Lane Library

Oliver Wendell Holmes

In 1834, at the age of 34, Oliver Wendell Holmes wrote a paper on The Contagiousness of Puerperal Fever in which he convincingly argued that the infection was often transmitted to the patient by her attendants. In addition he laid down sound procedures for preventing the spread of the contagion. In retrospect, his treatise stands as an historic landmark, not because of any original observations, but because of the clarity and forcefulness with which he addressed both the transmission and prevention of this devastating disease - a disease now known as "postpartum endometritis", and so well controlled by asepsis and antibiotics as to be rarely life-threatening. In modern form it bear no resemblance to the fierce and consuming pelvic sepsis of the pre-Listerian era. Then it was usually an overwhelming infection, commonly sudden in its onset within a few days after delivery with chill, fever and prostration, and often as rapidly lethal. It occurred both sporadically and in epidemics, mortality reaching near 100 percent in small clusters of "malignant" infection, and up to more than 35 percent in some epidemics. As a threat to all young mothers it was the destroyer of families, and a most dreaded pestilence. [28]

Holmes's interest in puerperal fever came about by chance. He graduated from Harvard Medical School in 1836 and served as Professor of Anatomy and Physiology at Dartmouth College from 1838 to 1840. He then returned to Boston where he went into general practice and became a member of the Boston Society for Medical Improvement. At one of the Society's meetings a report was read that concerned a physician who did a post mortem examination on the body of a woman who had died of puerperal fever. The physician himself died of infection in less than a week, apparently in consequence of a wound received while doing the autopsy. During the interval between receiving the wound and dying from it, the physician delivered several women, all of whom developed puerperal fever.

Based on his conviction that the physician had transmitted the contagion of puerperal fever from the deceased woman to the women he delivered, Holmes stated the following general principle: [29]

The disease, known as Puerperal Fever is so far contagious as to be frequently carried from patient to patient by physicians and nurses.

This concept was by no means new, as Holmes illustrated by citations from the medical literature, mainly British journals, which he reviewed thoroughly. He found numerous accounts of epidemics of puerperal fever, and many reports of multiple cases of puerperal sepsis occurring in the practice of a specific doctor, midwife or nurse. Frequently, the presumed carrier of the contagion had an immediately prior exposure to puerperal sepsis in another patient or at an autopsy, or to a patient with erysipelas or peritonitis. In addition to reviewing the literature, Holmes consulted older and more experienced practitioners in Massachusetts who informed him of similar outbreaks of puerperal fever in the practice of individual physicians in their areas.

As an example of the spread of puerperal fever by contagion, Holmes referred to a paper widely quoted in the medical literature of his day: "A Treatise on the Epidemic Puerperal Fever of Aberdeen" published in 1795 by Dr. Alexander Gordon who treated 77 cases of puerperal fever (with 28 deaths) in Aberdeen, Scotland, during the two year period from December 1789 to March 1792. Dr. Gordon wrote: [30]

(T)his disease seized such women only as were visited, or delivered, by a practitioner, or taken care of by a nurse, who had previously attended patients affected with the disease. I had evident proofs of its infectious nature, and that the infection was as readily communicated as that of the small-pox or measles and operated more speedily than any other infection with which I am acquainted.

I had evident proofs that every person who had been with a patient in the puerperal fever, became charged with an atmosphere of infection, which was communicated to every pregnant woman who happened to come within its sphere. This is not an assertion, but a fact, admitting of demonstrations, as may be seen by a perusal of the foregoing table (referring to a table in his paper of 77 cases in which the channel of propagation was evident).

(He adds.) It is a disagreeable declaration for me to mention, that I myself was a means of carrying the infection to a great number of women. (He then enumerates a number of instances in which the disease was conveyed by midwives and others to neighboring villages, and declares that) these facts fully prove, that the cause of puerperal fever, of which I treat, was a specific contagion, or infection, altogether unconnected with a noxious constitution of the atmosphere. . . .

(But his most terrible evidence is given in these words,) I arrived at that certainty in the matter, that I could venture to foretell what woman would be affected with the disease, upon hearing by what midwife they were to be delivered, or by what nurse they were to be attended, during their lying-in: and, almost in every instance, my prediction was verified.

