 has been struggling to type at work and use her mobile phone. Her symptoms have been 29 30 deteriorating gradually over the preceding months. 31 32 Answers 33 34 1 - F: this patient has peripheral vascular disease, as evident by his ankle-brachial 35 36 pressure index (ABPI). NICE guidance suggests that first line management should 37 include lifestyle modification such as smoking cessation, weight loss, lipid http://bmjopen.bmj.com/ 38 39 modification, optimisation of diabetes mellitus/hypertension and antiplatelet therapy. 40 41 A supervised exercise programme can also be arranged. Local guidelines vary on 42 43 when referral to specialist care is needed, but typically this would be where 44

45 conservative treatment fails after 3 months or the ABPI is below a defined threshold on September 24, 2021 by guest. Protected copyright. 46 47 (e.g. <0.6). 48 49 2 - E: this patient is likely to have carpal tunnel syndrome. This occurs due to 50 51 median nerve entrapment beneath the flexor retinaculum. It is more common in 52 53 pregnant women due to the increase in oedema. NICE clinical knowledge 54 summaries (CKS) recommend lifestyle measures (e.g. wrist ergonomic devices at 55 56 work), as well as wrist splints (usually prescribed by physiotherapists), 57 58 corticosteroid injections or referral for surgical management. Wrist splints can be 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 66 of 68

1 2 helpful for nighttime symptoms, as in her case. Corticosteroid injections require

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 local expertise that may or may not be present. 5 6 3 - D: this patient is likely to have degenerative cervical myelopathy. DCM is often 7 8 missed initially and there is a delay in the diagnosis of this condition by >2 years in 9 10 some studies [1]. Patients have predominantly upper motor neuron signs such as 11 12 increased toned, hyper-reflexia and pyramidal weakness. Neurological signs are 13 14 often subtle initially and easily missed, but as a progressive condition they are likely 15 16 to get worse [2]. Management of these patients should be by specialist spinal 17 18 services (neurosurgeryFor or peerorthopaedic review spinal surgery). only An MRI scan is required for 19 20 diagnosis. All patients should be assessed by a spinal surgeon. 21 22 23 References: 24 25 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 26 27 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 28 29 Focus. 2013 Jul;35(1):E1. 30 31 2. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 32 33 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 34 35 Supp1 1):S35-41. 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 67 of 68 BMJ Open

1 2 Select the best management option for the following clinical scenarios. Each option can be

3 chosen once, more than once or not at all. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 A) Arrange X-rays of the cervical spine 7 B) Refer to pain clinic 8 C) Trial of analgesia and re-review 9 10 D) Refer to spinal surgery 11 E) Refer for physiotherapy 12 F) Lifestyle modification and co-morbidity optimisation 13 G) Trial of a cervical collar 14 15 H) Refer for a minor ops procedure, division of flexor retinaculum 16 I) Refer to neurology 17 J) Refer to elderly care medicine 18 K) Refer to vascularFor surgery peer review only 19 20 21 1. A 60 year-old male presents with clumsy hands. He has been dropping cups around the 22 house. His wife complains he doesn’t answer his mobile as he struggles to use it. His 23 symptoms have been gradually deteriorating over the preceding months. 24 25 2. A 32 year-old female presents with a 3 day history of altered sensation on her left foot 26 and right forearm. She had an episode of visual blurring in her right eye a few months 27 ago which resolved after a few days. Examination reveals brisk reflexes. 28 3. A 45 year-old female presents with stiffness and pain in her left shoulder, which started 29 30 around a month ago. She had a similar episode that resolved by itself. Examination 31 reveals global restriction of shoulder movement, particularly external rotation. 32 33 Answers 34 35 1 - D: this patient is likely to have degenerative cervical myelopathy. DCM is often 36

37 missed initially and there is a delay in the diagnosis of this condition by >2 years in http://bmjopen.bmj.com/ 38 39 some studies [1]. Patients have predominantly upper motor neuron signs such as 40 41 increased toned, hyper-reflexia and pyramidal weakness. Neurological signs are 42 43 often subtle initially and easily missed, but as a progressive condition they are likely 44 to get worse [2]. Management of these patients should be by specialist spinal 45 on September 24, 2021 by guest. Protected copyright. 46 services (neurosurgery or orthopaedic spinal surgery). An MRI scan is required for 47 48 diagnosis. All patients should be assessed by a spinal surgeon. 49 50 2 - I: this patient is likely to have Multiple Sclerosis. Her visual loss was probably 51 52 secondary to optic neuritis, a common ophthalmological association with multiple 53 54 sclerosis. She should be referred to neurology. 55 56 3 - C: this patient is likely to have adhesive capsulitis (frozen shoulder). Patients 57 58 have global restriction of shoulder movements, in at least two axes, though external 59 rotation is classically described as the most affected and painful. Management of 60 frozen shoulder is controversial and there is not much evidence to inform practice.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 68 of 68

1 2 In general, alternative diagnoses should be excluded and pain relief optimised.

3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Gentle shoulder movement is encouraged and there is limited evidence for 5 6 physiotherapy. 7 8 9 10 References: 11 12 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 13 14 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 15 16 Focus. 2013 Jul;35(1):E1. 17 18 2. Baron EM, YoungFor WF. peer Cervical review spondylotic myelopathy: only a brief review of its 19 20 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 21 Supp1 1):S35-41. 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from

A quantitative analysis of medical students’ and physicians’ knowledge of degenerative cervical myelopathy

Journal: BMJ Open ManuscriptFor ID peerbmjopen-2018-028455.R1 review only Article Type: Research

Date Submitted by the 26-Apr-2019 Author:

Primary Subject Neurology Heading: http://bmjopen.bmj.com/ Neurology, Medical education and training, General practice / Family Secondary Subject Heading: practice

Cervical myelopathy, spondylosis, degenerative spine, physician Keywords: knowledge, gap analysis

on September 24, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 62 BMJ Open

37 http://bmjopen.bmj.com/ 38 Liverpool, UK 39 40 41 42 3Department of Academic Neurosurgery, Cambridge University Hospital, University 43 44 of Cambridge, Cambridge, UK

45 on September 24, 2021 by guest. Protected copyright. 46 47 4 48 WT MRC Cambridge Stem Cell Institute, Anne McLaren Laboratory, University of 49 50 Cambridge, UK 51 52 53 54 Corresponding author 55 56 Dr. Mark R.N. Kotter 57 58 WT MRC Cambridge Stem Cell Institute and Department of Clinical Neurosciences 59 60 University of Cambridge

Page 1 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 Acknowledgements 7 8 The authors would like to thank Passmedicine for their assistance with this work. 9 10 11 12 Declarations and conflicts of interest 13 14 Research in the senior author’s laboratory is supported by a core support grant 15 16 from the Wellcome Trust and MRC to the Wellcome Trust-Medical Research Council 17 18 Cambridge StemFor Cell Institute. peer MRNK review is supported only by a NIHR Clinician Scientist 19 20 Award. 21 22 23 24 Disclaimer: This report is independent research arising from a Clinician Scientist 25 26 Award, CS-2015-15-023, supported by the National Institute for Health Research. 27 28 The views expressed in this publication are those of the authors and not necessarily 29 30 those of the NHS, the National Institute for Health Research or the Department of 31 32 Health and Social Care. 33 34 35 36 Word count 37 http://bmjopen.bmj.com/ 38 39 2425 40 41 42 43 Data 44

45 Additional data is provided in supplementary table 1. on September 24, 2021 by guest. Protected copyright. 46 47 48 49 Author statement 50 51 Each author met ICMJE criteria for authorship: 52 53  MW: coordinated creation of questions, performed statistical analysis, wrote 54 55 up all manuscript drafts. 56  JW: created first draft of questions for question-bank, aided in manuscript 57 58 analysis and write-up, helped to check of final draft. 59 60

Page 2 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 62 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4  JG: aided in creation of first draft of questions for question-bank, aided in 5 manuscript analysis and checking of final draft. 6 7  BMD: checked questions, oversaw data analysis, aided in authoring 8 9 introduction and collecting key references, helped to devise project idea, 10 11 checked final manuscript draft. 12  MRNK: devised study idea, provided draft protocol, checked questions, 13 14 edited final manuscript draft. 15 16 17 18 Funding For peer review only 19 20 Research in the senior author’s laboratory is supported by a core support grant 21 22 from the Wellcome Trust and MRC to the Wellcome Trust-Medical Research Council 23 24 Cambridge Stem Cell Institute. MRNK is supported by a NIHR Clinician Scientist 25 26 Award. 27 28 Disclaimer: This report is independent research arising from a Clinician Scientist 29 30 Award, CS-2015-15-023, supported by the National Institute for Health Research. 31 32 The views expressed in this publication are those of the authors and not necessarily 33 34 those of the NHS, the National Institute for Health Research or the Department of 35 36 Health and Social Care.

37 http://bmjopen.bmj.com/ 38 39 40 Data availability statement 41 42 No additional data available in addition to that presented in the supplement. 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 3 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Abstract 5 6 7 8 Objectives. We have previously identified a delay in general practitioner (GP) 9 10 referrals for patients with degenerative cervical myelopathy (DCM). The aim of this 11 12 study was to evaluate whether an education gap existed for DCM along the GP 13 14 training pathway by quantitatively assessing training in, and knowledge of, this 15 16 condition. 17 18 For peer review only 19 20 Design. Gap analysis: comparison of DCM to other conditions. Comparators 21 22 selected on the basis of similar presentation/epidemiology (multiple sclerosis), an 23 24 important spinal emergency (cauda equina syndrome) and a common disease 25 26 (diabetes mellitus). 27 28 29 30 Subjects. Medical students, foundation doctors and GP trainees. 31 32 33 34 Primary and secondary outcome measures. (1) Assessment of training: 35 36 quantitative comparison of references to DCM in curricula

37 http://bmjopen.bmj.com/ 38 (undergraduate/postgraduate) and commonly used textbooks (Oxford Handbook 39 40 Series), to other conditions using modal ranks. (2) Assessment of knowledge: using 41 42 standardised questions placed in an online question-bank (PassMedicine). Results 43 44 were presented relative to the question-bank mean (+/-).

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 Results. DCM had the lowest modal rank of references to the condition in curricula 49 nd 50 analysis and 2 lowest modal rank in textbook analysis. In knowledge analysis 51 52 questions were attempted 127,457 times. Performance for DCM questions in 53 54 themes of presentation (+6.1%) , workup (+0.1%) and management (+1.8%) were 55 56 all greater than the question-bank mean and within one standard deviation. For 57 58 students and junior trainees, there was a serial decrease in performance from 59 60

Page 4 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 62 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 presentation and workup (-0.7% to +10.4% relative to question-bank mean) and 5 6 management (-0.6% to -3.9% relative to question-bank mean). 7 8 9 10 Conclusions. Although infrequently cited in curricula and learning resources, 11 12 knowledge relating to DCM was above average. However, knowledge relating to 13 14 its management was relatively poor. 15 16 17 18 Keywords For peer review only 19 20 Cervical myelopathy; spondylosis; degenerative spine; physician knowledge; gap 21 22 analysis 23 24 25 26 Strengths and limitations of this study 27 28  Strength: Search terms relating to DCM were queried from three UK specific 29 30 medical school curricula and relevant postgraduate curricula 31 32  Strength: a large number of responses were obtained by placing questions 33 34 in an online question-bank, relating to DCM 35 36  Limitation: A limited number of learning resources were searched to assess

37 http://bmjopen.bmj.com/ 38 references to DCM 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 5 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Introduction 5 6 7 Degenerative cervical myelopathy (DCM) is a common and insidious condition that 8 9 can lead to severe disability.1 It arises when degenerative changes of the spine 10 11 12 compress the spinal cord, causing a progressive spinal cord injury. Currently, 13 14 15 treatment is limited to surgical decompression, which is able to stop further injury 16 17 but due to the limited capacity of the spinal cord to repair, recovery is limited. The 18 For peer review only 19 20 timing of surgery is therefore crucial to recovery (‘time is spine’), and a recent 21 22 meta-analysis has demonstrated treatment within 6 months offers a greater chance 23 24 25 of making a full recovery.2 Unfortunately, few patients are diagnosed promptly, 26 27 with the majority waiting more than 2 years for a diagnosis.3 Consequently, most 28 29 30 patients retain life-long disabilities, contributing to quality of life scores lower than 31 32 4, 5 33 cancer, heart and lung diseases. 34 35 36

