Part 1. The Thymus and Sudden Death:
A Pervasive Fear in Medicine and in the Public
"A thymic death is one of the supreme tragedies of surgery. An apparently healthy child dies during the administration of an anesthetic, during or after an uncomplicated tonsil and adenoid operation, or, as recently happened, during a simple circumcision. Again, as reported by one of our medical examiners [coroners], a child was standing on the edge of the sidewalk. A runaway horse dashed by and the child dropped dead. At autopsy the condition known as status lymphaticus was found; that is, there was an enlarged thymus and a hypertrophy of all the lymphoid structures of the alimentary canal, these structures being the solitary follicles --- Peyer's Patches and the mesenteric glands. This slight pathology was all that was found to explain the unexpected death." This statement in 1926 is from Dr. Harris P. Mosher, Dr. Alexander S. MacMillan, Dr. Frederic E. Motley (Mosher 1926).
By no means was the concern about disorders of the thymus gland confined to infants under one year of age. Indeed, as we shall see in this chapter and the next, such concern was evident for all ages up through the 'teens. There were long-standing ideas concerning the thymus gland and sudden death; the availability of x-rays did not help to dismiss these long-standing worries, particularly in situations of stress, of which anesthesia and surgery were prime examples.
We continue with a quote from M.L. Janower and O.S. Miettenen, concerning policy at a major institution in Boston, Massachusetts (Janower 1971, p.753):
"From 1924 to 1946, it was the policy of the Massachusetts Eye and Ear Infirmary in Boston to apply prophylactic irradiation in every case in which an "enlarged" thymus gland was diagnosed in infancy. The assessment of the size of the thymus gland was based upon an anteroposterior roentgenogram of the chest taken in expiration with the patient in the supine position. Whenever the width of the superior mediastinum was at least half the width of the heart the gland was characterized as `enlarged' or `suspicious,' and the child was given radiation treatment; if the gland was less than half the width of the heart the child was not given radiation. On the basis of these criteria, 1,131 children received thymus gland radiation in the 22-year period." Despite one use of the phrase "in infancy," the irradiated children were an average of 4.7 years of age at exposure.
At the end of this chapter, we will evaluate annual average breast-dose from such treatment of children, ages 1 through 9, for Column M of our Master Table.Part 2. Enlarged Thymus, with and without Symptoms: Dr. O'Brien
The issues surrounding such treatments were succinctly and well stated by Dr. Frederick W. O'Brien in 1929, at the annual meeting of the American Roentgen Ray Society. We quote O'Brien (1929, p.271):
"For some years now, it has been the routine in certain hospitals to examine all children roentgenologically before submitting them to a general anesthetic. If there is found what is thought to be an enlarged thymus gland, a prophylactic roentgen treatment is given, because of the rather well-entrenched belief that an enlarged thymus gland is an integral factor of the syndrome of so-called status lymphaticus, a reputed cause of otherwise unexplained sudden death." He also described the controversy:
"An enlarged thymus as a symptom-producing organ, and the roentgenologist's ability to diagnose it, has not gone unchallenged. Status lymphaticus, indeed, as a pathologic entity is declared a misconception."
And the two propositions posed by Dr. O'Brien:
1. Is there a symptom-producing enlarged thymus in infants that can be diagnosed by the roentgen ray and relieved by irradiation?
2. Is there an enlarged thymus without symptoms in infants, children, and young adults, which represents objective evidence of status lymphaticus, and which can be diagnosed by the roentgen ray and should receive prophylactic radiation?"
Uncertainty about Where the Truth Lay
Dr. O'Brien appeared uncertain about where the truth lay. With respect to mortality, he divided the cases as follows (1929, p.273): "Cases of thymic death are readily referred to two distinct groups, those with and without symptoms." About cases with symptoms:
"Accepting the concept of the evolution of the thymus, I do not find any student of the subject who does not concede the existence of a symptom-producing thymus at least in infants. The groups with symptoms usually display the so-called syndrome of thymic asthma, characterized by attacks of inspiratory dyspnea, inspiratory stridor and the so-called Rehn's symptoms, the expiratory swelling of a tumor [meaning a fulness, not a real tumor], the cranial thymus-end in the jugulum." And about cases without symptoms (p.274):
"The other group of so-called thymic deaths occurring in older children in an apoplectiform manner without premonitory symptoms is the one responsible for the current hospital practice to which I have referred of roentgenographing all children before anesthesia." And:
"It has been suggested that these sudden deaths have been due to improper anesthetization and poor operative judgment and technique [see Part 3]. But they occur even without anesthesia and following minor or no surgical procedure at all. These are the cases that are the tragedies of practice and if personal, or under one's immediate authority, cause one to pause."
