From Holmes to House – 500 years of the diagnostic neurologist

Posted in Courses & Conferences,History on 3rd Apr 2018







Conference details: 20 March 2018, Royal College of Physicians, London, UK,

Report by: Michael Zandi, National Hospital for Neurology and Neurosurgery
Conflict of interest statement: None declared
Published online: 3/04/18

The Royal College of Physicians building, next to Regent’s Park, was designed by Denys Lasdun, opened in 1964 as the 5th home of the RCP since foundation in 1518, and represented a leap forward from previous buildings. The site marks an intersection of an almost straight line between Sherlock Holmes’ 221B Baker street address, and William R Gowers’ birthplace in Mare Street, Hackney. One of many events to mark 500 years of the RCP, David Nicholl (Birmingham) organised this event which took place on the 20th March 2018 (what would be Gowers’ 173rd birthday) jointly with the Association of British Neurologists. Speakers were asked to reflect on neurology past, present and future, with an eye on what would have been in the times of Sherlock Holmes and what could be in a near-future time of physician Dr Gregory House M.D. (portrayed by Hugh Laurie in the US television series created by David Shore). I enjoyed this meeting in its entirety which took a broad historical eye at progress in neurology, with discussion of the art of clinical information gathering, examination and reasoning, and also the impact of three major areas: neuroimmunology, neurogenetics and brain imaging. The meeting opened with two plenary lectures by Andrew Lees and Angela Vincent.

Eliminate the impossible, and whatever remains must be the truth” (Arthur Conan Doyle, The Sign of Four, 1890).

Andrew Lees (UCL) opened the meeting with William Gowers’ legacy in clinical method, and drew comparisons with Arthur Conan Doyle’s Sherlock Holmes. Advised to read The Complete Works of Sherlock Holmes by William Gooddy, Lees realised later that this was his introduction to the method of William Gowers. Edinburgh-born Conan Doyle was inspired by Joseph Bell, completed his MD in tabes dorsalis, and was likely to have attended some of Gowers’ lectures in London. Beyond his 1886 Manual of Diseases of the Nervous System, many lectures from Gowers are available still via the British Medical Journal on-line, transcribed by his students in shorthand. Lees reminded us how Gowers and Conan Doyle both teach us to develop our skills of observation (become an expert on cigarette ash, sleeves, ears, the art of seeing what others cannot see), put prejudices aside, and hone abductive reasoning (reasoning backwards from the end point of a crime scene, or clinical scenario). Only through practising this method can neurologists learn to recognise diagnoses hidden in plain sight, and not fall in error through use of pattern recognition or algorithmic thinking. Perhaps 10 or 20 thousand hours of training on the wards and in the clinics seeing patients are needed to develop this skill and to spot ‘black swans’.

Angela Vincent (Oxford) provided a historical tour of neuroimmunology, and in particular the story of myasthenia gravis from Thomas Willis’s description in De Anime Brotorum (1672) to contributions by Simpson, Chang and Lee, and working with her own mentor Ricardo Miledi (1927-2017) when her own academic life and contributions began. This talk covered publishing scoops in the 1970s, a long working partnership with colleague John Newsom Davies (1932-2007), innovations with plasma exchange, genetic and autoimmune seronegative myasthenia gravis, to recent insights from IgG1 and IgG4 antibody subtypes across a range of diseases. Highlighting the immunological contributions of Paul Ehrlich (Nobel prize 1908 shared with Mechnikov), including prediction of magic bullet therapies and studies of maternal autoantibody transfer (though Ehrlich had discounted the possibility of autoimmunity), Vincent discussed her own work on arthrogryposis with Leslie Jacobson, and maternal transfer of caspr2 autoantibodies with Ester Coutinho most recently. Vincent then described her approach to antibody mediated central nervous system diseases, where one neuroimmunology methodology which has stood the test of time has been application of the Witebsky modified-Koch’s postulates (latest iteration: Rose and Bona 1993). Finally Vincent speculated on a near-future in which House M.D. in the emergency room would use point-of-care rapid antibody assays allowing immediate prescription of plasma cell bespoke therapies.

In the remaining morning talks, Anu Jacob (Liverpool) spoke of the past, present and future of paraneoplastic disorders, from Trousseau’s migratory thrombophlebitis through to Denny Brown’s contributions in the 1940s and novel insights from genetics and immunotherapy-induced encephalitis. Current approaches in recognising paraneoplastic neurological disorders remain guided by advances in antibody assays and imaging including PET. Jacob discussed novel insights for pathogenesis and therapies from advanced-immunotherapy induced encephalitis. Belinda Lennox (Oxford) discussed whether schizophrenia could be considered a neurological disorder, described the ongoing fragmentation of care for patients with neuropsychiatric symptoms, and the problems of how psychiatrists and neurologists name similar phenomena differently. Quoting Tom Insel, in 2010 then head of the US National Institutes of Mental Health, “a century ago we had large public institutions for serious mental illness, tuberculosis and leprosy. Of these three, today only mental illness, especially schizophrenia, remains unchanged in prevalence and disability.” Lennox highlighted the work of Eve Johnstone (Glasgow) in CT scanning in schizophrenia (1976) and described her own work exploring the role of immunity in psychosis. Nick Wood (UCL) had the task to describe advances in genetics from Mendel to Venter. Wood spoke of the lack of opportunity compared to cancer in brain diseases until recently, to the marked emergence of data and advances in knowledge in the last 5 years, in which changing thinking from “diseases to pathways” had been most effective. Wood discussed new techniques to find the “genes beneath the peaks” in Manhattan plots, and the need to solve genetic problems first to allow us to build the right models of diseases, including examples from LRRK2, and GCH1 rare coding variants in Parkinson’s Disease. A highlight of this talk was a clear description of Mendelian Randomisation, and its power to act as an inexpensive surrogate for expensive pharmaceutical randomised trials, and reveal nature’s own randomised controlled trials, with examples including a causal link between raised BMI and reduced risk of Parkinson’s disease. Finally, highlighting one of the most impressive recent papers in neurogenetics and therapy (nusinersen for spinal muscular atrophy), reminding us all that a clear result is plain to see when found. After lunch, Doug Turnbull reminded the audience just how recent the 66 year history of mitochondrial genetics is, with a survey of work by Mitchell and Mitchell, Sanger, and the landmark of Anita Harding’s work with John Morgan Hughes and Ian Holt. Turnbull discussed the serendipity of being able to measure respiratory chain function sensitively, how NHS specialised services have enabled driving forward rare disease research, and the ethics, advances, and changes in law in mitochondrial donation.