In support of his thesis that puerperal fever is contagious Holmes also made reference to more than twenty other authors whose views conformed with his own and those of Dr. Gordon. [31] Among them was the distinguished James Blunder, Professor of Obstetrics and Lecturer on the Diseases of Women at Guy's Hospital. The following excerpt from Blundell's Lectures on Midwifery, as quoted by Holmes, reflects the lingering uncertainty in the 1840's, even among some of the highest authorities, as to the contagiousness of puerperal fever: [32]

Those who have never made the experiment, can have but a faint conception how difficult it is to obtain the exact truth respecting any occurrence in which feelings and interests are concerned. Omitting particulars, then, I content myself with remarking, generally, that from more than one district I have received accounts of the prevalence of puerperal fever in the practice of some individuals, while its occurrence in that of others, in the same neighborhood, was not observed. Some, as I have been told, have lost ten, twelve, or a greater number of patients, in scarcely broken succession; like their evil genius, the puerperal fever has seemed to stalk behind them wherever they went. Some have deemed it prudent to retire for a time from practice. In fine, that this fever may occur spontaneously, I admit; that its infectious nature may be plausibly disputed, I do not deny; but I add, considerately, that in my own family, I had rather that those I esteemed the most should be delivered, unaided, in a stable, by the manger-side, than that they should receive the best help, in the fairest apartment, but exposed to the vapors of this pitiless disease. Gossiping friends, wet nurses, monthly nurses, the practitioner himself, these are the channels by which, as I suspect, the infection is principally conveyed.

Blundell, in his textbook on The Principles and Practice of Obstetricy, has little more than this to say regarding the control of the spread of puerperal fever: "As to its prevention, I know of nothing certain." [33]

Holmes, having gathered exhaustive and thoroughly convincing evidence of the contagiousness of puerperal fever, was doubtless gratified to find himself in agreement with the majority of contemporary British authors on the subject. Nevertheless, he learned that some eminent obstetricians did not agree with his conclusion. For example, in the course of his research he discovered in the Quarterly Summary of the Transactions of the College of Physicians of Philadelphia for May, June and July of 1842 the report of an on-going epidemic of puerperal fever in Philadelphia in which there were egregious examples of cases traceable to single physicians. [34]

One of these Philadelphia physicians, a Dr. Rutter, had some 70 cases of puerperal fever with 15 deaths during a period of less than 12 months during 1842 - a number rivaling the 77 cases in two years in the Aberdeen epidemic reported by Dr. Gordon. Moreover, it immediately caught Holmes's eye that Dr. Charles D. Meigs, Professor of Obstetrics at Jefferson Medical College, had consulted on some of Dr. Rutter's patients. Professor Meigs was aware that Dr. Rutter had a far greater number of such cases than any other practitioner in Philadelphia, but considered this due to the fact that he had a large practice. [35] Holmes took special note of Meigs's viewpoint and made it plain in his paper that this was an outrageous conclusion to be reached by a professor of midwifery who, in the face of a raging epidemic of puerperal fever in Philadelphia, made no reference to the contagiousness of the disease, and attributed its grossly epidemic proportions in Dr. Rutter's private practice to coincidence. Meigs's failure to recognize the role of contagion in the epidemic is particularly surprising since he had recently (in 1842) edited a monograph on puerperal fever that included the treatises of Dr. Gordon and three other well known British obstetricians, all of whom commented on its communicable nature. [36]

In any event, Meigs refused to acknowledge the contagiousness of puerperal fever and took strong exception to Holmes's sharp criticism of his position on the matter. There followed an acrimonious exchange in which Meigs attacked Holmes in disparaging language to which Holmes replied: "I take no offense and attempt no retort. No man makes a quarrel with me over the counterpane that covers a mother, with her new-born infant at her breast! There is no epithet in the vocabulary of slight or sarcasm that can reach my personal sensibilities in such a controversy." [37] Holmes gave not an inch of ground in the dispute that continued for over a decade. The judgement of posterity has since been harsh on Professor Meigs who stated that "(I prefer) to attribute these cases (of puerperal fever) to accident, or Providence, of which I can form a conception, rather than to a contagion of which I cannot form any clear idea, at least as to this particular malady." [38] The 1842 edition of Meigs's widely acclaimed textbook entitled The Philadelphia Practice of Midwifery makes no mention of the contagiousness or the prevention of puerperal fever. It was as though the extensive and compelling contemporary literature on the subject did not exist. The mind of the Dean of American Obstetrics was completely closed. [39]

As was the mind of Hugh L. Hodge, Professor of Obstetrics at the University of Pennsylvania in Philadelphia, who also denied the contagiousness of puerperal fever and assured his students that they, as physicians, could never be the minister of evil to convey a horrible virus to their parturient patients. [40] Thus Holmes had the two most influential professors of obstetrics in America aligned against him.

