|
 
  
 
|
 |
|
Clinical
Contributions
Kaiser
Permanente Medicine 50 Years Ago: Acute Rheumatic Fever in Adults
By Alvin L Sellers,
MD; Eugene B Levine, MD
Commentary by
W
Jeffrey Fessel, MD, FACP, FRCP
In the 22-month
period from March 1943, to January 1945, twenty patients with acute rheumatic
fever have been studied and treated at the Permanente Foundation Hospital.
All cases occurred among workers in the Richmond shipyards. The diagnostic
criteria used were those set forth recently by Jones.1 The
major criteria for diagnosis were: 1) Active carditis, as evidenced by
the development of electrocardiographic changes, cardiac enlargement,
significant cardiac murmurs (grade two or louder systolic murmurs, all
diastolic murmurs), congestive heart failure, or pericarditis. 2) Arthralgia.
3) Chorea. 4) Subcutaneous nodules, and 5) Recurrence of rheumatic fever.
The minor manifestations include: 1) Fever. 2) Abdominal pain. 3) Precordial
pain. 4) Rash. 5) Epistaxis. 6) Pulmonary findings. 7) Laboratory findings
of leucocytosis, elevated erythrocyte sedimentation rate, and microcytic
anemia. The minimum requirement for inclusion in this study was the presence
of one major manifestation with at least two of the minor manifestations.
The age
of our patients ranged from 17 to 39 years, the average age being 28.3
years. This average age is somewhat higher than those reported by Master2
and Coburn3 and is probably due to the more advanced average
age of the population served by this hospi
tal as compared with Navy personnel. There were 16 males and four females.
This is approximately the same as the ratio of males to females in the
employed population served by this hospital.
Sixteen
of the patients spent the greater part of their lives in rural mid-western
or southern communities. The remaining four patients were natives of California.
The length of stay in the Bay area prior to the onset of rheumatic fever
in the former group ranged from six months to 24 months, the average being
twelve months. It is of interest that of the 16 out-of-state patients
that developed acute rheumatic fever, none came from large urban centers.
These facts will be alluded to later in considering the possible etiologic
importance of streptococcus infection in rheumatic fever.
Eleven patients
(55%) entered the hospital with a history suggestive of previous rheumatic
fever in childhood or established cardiac murmurs, or both. Two of these
patients had no murmurs but gave histories of migratory polyarthritis,
repeated bouts of epistaxis, or previous diagnosis of "leaky heart."
Three of these patients had significant established cardiac murmurs but
no history suggestive of childhood rheumatic fever. Six of the patients
had both significant murmurs and a suggestive history of previous childhood
rheumatic fever.
Nine of
our 20 patients (45%) had neither a history of rheumatic fever nor significant
cardiac murmurs and might, therefore, represent new cases of acute rheumatic
fever. It is understood that the absence of a significant past history
of rheumatic fever does not mean that attacks of childhood rheumatic fever
did not occur. This question has recently been discussed by Master.2
Cohn and Lingg4 in a study of 3125 patients with rheumatic
cardiac disease followed from onset to death, state that 70 percent of
patients have acquired the disease before the age of fifteen. Sixty-six
percent of their patients with rheumatic cardiac disease had no recollection
of previous attacks of rheumatic fever. Direct transference of these statistics
and inferences drawn from them to our experience is hazardous as it appears
that they were dealing with a stable population while we treated a dislocated
group. It may well be that a larger number of our cases than ordinarily
expected actually represent new infections.
Eleven patients
(55%) were found to have had either a severe sore throat or "tonsillitis"
from one to four weeks prior to the onset of symptoms of rheumatic fever.
Two additional patients gave histories of upper respiratory infections
two weeks prior to the onset of symptoms. Hence 13 patients (65%) gave
histories of disease of the upper respiratory tract one to four weeks
prior to the onset of rheumatic fever symptoms. The etiologic agents in
these infections were not identified. Coburn5 has emphasized
the importance of streptococcus infections in the etiology of rheumatic
fever. Indeed, he states that sensitization to antigens produced by the
alpha streptococcus is the underlying factor in acute rheumatic fever.
He has observed that people who come from environments poor in streptococci
to congested areas rich in these organisms are very susceptible to infection.
