S89
Recent Evaluation of Antimicrobial Resistance in /3-Hemolytic Streptococci
Edward L. Kaplan
From the World Health Organization Collaborating Center for
Reference and Research on Streptococci, and the Department of
Pediatrics, University of Minnesota Medical School,
Minneapolis, Minnesota
Although antimicrobial resistance among bacteria continues to increase and to be a clinical
problem, the /3-hemolytic streptococci have remained remarkably susceptible to most antibiotics.
For example, there has not been a single documented instance of a clinical isolate of a penicillinresistant group A streptococcus. Moreover, available data indicate that the minimal inhibitory
concentrations (MICs) of penicillin for group A streptococci have not changed during the past 4
decades. In one study, the MIC 90 for more than 300 strains of group A streptococci was only 0.012
iLg/mL. Resistance to the macrolide antibiotics, the tetracyclines, and the sulfa drugs remains more
clinically important. Outbreaks of macrolide resistance have been documented in several parts of
the world, but macrolide resistance in most countries of the world remains at <5% among group
A streptococci. Despite the fact that clinically significant antibiotic resistance has not emerged,
careful surveillance is required.
Two recent events have focused considerable attention on
the possibility of antimicrobial resistance in /3-hemolytic streptococci, especially those of Lancefield group A. First, there
has been a resurgence of serious infections due to group A
streptococci and their suppurative and nonsuppurative complications [1]. Outbreaks of acute rheumatic fever in middle class
populations of schoolchildren and in military recruits in North
America since the mid-1980s as well as the apparently increased incidence of serious systemic infections due to group
A streptococci, including streptococcal toxic shock syndrome,
have raised concerns regarding antimicrobial resistance.
Second, of the bacteria commonly responsible for acute pyogenic infections, only /3-hemolytic streptococci have appeared
to retain considerable susceptibility to both new and established
antibiotics. Staphylococci, Haemophilus influenzae, and Streptococcus pneumoniae are three examples of epidemiologically
and clinically important organisms that have changed in this
regard.
A second group of 0-hemolytic streptococci of considerable
interest has been the group B streptococci (e.g., Streptococcus
agalactiae), which remain considerably important causes of
neonatal sepsis, meningitis, and maternal infections. Although
0-hemolytic streptococci of Lancefield groups C and G are
less often associated with human infections, these organisms
cause lethal infections such as meningitis and even endocarditis.
In this article, the antimicrobial susceptibilities of 41-hemolytic streptococci are briefly reviewed and significant changes
Reprints or correspondence: Dr. Edward L. Kaplan, Department of Pediatrics, Division of Infectious Diseases, University of Minnesota, Box 296, 420
Delaware Street Southeast, Minneapolis, Minnesota 55455.
Clinical Infectious Diseases 1997; 24(Suppl 1):S89--92
© 1997 by The University of Chicago. All rights reserved.
1058-4838/97/2401 —0042$02.00
in antimicrobial resistance in wild strains of these organisms
that are capable of causing clinical infections are determined
and noted.
Group A /3-Hemolytic Streptococci
The early optimism that infections due to group A streptococci (e.g., Streptococcus pyogenes) could be controlled with
the sulfa drugs soon turned to disappointment, as resistance
developed in these bacteria over a very short period; therefore,
the sulfa drugs were of no use as therapy for these infections
[2]. (It is of interest that while the sulfonamides have been
ineffective as treatment of established pharyngitis due to group
A streptococci, they remain very effective as secondary prophylaxis for colonization of the upper respiratory tract with group
A streptococci in patients who have had rheumatic fever.)
The early enthusiasm for sulfa drugs can be understood when
it is recognized that, in the preantibiotic era, mortality rates of
up to 33% were associated with the relatively frequent outbreaks of scarlet fever in children. Epidemiological data from
the United Kingdom at the close of the last century documented
the frequency with which these epidemics occurred.
The accounting of how the introduction of penicillin into
clinical medicine changed the clinical and public health approach to group A streptococcus infections is well known. Not
only were these organisms exquisitely susceptible to penicillin,
but cure rates of >90% were frequently reported [3]. Whether
penicillin was used as primary prophylaxis for rheumatic fever
(e.g., following treatment of group A streptococcus pharyngitis)
or as prophylaxis for recurrent attacks (secondary prevention),
the results were uniformly excellent. Treatment failures (i.e.,
failure to eradicate the organism from the upper respiratory
tract) were relatively rare.