Prevention of puerperal fever in the pre-microbial era was based on the assumption that an unknown contagion existed in the lying-in premises, or was carried to the childbed by an attendant of the mother. Holmes did not indulge in speculation (as did Drake) regarding the nature of the contagion, but assumed the physical presence of an unseen, transmissible agent. Years later, in 1894, Holmes said he was pleased to remember that "I took my ground on the existing evidence before a little army of microbes was marched up to support my position." [41]

As to preventive measures within lying-in hospitals, the British medical literature of the day called for strict cleanliness of bedding and wards, and good ventilation to combat epidemics. If these measures failed, the ward should be closed and the patients relocated. Outbreaks of puerperal fever were not unusual in lying-in wards and, on that account, some obstetricians were convinced that the loss of life from puerperal fever occasioned by lying-in institutions completely defeated the object of their founders. Although he does not prescribe a specific regime for the decontamination of hospitals, Holmes stresses the danger of spread of contagion within that environment. [42] Among others, he refers to the observations of Dr. Edward Rigby, Physician to the General Lying-in Hospital and Lecturer on midwifery at St. Bartholomew's Hospital in London: [43]

That the discharges from a patient under puerperal fever are in the highest degree contagious, we have abundant evidence in the history of lying-in hospitals. The puerperal abscesses are also contagious, and may be communicated to healthy lying-in women by washing with the same sponge: this fact has been repeatedly proven at the Vienna Hospital; but they are equally communicable to women not pregnant; on more than one occasion the women engaged in washing the soiled bed linen of the General Lying-in Hospital have been attacked with abscesses in the fingers or hands, attended with rapidly spreading inflammation of the cellular tissue.

As to preventive measures applicable to personnel, many authors recommended procedures to be observed by accoucheurs and other attendants in order to avoid spreading the contagion. The following are examples of such recommendations.

In 1795 Alexander Gordon, MD, Obstetrician at Aberdeen, Scotland, suggested: [44]

With respect to the most effectual means of preventing infection from being communicated, I must speak with great uncertainty, because in this matter I have not experience for my guide. . . That fresh air and cleanliness are insufficient for the destruction of contagion, and that there is no certain antidote but fire and smoke, has been demonstrated . . . .(Therefore), the patient's apparel and bedclothes ought either to be burnt or thoroughly purified, and the nurses and physicians who have attended patients affected with puerperal fever ought carefully to wash themselves, and to get their apparel properly fumigated before it be put on again.

In 1817 William Hey, Esq., Surgeon of the General Infirmary at Leeds, England, wrote: [45]

It was my custom . . . to use such precautions in my attendance on patients, as to render it impossible for me to convey infection to them; and those who would take the same trouble might practice safely, were the disease as infectious as Dr. Gordon represents it to be. It was an invariable rule with me never to attend a patient in childbed in any article of clothing which had been in the presence of one affected with the puerperal fever; nor without washing repeatedly such parts of my person as could have been exposed to infection. This trouble I took for the satisfaction of my own mind, and the safety of my patients, though not convinced it was necessary. But in so important a matter I wished for perfect security under any supposition.

In about 1814 John Armstrong, MD, Obstetrician at Sunderland, England, had this brief comment: [46]

When puerperal fever is epidemical, the accoucheur should make it a point of duty to have the apartments of women who he is engaged to attend properly cleaned and ventilated before confinement; to prevent nurses and other persons who have been with those affected, from waiting upon or going near any patient about to be delivered; to pay the utmost scrupulous regard to the cleanliness of his own person, using daily ablutions of the whole body, and frequent changes of linen and dress.

In 1833 Robert Lee, MD, Obstetrician at the British Lying-in Hospital in London, proposed: [47]

These facts (affirming the contagiousness of puerperal fever) point out the necessity of adopting every precaution to prevent the extension of the disease, by careful and repeated ablution, and changing of clothes after attending patients who are affected with it. They show, also, whether they be conclusive or not as to the communicability of the affection from person to person, that we ought not to expose ourselves beyond what is necessary in examining the bodies of those who have been cut off by the complaint. When post mortem examinations are required, they should be conducted by those who are not engaged in the practice of midwifery. We certainly owe it as a duty to our patients to act as if the contagion always existed.