37 http://bmjopen.bmj.com/ 38 The diagnostic pathway for DCM almost exclusively starts with assessment and 39 40 triage by a community physician, termed in the UK a General Practitioner (GP).3 If 41 42 43 appropriate, the patient is then referred on for further investigation and 44

45 on September 24, 2021 by guest. Protected copyright. 46 management. Our analysis of this pathway has identified time to initial referral by 47 48 GP (6.4±7.7 months) as representing 51% of diagnostic delay.6 49 50 51 52 53 This period of the diagnostic pathway is difficult to examine in detail, and whilst 54 55 56 delayed patient presentation is likely to contribute, delayed detection measured by 57 58 3, 59 multiple consultations and patient perspective, is certainly a relevant component. 60

Page 6 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 62 BMJ Open

1 2 3 7 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 The cause for delayed detection is also likely to be multi-faceted, including subtle, 5 6 non-specific symptoms and incomplete clinical examination.1 However, it is 7 8 9 conceivable that a lack of understanding of the workup and management of DCM 10 11 may also contribute. Indeed, DCM would be included within the spectrum of 12 13 14 “neurophobia” – an aversion to the neurosciences due to perceived difficulty, which 15 16 8, 9 17 has been demonstrated in GPs and GP trainees. 18 For peer review only 19 20 21 22 Our objective therefore was to evaluate whether an education gap exists for DCM 23 24 amongst GPs, by quantitatively assessing their training in, and knowledge of, DCM. 25 26 27 28 29 30 Methods 31 32 33 34 35 Ethics 36

37 http://bmjopen.bmj.com/ 38 The present study design did not require ethical approval. 39 40 41 42 43 Patient and Public Involvement 44

45 on September 24, 2021 by guest. Protected copyright. 46 DCM patients were surveyed online and this confirmed that the diagnosis of DCM 47 48 is frequently delayed.10 The question of whether this is due to lack of knowledge 49 50 51 amongst health professionals and whether this was the consequence of a gap in 52 53 medical education was formulated with the input of DCM sufferers at the First 54 55 56 Cambridge Myelopathy day. 57 58 59 60

Page 7 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 Education Gap Analysis 7 8 9 A gap analysis is a process to identify gaps in existing systems such as education 10 11 curricula. It involves an assessment of existing knowledge or standards against pre- 12 13 14 determined standards that define core competency requirements.11, 12 This differs, 15 16 17 for example, from simple cross sectional knowledge assessments whose primary 18 For peer review only 19 outcome is usually assessment of knowledge beyond core requirements. 20 21 22 23 24 The study objective was therefore approached in two separate gap analyses 25 26 27 (Table 1): 28 29 30 1. Assessment of Training: quantification of DCM in curricula and commonly 31 32 used learning resources, including assessment of relative importance 33 34 35 through comparison to other conditions. 36 37 2. Assessment of Knowledge: formal assessment of trainees’ knowledge using http://bmjopen.bmj.com/ 38 39 40 questions placed in an online question-bank. 41 42 43 44 45 The GP Training Pathway on September 24, 2021 by guest. Protected copyright. 46 47 48 UK medical graduates complete the UK Foundation Programme to enter higher 49 50 specialty training as formal GP trainees. There are several assessments along this 51 52 53 route including: medical school final examinations to gain registration to practice 54 55 56 in the UK, or the equivalent Professional and Linguistic Assessments Board test 57 58 (PLAB) for international medical graduates; the Specialty Recruitment Assessment 59 60

Page 8 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 62 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 (SRA) – a written assessment taken towards the start of the second foundation year 5 6 prior to higher specialty applications; and the Membership of The Royal College of 7 8 9 General Practitioner’s (MRCGP), an exit exam for GP trainees. Whilst alternative 10 11 entry routes to General Practice exist, this pathway represents the most common 12 13 14 training route for GPs today. 15 16 17 18 For peer review only 19 Definition of DCM and comparators 20 21 22 DCM was chosen as an inclusive term for a variety of diseases resulting in 23 24 compressive myelopathy.1 Comparator diseases were also selected to compare 25 26 27 findings to: 28 29 30 31 32 1. Direct comparator to myelopathy: a disease that is a differential diagnosis 33 34 35 for DCM with equivalent or greater incidence. 36 37 2. Degenerative spine comparator: an alternative degenerative spine disease http://bmjopen.bmj.com/ 38 39 40 that is widely taught. 41 42 43 3. Generic non-neuroscience comparator: a common disease that all clinicians 44 45 would have some knowledge about and interaction with. on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 The a priori hypothesis was, if an education gap existed within DCM, we would 51 52 53 expect metrics of training and knowledge of DCM to rank inferior to controls. 54 55 56 Controls were identified through a consensus author meeting and finalised after 57 58 confirmation of epidemiological profiles through literature review. 59 60

Page 9 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 Assessment of Training: Learning Resource Analysis 7 8 9 We selected learning resources that were in common usage by trainees, available 10 11 in electronic format (to be amenable for searching) and stratifiable by training 12 13 14 stage. Specifically, training curricula, the Oxford Handbook series 13 and online 15 16 17 question-banks were selected. Alongside the United Kingdom Foundation 18 For peer review only 19 Programme and MRCGP, an example from each of the three main, UK medical 20 21 22 school teaching models 14 were included: problem-based learning (The University 23 24 of Manchester, UK); traditional lecture based (The University of Cambridge, UK); 25 26 27 and an integrated system including aspects of both (Imperial College London, UK). 28 29 30 The choice of medical school curricula was pragmatically selected, based on access. 31 32 The handbooks most pertinent to the GP training pathway were included in our 33 34 35 analysis: The Oxford Handbook of Clinical Medicine (OHCM), The Oxford Handbook 36 37 for the Foundation Programme (OHFP) and The Oxford Handbook of General http://bmjopen.bmj.com/ 38 39 40 Practice (OHGP). 41 42 43 44 45 Search terms for analysing learning resources were created for DCM and each on September 24, 2021 by guest. Protected copyright. 46 47 48 comparator using relevant terms from the search syntax of Cochrane Reviews, or if 49 50 absent, recent systematic reviews. Curricula were searched for the number of 51 52 53 references per disease, text books for the number of words per disease section and 54 55 56 question banks for the number of questions. 57 58 59 60

Page 10 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 62 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Assessment of Knowledge 5 6 The author panel includes GPs with an interest in education (JW), university 7 8 9 appointed educationalists (JG) and neurosurgeons (MW, BMD, MRK). The authors 10 11 devised a set of questions to be included in an online question bank, composed 12 13 14 of both multiple choice questions [MCQs] and extended matching questions 15 16 17 [EMQs], designed to cover the different components of medical assessment; 18 For peer review only 19 presentation, investigation and management. The final set consisted of 19 20 21 22 questions (13 multiple choice questions [MCQs] and 6 extended matching 23 24 questions [EMQs]). The questions together with answers are included in 25 26 27 Supplement 1. 28 29 30 31 32 Online question-banks including subsections devoted to each stage of the GP 33 34 35 training pathway were contacted with details of the study. Only one question-bank 36 37 - Passmedicine 15, responded and was therefore selected for this arm of the study. http://bmjopen.bmj.com/ 38 39 40 41 42 43 Analysis 44 45 Statistical analysis was performed using SPSS version 22 (SPSS Inc., Chicago, Illinois, on September 24, 2021 by guest. Protected copyright. 46 47 48 USA). For assessment of training, frequencies of search terms were described and 49 50 the mode and modal ranks determined. For assessment of knowledge, histograms 51 52 53 were constructed with either a condition or question-bank along the X-axis and 54 55 56 user performance relative to question-bank mean along the Y-axis. The user 57 58 performance relative to question-bank mean corresponded to the raw difference 59 60

Page 11 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 between the mean value for each theme and the mean value for a particular 5 6 question-bank. 7 8 9 10 11 Results 12 13 14 15 16 17 Selection of Comparators and Resources 18 For peer review only 19 20 21 22 Multiple sclerosis (MS), cauda equina syndrome (CES) and diabetes mellitus were 23 24 selected as the three disease comparators and list of search terms compiled 25 26 27 accordingly (Table 2).16-19 MS is a common differential for DCM,20 with overlapping 28 29 1 30 signs and symptoms. Whilst DCM is likely to be more prevalent in reality, currently 31 32 their characterised epidemiology is comparable.21, 22 CES is uncommon, but its 33 34 35 missed or delayed diagnosis carries significant consequences. Diabetes is relevant 36 37 to all medical fields. The question-bank also provided data on neurology as a http://bmjopen.bmj.com/ 38 39 40 theme, encompassing all questions relating to the central and peripheral nervous 41 42 system. 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 Assessment of training 49 50 51 52 53 Curricula analysis 54 55 DCM and cauda equina syndrome had the lowest modal rank in curricula search 56 57 58 analysis (Table 3). This was true for early stage undergraduate curricula and late 59 60

Page 12 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 62 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 stage curricula for the foundation programme and MRCGP. Multiple sclerosis was 5 6 mentioned less frequently in early versus late stage curricula. The opposite trend 7 8 9 was observed for diabetes mellitus. 10 11 12 13 14 Textbook analysis 15 16 17 Overall, DCM had the second lowest modal rank but above that of cauda equina 18 For peer review only 19 syndrome (Table 4). The relative word count attributed to neurological conditions 20 21 22 such as DCM, multiple sclerosis and cauda equina syndrome decreased with 23 24 advancing stage of textbook. The opposite trend was observed for diabetes 25 26 27 mellitus, for whom the total word count increased. 28 29 30 Question Bank Analysis 31 32 33 34 Assessment of Knowledge 35 36

37 Passmedicine was approached, on the basis it did not have any DCM questions http://bmjopen.bmj.com/ 38 39 previously which could influence analysis. Questions were introduced between the 40 41 42 period June 2017 – October 2017. Section editors for each question-bank ultimately 43 44 selected which of the 19 questions were relevant, such that finals/PLAB contained 45 on September 24, 2021 by guest. Protected copyright. 46 47 14, SRA – 19 and MRCGP – 19. 48 49 50 51 52 Overall, questions were attempted 127,457 times; finals/PLAB - 36706; SRA - 47530; 53 54 55 MRCGP – 43221 (Supplementary Table 1). Data on the number of unique attempts 56 57 or first-time viewers was not extractable. 58 59 60

Page 13 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 There were differences in user performance in the three DCM question themes – 5 6 presentation, workup and management, though average scores were all within 1 7 8 9 standard deviation of the mean (Fig. 1). Performance sequentially decreased across 10 11 these themes for the finals/PLAB group (presentation +4.7%, workup +4.2%, 12 13 14 management -0.6%; relative to question-bank mean), SRA group (presentation 15 16 17 +10.4%, workup -0.7%, management -3.9%; relative to question-bank mean) and 18 For peer review only 19 when all groups were considered together (presentation +6.1%, workup +0.1%, 20 21 22 management +1.8%; relative to question-bank mean). In the MRCGP group 23 24 however, performance clustered around the mean for the question bank with little 25 26 27 variability (presentation -0.1%, workup +1.0%, management +1.8%; relative to 28 29 30 question-bank mean). 31 32 33 34 35 User performance by disease showed large variation between question-banks (Fig. 36 37 2). The mean user performance decreased sequentially for DCM with advancing http://bmjopen.bmj.com/ 38 39 40 question-bank (finals/PLAB +4.2%, SRA +2.7%, MRCGP +0.9%; relative to question- 41 42 43 bank mean). However, the mean user performance for DCM was always greater 44 45 than the question-bank mean (+2.6% relative to question-bank mean). on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 Discussion 52 53 In this study, we found that DCM is relatively infrequently cited in curricula and 54 55 56 commonly used textbooks, even compared to diseases such as multiple sclerosis, 57 58 59 with similar epidemiology. Despite this however, user performance in some of the 60

