Sir William Osler was a Canadian physician with significant contribution to our modern knowledge of infective endocarditis. He studied medicine at McGrill University and obtained his medical degree in 1872. He was appointed professor of medicine at John Hopkins in 1889. Subsequently he was nominated the chairman of medicine at Oxford University in 1905.
In 1881, Osler published an article, "On some points in the etiology and pathology of ulcerative endocarditis," in which he recognized certain role to micrococci in the pathogenesis of endocarditis. He, however, expressed the opinion that these agents may not be the primary responsible for all cases of endocarditis:
"If we study the conditions under which endocarditis develops, we find almost invariably that the patients are the subject of some other constitutional affection which, as we say, predispose to it. What determines the precise form of the endocarditis, we do not know, but the soft endocardial vegetations form a suitable nidus for the development of micrococci."
He also raised the following question regarding the potential role of these agents in the seriousness of this disease:
"It seems a pertinent question to ask, if in the malignant form of endocarditis, the micrococci are so potent, why in other cases in which they are equally prevalent, should they be inert? Of course it may be urged that the micrococci may be of different kinds or possess diverse qualities, or that the resistance offered by the tissue to their penetration varies in different cases or that it is only in weakened and debilitated states that these little bodies thrive. There is, I think, something worthy of attention in this latter view."
It is important to stress that techniques of tissue culture and microscopic identification of microorganisms were at their early stage of development and still very imprecise in 1880's. Osler's view changed over a short period of time as reflected in his landmark series of paper that appeared in 1885.
Osler's Gulstonian Lectures on malignant endocarditis appeared in Lancet as a series of three articles and are regarded as the most important contribution pertaining to the field of infective endocarditis. They were based on Osler's personal experience which included more than 200 patients with clinico-pathological correlation.
In the introduction of the first article, he noted that "in discussing the subject of endocarditis we are met at the outset by difficulties of nomenclature and classification."
He then suggested a simpler classification mostly based on clinical presentation and rejected the use of terminologies based on anatomical features (e.g. verrucose endocarditis, ulcerative endocarditis). In his Gulstonian lecture, Osler remarked:
"The designation acute may be used to indicate those forms which are accompanied by proliferation [vegetation] of and exudation upon the endocardial surface, with or without loss of substance, as opposed to chronic, in which there are sclerotic changes without vegetation." He also mentioned that from anatomical point of view, there were no essential differences in the various forms of acute endocarditis.
From clinical point of view, he distinguished two forms that he described under the title simple and malignant: "the simple being those with few or slight symptoms, and which run a favorable course; the malignant the cases with severe constitutional disturbance and extensive valve lesions, whether ulcerative or vegetative, the term being more clinical than anatomical."
Finally he did not favor the use of numerous other terms such as septic, infectious, diphteric, mycosis endocardii, or arterial pyaemia, although he recognized that "each one of these terms expressed some special clinical feature. "
The distribution of valvular lesions among these 200 patients was as follows: aortic and mitral valve in 41, aortic valve in 53, mitral valve in 77,tricuspid valve in 19, pulmonary valve in 15, and the heart wall in 33. In his experience, Osler recorded that patients with "sclerotic or malformed valves" were more prone to endocarditis.
Osler provided a very accurate description of valvular vegetation:
"The study of a small fresh endocardial vegetation shows it to be made up of cells derived from the subendothelial layer, round and fusiform, which by their proliferation have produced a small nodular projection on the surface of the endocardium;...What part the endothelium plays in this growth has not been determined; usually before the mass attains any size the smooth surface is lost, and there is deposited upon it a cap of fibrin in the form of a granular sometimes stratified , material, of variable thickness. Though this resembles an ordinary coagulable exudation, it is probably deposited directly from the blood, and is of the nature of a thrombus. Upon and in this layer may be found, sometimes in large numbers, those remarkable little bodies which have so long been known, when collected together, as Schultze's granule masses, and which have of late become so prominent as the blood-plates of Bizzozero...,the connexion of which with fibrin formation has been so strongly insisted upon by Bizzozero."
In this microscopic analysis, Osler described for the first time the fact the blood platelets associated with fibrin were included in the composition of the vegetation. He also indicated that these elements were originated from the circulating blood and were not produced by endocarditic lesions. He stressed the fact that micrococci were constant elements in the vegetation and that Gram Staining was the most satisfactory method to identify them.
