There have been striking changes in the incidence of rheumatic fever in the United States over the past 50 years. Rheumatic fever was a serious health problem for young people during the 1930s and 1940s. This may be appreciated best by briefly sharing with you my own experience in 1946 when I worked for a year in a rheumatic fever convalescent hospital on the outskirts of New York City.
The hospital was filled with 90 children and adolescents, and there was even a waiting list. Almost all of the patients had heart disease and were convalescing from 6 months to 1 year after one or more attacks of acute rheumatic fever. This experience had a deep and lasting impression on me, and nurtured my interest in this disease over the last half century. There were a number of such convalescent hospitals near cities with large urban populations. For example, the House of Good Samaritan here in Boston cared for >2000 rheumatic fever patients between 1921 and 1970.
My associate, Dr Leon Gordis, conducted a study on the incidence of rheumatic fever in Baltimore City between 1960 and 1964. He found an incidence of 24 per 100 000—only a modest decline from earlier figures. He identified 261 patients with an attack of rheumatic fever during a 5-year period in the early 1960s. The majority of these patients lived in inner-city Baltimore. At that time, and for decades before that, rheumatic fever had almost always been concentrated in urban populations living in poor, crowded neighborhoods.
Dr Gordis repeated the rheumatic fever incidence survey in Baltimore for the years 1977 to 1981 and found a striking decline to 0.5 per 100 000 population, a remarkable 50-fold drop.1Incidence figures reported from other parts of the country were almost equally as low. The virtual disappearance of rheumatic fever in the United States seemed at hand. But it proved premature to write the epitaph for this disease.
Much to everyone's surprise, in 1987 a report appeared concerning an outbreak of rheumatic fever in the Salt Lake City area. Fifty patients with this disease were seen in 1985 in that area for a population-based incidence of 18 per 100 000 population.2
Over the next 5 years there were reports of nine outbreaks among children and two in military camps, all in widely separated parts of the country. Equally surprising was that the majority of the civilian patients involved in these outbreaks were among middle-class children living in suburban and rural areas. In contrast to our earlier experience, it would appear that no population is exempt from an attack of rheumatic fever under the right epidemiologic conditions.
Historically, streptococcal disease patterns, in terms of incidence and severity, have waxed and waned, much as had other infectious diseases. For example, during the last century there were times when outbreaks of scarlet fever had a very high death rate. The other times, however, scarlet fever was considered a relatively benign disease.3
The reasons for these cyclic changes in the past are unclear. Our experience with cyclic changes over the past 50 years has provided an opportunity to learn more about why these changes occur. Several possible reasons for the virtual disappearance of rheumatic fever in the 1970s include: a) changes in the host, b) improved living conditions, and c) changes in the pathogenic potential of streptococci.
Changes in the host in terms of immunity or genetics are not a likely explanation because such changes would only be expected to take place over generations, and this sharp decline occurred in a short period of time. I do believe improvement in access to health care for the high-risk population, and the widespread use of antibiotics played an important role.4 There is evidence for this conclusion from other studies that Dr Gordis and I performed in Baltimore.
In 1965 we established a federally funded, free clinic for children living in the high rheumatic fever incidence area of inner-city Baltimore. Comprehensive care was provided with a great deal of attention to the diagnosis and treatment of streptococcal infections. Several years after the clinic was in operation the incidence of rheumatic fever was again surveyed. There was a significant decline in the incidence of rheumatic fever in the population served by the clinic when compared with those who did not have access to the clinic.
We believe that it was the treatment of streptococcal infections that accounted for the differences in incidence. The decline in rheumatic fever occurred in patients with a history of a prior clinical infection, the only group that could have received treatment. A decline did not occur in the group without symptoms who obviously did not receive treatment.5
In subsequent years, with the introduction of Medicaid, disadvantaged children have had greater access to health care. As you know from your own practice experience, there has been a striking increase in physician visits by Medicaid recipients since 1967.
Although we believe that the recognition and treatment of streptococcal pharyngitis played a significant role in preventing rheumatic fever in this population, it could not have been the only reason. At least half of the children who have an attack of rheumatic fever give a history of no or few symptoms of preceding pharyngitis and, therefore, would have no reason to seek care. Yet, rheumatic fever was practically nonexistent in the 1970s in this group as well. This can only be explained by some unknown change in the biology of the streptococcus. The nature of this change is unclear. It is possible, however, that the streptococcus may have become less virulent during the period because antibiotics interfered with person-to-person transmission which, when it does occur, tends to increase virulence.