Page 14 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 62 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 DCM questions remained consistently above question-bank averages. There was a 5 6 sequential decrease in user performance across the themes of DCM presentation, 7 8 9 workup and management for early years’ trainees, whereas for senior trainees, 10 11 performance did not vary by theme of question. For other neuroscience themes 12 13 14 such as multiple sclerosis and cauda equina syndrome, the user performance was 15 16 17 below average. This was in contrast to user performance in questions relating to 18 For peer review only 19 diabetes mellitus, which was consistently above question-bank mean. 20 21 22 23 24 We observed a below average performance in questions grouped under the 25 26 27 neurology theme, which included all questions relating to the central and peripheral 28 29 30 nervous system. Reduced knowledge pertaining to neurosciences has previously 31 32 been linked to a term called neurophobia, though this is by no means a universally 33 34 35 accepted concept.8, 9, 23 In one questionnaire study for example, GPs rated 36 37 neurology as the most difficulty medical specialty and the one for which they had http://bmjopen.bmj.com/ 38 39 40 the least confidence compared to cardiology, endocrinology, gastroenterology, 41 42 9 43 geriatrics, respiratory medicine and rheumatology. However, the question-bank 44 45 data did not allow distinction between basic science and clinical questions, for on September 24, 2021 by guest. Protected copyright. 46 47 48 whom the performance may be different, as evidenced by the above average 49 50 performance in our clinically orientated DCM questions. 51 52 53 54 55 56 Secondly, our study supports the observation of poorer performance in 57 58 management themes by clinicians at an earlier training stage. Prior studies on 59 60

Page 15 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 medical students that have evaluated knowledge formally have shown similar 5 6 results in specialties such as ophthalmology.24 This also appears to translate into 7 8 9 patient encounters during clinical assessments in neurology.25, 26 In one study 10 11 evaluating student performance in neurology outpatients, students were more 12 13 14 likely to make errors regarding diagnostic tests or planning treatment than lesion 15 16 25 17 localisation. Another study evaluating student performance in neurology themed 18 For peer review only 19 objective structured clinical examinations (OSCEs) also reported similar results, 20 21 22 finding poorer student performance in supportive management of neurological 23 24 conditions versus diagnosis.26 25 26 27 28 29 30 Our original hypothesis attempting to correlate delays in diagnosis for DCM with 31 32 a deficiency in training and knowledge, was not conclusively demonstrated in the 33 34 35 present study. Earlier stage trainee doctors demonstrated comparably higher levels 36 37 of knowledge with regards to the presentation and work up of DCM and a potential http://bmjopen.bmj.com/ 38 39 40 gap with regards to management-related questions. These differences disappeared 41 42 43 in individuals studying for the GP exit exams. Nevertheless, the clinical argument 44 45 for an education gap is strong - our previous analysis of the referral pathway for on September 24, 2021 by guest. Protected copyright. 46 47 48 DCM identified a time to initial referral by the GP as representing 51% of the overall 49 50 diagnostic delay. 51 52 53 54 55 56 This study demonstrated DCM received relatively less training than other 57 58 conditions. In the West, the prevalence of DCM is underestimated but at least 60 59 60

Page 16 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 62 BMJ Open

1 2 3 21, 22 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 per 100,000, compared to around 100 per 100,000 for multiple sclerosis. 5 6 Despite this however, DCM was under-represented by eight-fold compared to 7 8 9 multiple sclerosis in curricula analysis. In late stage curricula (foundation 10 11 programme, MRCGP), DCM was not referred to at all. A similar trend was observed 12 13 14 in textbook analysis, though neither of these analyses were exhaustive. These 15 16 17 results were in contrast to our knowledge analysis, which showed consistently 18 For peer review only 19 above average performance for DCM unlike multiple sclerosis. 20 21 22 23 24 One potential way of reconciling the absence of a gap in formal knowledge with 25 26 27 the deficiencies in the diagnosis and management of DCM, is that factual 28 29 30 knowledge may not translate into recognition of the disease in the clinical context. 31 32 Causal association between training and clinical practice have not been clearly 33 34 35 demonstrated, albeit prior studies have found good correlation between exam 36 37 performance and clinical performance in a variety of medical and surgical http://bmjopen.bmj.com/ 38 39 40 specialties.27-29 Most of this data is based on the United States Medical Licensing 41 42 43 Examination (USMLE), which appears to correlate well with residency 44 45 performance.27, 28 Beyond medical school exams, in another study, internal medicine on September 24, 2021 by guest. Protected copyright. 46 47 48 clinicians failing their maintenance of certification exams had a more than double 49 50 chance of disciplinary action versus those who passed.29 An alternative explanation 51 52 53 is that questionnaire based methods may not be sufficiently sensitive to detect 54 55 56 poor clinical decision making in the context of DCM. Future studies should consider 57 58 59 60

Page 17 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 employing mock patients for the condition in undergraduate and postgraduate 5 6 OSCEs, to compare performance to other conditions. 7 8 9 10 11 There are several limitations to be considered in the context of these findings. 12 13 14 Firstly, only select learning resources were used in this study and questions were 15 16 17 inserted into a single online question bank. Secondly, question bank data was not 18 For peer review only 19 extractable per user, nor per first time answer. Whilst this was not a comprehensive 20 21 22 strategy, this is unlikely to have limited our results for at least two reasons: 23 24 questions were attempted more than 100,000 times and analysis was relative within 25 26 27 learning resources, and therefore limitations were applicable to each comparator. 28 29 30 The native question bank data extraction technique also did not provide data on 31 32 the number of attempts on question themes other than DCM. 33 34 35 36 37 The methodology employed in the present study may not have been sufficiently http://bmjopen.bmj.com/ 38 39 40 sensitive to detect a knowledge gap for DCM, for several reasons. Firstly, our 41 42 43 questions may not have been challenging enough given that they tested core 44 45 principles all clinicians would be expected to achieve. Although we did not employ on September 24, 2021 by guest. Protected copyright. 46 47 48 pilot testing of the questions in our target population, questions were designed by 49 50 an experienced author panel including educationalists. Furthermore, our questions 51 52 53 were subject to additional scrutiny by the question-bank editors, such that only 54 55 56 those questions deemed appropriate were included in a particular question-bank. 57 58 Secondly, there was no first-time answer data available for analysis. This means 59 60

Page 18 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 62 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 that ‘random effect answering’, users selecting an answer to read the explanation, 5 6 was not accounted for, though this was the case for both DCM and controls. Thirdly, 7 8 9 our controls may not have been appropriate and certainly, educational comparisons 10 11 between specialties is not always the best method of assessing knowledge 12 13 14 adequacy.30 Finally, our methodology assumed a direct correlation between “on 15 16 17 paper” knowledge and clinical care provided to patients – though this was not 18 For peer review only 19 proven in the present study and could still differ vastly. 20 21 22 23 24 25 Conclusion 26 27 In this study, we set out to evaluate whether there is a deficiency in training and 28 29 30 knowledge regarding DCM in the GP training pathway. Although DCM was 31 32 33 infrequently referenced in learning resources, trainees performed above average 34 35 for DCM questions in an assessment of knowledge using an online question-bank. 36

37 http://bmjopen.bmj.com/ 38 The clinical suggestion of an education gap is strong, and these contrasting 39 40 findings do not conclusively allay this concern. Future studies are required to better 41 42 43 understand these observations. 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 References 49 50 51 52 53 1. Davies BM, Mowforth OD, Smith EK, et al. Degenerative cervical myelopathy. BMJ 54 55 2018;360:k186. doi: 10.1136/bmj.k186 [published Online First: 2018/02/24] 56 2. Fehlings MG, Tetreault LA, Riew KD, et al. A Clinical Practice Guideline for the 57 58 Management of Patients With Degenerative Cervical Myelopathy: 59 60 Recommendations for Patients With Mild, Moderate, and Severe Disease and

Page 19 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Nonmyelopathic Patients With Evidence of Cord Compression. Global spine 5 journal 2017;7(3 Suppl):70S-83S. doi: 10.1177/2192568217701914 [published 6 7 Online First: 2017/11/23] 8 9 3. Behrbalk E, Salame K, Regev GJ, et al. Delayed diagnosis of cervical spondylotic 10 11 myelopathy by primary care physicians. Neurosurgical focus 2013;35(1):E1. 12 doi: 10.3171/2013.3.FOCUS1374 [published Online First: 2013/07/03] 13 14 4. Fehlings MG, Ibrahim A, Tetreault L, et al. A global perspective on the outcomes 15 16 of surgical decompression in patients with cervical spondylotic myelopathy: 17 results from the prospective multicenter AOSpine international study on 479 18 For peer review only 19 patients. Spine 2015;40(17):1322-8. doi: 10.1097/BRS.0000000000000988 20 21 [published Online First: 2015/05/29] 22 23 5. Oh T, Lafage R, Lafage V, et al. Comparing Quality of Life in Cervical Spondylotic 24 Myelopathy with Other Chronic Debilitating Diseases Using the Short Form 25 26 Survey 36-Health Survey. World Neurosurg 2017;106:699-706. doi: 27 28 10.1016/j.wneu.2016.12.124 [published Online First: 2017/01/10] 29 30 6. Hilton B T-MJ, Davies BM, Kotter MRN. Route to diagnosis of degenerative 31 cervical myelopathy in a UK healthcare system: a retrospective cohort study. 32 33 BMJ Open (in-press) 2018 34 35 7. Elemental Ideas – Cervical Spondylotic Myelopathy Part One [cited 2018 October 36

37 12th]. Available from: http://archive.cambridge-tv.co.uk/elemental-ideas-cervical- http://bmjopen.bmj.com/ 38 spondylotic-myelopathy-part-one/ accessed October 12th 2018. 39 40 8. McCarron MO, Stevenson M, Loftus AM, et al. Neurophobia among general 41 42 practice trainees: the evidence, perceived causes and solutions. Clinical 43 neurology and neurosurgery 2014;122:124-8. doi: 44 45 10.1016/j.clineuro.2014.03.021 [published Online First: 2014/05/06] on September 24, 2021 by guest. Protected copyright. 46 47 9. Loftus AM, Wade C, McCarron MO. Primary care perceptions of neurology and 48 49 neurology services. Postgraduate medical journal 2016;92(1088):318-21. doi: 50 10.1136/postgradmedj-2015-133683 [published Online First: 2016/01/23] 51 52 10. Davies BM, Munro CF, Kotter MR. A Novel Insight Into the Challenges of 53 54 Diagnosing Degenerative Cervical Myelopathy Using Web-Based Symptom 55 56 Checkers. J Med Internet Res 2019;21(1):e10868. doi: 10.2196/10868 57 [published Online First: 2018/10/10] 58 59 60

Page 20 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 62 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 11. Fater KH. Gap analysis: a method to assess core competency development in 5 the curriculum. Nurs Educ Perspect 2013;34(2):101-5. [published Online First: 6 7 2013/06/15] 8 9 12. O'Brien CE, Stafford R, Franks AM. Establishment of a patient-centered 10 11 communication course to address curricular gaps. Curr Pharm Teach Learn 12 2018;10(7):933-39. doi: 10.1016/j.cptl.2018.04.008 [published Online First: 13 14 2018/09/22] 15 16 13. Warriner D. The Oxford Handbook of Clinical Medicine. BMJ 17 2008;336(7640):393-93. doi: 10.1136/bmj.39451.675451.59 18 For peer review only 19 14. NHS careers. Teaching styles at medical school [27/05/2018]. Available from: 20 21 https://www.healthcareers.nhs.uk/sites/default/files/documents/Teaching styles at 22 medical school.pdf accessed 27/05/2018. 23 24 15. Passmedicine 2018 [cited 2018 12th October]. Available from: 25 26 https://www.passmedicine.com/ accessed 12 October 2018. 27 16. Kiely PD, Quinn JC, Du JY, et al. Posterior surgical treatment of cervical 28 29 spondylotic myelopathy: review article. HSS J 2015;11(1):36-42. doi: 30 31 10.1007/s11420-014-9425-5 [published Online First: 2015/03/05] 32 33 17. Khan F, Amatya B. Rehabilitation in Multiple Sclerosis: A Systematic Review of 34 Systematic Reviews. Arch Phys Med Rehabil 2017;98(2):353-67. doi: 35 36 10.1016/j.apmr.2016.04.016 [published Online First: 2016/05/25]

37 http://bmjopen.bmj.com/ 38 18. Chau AM, Xu LL, Pelzer NR, et al. Timing of surgical intervention in cauda equina 39 syndrome: a systematic critical review. World Neurosurg 2014;81(3-4):640- 40 41 50. doi: 10.1016/j.wneu.2013.11.007 [published Online First: 2013/11/19] 42 43 19. Richter B, Bandeira-Echtler E, Bergerhoff K, et al. Pioglitazone for type 2 diabetes 44