In the late 1970s, however, investigators became concerned
when reports indicating that there were a significant number
S90
Kaplan
of instances of failure to eradicate the organism from the upper
respiratory tract began to surface [3]. Bacteriologic failure rates
as high as 25%-30% were reported during the next several
years [4]. These incompletely explained observations were
made just before and at the time when the current resurgence
of group A streptococcus infections and their sequelae was
beginning to be observed. If this occurrence was not sufficient
to cause alarm among clinicians and patients alike, occasional
insufficiently documented (and quite incorrect) statements reporting the isolation of supposed penicillin-resistant group A
streptococci did [5]. As far as is known, there has never been
a penicillin-resistant group A streptococcus isolated from a
clinical source.
For completeness, it also must be noted that mutants resulting from treatment with ethyl methanesulfonate have been
produced in the laboratory [6], but these organisms have no
clinical relevance. The mechanism(s) by which these mutants
have been produced is not completely explained, but it has
been thought that the mechanism(s) might be related to an
effect on the penicillin-binding proteins. (A thorough review
of antibiotic resistance in group A streptococci has recently
been completed by Gerber [7]; the reader is referred to this
review for more details of antibiotic resistance in this serological group of /3-hemolytic streptococci.)
The issue of whether penicillin tolerance has any clinical
relevance should also be mentioned. Although penicillin tolerance has been demonstrated in studies of clinical isolates from
several countries around the world [8], the fact that numerous
laboratory variables may influence the MBCs makes interpretation of the results difficult. At the present time, the consensus
is that tolerance does not have clinical relevance with respect
to group A streptococci.
The confusion associated with possible antimicrobial resistance was enhanced by the inability to completely understand
the persistence of the organisms in the upper respiratory tract
after penicillin therapy. In vitro laboratory observations failed
to ever show a truly resistant group A streptococcus. Other
theories explaining this persistence after therapy were often
inadequately substantiated. The presence of bacteriocins produced by normal oral and/or upper respiratory tract flora, the
presence of normal oral flora producing /3-lactamase, and tolerance to penicillin were among the postulated but unproven
explanations offered.
Studies indicating that it was more difficult for penicillin to
eradicate the organism from the upper respiratory tract of true
streptococcal "carriers" were reported [9]. Review of the literature suggested that many of the reported bacteriologic failures
occurred in carriers. Yet, concern remained among clinicians.
The introduction of new antibiotics for treatment of group A
streptococcus pharyngitis was one result.
Because no one had examined the issue of possible antimicrobial resistance in the laboratory in detail during the past
several decades and because of the introduction of these newer
antimicrobial agents to clinical medicine, we examined >300
CID 1997;24 (Suppl 1)
Figure 1. MICs of 11 antibiotics for group A streptococci. Data
adapted from [10].
clinical isolates of group A /3-hemolytic streptococci recovered
from patients in the United States during the period from 1989
to 1992 [10]. The isolates studied were obtained from patients
with uncomplicated pharyngitis who lived in >31 states of the
United States. Included among the isolates were >20 different
distinct group A serotypes.
In addition, 43 isolates from patients with severe systemic
streptococcal infections were studied to explore the possibility
that these organisms were "more resistant" than those from
patients with uncomplicated pharyngitis. The MICs for individual strains were determined by serial dilutions of the antibiotics [10].
Eleven orally administered antibiotics were studied; these
agents were penicillin G; tetracycline; three macrolide antibiotics (erythromycin, azithromycin, and clarithromycin); clindamycin; representatives of the first, second, and third generation of cephalosporins (cephalothin, cefixime, cefaclor, and
cefpodoxime); and one representative of the quinolone family
(ciprofloxacin). The MIC 90 for each antibiotic are shown in
figure 1. Of importance, the MIC 90 of penicillin G was 0.012
11,g/ mL , thus indicating that it is extremely unlikely that there
has been a change in the in vitro effectiveness of penicillin
during the past three or four decades.
Because of the reports of a high proportion of erythromycin
resistance from Japan and other countries in Asia during the
1960s and 1970s [11] and because of the more recent data
from Finland (in some areas of the country the percent of
isolates resistant to erythromycin was >30%) that revealed
a similar finding [12], the MICs of the macrolides were of
considerable interest. Less than 5% of the group A streptococcus isolates studied were "resistant" to erythromycin.