Back to Conan Doyle’s Edinburgh for 3 of the last 4 speakers, Joanna Wardlaw (Edinburgh) then spoke of 123 years of imaging, a tale of accidental discoveries, rapid dissemination e.g. of the 1895 röntgenogram, ongoing issues around minimising exposure to radiation, and advances in brain vascular imaging. Paul Matthews (Imperial and UK Biobank) gave an update on the UK biobank imaging study (nearly up to a quarter of its target of 100K peope in the UK scanned, Rustam Al-Shahi Salman (Edinburgh) expanded a theme of preventing over-diagnosis and talked of the ongoing need to minimise the discovery of incidentalomas through practising defensive medicine by imaging, giving examples of where this ‘1 in 37 brain roulette’ has resulted in harm. He gave a notable example of arteriovenous malformation bleeding risk data and subsequent coiling in which the evidence (ARUBA) suggests more harm is done by intervention than medical management alone, and a familiar account of how such data can be met with hostility in the medical community at first. Finally, Andrew Elder (Edinburgh, and medical director of MRCP(UK)) talked with urgency that the ‘physical examination is in trouble’, citing examples from the US where physical examination has not been part of postgraduate medical or surgical training since 1972, and tied up the last theme of the day with evidence that physical examination beats unfocused ‘screening’ investigation hands down, and of its therapeutic and teaching value.

Several speakers gave clinical and research examples to support Gowers’ “It is always pleasant to be right, but it is generally a much more useful thing to be wrong” (Gowers, 1894). I found, as I walked back from this enjoyable day to the Bloomsbury present, the aphorisms and stories of revealed black swans most instructive, the advances in genetics the most full of promise, and the secure knowledge that the next scientific advances and next 500 years are (probably) unpredictable.


  2. Scott, A., Eadie, M.J., Lees, A., 2012. William Richard Gowers 1845-1915 : exploring the Victorian brain : a biography, Vol., Oxford University Press, Oxford.
  3. Lees, AJ. The strange case of Dr William Gowers and Mr Sherlock Holmes. Brain, Volume 138, Issue 7, 1 July 2015, Pages 2103–2108,
  4. Gowers WR. A post-graduate lecture on mistaken diagnosis: delivered in the National Hospital for the Paralysed and Epileptic. Br Med J 1894: 2: 1–3
  5. (video interview with Andrew Lees, April 2017, with a section on Gowers at the Queen Square Library, video produced by Ben Crowe @erafilms)
  6. Vincent A, et al. Ann N Y Acad Sci. 2018 Feb;1413(1):143-153. doi: 10.1111/nyas.13592. Serological and experimental studies in different forms of myasthenia gravis.
  7. Rose NR, Bona C. Immunol Today. 1993 Sep;14(9):426-30. Defining criteria for autoimmune diseases (Witebsky's postulates revisited)
  8. Noyce et al. PLoS Med. 2017 Jun; 14(6): e1002314. doi: 10.1371/journal.pmed.1002314 Estimating the causal influence of body mass index on risk of Parkinson disease: A Mendelian randomisation study
  9. Holt IJ, Harding AE, Morgan-Hughes JA. Nature. 1988 Feb 25;331(6158):717-9. Deletions of muscle mitochondrial DNA in patients with mitochondrial myopathies.
  10. Herbert M, Turnbull D. N Engl J Med. 2017 Jan 12;376(2):171-173. doi: 10.1056/NEJMcibr1615669. Epub 2016 Dec 28. Mitochondrial Donation - Clearing the Final Regulatory Hurdle in the United Kingdom.
  11. Ballerini L, et al. Sci Rep. 2018 Feb 1;8(1):2132. doi: 10.1038/s41598-018-19781-5. Perivascular Spaces Segmentation in Brain MRI Using Optimal 3D Filtering.
  12. Mohr JP et al. Lancet. 2014 Feb 15;383(9917):614-21. doi: 10.1016/S0140-6736(13)62302-8. Medical management with or without interventional therapy for unruptured brain arteriovenous malformations (ARUBA): a multicentre, non-blinded, randomised trial.
  13. Elder A et al.The Road Back to the Bedside JAMA. 2013;310(8):799-800. doi:10.1001/jama.2013.227195

Open Access, for medical professionals: Sign up to receive our email newsletter with links to the latest content. ACNR is free, thanks to the support of advertisers. The editorial content is peer reviewed and remains completely independent unless clearly specified. 

We may infrequently send you news from our sponsors which is relevant to the field of neurology, but you can opt out at any time.

This website is for medical professionals only.