In 1841 Dr. Rigby of St. Bartholomew's in London issued an emphatic warning: [48]

The contagiousness of puerperal fever has long since ceased to be a matter of doubt, and instances have repeatedly occurred of practitioners and nurses communicating the disease to several patients in succession. Dr. Gooch has recorded some striking instances of the kind, and we could enumerate many others if necessary. Where a practitioner has been engaged in the post mortem examination of a case of puerperal fever, we do not hesitate to declare it highly unsafe for him to attend a case of labour for some days afterwards. The peculiar smelling effluvia which arises from the body of a patient during life is quite, in our opinion, sufficient to infect the clothes; and every one who has made a minute dissection of the abdominal viscera, especially in fatal cases of puerperal fever, knows full well that it is almost impossible to remove the smell from the hands for many hours, even with the aid of repeated washing; it must be, therefore, self-evident, that, under such circumstances, it would be almost criminal to expose a lying-in patient to such risk.

Now consider, in relation to all of the above, the following protocol framed by Holmes: [49]

  1. A physician holding himself in readiness to attend cases of midwifery, should never take any active part in the post-mortem examination of cases of puerperal fever.

  2. If a physician is present at such autopsies, he should use thorough ablution, change every article of dress, and allow twenty-four hours or more to elapse before attending to any case of midwifery. It may be well to extend the same caution to cases of simple peritonitis.

  3. Similar precautions should be taken after the autopsy or surgical treatment of cases of erysipelas, if the physician is obliged to unite such offices with his obstetrical duties, which is in the highest degree inexpedient.

  4. On the occurrence of a single case of puerperal fever in his practice, the physician is bound to consider the next female he attends in labor, unless some weeks, at least, have elapsed, as in danger of being infected by him, and it is his duty to take every precaution to diminish her risk of disease and death.

  5. If within a short period two cases of puerperal fever happen close to each other, in the practice of the same physician, the disease not existing or prevailing in the neighborhood, he would do wisely to relinquish his obstetrical practice for at least one month, and endeavor to free himself by every available means from any noxious influence he may carry about with him.

  6. The occurrence of three or more closely connected cases, in the practice of one individual, no others existing in the neighborhood, and no other sufficient cause being alleged for the coincidence, is prima facie evidence that he is the vehicle of contagion.

  7. It is the duty of the physician to take every precaution that the disease shall not be introduced by nurses or other assistants, by making proper inquiries concerning them; and giving timely warning of every suspected source of danger.

  8. Whatever indulgence may be granted to those who have heretofore been the ignorant causes of so much misery, the time has come when the existence of a private pestilence in the sphere of a single physician should be looked upon not as a misfortune but a crime, and in the knowledge of such occurrences, the duties of the practitioner to his profession, should give way to his paramount obligations to society.

Holmes concluded his paper with these eight unambiguous rules of conduct for the accoucheur. Compared to guidelines in the existing literature, examples of which were cited above, Holmes's precepts were comprehensive, explicit and uncompromising. They were the most definitive standard yet published on the prevention of a fearsome and seemingly capricious disease. Respect for his protocol's eight enduring principles, ignored for decades by prisoners of false doctrines such as the Philadelphia professors, saved countless lives around the world.

There are other reasons for the historic significance of Holmes's essay. As we have seen, during the previous 50 years numerous epidemics of puerperal fever had been studied and reported by a new generation of British physicians. They had abandoned traditional medical dogma for a scientific approach involving correlation of clinical course with post mortem features. It was at this juncture that Holmes fortuitously became interested in the problem. Without burdensome preconceptions, he reviewed the available reports and recognized that they provided incontrovertible evidence of the contagiousness of puerperal fever. More importantly, he convincingly traced the contagion's common mode of epidemic spread to the physicians and others who attended the patient.

By this time the concept of a transmissible "contagion" of some kind as the agency of infection in puerperal fever had gained some but by no means general acceptance. There was still much equivocation and denial in high places, and widespread ignorance among practicing physicians of the risk of contagion. To Holmes's New England conscience, there was lacking in the medical community at large a proper sense of outrage and urgency over the propagation of a preventable calamity, and it was inexcusable.

