Sir William Osler's fatal trip to Scotland: “Mrs M” and the University Grants Committee

Sir William Osler (1849–1919), the most famous physician in the English-speaking world during the early twentieth century and an iconic figure in the history of medicine, died 29 December 1919 of complications from a respiratory infection acquired on his way home from Scotland, where he had gone to consult with three other physicians on a difficult case.

Harvey Cushing, Osler's first biographer, records from Osler's account book: “Left here (Oxford) the night of Sept. 22nd to see Mrs M. in Glasgow with Drs Ness, Cameron & Armstrong. Went on to Edinburgh, stayed with Lovell Gulland. Saw Harvey Littlejohn & others.” ¹ Michael Bliss, Osler's latest biographer, clarifies that “Mrs M.” was Mrs Fulton Martin and that his fee was £525, equivalent to “500 guineas.” ² John Ward recently reviewed Osler's relationship with Charles W. Dyson Perrins (1864–1958), whose telephone call to Osler allegedly prompted Osler to go to Scotland to see “Mrs M.” ³ Although he suggests why Osler might have felt obliged to accede to Dyson Perrins’ request to visit Mrs M, he was unable to determine why Dyson Perrins had become involved in the consultation request, nor who had approached him.

Our purpose is to discuss why Osler went to Scotland despite recurrent respiratory infections. We suggest that his role in the newly formed University Grants Committee (UGC) was a major factor in his decision. We also discuss the differential diagnosis of the case of “Mrs M.”

Dramatis personae

We now know that “Mrs M.” was Mrs Hugh Fulton Martin (hereafter, “Mrs Martin”) born Bethia Wylie in 1879,^4,5 having traced her great-grandson who confirms that she was the “Mrs Martin” who consulted Osler in September 1919 (Jonathan Wylie Martin, personal communication, 6 January 2021) (Figure 1). Her father, Adam Wylie (1856–1939), was a director of the firm Wylie & Lochhead, funeral directors and cabinet makers who branched into furniture manufacture, with several large department stores. Adam Wylie was also a noted yachtsman, undefeated in his class according to an obituary. 6

OPEN IN VIEWER

Figure 1.

Bethia Fulton Martin (1879–1920), possibly in her wedding dress in 1914, at which time she would have been 33 or 34 years of age. Used by permission, Jonathan Wylie Martin.

OPEN IN VIEWER

Figure 2.

Left: Sir Donald MacAlister (1854–1934), principal and vice-chancellor of the University of Glasgow. Right: Sir Henry Harvey Littlejohn (1862–1927), dean of the Faculty of Medicine of the University of Edinburgh. Both: Wikimedia Commons.

On 2 May 1913, Bethia Wylie married Hugh Fulton Martin (d. 1932), ⁷ a leather merchant. His firm, H.F. & G. Martin, Baltic Leather Works, Glasgow, manufactured harnesses and saddles principally for international markets. ⁸ Around the turn of the twentieth century, the firm pioneered a new chrome leather tanning process, and promoted its water-resistant quality as “Zug leather” …. “the toughest leather known” with products including “Zug uppers” for footwear and a “Zug football” that “always keeps its shape … (and) never becomes sodden.” ⁹

In the first few years of marriage the Martins lived in Lenzie, a town 7 miles northwest of Glasgow, where their son, William Wylie Martin was born on 10 June 1914. ¹⁰ The Martins later moved to Glasgow, where in May 1919 they relocated to 13 Windsor Circus. ¹¹ A family member surmises that this move was to aid Bethia with childcare as this new house was just round the corner from Hugh's widowed aunt, Jessie, who was already looking after Hugh's widowed brother and his children. Jessie is remembered as commenting that “Bethia tried to be a good mother but had to spend most of her day resting on a chaise longue” (Brenda Martin, personal communication, 8 January 2021).

The physicians with whom Osler consulted were Robert Barclay Ness (1863–1954), ¹² Agnes Wallace Cameron (1879–1975), ¹³ and William Buchanan Armstrong (1873–1935). ¹⁴ All were well-respected in the West of Scotland, but to the best of our knowledge none had a previous relationship with Osler. Barclay Ness was a physician associated with the Western Infirmary in Glasgow, as well as Professor of Materia Medica at the nearby Anderson's Medical School. William Buchanan Armstrong was a general practitioner in Kirkintilloch and Lenzie, and so could have been the family doctor to the Martins when they lived there. (Bethia's mother came from Lenzie and her maternal aunt lived there at the time of the consultation.)

Agnes W Cameron was the daughter of Murdoch Cameron (1847–1930), who was Regius Professor of Midwifery at the University of Glasgow and world famous as a pioneer of elective cesarean section under aseptic conditions. She appears to have been a close friend of Bethia Wylie. They were the same age and the Wylies and Camerons lived near each other in the Park area of Glasgow's affluent West End. It was Agnes Cameron who signed the marriage register as female witness to Bethia Wylie's marriage to Hugh Fulton Martin, which suggests she may have been the bridesmaid. ¹⁵

Dyson Perrins was a businessman, collector, philanthropist, and grandson of William Henry Perrins (1793–1867), codeveloper of the Lea & Perrins brand of Worcestershire sauce. Dyson Perrins managed the Lea & Perrins Company and also the Royal Worcester Porcelain Factory. Perrins and Osler were well-acquainted through mutual clubs and interests. ³

George Lovell Gulland (1862–1941), with whom Osler stayed in Edinburgh (Figure 2), was an Edinburgh physician best known as a pioneer in clinical hematology. He later served as president of the Royal College of Physicians of Edinburgh. ¹⁶

Sir Henry Harvey Littlejohn (1862–1927) was dean of the Faculty of Medicine at the University of Edinburgh. He is best known for innovations in public health; in 1891 he traced an outbreak of typhoid in Edinburgh to a batch of contaminated milk from a single farm. ¹⁷ Littlejohn and Osler were well-acquainted; their mutual interests included collecting books and the history of body-snatching.^18,19

Other persons who may have been involved in Osler's trip to Scotland, including members of the UGC, are introduced below.

Osler's decision to go to Scotland

Previous accounts of Osler's decision to go to Scotland emphasize a telephone call from Dyson Perrins. ^{1 ‒3,23} John Brett Langstaff (1889–1985) recalled 50 years later:

… I was sitting alone with Lady Osler when a long-distance call took her to a nearby telephone. It was a professional emergency for Sir William from Scotland, and so poor was the connection that Lady Osler could not distinguish the name of the person calling. After a struggle Lady Osler asked to have the name spelled out. Then I heard her exclaim, “I have it now. It's Perrins as in fish sauce!” When she returned from the phone she was flushed with confusion because it really was the Mr. Perrins who had made a fortune selling Lee (sic) and Perrins Sauce. Later I learned that Sir William's immediate response to this call—in spite of the weather and his own unfitness at the time—brought on the bronchopneumonia from which he died four days after Christmas 1919. ²⁰

Several questions remain unanswered.

First, although Osler had traveled as far north as Massachusetts, as far south as Florida, and as far west as Iowa, during his Baltimore period (1889–1905), in Britain his practice was mostly in Oxford and its environs with occasional day trips to London.

With the onset of war in 1914, the number of consultations Osler undertook fell significantly. He spent more time on military activities, often away from Oxford, visiting army camps and military hospitals, but continued to see civilians in consultation as he traveled around. ²¹ Also, to our knowledge, Osler had no previous relationship with any of Mrs Martin's physicians.

Second, Osler was not entirely well. Between December 1916 and July 1919 he had at least seven respiratory infections that caused him to be house-confined.^22,23 The most recent one had begun during a celebration of his seventieth birthday on 12 July 1919; he wrote a friend that “my Pneumococcus struck in & I had a high fever & have been in bed ever since.” ²⁴ Recovery was slow, but on 1 August the Oslers embarked on a 6-week vacation on Jersey, where they rented a “pink cottage” at St. Brelade's Bay.^25,26 He regained lost weight but there were still concerns about his health.

Osler could have told Perrins that he did not make overnight trips to see patients, reminded Perrins of his vulnerability to respiratory infections, and suggested another consultant such as Thomas Kirkpatrick Monro (1865–1955), Regius Professor of the Practice of Medicine at the University of Glasgow. Monro, like Osler, was a textbook author, editor, historian, and bibliophile with special interest in Sir Thomas Browne (1605–1682), and he and Osler corresponded. ²⁷ Bliss suggests Osler may have accepted the consultation because he was “perhaps eager to cover the costs of his holiday.” The Wylies and Martins were presumably well-off and could afford a large fee that would include travel mileage. We suggest a more compelling reason: Osler's involvement with the recently formed UGC.

The UGC was an advisory committee to the British government to make recommendations for funding universities. On 23 November 1918, an organizational meeting attended “by all of the representatives and university colleges of the United Kingdom and from the London medical schools” took place at which Sir Donald MacAlister (1854–1934), principal and vice-chancellor of the University of Glasgow (Figure 2), spoke on behalf of the universities. ²⁸ MacAlister was the longtime president of the General Medical Council, had attended a celebration of Osler's 70th birthday, ²⁹ and was the older brother of Osler's close friend Sir John Young Walker MacAlister (1856–1925), librarian and secretary of the Royal Society of Medicine. Sir William Symington McCormick (1859–1931) was named chairman of the UGC. McCormick was also chairman of the advisory council of the Board of Education. Although the UGC was conceived in the Board of Education, its members were appointed by the UK Treasury. When the membership was announced in June 1919, it was Osler who was to represent the medical schools.