Our data lend support to this observation for, as we have seen above,
80 percent of our patients have been dislocated from rural communities
to the congested environment of the Richmond-San Francisco Bay area. Coburn3
has stated that sulfonamide administration is of no value in the prophylaxis
of acute rheumatic fever once the upper respiratory infection has become
established and may even precipitate acute rheumatic fever in patients
that might not otherwise have suffered from the disease. Four of our patients
had been given sulfonamides during their precipitating upper respiratory
infection. Two of these patients had a past history of childhood rheumatic
fever. In view of its inefficacy and possible danger, it would appear
advisable to withhold sulfonamide therapy in upper respiratory infections,
particularly in persons with a past history suggestive of rheumatic fever.
Migratory
polyarthritis was an outstanding complaint in seventeen cases. All of
these patients showed involvement of one or more of the major weight-bearing
joints. The small joints of the hands or feet were involved in only five
cases. The joints were typically involved in migratory fashion and were
in most instances painful, hot and swollen. Joint symptoms subsided, as
a rule, within one week after the initiation of salicylate therapy and
bed rest.
Of the eleven
patients who entered the hospital without cardiac murmurs, six developed
significant murmurs while under observation. Five of the nine patients
who entered the hospital with significant murmurs developed changes in
the intensity of their murmurs or new murmurs. Thus eleven (55%) of the
patients developed significant new murmurs or showed an increase in the
intensity of already present murmurs while under observation.
Apical systolic
murmurs were heard most commonly but aortic systolic and diastolic murmurs
and mitral diastolic murmurs occurred frequently. In only two cases did
murmurs disappear once they had become of significant intensity.
Sinus tachycardia
occurred in 50 percent of cases and in all instances the heart rate became
normal during the hospital stay.
Erythema
nodosum was present on entry in two patients. It did not develop in any
patient during the period of hospitalization.
Seventeen
patients had temperatures over 100°F on entry. In all cases the temperature
fell to normal within one week after beginning salicylate therapy and
bed rest.
Chorea,
as a manifestation of rheumatic fever, is rarely seen after adolescence.
Levine6 has stated that chorea is never seen in adults over
20 years of age except in association with pregnancy. We have observed
one case in our series, that of a 19-year-old female who gave no history
of previous occurrence of chorea or rheumatic fever but had a grade three
apical systolic murmur on entry. This young woman had one child three
years old. She gave no history of prior or subsequent pregnancies.
Subcutaneous
nodules were not present in any of our cases.
Roentgenograms
of the chest were taken in 17 of the 20 patients and cardiac enlargement
was revealed in four. Two of these patients gave past histories of childhood
rheumatic fever. None of the four patients had significant cardiac murmurs.
Repeated roentgenograms taken on these four patients revealed no change
in the size of the heart during hospitalization.
Electrocardiographic
abnormalities, other than sinus tachycardia were presented by 13 patients
(65%). These included: extreme wandering of the pacemaker (two cases),
A-V nodal rhythm (one case), auricular fibrillation (one case), multiple
ventricular premature beats (two cases), incomplete A-V block (eight cases),
incomplete A-V block with dropped beats (two cases), Wendckebach type
(one case), complete A-V block (two cases), abnormally elevated ST segments
(one case), and T wave changes (two cases). Auriculo-ventricular block
of varying degree is clearly the most common electrocardiographic abnormality.
The electrocardiograms in ten of the 13 patients showing the above abnormalities
returned to normal during the hospital stay. Since the electrocardiographic
changes are often transient, tracings must be made at frequent intervals,
and it has been our practice to repeat electrocardiograms at least once
a week during hospitalization.
Wendkos
and Noll7 have recently reviewed the electrocardiographic changes
in 100 patients with polyarticular rheumatism and point out the value
of finding electrocardiographic changes in making the differential diagnosis
between rheumatoid arthritis and acute rheumatic fever. Their finding
of electrocardiographic changes in 70 percent of patients with acute rheumatic
fever agrees well with our experience.
The erythrocyte
sedimentation rate is the most valuable single laboratory aid in judging
the activity of the rheumatic process. Coburn and Kapp8 use
the sedimentation rate as a measure of the extent of inflammation in acute
rheumatic fever. Every one of our twenty patients showed an elevated sedimentation
on entry. The range varied from 14 millimeters to 128 millimeters per
hour (Westergren method).