Of interest, the MICs of the other two macrolides tested
were relatively high for all isolates that were resistant to erythromycin. However, none of these isolates demonstrated in vitro
resistance to clindamycin, even though rates of clindamycin
resistance of up to almost 5% have been reported among different populations during approximately the same period. How-
CID 1997;24 (Suppl 1)
Antimicrobial Resistance in 0-Hemolytic Streptococci
ever, strains for which tests reveal resistance to erythromycin
but not to clindamycin may mutate and become constitutively
resistant to all of the macrolides [13].
A resistance rate of <5% to macrolides is common in the
United States and in several other countries. On occasion, we
have found as high as 10% of recovered group A streptococci
demonstrating erythromycin resistance, but this finding has
been limited to small, specific populations. On the basis of this
admittedly limited sample size, it is unlikely that macrolide
resistance is a significant clinical issue at the present time in
North America and most parts of the world.
Previous reports have suggested that erythromycin resistance
has often been associated with specific serotypes, especially
isolates with T agglutination pattern 12. In a recent study [10],
this association was not the case. My colleague and I were
unable to associate any specific serotypes or T agglutination
patterns with macrolide resistance. Whether our finding is reflective of the entire population of group A streptococci cannot
be determined unless a much larger sample is investigated.
There was little, if any, difference in the range of MICs for
these 282 pharyngeal isolates of group A streptococci and the
43 isolates from patients with invasive group A streptococcus
infections, thereby strongly suggesting that the virulent clinical
course for these infections is unrelated to antimicrobial resistance.
The MICs of the representative of the quinolone family,
ciprofloxacin, suggested that it should not be used as therapy
for group A streptococcus infections. While ciprofloxacin is
obviously not a first line antibiotic for this purpose, it was
included in our study because my colleagues and I thought that
its use is more frequent than is recognized.
Ten percent of the group A streptococci described in the
recent study [10] were resistant to tetracycline. While this rate
is considerably less than the rates of 25% to even 90% that
were reported several decades ago, it does indicate that this
antibiotic class still should not be used as treatment of group
A streptococcal infections. This finding is important because
tetracycline still is frequently used by primary care physicians
in many countries as therapy for pharyngitis.
While there were differences in the MIC K, values of the
several different cephalosporins studied and while these differences suggested an advantage for some specific antibiotics, no
clinically significant resistance was observed.
Thus, while there have been relatively rare outbreaks of
group A streptococcus infections associated with antibiotic resistance during the past few decades, clinical problems associated with finding appropriate antimicrobial agents as treatment
of group A streptococcus infections have not occurred. The
available data suggest that many organisms have demonstrated
resistance because of the selection effects of specific antibiotics
that are used widely or inappropriately, as has been the case
with macrolides, tetracyclines, and sulfa drugs. Because of the
possibility of drug resistance, it is prudent that appropriate
surveillance programs monitor representative isolates of group
A f3-hemolytic streptococci.
S91
Group B f3-Hemolytic Streptococci
Clinically, /3-hemolytic streptococci of Lancefield group B
(e.g., S. agalactiae) are the second most important serological
group of f3-hemolytic streptococci. These organisms remain a
major cause of serious and life-threatening infections in neonates but also have the capacity to cause serious infections in
adults.
There are far fewer reports of studies of antimicrobial resistance in group B streptococci. In one study by Berkowitz and
colleagues [14] in 1990, the patterns of resistance in 156 cervical and vaginal isolates of group B streptococci to penicillin,
ampicillin, clindamycin, cefoxitin, gentamicin, and erythromycin were examined. While a high rate of gentamicin resistance
(>90%) was noted, resistance or what was called "intermediate susceptibility" to erythromycin, clindamycin, and cefoxitin
was demonstrated in 10%-15% of the isolates. No resistance
to either penicillin or ampicillin was reported in that study.
However, because —20% of the recovered organisms were
resistant to multiple antibiotics, the investigators recommended
continuing surveillance.
A more recent evaluation of penicillin resistance and tolerance in group B streptococci has been reported from Spain
[15]. The investigators reported that 2% and 8% of 100 strains
had "intermediate sensitivity" to penicillin and to ampicillin,
respectively. Seventeen percent of the strains were reported to
show tolerance to penicillin, but the clinical relevance was not
convincing.
Historically, a relative paucity of information about the antimicrobial susceptibilities in human isolates of group B streptococci has been reported worldwide. In fact, the magnitude of
the problem of group B streptococcus infections in neonates
has been only superficially addressed in many countries of the
world. Because of the frequency with which group B streptococcus is found when looked for and because of the significant
morbidity and mortality associated with it, considerably more
attention is required.