Prophylactic Irradiation: Claims of Success vs. the Skeptics
Dr. O'Brien continued (p.274):
"Mosher, well aware of the newer concepts, clings to the old idea that the thymus is involved in these sudden deaths, and he has the added argument of no sudden death in his great clinic since his study with MacMillan and Motley (Mosher 1926). They found out of a total of 2,344 children, a positive thymus shadow in 7.5 per cent. Of the positive cases treated by roentgen radiation all were successfully operated on and since employing the routine roentgen examination of the mediastinum, they have had no sudden unexplained death under anesthesia." And (p.274):
"This has been the practice for two years at the Boston City Hospital and the Cambridge Municipal Hospital with both of which I am connected. At the Boston City Hospital, out of a total of more than 2,000 children, there has been but one unexplained death under anesthesia, which occurred recently. There had been no prior roentgen examination and there was no autopsy. At the Cambridge Municipal Hospital, in 526 cases, there have been no deaths. A survey of the chests showed a `broadened mediastinal shadow'in +6 percent."
We will return to Dr. O'Brien's conclusions in Part 4.
Others were much more skeptical about the 1926 report by Dr. Mosher and colleagues, on the grounds that the absence of deaths was not proof that the roentgen treatment of ostensibly enlarged "silent" thymuses was the real reason for such absence.
Boston: "Nearly All Infants Are X-Rayed Promptly"
During the discussion of Dr. O'Brien's paper, Dr. R.D. Leonard of Boston commented as follows (in O'Brien 1929, pp.276-277):
"We have to construct some theory along which to work, and as our empirical results and our interest and enthusiasm in this theory develops we subconsciously forget that we are working simply on a hypothesis. I think this is something we need to bear in mind in relation to this thymus problem. As we see from these two papers, we know what is generally thought about the thymus gland, but very little has actually been proven." And:
"In Boston, as in various other places, the thymus gland is at the present time a popular subject for discussion. Our obstetricians are very much interested in the thymus gland. Nearly all infants are x-rayed promptly and if any sort of a shadow is seen in the mediastinum, treatment is instituted. In the sudden, unexplained deaths, we recognize the thymus gland as a probable cause. Therefore on account of the present popular interest in the thymus gland, I would add this word of caution that, as far as possible, we as roentgenologists should be sure we know what we are talking about."
The statement by Dr. Leonard strongly indicates that the practice of roentgenographing "nearly all infants" was not confined to research studies. One implication: Our estimate of breast-dose from the screening-process itself --- in Chapter 8, Part 2, Item 10 --- is almost certainly a serious underestimate.Part 3. The 1914 Edition of "Anesthesia": Dr. Gwathmey
In his 1914 Edition of Anesthesia, Dr. James Tayloe Gwathmey made some quite illuminating comments about the so-called Status Lymphaticus deaths which occurred in relation to anesthetic administration (p.331):
"Status Lymphaticus. --- Definition. --- Status lymphaticus or thymicus, or lymphatism, is a condition of infancy and childhood, marked by hyperplasia of the lymphatic structures, spleen and bone marrow, and persistence of the thymus gland (Stedman). It has also been defined as a condition of unstable equilibrium, coma, convulsions and vomiting accompanying hyperplasia of the persisting thymus (Gould); and as a morbid state due to excessive production or growth of lymphoid tissues, such as the thymus and thyroid glands, resulting in impaired development, lowered vitality, and sometimes death (Dorland)." And:
"History. --- As early as 1614 attention was called by Felix Plater to the fact that the thymus was enlarged in three cases of sudden death from dyspnea in one family. In 1823, and again in 1829, Kopp mentioned the association of the enlargement of the thymus gland with sudden death. Paltauf, in 1889 and 1890, collected, for the first time, a large number of cases of sudden death in adults, in which there was enlargement of the tonsils, lymphatic gland system, the follicles at the base of the tongue, the spleen, and the thymus gland, with narrowing of the aorta. Kundrat, in 1895, published ten cases of death immediately after anesthesia by chloroform or some mixture containing it, also one case in which ether was the anesthetic. Sudden deaths were noted after this time in many cases in which no anesthetics had been administered. Lymphatic hyperplasia had been found to occur in every chloroform fatality for the past twenty years in the children's clinic at Gratz. The first case recorded in England was reported by Wolff in 1905." And, in commenting on some features of such patients (p.332):