A white
blood cell count of between 10,000 and 20,000 per cubic millimeter was
found in 19 of the 20 patients. The leucocyte count returned to normal
in all cases during the hospital stay.
Treatment
The
treatment of acute rheumatic fever in this hospital follows the principles
outlined by Coburn.3 Patients are put at absolute bed rest
as soon as the diagnosis is made. Sodium salicylate in enteric coated
tablets is given in dosage of 1.7 grams every four hours, day and night.
The erythrocyte sedimentation rate and white blood cell count are determined
weekly and the sodium salicylate is administered until the sedimentation
rate has been normal for one week. Salicylates are then discontinued.
At the end of one week without sodium salicylate the sedimentation rate
is again checked and if found to be normal the patient is allowed to sit
in a chair at his bedside for 15 minutes two times daily. If after one
week the sedimentation rate has become abnormally rapid the patient is
returned to strict bed rest and the initial treatment is resumed. If the
sedimentation rate is normal after one week of sitting at the bedside,
the patient is allowed bathroom privileges. If after one week of bathroom
privileges the sedimentation rate remains normal, the patient's activities
are increased and he is shortly released from the hospital to continue
his convalescence at home. If at any time signs of active rheumatic fever
return, strict bed rest and sodium salicylate therapy are reinstituted.
High caloric,
high vitamin diets are provided and supplemental vitamins are added, including
100 milligrams of ascorbic acid daily.9
Methyl salicylate
ointment, cotton batting, splints, heat and physiotherapy are used for
symptomatic treatment of joint pains and stiffness.
It is felt
that some sort of occupational therapy involving minimal activity is of
great value to the patient's morale during this usually protracted hospital
stay.
Sodium bicarbonate
was used in treatment of some of the earlier cases, but is no longer given
due to its property of lowering the serum salicylate level.10
Hypoprothrombinemia
due to salicylate therapy has been reported by Shapiro.11 We
routinely check prothrombin blood levels and have not found significant
depression of the prothrombin concentration in any of our patients. Solley12
has observed that the prothrombin levels in the blood frequently dropped
after the first few days of salicylate therapy and then slowly returned
to normal without the aid of vitamin K preparations and with continued
salicylate administration.
Patients
under treatment with ten grams of sodium salicylate daily complain of
tinnitus, decreased auditory acuity, dizziness, headache, nausea, and
occasionally vomiting. These symptoms are ordinarily not very troublesome
and disappear with the cessation of salicylate therapy. We have not had
to discontinue salicylate treatment for any of the above complaints.
Discussion
Rheumatic
fever as we have observed it in adults differs in none of its essentials
from the classical rheumatic fever of childhood. The age of the patient
suspected of having this disease is not of great importance when characteristic
or suspicious signs and symptoms are present. The greater frequency of
rheumatoid arthritis among adults, and the general feeling that acute
rheumatic fever is a disease of children will lead to error and missed
diagnoses unless a high index of suspicion is maintained.
The literature
prior to 1941 dealt primarily with the problems of childhood rheumatic
fever. However, with the mobilization of large numbers of men into the
armed forces, numerous reports appeared describing outbreaks of acute
rheumatic fever among army and navy personnel. Our study of rheumatic
fever was stimulated in large part by these later reports, and it was
found that the average age of our patients was even higher than those
reported from the navy. This experience emphasizes the fact that acute
rheumatic fever occurs commonly during the third and fourth decades.
As has been
already pointed out, an attack during this period of life may well be
the patient's initial episode. A history of previous rheumatic fever is
not essential to this diagnosis in adult patients.
It is thought
that patients who have their initial attack of rheumatic fever after the
age of 20 suffer less permanent heart damage than those who acquire their
disease in childhood. We have not followed our patients long enough at
the present time to be able to confirm or deny this statement. The electrocardiographic
changes are typically transient. Cardiac murmurs, once they become of
significant intensity, tend to remain. On the other hand, we have observed
the complete disappearance of a mitral diastolic murmur which was doubtless
that of relative mitral stenosis due to dilatation of the left ventricle.
In another patient the Graham Steele murmur of relative pulmonic insufficiency
disappeared completely.