Other Serological Groups of /3-Hemolytic Streptococci
The two other serological groups of /3-hemolytic streptococci
that are most frequently associated with infection in humans
are 0-hemolytic streptococci of Lancefield groups C and G.
There have been studies that indicate that resistance to frequently used antibiotics can be induced in these groups, but
significant antimicrobial resistance in clinical isolates has not
been reported. However, studies have suggested that the MICs
of penicillin for these organisms, as well as those for Lancefield
group B, tend to be higher than those for group A streptococci.
Summary
In contrast to antimicrobial resistance in many of the other
important pyogenic bacteria, antimicrobial resistance in
S92
Kaplan
0-hemolytic streptococci has not yet constituted a major problem for either clinicians or public health authorities up to the
mid-1990s. More data about antibiotic susceptibilities in group
A streptococci appear to be known, probably because more of
these studies have been carried out. While these organisms
have clearly demonstrated both in the laboratory and in the
clinical setting that they have the capacity to become resistant to some antimicrobial agents, clinical and epidemiological
consequences have been relatively minor to date. It appears
prudent that active surveillance of the /3-hemolytic streptococci
for antibiotic resistance be implemented since there are no
currently effective vaccines or other methods for controlling
the spread of infections due to these virulent organisms.
References
1. Kaplan EL. Global assessment of rheumatic fever and rheumatic heart
disease at the close of the century. The influences and the dynamics of
population and pathogens: a failure to realize prevention? Circulation
1993; 88:1964-72.
2. Stevens DL, Gibbons AE, Bergstrom R, Winn V. The Eagle effect revisited: efficacy of clindamycin, erythromycin, and penicillin in the treatment of streptococcal myositis. J Infect Dis 1988; 158:23-8.
3. Kaplan EL. Benzathine penicillin G for treatment of group A streptococcal
pharyngitis: a reappraisal in 1985. Pediatr Infect Dis 1985; 4:592-6.
4. Smith TD, Huskins WC, Kim KS, et al. Efficacy of a beta-lactamase
resistant penicillin and influence of penicillin tolerance in eradicating
streptococci from the pharynx after failure of penicillin therapy for
group A streptococcal pharyngitis. J Pediatr 1987; 110:777-82.
CID 1997;24 (Suppl 1)
5. Still JG, Hubbard WC, Poole JM, Sheaffer CI, Chartrand S, Jacobs R.
Comparison of clarithromycin and penicillin VK suspensions in the
treatment of children with streptococcal pharyngitis and review of currently available alternative antibiotic therapies. Pediatr Infect Dis J
1993; 12(suppl no 3):5134-41.
6. Gutmann L, Tomasz A. Penicillin-resistant and penicillin-tolerant mutants
of group A streptococci. Antimicrob Agents Chemother 1982; 22:
128-36.
7. Gerber MA. Antibiotic resistance in group A streptococci. Pediatr Clin
North Am 1995;42:539-51.
8. Kim KS, Kaplan EL. Association of penicillin tolerance with failure to
eradicate group A streptococci from patients with pharyngitis. J Pediatr
1985; 107:681-4.
9. Kaplan EL, Gastanaduy AS, Huwe BB. The role of the carrier in treatment
failures after antibiotic for group A streptococci in the upper respiratory
tract. J Lab Clin Med 1981; 98:326-35.
10. Coonan KM, Kaplan EL. In vitro susceptibility of recent North American
group A streptococcal isolates to eleven oral antibiotics. Pediatr Infect
Dis J 1994; 13:630-5.
11. Nakae M, Murai T, Kaneko Y, Mitsuhashi S. Drug resistance in Streptococcus pyogenes isolated in Japan. Antimicrob Agents Chemother 1977;
12:427-8.
12. Seppala H, Nissinen A, Jarvinen H, et al. Resistance to erythromycin in
group A streptococci. N Engl J Med 1992; 326:292-7.
13. Weisblum B. Insights into erythromycin action from studies of its activity
as inducer of resistance. Antimicrob Agents Chemother 1995; 39:
797-805.
14. Berkowitz K, Regan JA, Greenberg E. Antibiotic resistance patterns of
group B streptococci in pregnant women. J Clin Microbiol 1990; 28:
5-7.
15. Betriu C, Gomez M, Sanchez A, et al. Antibiotic resistance and penicillin
tolerance in clinical isolates of group B streptococci. Antimicrob Agents
Chemother 1994; 38:2183 -6.