"Pasty complexion, a large amount of subcutaneous fat, and, in adults, a scant amount of axillary or pubic hair are usual; also the hair of the head has a peculiar dry brittle character ..." And (p.332):
"Most patients dying during or immediately after anesthesia have been young people or children, of flabby type, with enlarged adenoids, tonsils, thyroid (usually), and thymus; with narrow, high-arched palate, small mouth and throat, and weak heart sounds. During anesthesia a grayness of complexion or pallor is witnessed, with weak heart action and shallow breathing. Enlargement of the thyroid is said to exist in more than 50% of cases. Enlargement of the tongue is an important factor in diagnosis. The spleen has been found to be greatly enlarged in many cases, also the mesenteric, popliteal, axillary, and inguinal glands. Exophthalmic goiter may also be present, in which event heart failure under the anesthetic is probable. Congenital defects such as cleft palate and cleft kidney are sometimes associated with status lymphaticus. All patients have a pale, thin skin, pasty complexion, and usually subcutaneous fat. The glands of the neck are also sometimes enlarged. The above complex symptoms are noted when, given chloroform for any length of time, much of the anesthetic is absorbed and less secreted than is usual, with a consequent continual poisoning of the system until death occurs several days after the anesthetic. Sometimes delayed chloroform poisoning is mistaken for status lymphaticus. In status lymphaticus, especially in children, patients seem to dread the anesthetic more than is usually the case ..."
And Dr. Gwathmey offers a strong warning against using chloroform (p.333):
"From the study of a large number of statistics, the fact that chloroform is contraindicated cannot be questioned. Roberts concludes that ether is the safest anesthetic for all of these cases. Unquestionably, chloroform should be avoided in all suspected cases."
A Case Description from Hilliard (1908), as Cited by Dr. Gwathmey (p.334)
"Mortality. --- Harvey Hilliard gives a very complete history of a fatal case of status lymphaticus in a young man aged twenty-one, six feet two inches in height, very thin, and of a highly neurotic temperament. Operation: Circumcision." And:
"The patient was a great smoker of cigarettes and subject to attacks of faintness. The patient had the usual preparation, but was allowed to smoke during the morning, the operation being at twelve o'clock. Hilliard found on examination a rapid pulse, poor chest expansion, and considerable enlargement of the thyroid gland. Chloroform-ether mixture was the anesthetic. During the induction period, the heart beat very violently. A light anesthesia was maintained. When the prepuce was severed, the patient turned an ashen color and stopped breathing. Rhythmic tongue traction was employed and amyl nitrite vapor, the administrator pressing the lower ribs to restore respiration. This brought the patient round. The anesthetic was discontinued with the idea of discontinuing the operation, when the patient immediately stopped breathing. The usual restorative methods were resorted to, but proved unavailing. Artificial respiration was kept up for forty-five minutes, but the patient did not again come around."
An Alternative Hypothesis about Deaths during Anesthesia
Dr. Gwathmey appeared skeptical about some of the cases alleged to be deaths due to status lymphaticus. At p.336, he cites Yandell Henderson (in Surgery, Obstetrics, and Gynecology, August 1911), who felt that unskillful anesthesia is more often the cause of death, and especially in adenoid and tonsil cases, than the status lymphaticus or heart disease. According to Dr. Gwathmey, Dr. Henderson wrote in 1911:
"Writers assume that status lymphaticus was the cause of death, although there may have been no autopsy. Even in those cases in which an autopsy was performed, the pathologist's report sometimes indicates that if he had not been told what to find, he would scarcely have found it."
And Dr. Henderson predicted as follows (according to Gwathmey):
"In many of the very best text-books of pharmacology ... the practice of occasionally interrupting the administration of ether, and of allowing the patient to come for a few moments pretty well out of anesthesia, is expressly recommended. If anesthetists will only realize that this is a procedure which, above all others, should be shunned, the number of cases of so-called status lymphaticus fatalities, under anesthesia will, I believe, show a sudden and marked decrease."