45 mellitus. Cochrane Database Syst Rev 2006(4):CD006060. doi: on September 24, 2021 by guest. Protected copyright. 46 10.1002/14651858.CD006060.pub2 [published Online First: 2006/10/21] 47 48 20. Davies BM, Munro CF, Kotter M. A Novel Insight into the Challenges of 49 50 Diagnosing Degenerative Cervical Myelopathy using Online Symptom 51 52 Checkers. J Med Internet Res 2018 doi: 10.2196/10868 [published Online 53 First: 2018/10/10] 54 55 21. Nouri A, Tetreault L, Singh A, et al. Degenerative Cervical Myelopathy: 56 57 Epidemiology, Genetics, and Pathogenesis. Spine 2015;40(12):E675-93. doi: 58 10.1097/BRS.0000000000000913 [published Online First: 2015/04/04] 59 60

Page 21 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 22. Dilokthornsakul P, Valuck RJ, Nair KV, et al. Multiple sclerosis prevalence in the 5 United States commercially insured population. Neurology 2016;86(11):1014- 6 7 21. doi: 10.1212/WNL.0000000000002469 [published Online First: 8 9 2016/02/19] 10 11 23. Flanagan E, Walsh C, Tubridy N. 'Neurophobia'--attitudes of medical students 12 and doctors in Ireland to neurological teaching. Eur J Neurol 13 14 2007;14(10):1109-12. doi: 10.1111/j.1468-1331.2007.01911.x [published 15 16 Online First: 2007/09/21] 17 24. Esparaz ES, Binder SB, Borges NJ. How prepared are medical students to 18 For peer review only 19 diagnose and manage common ocular conditions. J Educ Eval Health Prof 20 21 2014;11:29. doi: 10.3352/jeehp.2014.11.29 [published Online First: 22 23 2014/11/25] 24 25. Davis LE, King MK. Assessment of medical student clinical competencies in the 25 26 neurology clinic. Neurology 2007;68(8):597-9. doi: 27 28 10.1212/01.wnl.0000254490.24555.f7 [published Online First: 2007/02/21] 29 30 26. Lukas RV, Blood A, Park YS, et al. Assessment of neurological clinical 31 management reasoning in medical students. Journal of clinical neuroscience 32 33 : official journal of the Neurosurgical Society of Australasia 2014;21(6):919- 34 35 22. doi: 10.1016/j.jocn.2013.09.006 [published Online First: 2013/12/07] 36

37 27. Thundiyil JG, Modica RF, Silvestri S, et al. Do United States Medical Licensing http://bmjopen.bmj.com/ 38 Examination (USMLE) scores predict in-training test performance for 39 40 emergency medicine residents? J Emerg Med 2010;38(1):65-9. doi: 41 42 10.1016/j.jemermed.2008.04.010 [published Online First: 2008/10/28] 43 28. Sutton E, Richardson JD, Ziegler C, et al. Is USMLE Step 1 score a valid predictor 44 45 of success in surgical residency? Am J Surg 2014;208(6):1029-34; discussion on September 24, 2021 by guest. Protected copyright. 46 47 34. doi: 10.1016/j.amjsurg.2014.06.032 [published Online First: 2014/12/03] 48 49 29. McDonald FS, Duhigg LM, Arnold GK, et al. The American Board of Internal 50 Medicine Maintenance of Certification Examination and State Medical Board 51 52 Disciplinary Actions: a Population Cohort Study. J Gen Intern Med 2018 doi: 53 54 10.1007/s11606-018-4376-z [published Online First: 2018/03/09] 55 56 30. Di Paola M, Bennet GC, Shearer JR. Orthopaedic teaching in United Kingdom 57 medical schools. Med Teach 1986;8(2):155-8. [published Online First: 58 59 1986/01/01] 60

Page 22 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 62 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 23 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Figure captions 5 6 7 8 9 Figure 1. Trainee knowledge analysis: user performance by DCM question 10 11 12 theme. DCM questions assessed three themes - presentation, workup and 13 14 15 management. Early stage trainees, those sitting finals/PLAB and the SRA, performed 16 17 worse in questions relating to management versus more senior trainees taking the 18 For peer review only 19 20 MRCGP. Performance for MRCGP trainees was more consistent. The average 21 22 performance for each question theme was within 1 standard deviation (SD) of the 23 24 25 mean. 26 27 28 29 30 Figure 2. Trainee knowledge analysis: user performance by disease. The mean 31 32 33 user performance for DCM, neurology and cauda equina syndrome decreased with 34 35 advancing question-bank. This trend was observed in most neurological 36

37 http://bmjopen.bmj.com/ 38 pathologies. User performance of diabetes mellitus was more variable. * = there 39 40 were no questions on cauda equina syndrome present in the SRA question-bank. 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 62 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Table 1. Summary of gap analysis methods. This table shows the methods used 5 in this study. Curricula and textbooks were screened by training stage to assess 6 7 references to key search terms. An online question bank was used in knowledge 8 9 assessment. 10 11 12 Medical Foundation GP Training Metric 13 14 School 15 16 Curricula UoM Foundation MRCGP References to 17 18 ForUoC peerprogramme review curriculumonly search terms 19 Imperial curriculum 20 21 Text Book OHCM OHFP OHGP References to 22 23 search terms 24 25 26 Online PLAB SRA AKT Performance 27 28 Question Medical finals in questions 29 30 Bank 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 25 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Table 2. Selected diseases for textbook and curricula searches. The reasoning 5 for each of the comparators is explained in this table. The search terms employed 6 7 for the curricula searches are shown below this. 8 9 10 11 12 Degenerative Cauda equina Multiple sclerosis Diabetes mellitus 13 cervical myelopathy syndrome 14 Disease of interest A disease with a Widely taught spinal Non-spinal control 15 16 similar incidence and emergency, selected 17 morbidity to as a spinal control. 18 For peerdegenerative review cervical only 19 Reasoning myelopathy. Selected 20 as a clinico- 21 epidemiological 22 control 22 23 24 25 • Cervical myelopathy • Multiple sclerosis • Cauda equina • Diabetes mellitus 26 • Cervical • Demyelinating • Saddle anaesthesia • Insulin dependent 27 myeloradiculopathy disease diabetes mellitus 28 • Cervical stenosis • Demyelination • Type 1 diabetes 29 • Cervical mellitus 30 31 compression • Non insulin 32 • Cervical herniation dependent diabetes 33 • Cervical mellitus 34 degeneration • Type 2 diabetes 35 • Ossification of mellitus 36 Search terms posterior • Maturity onset 37 http://bmjopen.bmj.com/ 38 for curricula longitudinal diabetes of young 39 searches ligament • Gestational diabetes 40 • Spinal mellitus 41 osteophytosis • Late onset diabetes 42 • Spinal cord • Maturity onset 43 compression diabetes 44 • Spondylosis • Insulin resistance 45 on September 24, 2021 by guest. Protected copyright. 46 • Diabetic ketoacidosis 47 • Hyperosmolar 48 hyperglycaemic state 49 • Hyperosmolar non 50 ketotic coma 51 52 53 54 55 56 57 58 59 60

Page 26 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 62 BMJ Open

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Table 3. Curricula analysis. Electronic copies of curricula were queried with 5 relevant search terms (as shown in Table 2). 6 7 8 9 Number of references to term 10 Degenerative 11 Curriculum Multiple Cauda equina Diabetes cervical 12 sclerosis syndrome mellitus 13 myelopathy 14 15 Undergraduate - PBL 0 3 1 43 16 Undergraduate – traditional 1 13 1 66 17 Undergraduate – integrated 3 15 0 30 18 For peer review only 19 Foundation Programme 0 0 0 0 20 MRCGP 0 4 2 113 21 22 Total 4 32 4 252 23 Rank 3 2 3 1 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 27 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 62

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Table 4. Learning resource analysis. The number of words devoted to DCM and 5 other diseases were determined. Importantly, only the words contained within the 6 7 main section for the particular disease were considered. 8 9 10 11 Word count devoted to 12 Degenerative 13 Resource Multiple Cauda equina Diabetes 14 cervical sclerosis syndrome mellitus 15 myelopathy 16 17 OHCM 870 1104 567 5165 18 For peer review only 19 OHFP 0 112 58 3599 20 OHGP 252 679 120 5736 21 22 Cumulative 1122 1895 745 14500 23 Modal rank 3 2 4 1 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 28 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 62 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 Figure 1 29 30 364x239mm (300 x 300 DPI) 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 on September 24, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 62 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from

Page 31 of 62 BMJ Open

1 2 3 Supplementary Table 1. Trainee knowledge analysis. A = number of questions used in question-bank; B = number of correct responses / 4 5 number of attempts; C = correct response rate (%). Note that each individual EMQ had 3 distinct clinical scenarios. 6 7 8 9 Question-bank 10 Cumulative results by theme 11 Theme Type Finals/PLAB (beginner) SRA (intermediate) MRCGP (advanced) 12 A ForB peerC A reviewB C A onlyB C A B C 13

1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 Passmedicine questions 5 6 Authors 7 1. Mueez Waqar 8 2. Jane Wilcock 9 10 3. Jayne Garner 11 4. Benjamin M Davies 12 5. Mark RN Kotter 13 14 Introduction 15 The following is a set of questions prepared for the degenerative cervical myelopathy 16 education initiative. We would request a link to our website - www.myelopathy.org, our 17 18 website to promote thisFor initiative, peer below thesereview questions. only 19 20 21 22 23 24 25 26 Methods 27 28 Questions have been prepared as follows: 29 • Clinical presentation - focussing on differential diagnosis: 5 MCQs, 2 EMQs 30 • Assessment - focussing on examination and workup: 5 MCQs, 2 EMQs 31 • Management - focussing on referral pathways and follow-up: 5 MCQs, 2 EMQs 32 33 34 These questions are targeted to the following question banks: 35 • MRCGP 36 • MSRA 37 • Medical finals http://bmjopen.bmj.com/ 38 39 Output 40 41 The aim of this initiative is to provide data on clinician knowledge on degenerative cervical 42 myelopathy. We would like to analyse the following metrics based on these questions: 43 • Distribution of answers on 1st attempt 44 • Distribution of answers on 2nd attempt 45 on September 24, 2021 by guest. Protected copyright. 46 In addition, we would like to compare the correct response rate of this sample to existing 47 48 questions in the question-bank along the following themes: 49 • Multiple sclerosis 50 • Cauda equina syndrome 51 • Diabetes mellitus 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 33 of 62 BMJ Open

1 2 Theme 1: Clinical presentation 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 A 65-year-old female reports progressive unsteadiness, mild weakness, pain and 6 paraesthesias in her legs over the last 2-3 months. She reports no symptoms in her 7 upper limbs and there is no history of trauma. Examination reveals increased lower 8 limb tone with brisk reflexes. Which of the following is the most likely explanation for 9 10 her symptoms? 11 12 A. Motor neuron disease 13 B. Degenerative cervical myelopathy 14 15 C. Multiple sclerosis 16 D. Syringomyelia 17 E. Cauda equina syndrome 18 For peer review only 19 20 Answer: B 21 22 Degenerative cervical myelopathy (DCM) refers to cervical spinal cord compression 23 due to cervical spondylosis. The term replaces CSM (Cervical Spondylotic 24 25 Myelopathy). Patients with DCM can present with predominantly lower limb 26 symptoms. Examination findings, especially early on, are often subtle. DCM is the most 27 common cause of non-traumatic paraparesis in developed countries and becomes 28 29 more common with age [2]; it is estimated to have a prevalence of 5% in over 50-year- 30 olds [3, 4]. In a series of patients presenting to Neurology outpatients, it was twice as 31 common as Multiple Sclerosis [2]. Early diagnosis is key to ensuring good clinical 32 33 outcomes. Currently most patients wait over 2 years for a diagnosis. 34 35 Incorrect options: 36 Motor neuron disease is a disease of motor neurons, with mixed upper and lower 37 • http://bmjopen.bmj.com/ 38 motor features. Patients are classically described to have brisk reflexes in wasted 39 and fasciculating limbs. As a rule, motor symptoms of weakness predominate and 40 patients do not have sensory findings. 41 42 • Multiple sclerosis can also cause paraparesis. As a disease of the central nervous 43 system, the paraparesis is usually associated with upper motor neuron signs. A first 44 presentation of multiple sclerosis is most common between the ages of 20 and 40, 45 on September 24, 2021 by guest. Protected copyright. 46 younger than this patient. Degenerative Cervical Myelopathy is also more common 47 with age and more common than MS overall, making that diagnosis more likely. 48 • Syringomyelia refers to the development of a syrinx in the spinal cord. It presents 49 50 with a central cord syndrome, with predominantly upper limb signs. It is a relatively 51 uncommon condition. 52 • Cauda equina syndrome results from compression of the cauda equina and 53 54 classically includes leg weakness, saddle anaesthesia and sphincter disturbance. It 55 is usually an acute syndrome with progressive signs. It is also a relatively uncommon 56 condition. 57 58 59 References: 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 34 of 62