Far from avoiding the implications of this conclusion, Holmes analyzed existing evidence and, in a persuasive treatise that for cogency and eloquence is at once both a medical and a literary classic, he defined the obligations of all who attend at childbirth. It has been rightly observed that Holmes was not an obstetrician nor had he done independent research on his subject, but he was the first to give unmistakably clear and credible voice to the emerging consensus that puerperal fever was contagious, a specific infection often conveyed by doctors and nurses. His achievement was to create a synthesis of existing observations and ideas from which he evoked a momentous conclusion - no longer could there be any question of the contagiousness of this terrible affliction, or of the human agency in its dissemination. For this historic contribution Holmes deserves to be honored as an illustrious pathfinder in world medicine.

Unfortunately, circumstances prevented the early and wide distribution of Holmes's paper that its importance merited. The paper was originally an essay read before the Boston Society for Medical Improvement. At the request of the Society, the essay was printed as a paper in the New England Quarterly Journal of Medicine and Surgery for April 1843. As this journal never had a large circulation and was discontinued after one year, the paper was not brought fully to the attention of physicians or the public. That it was not entirely unnoticed is shown by favorable reference to it in 1852 in the highly regarded Dictionary of Practical Medicine by James Copland, MD, Consulting Physician to Queen Charlotte's Lying-in Hospital in London. In affirming his belief in the infectiousness of puerperal fever, Dr. Copland pointed out that "Dr. Holmes has forcibly and eloquently brought this much neglected subject before the profession." But Copland also reminded his readers that the contagiousness of puerperal fever was still denied by such established authorities as Hulme, Leake, Hull, Beaudeloque, Tonnellé, Dugé, Dewees and others. [50]

In 1855, twelve years after its original appearance in the New England Quarterly, Holmes reprinted his essay ,"without the change of a word or syllable", as a private publication under the title of Puerperal Fever, as a Private Pestilence. He was led to do so by his disappointment over its limited distribution originally, and by his conviction as to the continuing importance of warning refractory members of the profession of the contagiousness of puerperal fever. On a more personal level he was offended by the disparaging remarks of Dr. Meigs, and appalled by the pompous denial by the Philadelphia professors of the infectious nature of this terrible disease, a truth that the "commonest exercise of reason" should reveal. Holmes prefaced the reprint with a masterful Introduction in which he aired all these issues, deflated the pretensions of the professors, and warned medical students of the sophistry in their arguments. [51]

By the time his essay was reprinted in 1855, Holmes had joined the Harvard medical faculty as Parkman Professor of Anatomy and Physiology, a post that he held for 35 years from 1847 to 1885 (the chair of Physiology was separated in 1871), after which he continued for 12 more years (1882-1894) as Emeritus Professor. He was Dean of the Medical School from 1847 to 1853. [52] After his appointment to the Parkman professorship, Holmes gradually withdrew from the practice of medicine, but he is warmly remembered as a legendary teacher of Anatomy. His engaging style and captivating wit made him, it is said, the only professor who could keep the students awake during a 1 p.m. lecture. As the years passed, his literary affinities increasingly claimed his interest, and he became better known as a conversationalist and author than as a physician. His graceful pen earned him a respected place as poet (Chambered Nautilus) and essayist (Autocrat of the Breakfast Table) among such contemporary writers of the New England Renaissance as Emerson, Hawthorne, Longfellow and Whittier. [53] [54] [55]

Nevertheless, Holmes will also be well remembered by distant posterity for his Thesis of 1843 on puerperal fever and the passion with which he defended it against all sceptics and against the entrenched error of "the teachings of two Professors in the great schools of Philadelphia." [56]

If I am wrong (he wrote), let me be put down by such rebuke as no rash declaimer has received since there has been a public opinion in the medical profession of America; if I am right, let doctrines which lead to professional homicide be no longer taught from the chairs of those two great Institutions. Indifference will not do here; our Journalists and Committees have no right to take up their pages with minute anatomy and tediously detailed cases, while it is a question whether or not the "black-death" of child-bed is to be scattered broadcast by the agency of the mother's friend and adviser. Let the men who mould opinions look to it; if there is any voluntary blindness, any interested oversight, any culpable negligence, even, in such a matter, and the fact shall reach the public ear; the pestilence-carrier of the lying-in chamber must look to God for pardon, for man will never forgive him.

Never had the rites of motherhood been so ably defended.

Lane Library