We do not know how many anesthetists looked at Dr. Henderson's advice. But we do know that concern about sudden death in childhood, especially during anesthesia, caused decades of radiation screening and treatment for "Status Lymphaticus" and "enlarged thymus."
Differentiation of Thymus Disorders from Others ... with "Happy Results"
A number of writers (Dr. Henry Pancoast, 1930, in particular) have recorded their opinions that a variety of disorders in the thorax need differentiation from a possible enlarged thymus. Especially has this been true of bronchitis, bronchopneumonia, tuberculous adenitis, sinusitis with associated bronchitis, non-tuberculous lymphadenitis, and possibly other disorders.
Dr. C. Winfield Perkins is another who alluded to possible mis-identifications (1925, p.219): "Many sudden deaths of children formerly supposed to be due to bronchopneumonia [emphasis in original], congenital anomaly of the heart or acute intestinal indigestion, in which autopsies have given little information as to the cause of death, may have been due to unrecognized thymic hypertrophy. Such types of cases have recently been examined, considering the possibility of thymic enlargement, in spite of possible negative roentgen findings, [and have been] treated as such with the roentgen ray with happy results and the disappearance of the cyanosis and dyspnea."Part 4. The Belief That the Radiation Did No Harm
In Part 2, we featured two pertinent questions of Dr. O'Brien. The second question concerned prophylactic use of radiation therapy for "enlarged thymus." In closing the discussion at the 1929 presentation, Dr. O'Brien made a very strong statement about the safety of thymus irradiation (p.280):
"As to the danger connected with treating the thymus, Hammar has studied the thymus for twenty years and should know something about it. He says there is absolutely no danger from roentgen treatment. The cases he has examined show an emigration of lymphocytes which return rather promptly after the treatment has ceased. That emigration of lymphocytes accounts for the decrease in the shadow" [when it does occur, of course].
This is not the first time we have reported assurances that such roentgen treatments appeared to be harmless (see Index: Safety assurances). And in an earlier paper than O'Brien's, Dr. Roy M. Greenthal was arguing as follows (1922, p.438):
"On the other hand, it seems reasonable to give these patients the benefit of a treatment which we know will reduce the size of the thymus gland. We are aware of the marked reduction in the size of the thymus that can be secured when patients with thymic symptoms are exposed to therapeutic roentgen rays or to radium emanations. We know of no reports of harmful effects following this form of treatment of the thymus and we have never observed any in this clinic."
Possible long latency in development of radiation-induced cancer simply was not part of the discourse on the presumed harmlessness of roentgen-ray exposure in 1922. Nor did radiation-induction of cancer receive widespread attention for several additional decades.
Prophylactic Irradiation: "Not Only Desirable, but Requisite"
How did Dr. O'Brien answer his second question about screening for enlarged thymus and about prophylactic radiation therapy in symptom-free infants, children, and young adults?
"Since our second query concerns the enlarged thymus without symptoms in children, it is not necessary to consider here tracheobronchial adenitis or lymphosarcoma or thymoma except to say that any of these conditions sufficiently advanced to give a broadened mediastinal shadow would carry with them a very definite clinical as well as roentgenological picture." And Dr. O'Brien continued (p.276):
"I am therefore presuming that the 6 to 7 percent of cases of `broadened mediastinal shadow' seen in children without symptoms represent at least relatively enlarged thymus glands. No one who is informed thinks for a moment that all of this group represent pathological glands. This group comprises, undoubtedly, those which have not undergone accidental involution from disease, those in whom the rate of chest growth has not kept pace with the thymus, as well as those glands considered potentially a menace."
Dr. O'Brien's emphatic recommendation, about thymic irradiation prior to anesthesia, was tied to the presumed harmlessness of such irradiation (p.276):
"Since there is no evidence that the thymus is not an integral causative factor in the type of death under discussion, and since it is known that involution of the thymus takes place rapidly and without harm (Hammar) following roentgen or radium treatment, it would appear not only desirable but requisite, until such time as more exact knowledge or experience shall warrant a contrary opinion, to prescribe radiation therapy for those children presenting roentgen evidence of `broadened mediastinal shadow' without symptoms in whom general anesthesia or surgery is contemplated."Part 5. Quantitative Analysis of
the Massachusetts Eye and Ear Infirmary Data
Here, we shall evaluate the breast-dose received by children of ages about 1 to 9 years old, from thymic irradiation administered due to the fear of sudden death when such children had a variety of chest problems. This will become the entry for Column M of our Master Table. While some of the children had surgical procedures, we shall treat the cases of tonsillectomy and adenoidectomy separately in the next chapter.