1 2 1. Kjaer P, Leboeuf-Yde C, Korsholm L, Sorensen JS, Bendix T. Magnetic resonance 3 imaging and low back pain in adults: a diagnostic imaging study of 40-year-old BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 men and women. Spine. 2005 May 15;30(10):1173–80. 6 2. Moore AP, Blumhardt LD. A prospective survey of the causes of non-traumatic 7 spastic paraparesis and tetraparesis in 585 patients. Spinal Cord. 1997 8 9 Jun;35(6):361-7. 10 3. Kovalova I, Kerkovsky M, Kadanka Z, Kadanka Z, Nemec M, Jurova B, et al. 11 Prevalence and Imaging Characteristics of Non-Myelopathic and Myelopathic 12 13 Spondylotic Cervical Cord Compression. Spine. 2016 Aug 9. 14 4. Bednarik J, Kadanka Z, Dusek L, Kerkovsky M, Vohanka S, Novotny O, et al. 15 Presymptomatic spondylotic cervical myelopathy: an updated predictive model. 16 17 Eur Spine J. 2008 Mar;17(3):421–31. 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 35 of 62 BMJ Open

1 2 A 73-year-old male presents with progressively worsening gait and urinary urgency. 3 He is diagnosed with degenerative cervical myelopathy. Which ONE of the BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 following is true regarding this condition? 6 7 A. Smoking is not a risk factor in isolation 8 B. Asians can have a different underlying aetiology than caucasians 9 10 C. Bowel and bladders symptoms are rare and should prompt consideration of 11 cauda equina syndrome 12 D. Most patients present with a classic triad of neck pain, finger paraesthesias and 13 weak legs 14 15 E. Family history is of limited value 16 17 Answer: B 18 For peer review only 19 20 Asian populations have a higher rate of ossification of the posterior longitudinal 21 ligament (OPLL), which can result in myelopathy. 22 23 24 Degenerative cervical myelopathy (DCM) has a number of risk factors, which include 25 smoking due to its effects on the intervertebral discs (A, false), genetics (option E, 26 false) and occupation - those exposing patients to high axial loading [1]. 27 28 29 The presentation of DCM is very variable (option D, false). Early symptoms are often 30 subtle and can vary in severity day to day, making the disease difficult to detect 31 initially. However as a progressive condition, worsening, deteriorating or new 32 33 symptoms should be a warning sign. 34 35 DCM symptoms can include any combination of [1]: 36

37 • Pain (affecting the neck, upper or lower limbs) http://bmjopen.bmj.com/ 38 • Loss of motor function (loss of digital dexterity, preventing simple tasks such as 39 holding a fork or doing up their shirt buttons, arm or leg weakness/stiffness leading to 40 41 impaired gait and imbalance 42 • Loss of sensory function causing numbness 43 • Loss of autonomic function (urinary or faecal incontinence and/or impotence) - these 44

45 can occur and do not necessarily suggest cauda equina syndrome in the absence of on September 24, 2021 by guest. Protected copyright. 46 other hallmarks of that condition 47 48 The most common symptoms at presentation of DCM are unknown, but in one series 49 50 50% of patients were initially incorrectly diagnosed and sometimes treated for carpal 51 tunnel syndrome [2]. 52 53 54 References 55 1. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 56 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 Supp1 57 58 1):S35-41. 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 36 of 62

1 2 2. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 3 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 Focus. 2013 Jul;35(1):E1. 6 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 37 of 62 BMJ Open

1 2 Stem 1: A 60-year-old gentleman with a background of hypertension and high 3 cholesterol presents with a long history of neck pain, lower limb stiffness and BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 urinary hesitancy. Which of the following is most likely? 6 7 Stem 2: A 56-year-old gentleman presents with lower limb stiffness and imbalance. 8 His only past medical history of note is carpal tunnel syndrome that was diagnosed 9 10 a year ago on clinical grounds and has been refractory to treatment with splints and 11 steroid injections. Which of the following is most likely? 12 13 A. Cauda equina syndrome 14 15 B. Subacute combined degeneration of the cord 16 C. Degenerative cervical myelopathy 17 D. Parkinson’s disease 18 E. Multiple sclerosisFor peer review only 19 20 21 Answer: C 22 23 The presentation of degenerative cervical myelopathy [DCM] is variable. Early 24 25 symptoms are often subtle and can vary in severity day to day, making the disease 26 difficult to detect initially. However as a progressive condition, worsening, deteriorating 27 or new symptoms should be a warning sign. 28 29 30 DCM symptoms can include any combination of [1]: 31 • Pain (affecting the neck, upper or lower limbs) 32 33 • Loss of motor function (loss of digital dexterity, preventing simple tasks such as 34 holding a fork or doing up their shirt buttons, arm or leg weakness/stiffness leading to 35 impaired gait, imbalance and 36 Loss of sensory function causing numbness 37 • http://bmjopen.bmj.com/ 38 • Loss of autonomic function (urinary or faecal incontinence and/or impotence). 39 40 The most common symptoms at presentation of DCM are unknown, but in one series 41 42 50% of patients were initially incorrectly diagnosed and sometimes treated for carpal 43 tunnel syndrome [2]. 44 45 on September 24, 2021 by guest. Protected copyright. 46 Other answers: 47 • Cauda equina syndrome results from compression of the cauda equina and 48 classically includes leg weakness, saddle anaesthesia and sphincter disturbance. It 49 50 is usually an acute syndrome with progressive signs. It does not cause leg stiffness. 51 • Subacute combined degeneration of the cord results from long-standing vitamin B12 52 deficiency, classically presenting as a posterior cord syndrome – with impaired 53 54 proprioception. It can feature both upper and lower motor neuron signs. B12 55 deficiency can be associated with several neurological features. These include a 56 myelopathy (classically the subacute combined degeneration of the cord), 57 neuropathy and paraesthesias without neurological signs [3]. Subacute combined 58 59 degeneration is extremely rare in developed countries, though in tropical countries it 60 is frequently the commonest cause of non-traumatic myelopathy [4].

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 38 of 62

1 2 • Idiopathic Parkinson’s disease is a tetrad of Tremor, Rigidity, Akinesia and Postural 3 Instability (this can be remembered using the TRAP mneumonic). In the early stages BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 pain is not a typical feature and it does not cause numbness. 6 • Multiple Sclerosis [MS] can have a variable presentation, with both sensory and 7 motor symptoms and signs. Inflammatory changes are often present at multiple 8 9 sites, which can cause symptoms at more than one site; a ‘dissociated sensory loss’, 10 that is numbness at different and unlinked sites, is a hallmark of MS. Often patients 11 will recall previous episodes of odd neurological deficits, which resolved. MS 12 13 predominantly affects woman (3-4 times common) and usually presents before the 14 age of 45. 15 16 17 References: 18 1. Baron EM, YoungFor WF. peer Cervical reviewspondylotic myelopathy: only a brief review of its 19 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 20 21 Supp1 1):S35-41. 22 2. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 23 diagnosis of cervical spondylotic myelopathy by primary care physicians. 24 Neurosurg Focus. 2013 Jul;35(1):E1. 25 26 3. Kumar N1. Neurologic aspects of cobalamin (B12) deficiency. Handb Clin Neurol. 27 2014;120:915-26. 28 4. Pinto WB, de Souza PV, de Albuquerque MV, Dutra LA, Pedroso JL, Barsottini OG. 29 30 Clinical and epidemiological profiles of non-traumatic myelopathies. Arq 31 Neuropsiquiatr. 2016 Feb;74(2):161-5. 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 39 of 62 BMJ Open

1 2 Select the most likely diagnosis for the following clinical scenarios. Each option can 3 be chosen once or not at all. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 A. Multiple sclerosis 7 B. Aortic aneurysm 8 C. Fractured clavicle 9 10 D. Peripheral neuropathy 11 E. Degenerative cervical myelopathy 12 F. Median nerve entrapment 13 G. Ulnar nerve entrapment 14 15 H. Lumbar canal stenosis 16 I. Saturday night palsy 17 J. Adhesive Capsulitis 18 For peer review only 19 20 1. A 70 year-old man has pain and weakness in both legs on walking. It settles 21 with rest. 22 2. A 54 year-old female complains of right hand pain radiating into her thumb, 23 index and middle finger. It often wakes her up from sleep. 24 25 3. A 54 year-old female presents with a loss of dexterity in both hands. She has 26 been struggling to type at work and use her mobile phone. Her symptoms have 27 been deteriorating gradually over the preceding months. 28 29 30 Answers 31 • 1H - lumbar spinal canal stenosis or vascular claudication from peripheral vascular 32 disease are likely. Peripheral vascular disease is more common, but not an option 33 here. 34 35 • 2F - Carpal tunnel syndrome results from median nerve compression at the wrist, 36 within the carpal tunnel, and results in lower motor neuron signs, with thenar muscle 37 wasting and weakness of the LOAF muscles (lateral lumbricals, opponens pollicis, http://bmjopen.bmj.com/ 38 39 abductor pollicis brevis and flexor policis brevis). Patients can have paraesthesias in 40 the median nerve distribution, classically at night. Tinel’s test and Phalen’s test can 41 be positive. 42 43 • 3E - Degenerative cervical myelopathy leads to loss of fine motor function in both 44 upper limbs. There is a delay in diagnosis of degenerative cervical myelopathy, which 45 is estimated to be >2 years in some studies [1]. It is most commonly misdiagnosed on September 24, 2021 by guest. Protected copyright. 46 47 as carpal tunnel syndrome and in one study, 43% of patients who underwent surgery 48 for degenerative cervical myelopathy, had been initially diagnosed with carpal tunnel 49 syndrome [1]. 50 51 52 References: 53 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 54 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 55 56 Focus. 2013 Jul;35(1):E1. 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 40 of 62

1 2 Select the most likely diagnosis for the following clinical scenarios. Each option can 3 be chosen once or not at all. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 A. Multiple sclerosis 7 B. Aortic aneurysm 8 C. Fractured clavicle 9 10 D. Peripheral neuropathy 11 E. Degenerative cervical myelopathy 12 F. Median nerve entrapment 13 G. Ulnar nerve entrapment 14 15 H. Lumbar canal stenosis 16 I. Saturday night palsy 17 J. Adhesive Capsulitis 18 For peer review only 19 20 1. A 60 year-old male presents with clumsy hands. He has been dropping cups 21 around the house. His wife complains he doesn’t answer his mobile as he 22 struggles to use it. His symptoms have been gradually deteriorating over the 23 preceding months. 24 25 2. A 32 year-old female presents with a 3 day history of altered sensation on her 26 left foot and right forearm. On examination she has clonus in both legs and has 27 hyperreflexia in all limbs. 28 3. A 45 year-old female presents with stiffness and pain in her left shoulder, which 29 30 started around a month ago. She had a similar episode that resolved by itself. 31 Examination reveals limited external rotation. 32 33 Answers 34 35 • 1E - Degenerative cervical myelopathy leads to loss of fine motor function in both 36 upper limbs. There is a delay in diagnosis of degenerative cervical myelopathy, which 37 is estimated to be >2 years in some studies [1]. It is most commonly misdiagnosed http://bmjopen.bmj.com/ 38 39 as carpal tunnel syndrome and in one study, 43% of patients who underwent surgery 40 for degenerative cervical myelopathy, had been initially diagnosed with carpal tunnel 41 syndrome [1]. 42 2A - Multiple sclerosis (MS) can have a variable presentation, affecting both the 43 • 44 sensory and/or motor systems. Inflammatory changes are often present at multiple