Let us next turn to some of the possible reasons for the resurgence of rheumatic fever in the mid-1980s. There was about a 15-year period when pediatricians-in-training almost never encountered a patient with rheumatic fever. Complacency with respect to following recommendations for the diagnosis and treatment of suspected streptococcal pharyngitis may have occurred, but this is unlikely to have played a significant part in the resurgence of rheumatic fever.
Secondly, have group A streptococci become resistant to penicillin? Studies of streptococcal strains isolated from clinical infections over the past 50 years have shown that group A streptococci have remained exquisitely sensitive to penicillin. Yet, there are those who continue to claim that penicillin is less effective than in the past. These claims are based on reports of an apparent increase in treatment failures in recent years after penicillin therapy.
We reviewed all the studies using oral penicillin for streptococcal pharyngitis published in the United States over the past 40 years that met certain strict, methodologic criteria. We compared the results during the earlier years (1953 to 1979) to the more recent ones (1980 to 1993). We were unable to document a significant increase in treatment failures.6 Treatment failures as high as 30% were reported in some studies, because of the inclusion of large numbers of asymptomatic carriers in the study population. Penicillin has never been effective for the treatment of carriers and, in most cases, carriers do not need to be treated.
I wish to emphasize that penicillin remains the drug of choice for patients with acute streptococcal upper respiratory infections, and pediatricians should not be taken in by those who tout the newest and very expensive super antibiotic.
Because appropriate treatment with penicillin of the streptococcal infections previously described can prevent the majority of first attacks of rheumatic fever, it is not unreasonable to ask why the rheumatic fever attacks during these outbreaks were not prevented? Three-quarters of the 173 patients for whom we have data from these outbreaks, had few or no symptoms and, therefore, had no reason to seek medical care. About 20% were treated, but there is insufficient information available on the antibiotics prescribed, the length of treatment, or compliance. Some may have been treated appropriately and still developed rheumatic fever because a small number of failures have been known to occur.
There is good evidence that the reappearance of rheumatogenic strains and an increase in the virulence of group A streptococci account for many of the changes in streptococcal epidemiology during the latter half of the 1980s. A Centers for Disease Control and Prevention study shows an increase in streptococcal serotypes type 1, 3, and 18 during this period, which are strains known to have triggered attacks of rheumatic fever in the past.7 These serotypes were isolated from a few of the rheumatic fever patients and from some of their siblings during the outbreaks. These findings have strengthened the concept of the existence of rheumatogenic streptococci; namely, that a limited number of serotypes have the propensity to trigger an attack of rheumatic fever.
During the years of resurgence of rheumatic fever, there was also an increase in streptococcal invasive disease, as well as the occurrence of streptococcal toxic shock syndrome, which has a 30% mortality. These illnesses are because of an increase in the virulence of streptococci. Streptococcal pyogenic exotoxin is a virulence marker, and strains isolated from patients with these serious illnesses have been shown to have an increased production of these toxins.
There has been a striking increase in the incidence of streptococcal invasive disease both in children and in adults.8 The incidence of rheumatic fever has fallen to low levels once again, but it remains very prevalent in developing countries. However, there continue to be reports of serious infections caused by streptococci in our country and around the world.
Finally, what can we look forward to in the future? There are some obvious limitations in our ability to completely eradicate rheumatic fever with antibiotics. A vaccine would be the ultimate preventive measure. Time does not permit going into details regarding the status of the streptococcal vaccine. There has been progress in purifying and characterizing streptococcal M protein, the antigen that elicits antibodies that confer immunity to the streptococcus.3However, the availability of a vaccine is not imminent.
Progress has also been made in identifying a genetic marker for rheumatic fever susceptibility. If and when a streptococcal vaccine does become available, the genetic marker could be useful to identify susceptible individuals in families with a history of rheumatic fever.
The likelihood of a highly sensitive rapid strep test is of more immediate relevance to practitioners. The availability of such a test would promote even earlier detection and even more prompt treatment, which would shorten the course of the illness and lessen the chances of spread to other children.9
Unfortunately, neither the traditional throat culture nor the rapid strep test can distinguish between the carrier state and infection. I began doing office cultures in my pediatric practice in 1951, 1 year after it was shown that adequate treatment of a strep sore could prevent rheumatic fever and heart disease. Now, 45 years later, bacteriologic confirmation in patients suspected clinically of having a strep infection is even more important to curb the inappropriate use of antibiotics for viral pharyngitis. This has contributed to the growing problem of antibiotic resistance as has already occurred with the pneumococcus. The Academy and its members have to do everything they can to support legislation currently pending in Congress to exempt tests that detect streptococcal infections from the Clinical Laboratories Improvement Act regulations.
- Copyright © 1998 American Academy of Pediatrics