We begin with the 1971 study by Janower and Miettenen entitled "Neoplasms after Childhood Irradiation of the Thymus Gland." The average age of the children in their study was 4.7 years old at the time of irradiation, as mentioned already in Part 1. Thus, these children were considerably older than the children evaluated in the previous chapter, about 90 percent of whom were under 6 months of age. The children in the Janower/Miettenen Study represent children with a variety of chest problems plus some head and neck problems, including bronchitis, lymphadenitis, and other disorders which brought them into the Massachusetts Eye and Ear Infirmary.
We shall use almost the same checklist of "items" used in the previous chapter.
- Item 1: What was the place of study? The Massachusetts Eye and Ear Infirmary in Boston, Massachusetts --- which is Suffolk County. This is a single well-defined facility in a stable location in which essential population information will be available.
- Item 2: Can we regard the study's participants as representative of Suffolk County for the relevant period? Unfortunately, we have no basis whatever for assuming that all the hospitals and private practitioners referred their patients to this one Infirmary. Nonetheless, we shall make our dose-estimate as if only the children in the Janower/Miettenen Study received such treatment in Suffolk County. For any children of this age-bracket who were treated in other institutions of Suffolk County, we assign zero dose.
This means that we shall definitely be underestimating the person-rads of breast-dose for Column M of our Master Table. But our intention is for doses in the Master Table to represent a credible lower limit of annual average breast-dose.
- Item 3: How many persons were treated? Over the 22 year period (1924-1946), there were 1,131 children treated for thymic enlargement.
- Item 4: What ages were represented in the treated group? Only the mean age was given in the report, a value of 4.7 years. We shall assume that the age-range of those irradiated was from 1 through 9 years of age. We shall assign equal numbers of children to each age-year of the 1-9 year age-range.
- Item 5: What was the period over which the treatments continued? The total period was about 22 years, starting in 1924 and ending in 1946.
Females of Each Age-Year Irradiated per Calendar-Year, Suffolk County.
- Item 6: We need to know how many children were present in each age-year for each year of the study. Since 1,131 children were irradiated in the course of 22 years, the number of children treated per year was (1,131 / 22), or 51.4 children per year. We shall make the approximation that one-half of the children were female, so there were 25.7 female children treated per year of the study.
If we divide this number equally into nine age-years, we arrive at (25.7 / 9), or 2.86 female children in each age-year who received therapeutic radiation at this Infirmary, in a calendar-year.
Total Females Ages 1-9, per Year, USA and Suffolk County
- Item 7: We need to know the total population of female children in each age-year (1-9) in Suffolk County for the average year in the period 1920-1960. We do this as we did it in Chapter 8 (Item 6).
In 1960, Suffolk County had 791,329 persons.
In 1960, the U.S. population was 179,333,000 persons.
The ratio, Suffolk / USA = (791,329 / 179,333,000).
In the Master Table, Column A, we have the number of females (age-year 1): (892,820 national) x (the ratio of 791,329 / 179,333,000) = 3,940 age-1 females.
And in this way, the full nine-year tabulation is built.
Age-Year Females, National: Females, Suffolk Cy.: Number in Age-Year Number in Age-Year 1 892,820 3,940 2 892,097 3,936 3 891,518 3,934 4 890,657 3,930 5 890,332 3,929 6 890,051 3,927 7 889,806 3,926 8 889,589 3,925 9 889,390 3,925
Breast-Dose per Treated Child
- Item 8: We need to know the radiation dose absorbed in this "enlarged thymus" therapy. The treated children generally had a cumulative air dosage of 400 Roentgens (4 doses of 100 R each), but we cannot be sure what the distribution was to the breasts. A phantom study, as done by Dr. Rosenstein for the Hildreth Study in Rochester, would have been desirable. Absent that, we shall assume the same irradiation techniques were used in Boston as in Rochester, and that the absorbed breast-dose was the same: About 32.62 rads per child, after adjusting for supra-linear bending of the dose-response curve (see Chapter 8, Item 8). We will use 32.6 rads, below. If most of the Boston children received 400 R (air dosage), our use of 32.6 breast-rads may underestimate the true dose, but this is not provable within the data.