45 sites, which can cause symptoms at more than one site; a ‘dissociated sensory loss’, on September 24, 2021 by guest. Protected copyright. 46 that is numbness at different and unlinked sites, is a hallmark of MS. Often patients 47 48 will recall previous episodes of odd neurological deficits, which resolved. MS 49 predominantly affects woman (3-4 times common) and usually presents before the 50 age of 45. 51 52 • 3J - Adhesive capsulitis or ‘frozen shoulder’ is most common in the fifth or sixth 53 decade of life. Women are more likely to be affected than men. It is also more 54 common in patients with diabetes mellitus. 55 56 57 References: 58 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 59 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 60 Focus. 2013 Jul;35(1):E1.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 41 of 62 BMJ Open

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 42 of 62

1 2 Theme 2: Assessment 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 A 58 year old gentleman presents with left sided paraesthesias affecting his thumb 6 and first finger. He complains of grip weakness and dropping objects 7 unintentionally. On examination, there is wasting over the thenar eminence. Which 8 of the following signs would suggest a diagnosis other than carpal tunnel 9 10 syndrome? 11 12 A. Positive Hoffman’s sign 13 B. Thenar muscle wasting 14 15 C. Unilateral weakness of pincer grip 16 D. Positive Phalen’s test 17 E. Positive Tinnel’s test 18 For peer review only 19 20 Answer: A 21 22 A positive Hoffman’s sign is a sign of upper motor neuron dysfunction and points to a 23 disease of the central nervous system - in this case from the history degenerative 24 25 cervical myelopathy [DCM] affecting the cervical spinal cord is most likely. To elicit it, 26 the examiner should flick the patient’s distal phalanx (usually of the middle finger) to 27 cause momentary flexion. A positive sign is exaggerated flexion of the thumb. 28 29 30 DCM is often missed initially and there is a delay in the diagnosis of this condition by 31 >2 years in some studies [1]. This is a problem as delayed treatment limits recovery. It 32 33 is most commonly misdiagnosed as carpal tunnel syndrome and in one study, 43% of 34 patients who underwent surgery for degenerative cervical myelopathy, had been 35 initially diagnosed with carpal tunnel syndrome [1]. DCM is therefore an important 36

37 differential in patients suspected to have Carpal Tunnel Syndrome [CTS]. http://bmjopen.bmj.com/ 38 39 CTS is a disease of the peripheral nervous system, resulting from median nerve 40 41 compression at the wrist inside the carpal tunnel. It therefore affects only the aspects 42 of the hand innervated by the median nerve: 43 • Sensation; Thumb / Index / Middle Finger. This typically manifests as intermittent 44 pain or parasthesiae. 45 on September 24, 2021 by guest. Protected copyright. 46 • Motor; ‘LOAF Muscles’(lateral lumbricals, opponens pollicis, abductor pollicis brevis 47 and flexor policis brevis). Motor signs are less commonly seen with presentations of 48 CTS, but wasting of the thenar eminence may be present. 49 50 51 Tinel’s test and Phalen’s test can be positive, but not always. Both tests aim to 52 increase the pressure within the carpal tunnel, to try to exacerbate symptoms; Tinel’s 53 54 test via tapping on it and Phalen’s test by sustained full flexion of the wrist. 55 56 In focal central nervous system disorders, like DCM, examination features are known 57 58 to have low sensitivity but high specificity [2]. As a disease of the cervical spinal cord, 59 DCM can affect the sensory, motor and autonomic nervous systems from the neck 60 downwards. Motor signs will be upper motor neuron signs such as increased toned, hyper-reflexia and pyramidal weakness. Note that the neurological signs of DCM are

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 43 of 62 BMJ Open

1 2 often subtle initially and easily missed, but as a progressive condition they are likely to 3 get worse [3]. Therefore detecting early DCM can be challenging. A high index of BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 suspicion, alongside a comprehensive neurological examination and monitoring for 6 progression is required. 7 8 9 References: 10 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 11 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 12 13 Focus. 2013 Jul;35(1):E1. 14 2. Nicholl DJ, Appleton JP. Clinical neurology: why this still matters in the 21st century. 15 Journal of Neurology, Neurosurgery & Psychiatry 2015;86:229-33. 16 17 3. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 18 pathophysiology, clinicalFor course, peer and reviewdiagnosis. Neurosurgery. only 2007 Jan;60(1 Supp1 19 1):S35-41. 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 44 of 62

1 2 Which of the following statements is most accurate regarding the usefulness of 3 cervical spine radiographs (X-rays) in the assessment of degenerative cervical BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 myelopathy (DCM)? 6 7 A. Cervical spine radiographs should be obtained in all patients suspected of 8 having DCM. 9 10 B. Where DCM is suspected, AP (anteroposterior), lateral and oblique cervical 11 spine radiographs should be requested 12 C. Cervical spine radiographs are a useful first line investigation where a diagnosis 13 of DCM is suspected 14 15 D. Cervical spine radiographs have a low sensitivity but high specificity for DCM 16 E. Cervical spine radiographs cannot diagnose DCM 17 18 Answer: E For peer review only 19 20 21 Radiographs are of limited value where a diagnosis of degenerative cervical 22 myelopathy is suspected [1] as they cannot visualise the soft tissue, such as the spinal 23 cord. 24 25 26 Spine radiographs have a high sensitivity, but limited specificity to diagnose most 27 spinal conditions. Oblique spine radiographs are usually requested in the lumbar spine 28 29 region to pick up defects in the pars interarticularis. They have no value in setting of 30 DCM. 31 32 33 The finding of spondylosis is common in spinal x-rays of adults over 40 [2]. Its 34 absence does not exclude neural compression. 35 36

37 Degenerative Cervical Myelopathy [DCM] is spinal cord compression due to http://bmjopen.bmj.com/ 38 degenerative changes of the surrounding spinal structures; e.g. from disc herniation, 39 ligament hypertrophy or calcification, or osteophytes. Therefore in order to visualise 40 41 these structures, a MRI is gold standard and first line. 42 43 Again the presence of such degenerative changes is common on MRI; in one study, 44 57% of patients older than 64 years of age had disc bulging, though only 26% had 45 on September 24, 2021 by guest. Protected copyright. 46 spinal cord compression [3]. Therefore a diagnosis of DCM requires the finding of MRI 47 compression in concert with appropriate signs and symptoms. 48 49 50 References 51 1. Nouri A, Tetreault L, Singh A, Karadimas SK, Fehlings MG. Degenerative Cervical 52 Myelopathy: Epidemiology, Genetics, and Pathogenesis. Spine (Phila Pa 1976). 53 54 2015 Jun 15;40(12):E675-93. 55 2. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 56 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 57 58 Supp1 1):S35-41. 59 3. Teresi LM, Lufkin RB, Reicher MA, Moffit BJ, Vinuela FV, Wilson GM, Bentson JR, 60 Hanafee WN: Asymptomatic degenerative disk disease and spondylosis of the cervical spine: MR imaging. Radiology 164:83–88, 1987.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 45 of 62 BMJ Open

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 46 of 62

1 2 A 67-year-old male undergoes investigations for bilateral paraesthesia in the radial 3 aspects of both hands, over the thumbs and first fingers. He also has paraesthesia BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 in the lateral aspects of both forearms and lower limb spasticity. Blood tests reveal 6 a HBA1c of 46. He undergoes nerve conduction studies and EMG with evidence of 7 denervation. Which ONE of the following diagnoses is most likely? 8 9 10 A. Bilateral carpal tunnel syndrome 11 B. Degenerative cervical myelopathy 12 C. Multiple sclerosis 13 D. Syringomyelia 14 15 E. Diabetic neuropathy 16 17 Answer: B 18 For peer review only 19 20 This patient’s twitches are probably fibrillations, a sign of lower motor neuron 21 dysfunction. This is confirmed on the neurophysiology report, with evidence of 22 denervation. His symptoms are predominantly in the C6 dermatome distribution 23 24 bilaterally. Although median nerve compression at the elbow bilaterally could in theory 25 produce his symptoms, it would be less likely to explain his symptoms given his age. 26 He is likely to have degenerative cervical myelopathy. This condition is associated with 27 28 a delay in diagnosis, estimated to be >2 years in some studies [1]. 29 30 Patients with degenerative cervical myelopathy can present with a number of problems 31 [2]: 32 33 • Pain/stiffness: affecting the neck, upper and/or lower limbs. L’hermitte’s sign is a 34 sharp pain radiating down the spine on flexion of the neck, which is classically 35 associated with multiple sclerosis, though it can occur in cervical myelopathy. 36

37 • Loss of function: Clumsiness (e.g. can’t do shirt buttons, hold cup), leg weakness http://bmjopen.bmj.com/ 38 leading to impaired gait, imbalance and falls. 39 • Sphincter disturbance: this can range from frequency and urgency to incontinence. 40 41 42 Neurological examination can reveal lower motor neuron signs at the level of the lesion 43 and upper motor neuron signs below. Note that neurological signs can be subtle and a 44

45 high degree of suspicion is needed [2]. on September 24, 2021 by guest. Protected copyright. 46 47 The other answers in this question are unlikely for the following reasons: 48 49 • A: bilateral carpal tunnel syndrome would not cause forearm symptoms. Carpal 50 tunnel syndrome results from median nerve compression at the wrist and results in a 51 lower motor neuron picture, with thenar muscle wasting and weakness of the LOAF 52 muscles (lateral lumbricals, opponens pollicis, abductor pollicis brevis and flexor 53 54 policis brevis). Tinel’s test and Phalen’s test can be positive. 55 • C: multiple sclerosis (MS) is rare in this age group. MS predominantly affects woman 56 (3-4 times common) and usually presents before the age of 45. It can have a variable 57 58 presentation, affecting both the sensory and/or motor systems. Inflammatory 59 changes are often present at multiple sites, which can cause symptoms at more than 60 one site; a ‘dissociated sensory loss’, that is numbness at different and unlinked

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 47 of 62 BMJ Open

1 2 sites, is a hallmark of MS. Often patients will recall previous episodes of odd 3 neurological deficits, which resolved. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 • D: Syringomyelia refers to the development of a syrinx in the spinal cord. It presents 6 with a central cord syndrome, with predominantly upper limb signs. It is a relatively 7 uncommon condition. 8 9 • E: His HBA1c is not within the diagnostic range of diabetes mellitus. Diabetes 10 mellitus can cause a peripheral neuropathy presenting in a glove and stocking 11 distribution, as well as neuropathy of peripheral nerves - mononeuritis multiplex. 12 13 14 References: 15 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 16 17 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 18 Focus. 2013 Jul;35(1):E1.For peer review only 19 2. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 20 21 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 Supp1 22 1):S35-41. 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 48 of 62

1 2 Which of the following investigations is the most important for diagnosing 3 degenerative cervical myelopathy ? BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 A. Nerve conduction studies and EMG 7 B. MRI Cervical spine 8 C. CT myelogram 9 10 D. CT C-spine 11 E. AP and lateral C-spine radiographs 12 13 Answer: B 14 15 16 An MRI of the cervical spine is the gold standard test where cervical myelopathy is 17 suspected. It may reveal disc degeneration and ligament hypertrophy, with 18 For peer review only 19 accompanying cord signal change. 20 21 Other answers: 22 23 • CT imaging is reserved for patients with contraindications to magnetic resonance 24 imaging. A CT myelogram is the first line investigation in this case. 25 • Radiographs are not clinically useful in the workup of these patients, though 26 27 osteoarthritic changes (e.g. osteophytes) can be visible if they are performed. 28 • Other investigatons (e.g. nerve conduction studies, EMG) may be performed when 29 the clinical picture is unclear. These can help to exclude mononeuropathies and 30 other lower motor neuron disorders. However, where there is strong clinical 31 32 suspicion and the diagnosis is suspected, an MRI of the cervical spine should be 33 performed. 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 49 of 62 BMJ Open

1 2 A 60-year-old gentleman with a background of lumbar spondylosis and chronic 3 back pain presents with gradually worsening bilateral upper limb paraesthesias and BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 leg stiffness. Which one of the investigations below is diagnostic for his likely 6 condition? 7 8 A. Nerve conduction studies and EMG 9 10 B. MRI Cervical spine 11 C. MRI Lumbar Spine 12 D. CT C-spine 13 E. AP and lateral C-spine radiographs 14 15 16 Answer: B 17 18 The presence of upperFor limb peer neurological review symptoms onlyindicates that there is pathology 19 20 either within his cervical spinal cord or brain. Brain disease is more likely to cause 21 unilateral problems. 22 23 24 A MRI lumbar spine would therefore not provide a unifying diagnosis here. 25 26 In the context of known lumbar degenerative spine, degenerative cervical myelopathy 27 28 is the number one differential for this presentation. An MRI of the cervical spine is the 29 gold standard test where cervical myelopathy is suspected. It may reveal disc 30 degeneration and ligament hypertrophy, with accompanying cord signal change. It is 31 32 not uncommon for patients to suffer from tandem (cervical and lumbar) stenosis. 33 34 Other answers: 35 CT imaging is reserved for patients with contraindications to magnetic resonance 36 •