The treatment
of acute rheumatic fever is still far from satisfactory. We feel that
sodium salicylate should be used as outlined above, and together with
bed rest is the best treatment available at the present time. There are
many patients who require two to three months of this regime to secure
a satisfactory result, and there are a few who do not respond at all to
intensive salicylate therapy. As Solley12 has pointed out,
it may be that the changes which occur in the endothelium of the arterioles
are irreversible in these patients before treatment is initiated and that
this limits or prevents any therapeutic effect.
Summary and Conclusions
- Twenty
cases of acute rheumatic fever have been observed in adults during a
twenty-two month period.
- Acute
rheumatic fever in adults does not differ in its clinical manifestation
from acute rheumatic fever of childhood.
- Nine
of our patients had neither history nor signs indicative of previous
attacks of rheumatic fever and might, therefore, represent new cases.
- The therapeutic
program followed at this hospital has been presented.
Bibliography
1. Jones TD. The Diagnosis of Rheumatic Fever. JAMA 1944 Oct 21; 126:481.
2. Master AM. The Problem of Rheumatic Fever in the Navy. Am Heart J 1944
May; 27:634.
3. Coburn AF. The Management of Navy Personnel with Rheumatic Fever. US
Naval Med Bull 1943 Sep; 41:1324.
4. Cohn AE; Lingg C. The Natural History of Rheumatic Cardiac Disease,
A Statistical Study. JAMA 1943 Jan 2; 121:1.
5. Coburn AF. Epidemiology of Streptococcus Hemolyticus Infections at
Naval Training Stations. US Naval Med Bull 1943 July; 41:1012.
6. Levine SA. Clinical Heart Disease (2d ed). WB Saunders Co. Philadelphia,
1942, p 31.
7. Wendkos M; Noll J. A Survey of Rheumatic Fever in a Large Station Hospital.
Med Clin of North Am 1944 Jan.
8. Coburn AF; Kapp EM. Blood Sedimentation Rate in Rheumatic Carditis.
J Clin Invest 1936 Nov; 15:715.
9. Reinhart J; Mettier S. Am J Path 1934 Jan; 10:61.
10. Smull K; Wegria R; Leland J. The Effect of Sodium Bicarbonate on the
Serum Salicylate Level (in Acute Rheumatic Fever). JAMA 1944 Aug 26; 125:1173.
11. Shapiro S. Studies on Prothrombin. JAMA 1944 June 24; 125:546.
12. Solley R. Paper Presented at the San Francisco Heart Association Meeting,
Oct 28, 1944.
|
Commentary
by W
Jeffrey Fessel, MD, FACP, FRCP
The
accompanying report by Sellers and Levine in the 1945 Permanente
Foundation Medical Bulletin makes for fascinating reading more
than a half century after its initial publication, because we may
compare what was known then with today's understanding of the disease.
Also of interest is that acute rheumatic fever is a disease no longer
seen in the developed countries of the Western world. Although Sellers
and Levine saw approximately one new case of rheumatic fever per
month, I doubt whether any reader who graduated from medical school
in the past 40 years has ever seen a patient with acute rheumatic
fever--yet the disease in its acute phase or in its eventual, chronic
manifestation as valvular heart disease was once responsible for
filling 10% of all hospital beds.
Today,
as I read the reprinted article by these two researchers, I am reminded
of the year 1955, when I was an intern in a hospital in England.
The 45-bed medical ward for which I was responsible would have contained,
at any one time, a half dozen patients with congestive heart failure
caused by chronic rheumatic valvular (mostly tight mitral stenosis)
heart disease. The only diuretic available at that time was a mercurial
agent, which was given intravenously two or three times weekly;
after a while, the patient became resistant to mercurial agents,
and we had to use Southey's tubes--silver tubes of approximately
18-gauge diameter with perforations along the shaft. Four of these
tubes would be inserted along the length of each edematous leg;
thin latex rubber tubes were attached to the needles, and these
tubes conducted the drained edema fluid to a garden bucket. That
procedure could drain a gallon of edema at a time and was repeated
two or three times weekly as needed.
I
arrived in the United States in 1957--just in time for the prerelease
of chlorothiazide, which revolutionized the treatment of congestive
heart failure. Because we were given stern warnings about the dangers
of overly rapid depletion of potassium, we drew blood daily to analyze
electrolyte levels! The first mitral valvulotomy--considered a medical
miracle--was performed in the late 1950s and resulted in deeper
understanding of the pulmonary vascular changes that underlie the
pulmonic hypertension seen in chronic mitral stenosis. After mitral
valvulotomy became an established procedure (but before the use
of valve replacement), fewer patients suffered end-stage pulmonary
hypertension and right-sided heart failure from tight mitral stenosis.
If
all this seems strange from today's perspective, it is yet stranger
that in those days, the medical wards were replete with patients
having the consequences of severe hypertension (also virtually never
seen today), for which the mainstays of treatment were Rauwolffia
alkaloids and a diet of only rice and honey (to provide calories
and no sodium)!
That
Sellers and Levine saw fit to expound upon probable streptococcal
involvement in the etiology of rheumatic fever--a fact today taken
for granted--is interesting. I am reminded, however, that not until
the mid-1950s was the streptococcal etiology of rheumatic (Sydenham's)
chorea established by long-term follow-up studies that related streptococcal
sore throats with the delayed onset of chorea--a delay as much as
several months long. How many physicians today have ever seen a
patient with rheumatic chorea? Even in those days, it was said that
before the diagnosis of chorea was made, children with this condition
were scolded for being fidgety. Aside from the rare Huntington's
disease, chorea today is mainly a manifestation of systemic lupus
erythematosus (SLE). In fact, SLE used to be exceptionally rare--I
saw one case in my entire medical school experience, whereas rheumatic
fever was common; today, the prevalence of these two diseases is
reversed, with SLE affecting 1 in 250 black women and 1 in 1000
white women.
At
one time, Streptococcus was held responsible for many diseases
of uncertain cause. In fact, the belief that Streptococcus causes
rheumatoid arthritis led to development of the serologic tests now
used to detect rheumatoid factor: these current serologic tests
were originally developed as streptococcal agglutinin reactions.
Of interest also is that rheumatoid arthritis was so named after
becoming differentiated from the polyarthritis of rheumatic fever.
However, so as not to make a complete nosologic break from rheumatic
fever, rheumatoid arthritis was labeled "rheumatic fever-like,"
ie, rheumatoid. A curious fact about rheumatic fever is that it
rarely affects children younger than four years of age, possibly
because of either immaturity of the immune mechanisms or, more likely,
lack of exposure to rheumatogenic strains of Streptococcus.
Recognition of this pattern showed that, generally, children under
age four years who had arthritis had juvenile rheumatoid arthritis
and not rheumatic fever.
Sellers
and Levine cite findings that assign no value to sulfonamide administration
in prophylaxis of acute rheumatic fever after the upper respiratory
infection becomes established. We now understand the reason for
this lack of effect: the antigenic stimulus has already triggered
the disease. Besides, sulfonamides are less effective antistreptococcal
agents than is penicillin. Sulfonamides were introduced into medicine
in 1936; in 1945, penicillin would have been available only to the
armed forces and only in doses of a few thousand units. My aunt
was a pathologist in those days and recalls recovering penicillin
from urine so that it might be reused in other patients!
Apical
systolic murmurs were heard commonly in the series of patients described
by Sellers and Levine. Not until the advent of sonocardiography
did the pansystolic murmur become recognized as the physical sign
of mitral regurgitation; and not until the mid-1950s was the late
apical systolic murmur recognized--after prolonged follow-up studies--as
the precursor of mitral regurgitation. Until then, the late apical
systolic murmur was regarded as benign. Also worth remembering is
that Barlow did not describe mitral valve prolapse as the click-murmur
syndrome until the mid-1960s; undoubtedly, many children with mitral
valve prolapse who happened to have minor joint pains or tonsillitis
were misdiagnosed as having rheumatic fever.