Regarding the localization of these vegetations, he noted: "The valves are most often attacked along lines of closure, as in the simple endocarditis, the auricular faces of the mitral flaps and the ventricular surfaces of the aortic cusps suffering most severely." He also mentioned that the anterior leaflet of the mitral valve was most frequently affected by aneurysmal formation.
In the second part of his Gulstonian Lecture, Osler began by commenting on the difficulty of the diagnosis and the variability of the symptoms:
"The different modes of onset, and the extraordinary diversity of symptoms which may arise, render it very difficult to present a satisfactory clinical picture."
He then highlighted the principle symptoms which consisted of:
Fever of variable intensity accompanied by rigor, vomiting and headache.
Cardiac symptoms with pain, palpitation, sense of distress and murmur.
Signs specific to embolization events:
Brain: delirium, coma and paralysis
Kidneys: hematuria, abdominal pain
Retinal hemorrhage with loss of vision
After presenting the general symptoms, Osler described a variety of subclinical groups and discussed the specific symptomatology and diagnosis for each one of them.
Osler described two forms of malignant endocarditis that he called "pyemic" type and "typhoid" type. He characterized the "pyemic" type as severe septicemia. He commented that the typhoid type was the most common and characterized by irregular temperature, early prostration and the involvement of the nervous system, delirium, somnolence and coma, and petechial and skin rashes. With regard to that particular subgroup, Osler stressed the fact that "Few diseases present greater difficulties in the way of diagnosis than malignant endocarditis, difficulties which in many cases are practically insurmountable."
He then defined the "cardiac group" which was composed of patients with chronic valve disease with recent endocarditis and commented that "these are the cases of ulcerative endocarditis which present fewest difficulties in diagnosis." He reported that the evolution of the disease in that group was slow in most instances: "the term malignant seems not at all applicable to them."
This group of patients would be further studied by Osler and would be characterized by a new entity called "chronic bacterial endocarditis."
He finally described the "cerebral group" in which the neurological manifestations were predominant. Regarding the cardiac manifestations in that group he wrote
"The heart symptoms may early attract attention, from the complaints of pain and palpitation; but, as a rule, they are latent, and unless looked for are likely to be overlooked." According to Osler, the existence of an irregular fever and the occurrence of embolic events were sufficient to make the diagnosis.
In the third Gulstonian lecture, Osler discussed the etiology of bacterial endocarditis and wrote:
"the theory of acute endocarditis which at present prevails, and the only one to which I shall refer, is that it is in all its forms an essentially mycotic process. This very attractive theory can be adjusted to meet every requirement of the case." He, however, added the need for further evidence to confirm this hypothesis.
The entire text of Osler's Gulstonian lectures is reproduced here.
In 1893, Osler published an article entitled, "The chronic intermittent fever of endocarditis", in which he reported two cases of endocarditis with prolonged duration ( 10 and 11 months respectively).
He summarized the clinical characteristics of these 2 cases as follows: "daily intermittent fever for many months, progressive failure of strength, physical signs of cardiac disease with systolic murmur and hypertrophy of the left heart, late development of embolic symptoms, and cutaneous ecchymoses." The autopsy of the heart in both cases showed mitral valve lesions of rheumatic etiology and secondary infective endocarditis.
In the conclusion of this article, he remarked:
"The special interest of the group illustrated by these cases is the chronic intermittent fever with progressive failure of health and strength, without dyspnea, anasarca, or other features of valvular disease."
In 1908, Osler published his extended experience reporting 10 patients with prolonged endocarditis that he had seen over a period of two decades. The title of this article which first appeared in French was "Endocardites infectieuses chroniques."
In this work, Osler described the clinical presentation of these ten cases including the painful red cutaneous nodules ("nodosites cutanees eryhemateuses douloureuses ephemeres"). The duration of the disease varied from 4 to 13 months. The most common microorganism identified on cultures was streptococcus.
This paper was the first description of subacute bacterial endocarditis (or so-called "Osler's endocarditis") and these cutaneous lesions which are commonly encountered in the subacute forms are now known as "Osler's nodes."
This paper was published in English language one year later.