37 imaging. A CT myelogram is the first line investigation in this case http://bmjopen.bmj.com/ 38 • Radiographs are not clinically useful in the workup of these patients, though 39 osteoarthritic changes (e.g. osteophytes) can be visible if they are performed. 40 41 • Other investigatons (e.g. nerve conduction studies, EMG) may be performed when 42 the clinical picture is unclear. These can help to exclude mononeuropathies and 43 other lower motor neuron disorders. However, where there is strong clinical 44

45 suspicion and the diagnosis is suspected, an MRI of the cervical spine should be on September 24, 2021 by guest. Protected copyright. 46 performed. 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 50 of 62

1 2 3 Select the most likely positive examination finding for the following clinical BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 scenarios. Each option can be chosen once, more than once or not at all. 5 6 7 A. Kernig’s sign 8 B. Ankle Brachial Pressure Index 9 C. Tinel’s test 10 11 D. Straight leg raise 12 E. Tongue fasciculations 13 F. Hoffman’s sign 14 G. Limited external rotation of the shoulder 15 16 I. Hypothenar wasting 17 J. Limited internal rotation of the shoulder 18 For peer review only 19 1. A 70 year-old male with a background of diabetes and hypertension presents 20 21 with pain and weakness in both legs on walking. It settles with rest. 22 2. A 54 year-old female complains of right hand pain radiating into her thumb, 23 index and middle finger. It often wakes her up from sleep. 24 3. A 65 year-old female presents with a loss of dexterity in both hands. She has 25 26 been struggling to type at work and use her mobile phone. Her symptoms have 27 been deteriorating gradually over the preceding months.3 28 29 Answers 30 31 1 - B: this patient is likely to have peripheral vascular disease [PVD] given his 32 background risk factors for this condition. The ankle brachial pressure index [ABPI] is 33 a simple method of assessing the peripheral circulation. It is calculated by dividing 34 35 systolic blood pressure in the ankle by the the systolic blood pressure in the arm. 36 These are equal in health (ABPI = 1). The ABPI is reduced in PVD. 37 2 - C: this patient is likely to have carpal tunnel syndrome. This occurs due to median http://bmjopen.bmj.com/ 38 nerve entrapment beneath the flexor retinaculum. Clinical tests to raise carpal tunnel 39 40 pressure can exacerbate symptoms and support a diagnosis. One such example is 41 Tinel’s test, includes tapping over the volar surface of the wrist joint i.e. over the carpal 42 tunnel, may reproduce paraesthesias. A normal Tinel’s test does not exclude carpal 43 44 tunnel syndrome.

45 3 - F: this patient is likely to have degenerative cervical myelopathy [DCM], which is on September 24, 2021 by guest. Protected copyright. 46 associated with upper motor neuron signs. Hoffman’s sign is elicited by flicking the 47 48 distal phalaynx of the middle finger to cause momentary flexion. A positive result is 49 exaggerated flexion of the terminal phalanyx of the thumb. Patients with DCM often 50 have subtle signs that are easily missed [1], but as a progressive condition, these are 51 52 likely to get worse [2]. Whilst the sensitivity of signs is low (i.e. their absence does not 53 rule out a problem), their specificity is high (i.e. there will be a problem). Therefore, in 54 order to diagnose early DCM and improve patient outcomes, a high index of suspicion, 55 56 alongside a 4comprehensive neurological examination and monitoring for progression 57 is required. 58 59 Other signs mentioned: 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 51 of 62 BMJ Open

1 2 • Kernig’s sign refers to painful knee extension, from a position of hip flexion and knee 3 flexion. It suggest meningeal irritation e.g. meningitis, subarachnoid haemorrhage. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 • Straight leg raise: this is positively associated with radicular pathology such as disc 6 herniation. The patient feels pain in the back when the leg is raised between 30-60 7 degrees. 8 9 • Limited external rotation is classically found in adhesive capsulitis. Patients have 10 global restriction of shoulder movements, in at least two axes, though external 11 rotation is usually the most affected and painful. 12 13 14 References: 15 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 16 17 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 18 Focus. 2013 Jul;35(1):E1.For peer review only 19 2. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 20 21 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 Supp1 22 1):S35-41. 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 52 of 62

1 2 Select the most likely positive examination finding for the following clinical 3 scenarios. Each option can be chosen once, more than once or not at all. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 A. Kernig’s sign 7 B. Ankle Brachial Pressure Index 8 C. Tinel’s test 9 10 D. Straight leg raise 11 E. Tongue fasciculations 12 F. Hoffman’s sign 13 G. Limited external rotation of the shoulder 14 15 I. Hypothenar wasting 16 J. Limited internal rotation of the shoulder 17 18 1. A 60 year-old maleFor presents peer with review clumsy hands. only He has been dropping cups 19 20 around the house. His wife complains he doesn’t answer his mobile as he 21 struggles to use it. His symptoms have been gradually deteriorating over the 22 preceding months. 23 2. A 32 year-old female presents with a 3 day history of altered sensation of her 24 25 left foot and right forearm. She had an episode of visual loss a few months ago 26 and says her friends have noted her eyes be flickery and jerky. 27 3. A 45 year-old female presents with stiffness and pain in her left shoulder, which 28 started around a month ago. She had a similar episode that resolved by itself. 29 30 31 Answers 32 1 - F: this patient is likely to have degenerative cervical myelopathy [DCM], which is 33 associated with upper motor neuron signs. Hoffman’s sign is elicited by flicking the 34 35 distal phalaynx of the middle finger to cause momentary flexion. A positive result is 36 exaggerated flexion of the terminal phalanyx of the thumb. Patients with DCM often 37 have subtle signs that are easily missed [1], but as a progressive condition, these are http://bmjopen.bmj.com/ 38 39 likely to get worse [2]. Whilst the sensitivity of signs is low (i.e. their absence does not 40 rule out a problem), their specificity is high (i.e. there will be a problem). Therefore, in 41 order to diagnose early DCM and improve patient outcomes, a high index of suspicion, 42 43 alongside a comprehensive neurological examination and monitoring for progression is 44 required. 45 2 - F: this patient is likely to have Multiple Sclerosis (MS). As a disease of the central on September 24, 2021 by guest. Protected copyright. 46 nervous system, MS is usually associated with only upper motor neuron signs such as 47 48 Hoffman’s sign (see above). The patient’s visual loss was probably secondary to optic 49 neuritis, a common presentation of MS. Cerebellar signs are particularly common with 50 MS and include nystagmus, which is likely to be the jerky eye movements noted by her 51 52 friends. 53 3 - G: this patient is likely to have adhesive capsulitis. Patients have global restriction 54 of shoulder movements, in at least two axes, though external rotation is classically 55 56 described as the most affected and painful. 57 58 Other signs mentioned: 59 60 • Kernig’s sign refers to painful knee extension, from a position of hip flexion and knee flexion. It suggest meningeal irritation e.g. meningitis, subarachnoid haemorrhage.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 53 of 62 BMJ Open

1 2 • Straight leg raise: this is positively associated with radicular pathology such as disc 3 herniation. The patient feels pain in the back when the leg is raised between 30-60 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 degrees. 6 • The ankle brachial pressure index [ABPI] is a simple method of assessing the 7 peripheral circulation. It is calculated by dividing systolic blood pressure in the ankle 8 9 by the the systolic blood pressure in the arm. These are equal in health (ABPI = 1). 10 The ABPI is reduced in peripheral vascular disease. 11 • Tinel’s test includes tapping over the volar surface of the wrist joint i.e. over the 12 13 carpal tunnel. This can reproduce paraesthesias in patients with carpal tunnel 14 syndrome. 15 16 17 References: 18 1. Behrbalk E, SalameFor K, peer Regev GJ, review Keynan O, onlyBoszczyk B, Lidar Z. Delayed 19 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 20 21 Focus. 2013 Jul;35(1):E1. 22 2. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 23 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 Supp1 24 1):S35-41. 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 54 of 62

1 2 Theme 3: Management and follow-up 3 BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 A 75-year old gentleman presents with a short history of neck pain, paraesthesia in 6 his finger tips and progressive leg weakness. Following a MRI scan of his spine, he 7 is diagnosed with degenerative cervical myelopathy due to a C4/5 disc prolapse. 8 Which of the following is the most appropriate management? 9 10 11 A. Cervical decompressive surgery 12 B. Cervical nerve root injection 13 C. Analgesia and referral to physiotherapy 14 15 D. Analgesia and review in 4 weeks time 16 E. Analgesia, a hard cervical collar and review in 4 weeks 17 18 Answer: A For peer review only 19 20 21 All patients with degenerative cervical myelopathy should be urgently referred for 22 assessment by specialist spinal services (neurosurgery or orthopaedic spinal surgery). 23 24 This is due to the importance of early treatment. The timing of surgery is important, as 25 any existing spinal cord damage can be permanent. Early treatment (within 6 months 26 of diagnosis) offers the best chance of a full recovery but at present, most patients are 27 presenting too late. In one study, patients averaged over 5 appointments before 28 29 diagnosis, representing >2 years [1]. 30 31 Currently, decompressive surgery is the only effective treatment. It has been shown to 32 33 prevent disease progression. Close observation is an option for mild stable disease, 34 but anything progressive or more severe requires surgery to prevent further 35 deterioration. Physiotherapy should only be initiated by specialist services, as 36

37 manipulation can cause more spinal cord damage. http://bmjopen.bmj.com/ 38 39 Prompt diagnosis and onward referral are therefore key to ensuring good outcome for 40 41 your patients. There are national initiatives to raise awareness of the condition to try 42 and improve referral times (www.myelopathy.org). All of the other listed options in this 43 question do not control the patient’s primary pathology. 44 45 on September 24, 2021 by guest. Protected copyright. 46 References: 47 1. Behrbalk E, Salame K, Regev GJ, et al. Delayed diagnosis of cervical spondylotic 48 myelopathy by primary care physicians. Neurosurg Focus 2013;35:E1. 49 50 doi:10.3171/2013.3.FOCUS1374 51 52 53

54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 55 of 62 BMJ Open

1 2 A 65-year old gentleman with a background of osteoarthritis and previous cervical 3 laminectomy for degenerative cervical myelopathy presents with a 2-month history BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 of worsening gait instability and urinary urgency. Which of the following is the most 6 likely explanation for his symptoms? 7 8 A. Transverse myelitis 9 10 B. Recurrent degenerative cervical myelopathy 11 C. Multiple sclerosis 12 D. Cauda equina syndrome 13 E. Spinal metastases 14 15 16 Answer: B 17 18 • Postoperatively, patientsFor peerwith cervical review myelopathy only require ongoing follow-up as 19 20 pathology can “recur” at adjacent spinal levels, which were not treated by the initial 21 decompressive surgery. This is called adjacent segment disease. Furthermore, 22 surgery can change spinal dynamics increasing the likelihood of other levels being 23 24 affected. Patients sometimes develop mal-alignment of the spine, including kyphosis 25 and spondylolisthesis, and this can also affect the spinal cord. All patients with 26 recurrent symptoms should be evaluated urgently by specialist spinal services. 27 28 • Transverse myelitis usually presents more acutely than in this case, with a sensory 29 level and upper motor neuron signs below the level affected. It can occur in patients 30 with multiple sclerosis or Devic’s disease (neuromyelitis optica). These patients tend 31 to also have features such as optic neuritis. 32 33 • Cauda equina syndrome results from compression of the cauda equina and 34 classically includes leg weakness, saddle anaesthesia and sphincter disturbance. 35 This gentleman’s history is much more likely to be in keeping with recurrent cervical 36

37 myelopathy, given his background and given the subacute presentation http://bmjopen.bmj.com/ 38 • Spinal metastases are uncommon, especially in a patient without a known primary. 39 Given previous DCM, recurrence is more likely. 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 56 of 62