Misdiagnosing
acute rheumatic fever with carditis had serious consequences. As
is emphasized in the article by Sellers and Levine, absolute bedrest
was important for preventing major cardiac damage and meant use
of bedpans and spoon feeding for the first week; thereafter, the
regimen was gradually relaxed until normal activities were allowed--usually
six weeks later. Patients were allowed to sit out of bed for 15
minutes twice daily only after the erythrocyte sedimentation rate
(ESR) had been normal for one week--an event which might not happen
for three or four weeks. If the ESR again became fast, the patient
was returned to absolute bed rest. (One of my tasks as a house officer
on the rheumatic fever wards was to draw blood from each of the
15 or 20 patients once weekly and to set up the ESR in Westergren
tubes. We did this by mouth-pipetting; hepatitis viruses and HIV
were unknown.) Absolute bedrest was the prescribed regimen not only
for the carditis of rheumatic fever but also for acute myocardial
infarction with the result that as many patients died from pulmonary
embolism as from the myocardial infarction. An interesting observation--one
which has never been adequately explained--is that pulmonary embolism
was rare in the children with acute rheumatic carditis and also
in patients treated with bed rest for active pulmonary tuberculosis.
A revolutionary announcement came about 25 years ago: Patients who
have had heart attacks can safely be discharged from the hospital
after three weeks. Equally revolutionary was the subsequent announcement
that these patients may leave the hospital after only seven days;
and discharge after only two days is now proposed for some patients
with myocardial infarct. We may imagine the outrage that such suggestions
might have provoked in earlier years! Some children with rheumatic
carditis were in the hospital for six months because the ESR might
not yet have become normal. Sellers and Levine indicate that "some
sort of occupational therapy involving minimal activity is of great
value to the patient's morale during this usually protracted hospital
stay." The occupational therapist taught basket weaving, sewing,
and simple activities that helped pass the time; much later, the
occupational therapist's focus shifted to physical therapy for the
upper limbs.
The
statement of Sellers and Levine that "rheumatic fever as we
have observed it in adults differs in none of its essentials from
the classical rheumatic fever of childhood" would not have
been made ten years after these authors' article appeared. In fact,
although the cardiac manifestation in adults is usually slight,
effects on joints are much more severe and long-lasting. Indeed,
as is said proverbially, rheumatic fever in children "licks
the joints and bites the heart," whereas in adults it "licks
the heart and bites the joints."
For
this issue of The Permanente Journal, I was asked to provide
both a Commentary on the paper by Sellers and Levine and an update
on the subject of rheumatic fever. I have given a personal perspective
with some reminiscences that might interest younger physicians;
however, little update is available, because the disease is now
virtually unseen in the United States. (Rheumatic fever is still
seen in Latin America.) We understand the immunologic basis of the
carditis, which results from cross-antigenicity between epitopes
in the streptococcal cell wall and cardiac myocytes. We also understand
why rheumatic fever was rarely accompanied by poststreptococcal
nephritis: The nephritogenic strain of Streptococcus has
cross-antigenicity with glomerular epitopes but not with epitopes
in the cardiac myocyte; and the Streptococcus responsible
for rheumatic fever has cross-antigenicity with epitopes on cardiac
myocytes but not with glomerular epitopes. Treatment today would
differ substantially from that offered in 1945, when the mainstays
of treatment were strict bed rest for carditis and use of salicylates
for inflammation. Although an international study done in the 1950s
showed no benefit of cortisone over salicylate treatment, the dosages
of cortisone used (ie, 60 mg daily, tapered over six weeks) would
today be regarded as inadequate, and carditis would be treated with
high dosages (60 to 120 mg daily) of prednisone. Death from acute
rheumatic pancarditis (endocarditis with valvulitis plus myocarditis
plus pericarditis) should be rare, and chronic valve disease should
be less likely to occur. Fortunately, today's rarity of rheumatic
carditis means that we will not have the opportunity to conduct
a clinical trial of TNF-a antagonists in patients with rheumatic
carditis, although I suspect this treatment might be beneficial
because part of the myocardial damage could be mediated by cytokines.
Rheumatic chorea is certainly mediated by cytokines, because no
neuropathologic changes are associated with Sydenham's chorea.
In
conclusion, rereading Sellers and Levine's article reminds us of
a scourge that is, fortunately, no longer with us, at least in the
developed world. Their work also allows us to reflect on the undreamed-of
advances made since those days toward our understanding of both
disease pathogenesis and therapeutics.
To
top of page >>
|
To
Clinical Contibutions index >>
| To
next Clinical Contributions article >>
|
|
|
|