The UGC was an elite, eight-member committee. The members besides Osler and McCormick were the biologist William Bateson (1861–1926), who popularized the ideas of Gregor Mendel (1822–1884) and coined the term “genetics”; the engineer Sir Dugald Clark (1854–1932), who developed the two-stroke engine; the chemist Sir James Johnston Dobbie (1854–1922), who isolated and characterized the alkaloids; Sir Stanley Mordaunt Leathes (1861–1938), the First Civil Service Commissioner who was also a poet, historian, and economist; Sir Frederic George Kenyon (1863–1952), the director of the British Museum who was also a biblical and classical scholar and a paleographer; and Sara Margery Fry (1874–1958), who in 1919 was best known as an advocate for prison reform.

Osler promptly became involved in the UGC on behalf of medical schools and teaching hospitals. On 24 July 1919, he informed Sir George Newman (1870–1948), the newly appointed Chief Medical Officer to the Ministry of Health, that he had urged McCormick to approve the immediate formation of clinical units at St. Bartholomew's Hospital, St. Thomas's Hospital, and University College, all in London. ³⁰

A letter from Osler to John George Adami (1862–1926), dated 4 October 1919, supplements the information provided by Cushing about Osler's trip to Scotland to see Mrs Martin. Osler told Adami that he had discussed “Treasury Committee Business” with Littlejohn in Edinburgh and MacAlister in Glasgow. ³¹

In conclusion, Osler had high-level meetings in Glasgow and Edinburgh on behalf of the UGC. Did he arrange these meetings on short notice after agreeing to go to Glasgow to see Mrs Martin after Dyson Perrins's telephone call? Or did Perrins somehow get wind of Osler's planned trip to Scotland and suggest that, while in Glasgow, he see Mrs Martin? (Perrins's involvement in this matter is unclear, since neither we nor John Ward has been able to establish a relationship between Perrins and the Wylies, the Martins, or any of the three physicians who attended Mrs Martin.) We suggest that Osler's interest in medical education in the Scottish universities and teaching hospitals may have been the deciding factor.

Differential diagnosis in the “case of Mrs M.”

The only surviving record of the consultation consists of a letter from Osler to three former colleagues at Johns Hopkins: “On September 23rd I went to Glasgow to see one of those remarkable Erythema cases (all sorts of skin lesions and three months on and off consolidations of both lower lobes).” ³² Mrs Martin died on 15 January 1920, 122 days after Osler saw her. The death certificate lists “angioneurotic edema with chronic nephritis” and “tuberculous enlargement of the mediastinal lymph nodes.” ³³ There is no record of a Bethia Martin or Wylie being admitted to the Western Infirmary, no record of a chest x-ray, and no record of a postmortem examination recorded in the pathology records at Glasgow University archives.

We thus have five clues: erythematosus skin lesions of various morphologies; chronic nephritis; intermittent consolidation of both lower lobes; angioneurotic edema; and “tuberculous enlargement of the mediastinal lymph nodes.” There is no mention of fever, shortness of breath, productive cough, pleuritic chest pain, hemoptysis, abdominal pain, arthritis or arthralgia, hematuria, gastrointestinal bleeding, weight loss, pruritus, headache, or neurologic deficit. She had ostensibly been in good health five years earlier, as evinced by a successful pregnancy. Her family history is remarkable only for chronic nephritis (her father) and tuberculosis (paternal grandfather). ³⁴

With this limited information, let us review what Osler meant by “those remarkable Erythema cases” and the potential significance of each clue.

Between 1895 and 1914, Osler wrote at least five papers describing cases now lumped as “Osler's erythema group of skin diseases with visceral manifestations.” ^{35 ‒ 39} He found these cases diverse and confusing: “What is needed, in truth, is a dermatological Linnæus, to bring out of the chaos at present existing in the group of erythemas.” ³⁹ Osler's 29 cases have been analyzed through the years by physician-historians including, most recently, Benedek ⁴⁰ and Scofield and Oates. ⁴¹ The latter researchers concluded that Henoch-Schönlein purpura (HSP) is the most likely diagnosis in 16 of Osler's 29 patients. They agree with Benedek that two of Osler's patients probably had systemic lupus erythematosus (SLE).