1 2 A 70 year old man has decompressive surgery for degenerative cervical 3 myelopathy. Three years later he presents with neck pain and hand paraesthesias. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 Which one of the following management strategies is recommended? 6 7 A. Trial of neuropathic analgesia and cervical nerve root injections 8 B. Investigate with nerve conduction studies and EMG in the first instance 9 10 C. Urgent AP/lateral cervical spine radiographs as an MRI scan is contraindicated 11 D. Urgent referral to spinal surgery or neurosurgery 12 E. Refer to physiotherapy services 13 14 15 Answer: D 16 17 Postoperatively, patients with cervical myelopathy require ongoing follow-up as 18 pathology can “recur”For at adjacent peer spinal review levels, which only were not treated by the initial 19 20 decompressive surgery. 21 22 Recurrent symptoms should be treated with a high degree of suspicion. Although 23 24 peripheral neuropathy can occur in any patient, this should not be the diagnosis that is 25 the most strongly suspected as delays in diagnosis and treatment of DCM affect 26 outcomes. Therefore, B is false. 27 28 29 All patients with recurrent symptoms should be evaluated urgently by specialist spinal 30 services (A and E, false). Axial spine imaging is necessary and a MRI scan is first line. 31 32 In patients unable to to have a MRI, CT or CT myelogram may be considered. AP and 33 lateral radiographs are of limited use when myelopathy is suspected (C, false). 34 35 References 36

37 1. Kong L, Cao J, Wang L, Shen Y. Prevalence of adjacent segment disease following http://bmjopen.bmj.com/ 38 cervical spine surgery: A PRISMA-compliant systematic review and meta-analysis. 39 Medicine (Baltimore). 2016 Jul;95(27):e4171. 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 57 of 62 BMJ Open

1 2 A 65-year-old gentleman is referred to neurology outpatients with arm pain, 3 stiffness and imbalance. Following investigations he is diagnosed with degenerative BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 cervical myelopathy. Unfortunately, he misses his next outpatient clinic due to 6 admission with acute coronary syndrome. He attends his GP 2 months later and 7 mentions his ongoing neurological symptoms. Which of the following is the most 8 important next step in his care? 9 10 11 A. Refer to spinal surgery or neurosurgery 12 B. Refer for cervical nerve root injections 13 C. Commence neuropathic analgesia 14 15 D. Reassure the patient of his diagnosis 16 E. Refer for physiotherapy 17 18 Answer: A For peer review only 19 20 21 Management of patients with cervical myelopathy should be by specialist spinal 22 services (neurosurgery or orthopaedic spinal surgery). Decompressive surgery is the 23 mainstay of treatment and has been shown to stop disease progression (B, false). 24 25 Close observation is an option for mild stable disease, but anything progressive or 26 more severe requires surgery to prevent further deterioration. Pre-operative 27 physiotherapy should only be initiated by specialist services, as manipulation can 28 29 cause more spinal cord damage. 30 31 The timing of surgery is important, as any existing spinal cord damage can be 32 33 permanent. Treatment within 6 months offers the best chance of making a full 34 recovery. At present most patients wait more than 2 years for a diagnosis [1]. 35 36

37 Other incorrect options: http://bmjopen.bmj.com/ 38 • Neuropathic analgesia is important for symptomatic relief but will not prevent further 39 cord damage. 40 41 • Physiotherapy does not replace surgical opinion, it can in fact cause more spinal 42 cord damage in patients yet to receive surgical treatment. It should therefore only 43 be initiated by specialist services. 44 45 on September 24, 2021 by guest. Protected copyright. 46 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 47 diagnosis of cervical spondylotic myelopathy by primary care physicians. 48 Neurosurg Focus. 2013 Jul;35(1):E1. 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 58 of 62

1 2 A 67-year old male recently attended A&E, with a 3 month history of bilateral 3 paraesthesias and twitching affecting the thumb, first finger and lateral forearm. He BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 denied any trauma. A MRI scan of his spine was performed and revealed cervical 6 canal stenosis with mild cord compression. He was discharged and advised to see 7 his GP for follow-up. Which of the following is the most appropriate initial step in 8 management? 9 10 11 A. Refer to spinal surgery services 12 B. Refer for locally commissioned cervical root injections and review after 6 weeks 13 C. Enlist on the weekly minor ops clinic for carpal tunnel decompression 14 15 D. Commence neuropathic analgesia in the first instance and consider surgical 16 evaluation if this does not work 17 E. Refer to physiology services and review in 6 weeks 18 For peer review only 19 20 Answer: A 21 22 Bilateral median nerve dysfunction is very suggestive of a diagnosis of degenerative 23 cervical myelopathy (DCM) rather than bilateral carpal tunnel syndrome (option C). 24 25 DCM should be suspected in elderly patients presenting with limb neurology. His 26 twitches are probably fibrillations, a sign of lower motor neuron dysfunction. 27 28 29 Degenerative cervical myelopathy is associated with a delay in diagnosis, estimated to 30 be >2 years in some studies [1]. It is most commonly misdiagnosed as carpal tunnel 31 syndrome and in one study, 43% of patients who underwent surgery for degenerative 32 33 cervical myelopathy, had been initially diagnosed with carpal tunnel syndrome [1]. 34 Management of these patients should be by specialist spinal services (neurosurgery or 35 orthopaedic spinal surgery). Decompressive surgery is the mainstay of treatment and 36 has been shown to stop disease progression. Physiotherapy and analgesia does not 37 http://bmjopen.bmj.com/ 38 replace surgical opinion, though they may be used alongside (options D and E). Nerve 39 root injections do not have a role in management (option B). 40 41 42 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 43 diagnosis of cervical spondylotic myelopathy by primary care physicians. 44 Neurosurg Focus. 2013 Jul;35(1):E1. 45 on September 24, 2021 by guest. Protected copyright. 46 47 5 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 59 of 62 BMJ Open

1 2 Select the best management option for the following clinical scenarios. Each option 3 can be chosen once, more than once or not at all. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 A) Arrange X-rays of the cervical spine 7 B) Refer to pain clinic 8 C) Trial of analgesia and re-review 9 10 D) Refer to spinal surgery 11 E) Refer for physiotherapy 12 F) Lifestyle modification and co-morbidity optimisation 13 G) Trial of a cervical collar 14 15 H) Refer for a minor ops procedure, division of flexor retinaculum 16 I) Refer to neurology 17 J) Refer to elderly care medicine 18 K) Refer to vascularFor surgery peer review only 19 20 21 1. A 70 year-old male with a background of diabetes and hypertension presents 22 with pain and weakness in both legs on walking. It settles with rest. His most 23 recent HBA1c is 56 and an ankle-brachial pressure index is calculated as 0.7. 24 25 2. A 35 year-old female who is 30 weeks pregnant complains of right hand pain 26 radiating into her thumb, index and middle finger. It often wakes her up from 27 sleep. 28 3. A 65 year-old female presents with neck pain and loss of dexterity in both 29 30 hands. She has been struggling to type at work and use her mobile phone. Her 31 symptoms have been deteriorating gradually over the preceding months. 32 33 Answers 34 35 1 - F: this patient has peripheral vascular disease, as evident by his ankle-brachial 36 pressure index (ABPI). NICE guidance suggests that first line management should 37 include lifestyle modification such as smoking cessation, weight loss, lipid modification, http://bmjopen.bmj.com/ 38 optimisation of diabetes mellitus/hypertension and antiplatelet therapy. A supervised 39 40 exercise programme can also be arranged. Local guidelines vary on when referral to 41 specialist care is needed, but typically this would be where conservative treatment fails 42 after 3 months or the ABPI is below a defined threshold (e.g. <0.6). 43 44 2 - E: this patient is likely to have carpal tunnel syndrome. This occurs due to median

45 nerve entrapment beneath the flexor retinaculum. It is more common in pregnant on September 24, 2021 by guest. Protected copyright. 46 women due to the increase in oedema. NICE clinical knowledge summaries (CKS) 47 48 recommend lifestyle measures (e.g. wrist ergonomic devices at work), as well as wrist 49 splints (usually prescribed by physiotherapists), corticosteroid injections or referral for 50 surgical management. Wrist splints can be helpful for nighttime symptoms, as in her 51 52 case. Corticosteroid injections require local expertise that may or may not be present. 53 3 - D: this patient is likely to have degenerative cervical myelopathy. DCM is often 54 missed initially and there is a delay in the diagnosis of this condition by >2 years in 55 56 some studies [1]. Patients have predominantly upper motor neuron signs such as 57 increased toned, hyper-reflexia and pyramidal weakness. Neurological signs are often 58 subtle initially and easily missed, but as a progressive condition they are likely to get 59 worse [2]. Management of these patients should be by specialist spinal services 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 60 of 62

1 2 (neurosurgery or orthopaedic spinal surgery). An MRI scan is required for diagnosis. 3 All patients should be assessed by a spinal surgeon. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 References: 7 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed 8 9 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 10 Focus. 2013 Jul;35(1):E1. 11 2. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 12 13 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 Supp1 14 1):S35-41. 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 61 of 62 BMJ Open

1 2 Select the best management option for the following clinical scenarios. Each option 3 can be chosen once, more than once or not at all. BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 6 A) Arrange X-rays of the cervical spine 7 B) Refer to pain clinic 8 C) Trial of analgesia and re-review 9 10 D) Refer to spinal surgery 11 E) Refer for physiotherapy 12 F) Lifestyle modification and co-morbidity optimisation 13 G) Trial of a cervical collar 14 15 H) Refer for a minor ops procedure, division of flexor retinaculum 16 I) Refer to neurology 17 J) Refer to elderly care medicine 18 K) Refer to vascularFor surgery peer review only 19 20 21 1. A 60 year-old male presents with clumsy hands. He has been dropping cups 22 around the house. His wife complains he doesn’t answer his mobile as he 23 struggles to use it. His symptoms have been gradually deteriorating over the 24 25 preceding months. 26 2. A 32 year-old female presents with a 3 day history of altered sensation on her 27 left foot and right forearm. She had an episode of visual blurring in her right eye 28 a few months ago which resolved after a few days. Examination reveals brisk 29 30 reflexes. 31 3. A 45 year-old female presents with stiffness and pain in her left shoulder, which 32 started around a month ago. She had a similar episode that resolved by itself. 33 Examination reveals global restriction of shoulder movement, particularly 34 35 external rotation. 36

37 Answers http://bmjopen.bmj.com/ 38 1 - D: this patient is likely to have degenerative cervical myelopathy. DCM is often 39 40 missed initially and there is a delay in the diagnosis of this condition by >2 years in 41 some studies [1]. Patients have predominantly upper motor neuron signs such as 42 increased toned, hyper-reflexia and pyramidal weakness. Neurological signs are often 43 44 subtle initially and easily missed, but as a progressive condition they are likely to get

45 worse [2]. Management of these patients should be by specialist spinal services on September 24, 2021 by guest. Protected copyright. 46 (neurosurgery or orthopaedic spinal surgery). An MRI scan is required for diagnosis. 47 All patients should be assessed by a spinal surgeon. 48 49 2 - I: this patient is likely to have Multiple Sclerosis. Her visual loss was probably 50 secondary to optic neuritis, a common ophthalmological association with multiple 51 sclerosis. She should be referred to neurology. 52 53 3 - C: this patient is likely to have adhesive capsulitis (frozen shoulder). Patients have 54 global restriction of shoulder movements, in at least two axes, though external rotation 55 is classically described as the most affected and painful. Management of frozen 56 57 shoulder is controversial and there is not much evidence to inform practice. In general, 58 alternative diagnoses should be excluded and pain relief optimised. Gentle shoulder 59 movement is encouraged and there is limited evidence for physiotherapy. 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 62 of 62

1 2 References: 3 1. Behrbalk E, Salame K, Regev GJ, Keynan O, Boszczyk B, Lidar Z. Delayed BMJ Open: first published as 10.1136/bmjopen-2018-028455 on 12 January 2020. Downloaded from 4 5 diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg 6 Focus. 2013 Jul;35(1):E1. 7 2. Baron EM, Young WF. Cervical spondylotic myelopathy: a brief review of its 8 9 pathophysiology, clinical course, and diagnosis. Neurosurgery. 2007 Jan;60(1 Supp1 10 1):S35-41. 11 12

13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 24, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml