David Alan Pyke (1921-2001)
David Alan Pyke (1921-2001) was best known for his long-running study of diabetes in identical twins and for an unfortunate series of papers on chlorpropamide alcohol flushing. He was also a well-known personality on the diabetes lecture circuit, celebrated for his style, knowledge and mordant wit.
He was the son of Margaret Pyke, a pioneer in birth control, and the eccentric Geoffrey Pyke. Geoffrey decided that his son needed an education which stimulated intellectual curiosity. To this end he founded the The Malting House School in Cambridge which ran from 1924 to 1927, when it folded because Geoffrey Pyke lost all his money. David told me that while he was at this school he was psychoanalysed by Melanie Klein who asked him questions about his parents' sex life!
He qualified in medicine in 1945 and became interested in diabetes as registrar to Alec Cooke (1899-1999) in Oxford from 1952 to 1957. Cooke influenced him greatly, particularly in how to write and present papers; many bon mots attributed to David came from Cooke such as: ‘you can’t speak too slowly’ or ‘remember you are speaking to deaf morons’. Before presenting his first paper in public David had to practice it many times in front of Cooke in an empty lecture theatre, an experience which I and many others later endured at David’s hands.
One of his first papers in 1954 pointed out the subjective nature of many physical signs such as finger clubbing. In his study there was wide disagreement between experienced observers with unanimity only when it was unequivocally absent.
In 1959 he was appointed to the staff of King's College Hospital, London, where from 1971 until his retirement in 1986 he was head of the Diabetic Department. The famous study of identical twins originally began as a project of the Rare Diseases Subcommittee of the Medical and Scientific Section of the British Diabetic Association but was taken over by David, a member of the committee. For the first few years of the study David drove to the twins' homes and did glucose tolerance tests there. Later this task was delegated to a succession of research fellows. In the first publication in 1972, 65 of 96 pairs of identical twins were concordant for diabetes and 31 discordant. When diabetes developed before age 40 half the pairs were discordant ; in those who developed diabetes after age 40 almost all were concordant. A family history of diabetes was common in concordant but rare in discordant pairs. These findings suggested that diabetes in older patients was “more genetic” and were one of the first indications that diabetes in the young and old were separate diseases, later to be recognised as type 1 and type 2 diabetes 
He was one of the first to use immunosuppression in the attempt to induce remission (unsuccessfully) in type 1 diabetes.
In 1978 he suggested that chlorpropamide alcohol flushing (CPAF) was a genetic marker for some cases of non-insulin-dependent diabetes and protective against retinopathy, proteinuria and large vessel disease. His group published 9 papers on CPAF between 1978 and 1981, 7 of which were in the British Medical Journal or Lancet. These findings were taken seriously, especially in the UK. Most continental Europeans were quizzical and Americans simply ignored it.
Pyke’s ‘simple’ test was to give the subject 250mg chlorpropamide and, 12 hours later, have them drink 40ml of sherry. On the face of it, the idea that a drug side effect can be a marker of a major gene is highly improbable and the reports did not make it clear that many of the patients who were given the test were on chronic chlorpropamide treatment (86% of those in the retinopathy study), and that this long-acting agent had not been adequately washed out of the circulation. Critics also pointed out that flushing is not an all or none phenomenon, that the findings were not reproducible and that a standard quantity of absolute alcohol should have been used rather than sherry. Eventually it was found that everyone, whether diabetic or not, will flush if their chlorpropamide levels are high enough.
Between 1975 and 1992 he was registrar of the Royal College of Physicians and wrote a quarterly column in which he wittily relayed gossip, scandal and apocryphal stories. His comment about it was, “The final text may not all be literally true, but it is not all untrue either.” It was certainly essential reading and in marked contrast to much of the turgid matter which emanated from the college. His Commentaries were later published separately, and are greatly treasured by those fortunate enough to own a copy.
David was widely read, as demonstrated by his letter on Vronsky's teeth  which noted that Tolstoy repeatedly emphasised the dental development of this character in Anna Karenina and expressed his own preference for edentulous heroes. He was a pioneer of family planning, and edited a journal on the topic with Jean Medawar for 24 years. After Peter Medawar's death, he co-authored a book with her entitled Hitler's Gift, which described the enormous contribution made by Jewish scientists expelled from Germany.
^ Pyke D. A.M. Cooke: an informal memoir. Quarterly Journal of Medicine 2000;93:693-698.
^ Pyke DA. Finger clubbing: validity as a physical sign. Lancet. 1954;2:352–54.
^ Pyke DA, Taylor KW. Glucose tolerance and serum insulin in unaffected identical twins of diabetics. Br Med J 1967;4:21-22.
^ Tattersall R B, Pyke D A. Diabetes in Identical Twins. Lancet 1972;2:1120-1125
^ Leslie RDG, Pyke DA, Denman AM. Immunosuppressive therapy in diabetes. Lancet 1985;325:516.
^ Hillson RM, Hockaday TDR. Chlorpropamide-alcohol flush: a critical reappraisal. Diabetologia 1984; 25: 6-11.
^ Pyke DA. Vronsky's teeth. Lancet 1994;334:1784