Conversion of Individual Dose to Population-Dose
- Item 9: We need the average population-dose from this irradiation, rather than the raw individual doses per treatment. We calculated in Item 6 that there were 2.86 female children in each age-year who received therapeutic radiation in Suffolk County, in a calendar-year. We use the same two-step process shown in Chapter 8, Item 9. First, we obtain "person-rads." So for each age-year:
(2.86 persons) x (32.6 rads) = 93.2 person-rads.
The second step is to distribute these person-rads, received by only 2.86 persons, into all the children of the same age-year in Suffolk County. We illustrate with the age-1 group:
93.2 person-rads Population Exposure, rads = ----------------- = 0.02365 medical rads 3,940 females per breast-pair.
And we must do this calculation for all nine age-groups.
Age- Females in Person- Mean Population Dose, Year Suffolk County Rads medical rads per breast-pair 1 3,940 93.2 0.02365 2 3,936 93.2 0.02368 3 3,934 93.2 0.02369 4 3,930 93.2 0.02372 5 3,929 93.2 0.02372 6 3,927 93.2 0.02373 7 3,926 93.2 0.02374 8 3,925 93.2 0.02375 9 3,925 93.2 0.02375
- Item 10: We have considered, so far, only the dose received by those children chosen to get radiation therapy to the thymus gland. This does not tell us what dose was received for those who were roentgenographed, but who did not qualify for radiation therapy. The Janower paper indicates that the criterion used was a certain width of thymic shadow in an antero-posterior roentgenogram. If the Infirmary took just one film and never used diagnostic fluoroscopy, the diagnostic radiation dose to breasts --- distributed over the whole population of those ages --- could have been quite low.
We shall assign a zero dose for the "entrance exam" for this series. Of course, this underestimates the doses received, but we prefer an underestimate where there are no usable data.
- Item 11: Duration. Janower and Miettenen say explicitly that such treatments were given for the years 1924-1946. They do not explicitly say that no such therapy was given in any of the other years of 1920-1960. We are consistently trying to develop a credible lower limit of dose, so we will not assume treatments in all forty years. For the period before 1924, we will assume 2 years without activity. And for the period beyond 1924-1946, we will assume no activity during 7 years. So, we will approximate that treatment occurred for 31 years, and no treatment occurred at all for 9 years. Therefore, we adjust the average annual dose downwards, as follows:
((31 x 0.02368) + (9 x 0))/40 = 0.01835 medical rads for age-years 1 through 3, and
((31 x 0.02374) + (9 x 0))/40 = 0.01840 medical rads for age-years 4 through 9.
This completes the analytical work for this group, and we make nine entries for nine age-years into Column M in the Master Table.
Use of These Data As Typical for Nationwide Practice
There are sometimes regional differences in medical practice. In our study of the literature, we have looked for evidence of such differences with respect to all chapters of this work. So far, we have not found any reason to think that the Boston data used above were atypical. But if we just suppose that treatment for "enlarged thymus" was more popular in Boston than elsewhere, for age-years 1-9, we should still not worry about any overestimate of national breast-dose in Column M. Why not? Because of the undeniable underestimate discussed in Item 2, above.
# # # # #
1922: No Surgery Planned?
Reduce The Thymus Anyway.
In the debate as to whether infants and children with evidence of enlarged thymus should be irradiated, we have the following (at p.438):
"Because some patients with thymic hyperplasia go through operations or severe illnesses without trouble, does not mean that all will do so. We have no means of knowing beforehand who will be the fortunate ones. It would seem, therefore, that the reduction of an enlarged thymus is indicated in all patients before operation."
As for those not being considered for surgery, Dr. Greenthal concluded that the following should occur:
"How shall we treat nonoperative cases which show thymic enlargement? We have given all these patients roentgen-ray treatments in order to reduce the size of the gland. It is our belief that this, too, is a beneficial procedure.
Why, we ask, did Dr. Greenthal think it was beneficial? The answer centered on one statement:
"It has long been known that some patients with enlargement thymus react to illnesses in a violent manner."
Roy M. Greenthal, "The Incidence of Thymic Enlargement Without Symptoms in Infants and Children," American Journal of Diseases of Children Vol.24: 433-440. 1922.