The major differential diagnosis is between SLE and HSP. Erythematous skin lesions and renal involvement are two of the ten major criteria for SLE as developed by the American College of Rheumatology. ⁴² The extent to which Osler understood SLE as a distinctive clinical entity has been debated through the years, with more recent authors being less enthusiastic in this regard.^40,41,43 About 90 percent of patients with SLE are women, with onset usually during the childbearing age. Erythematous skin lesions and renal involvement (“nephritis”) are also two components of the classic clinical tetrad of HSP (the others being arthralgia and abdominal pain). HSP (currently known as “IgA vasculitis”) is mainly a disease of childhood, with 90 percent of cases occurring between the ages of two and 11 years and with a male-to-female ratio of about 1.5 to one. Affected adults are more likely to develop severe kidney disease. ⁴⁴ Osler suggested that HSP is caused by “anaphylaxis,” ³⁹ and on this basis sometimes receives credit for being the first to suggest an immunologic basis for the disease.

The “on and off consolidation of both lower lobes” could represent pleural effusions, pulmonary alveolar hemorrhage, lupus pneumonitis, pulmonary embolism with infarction, or, conceivably, heart failure. We lack supporting evidence for any of these possibilities.

The prominence of angioneurotic edema (now called “angioedema”) on the death certificate possibly favors SLE over HSP. Osler coined the term “hereditary angio-neurotic edema” and described the first case with clear-cut familial occurrence. ⁴⁵ Hereditary angioedema results from decreased or dysfunctional C1 esterase inhibitor (C1-INH), and SLE is the most common co-occurring autoimmune disease. ⁴⁶ Acquired angioedema is less common and is most frequently associated with lymphoproliferative disorders. However, the association of acquired angioedema with SLE is well-described, and some cases have been attributed to decreased C1-INH levels caused by increased catabolism or autoantibodies.^47,48 Autoantibodies against complement components in patients with SLE can also cause hypocomplementemic urticarial vasculitis,^49,50 a distinctive type of vasculitis with multi-organ involvement, which could explain “all sorts of skin lesions” and “chronic nephritis” in the case of Mrs M. (Current opinion supports Osler's statement that “there is really no warrant for separating too sharply angio-neurotic œdema and urticaria.” ³⁶ ) The association of angioedema with HSP is less prominent, although both conditions were present in one of Osler's cases ⁴¹ :

“Tuberculous enlargement of the mediastinal lymph nodes” could represent misdiagnosed sarcoidosis if cultures were not performed. Sarcoidosis would account for pulmonary and skin involvement, and recent reports suggest that kidney involvement including acute renal failure may be more common in sarcoidosis than was previously thought to be the case.^51,52 Sarcoidosis presenting as bilateral hilar lymphadenopathy and multiple erythematosus skin lesions (erythema nodosum) is known as Löfgren's syndrome, which occurs most often in Caucasian woman. This presentation of sarcoidosis is nearly always self-limited, but multisystem disease and renal failure have been reported.^52,53

A fourth potential diagnosis would be an antineutrophil cytoplasmic antibody-associated vasculitis. Granulomatosis with polyangiitis (formerly, Wegener's granulomatosis), eosinophilic granulomatosis with polyangiitis (formerly, Churg-Strauss syndrome), and microscopic polyangiitis all cause rapidly progressive glomerulonephritis and diffuse alveolar hemorrhage (the so-called pulmonary renal syndrome) and can cause various types of erythematosus skin lesions. ⁵⁴ However it is unlikely that these entities would cause recurrent (as opposed to progressive) pulmonary consolidation in the absence of immunosuppressive therapy.

We reviewed the case with six American professors of rheumatology. Five favored SLE and one favored HSP, but all agreed the evidence is insufficient for a definitive diagnosis. Today the diagnosis would probably be made easily through some combination of serologic testing, biopsies, and imaging.

Aftermath

Osler's return to Oxford was delayed by a railway strike that necessitated travel in an “old and slow” automobile from Newcastle to Oxford. He arrived home on 28 September with an “awful cold.” On 1 October he saw his last patient, the daughter of a physician. ²¹ He was soon in bed. On 6 October he informed McCormick and Newman that he was ill with bronchitis and wished to resign from the UGC. ^{55 , 56} His resignation was accepted and Sir Wilmot Herringham (1855–1936) was appointed as his replacement. ⁵⁷ Osler developed bacterial pneumonia and then lung abscess with empyema, from which he died on 29 December 1919 from hemorrhage into the empyema cavity. Mrs Martin died 18 days later.

The UGC remained a force in British higher education until 1989, when its functions were transferred to a new body, the Universities Funding Council, which was later divided into three Higher Education Funding Councils. Among the eight original members, perhaps none was more effective than Margery Fry, who as the only woman on the committee from 1919 to 1947 championed gender equity including adequate residencies for women at universities. ⁵⁸ Osler, had he lived and remained reasonably healthy, would surely have thrived in this role as an advocate for government funding of medical schools and teaching hospitals.