Department of Pathology, Case Western Reserve
University and University Hospitals of Cleveland, Cleveland, Ohio
44106,1 and Department of Pathology,
Hershey Medical Center, Hershey, Pennsylvania
170332
Received 20 August 1998/Returned for modification 31 January
1999/Accepted 25 May 1999
 |
INTRODUCTION |
The ubiquitous pathogens
Streptococcus pneumoniae and Haemophilus
influenzae cause a wide spectrum of pediatric and adult infections, including acute otitis media, sinusitis, acute
exacerbations of chronic bronchitis, pneumonia, bacteremia, and
meningitis. The advent and widespread use of the protein-conjugated
type b capsular polysaccharide H. influenzae vaccine has
largely eliminated the risk of life-threatening infections due to
encapsulated type b strains (5), but localized infections
caused by nonencapsulated H. influenzae strains remain
common. Antimicrobial resistance has emerged in both H. influenzae and S. pneumoniae, and effective patient
management requires physicians to be aware of the patterns and clinical
significance of antibiotic resistance in these pathogens. This
knowledge is gained, in large measure, from periodic systematic epidemiological surveillance studies.
-Lactamase-mediated ampicillin resistance in H. influenzae, which first emerged in the United States in 1974 (26, 41), has evolved into an obstacle to effective
treatment with older -lactams. Its prevalence has risen steadily,
from 16% in 1986 (14) to 33% in 1993 (33) and
36% in 1994 and 1995 (25). Notably, the highest frequency
of -lactamase production
45.8%has been documented in children
aged younger than 5 years, the same patient population that experiences
the peak incidence of H. influenzae infection, primarily
otitis media (25). Additionally, investigators have shown
higher MICs of ampicillin, amoxicillin, cefaclor, loracarbef, and
cefprozil for -lactamase-producing strains (12, 25, 40). Far less common is non--lactamase-mediated ampicillin resistance (24). -Lactamase-negative, ampicillin-resistant strains
are thought to possess altered penicillin-binding proteins that have decreased affinity for many -lactam agents. However,
spheroplast-producing strains also appear resistant to ampicillin and
other -lactams as a result of the high osmolality of media which
currently are used routinely, but there is no evidence that
spheroplast-producing strains are resistant in vivo (24, 34,
39).
Antibiotic-resistant strains of S. pneumoniae have been
reported on all continents in the 2 decades since resistant
pneumococcal strains were identified in the United States
(23). In some regions, drug-resistant S. pneumoniae strains predominate, with numerous strains resistant to
multiple agents (23, 40). Changes in the affinity of
penicillin-binding proteins, which is chromosomally mediated, result in
-lactam resistance. Penicillin-resistant strains (penicillin MICs of
2 µg/ml) are common in France, Spain, Romania, Japan, Korea, and
Taiwan and are being identified with increasing frequency in the United
States (23). Early in this decade, clinical failure of
treatment for patients with meningitis led to the detection of S. pneumoniae strains highly resistant to expanded-spectrum
cephalosporins (6, 38). The MICs of these cephalosporins
exceeded those of penicillin, in contrast to previous experience, which
showed the MICs of these cephalosporins to be 1 to 2 doubling dilutions
lower than that of penicillin.
Current clinical practice generally involves treating
community-acquired respiratory tract infections empirically. In 1997, over 130 million courses of antibiotics were prescribed for outpatient treatment of such infections (including pharyngitis), with almost 90 million courses being for adults and almost 42 million being for
children (31). Twenty-one million antibiotic prescriptions were issued for otitis media in children. As empiric antibiotics should
be active against both S. pneumoniae and H. influenzae for respiratory tract infections other than
pharyngitis, periodic surveillance of respiratory tract isolates for
changes in the susceptibility patterns of these pathogens is therefore
essential to the effective management of community-acquired S. pneumoniae and H. influenzae infections (17,
20). Accordingly, the present study sought to characterize
current levels of resistance in S. pneumoniae and H. influenzae to 10 oral antimicrobial agents by evaluating the
susceptibility of isolates from outpatients with community-acquired
infections in the United States.
As current susceptibility breakpoints for many oral antimicrobial
agents no longer correspond to more recent clinical, microbiological, pharmacokinetic, and investigational experience (8-11),
investigators have proposed a new approach based on
pharmacokinetic-pharmacodynamic (PK/PD) modeling and on clinical
studies that have measured bacteriologic outcome and evaluated this in
relation to drug susceptibilities (7, 9, 11, 18, 19). The
activity of -lactams and macrolides has been shown to depend on the
time the drug concentration in serum exceeds the MIC of the agent, with
clinical success occurring in more than 80% of the cases in which the
concentration of the agent exceeds the MIC for an infecting strain for
more than 40 to 50% of the dosing interval (8). Using
standard dosing regimens and the serum pharmacokinetics of these
agents, the concentrations in serum that are maintained for at least 40 to 50% of the dosing interval can be determined and used as PK/PD
breakpoints. Different PK/PD parameters correlate with clinical outcome
with fluoroquinolones and with azalides such as azithromycin, and
breakpoints can be derived from one of two ratiosthe ratio of the
peak concentration in serum to the MIC or the ratio of the area under
the 24-h serum concentration-time curve (AUC) to the MIC (8,
29). Clinical cure correlates best when the AUC/MIC ratio exceeds
25 for these agents in immunocompetent-animal models (8),
and MIC breakpoints for susceptibility can therefore be derived from
the formula AUC/25.
| |
MATERIALS AND METHODS |
Study centers.
Between January and December 1997, strains of
H. influenzae and S. pneumoniae isolated from
outpatients were collected by eight large regional commercial
laboratories from upper and lower respiratory tract sources and other
sources such as blood. Patients whose samples were submitted lived in
communities representative of six major regions of the United States.
Isolates from hospitalized patients, duplicate patient isolates, and
isolates referred by other laboratories were excluded. After isolation
by the commercial laboratories, strains were frozen at 
70°C and
transported to reference laboratories at Case Western Reserve
University, Cleveland, Ohio (M.R.J.), and the Hershey Medical Center,
Hershey, Pa. (P.C.A.). The demographical information submitted for each
strain included patient age and gender, specimen collection date,
specimen source, and the state and zip code of the submitting
physician. The reference laboratories checked the compliance of the
isolates with the criteria for inclusion in the study and excluded
isolates not meeting these criteria.
Identification of isolates.
The reference laboratories
confirmed the identity and purity of every strain. S. pneumoniae isolates were confirmed by inhibition by optochin and
positive bile solubility tests. Haemophilus strains were
identified by X and V factor requirement and reaction to polyvalent
antiserum (Difco Laboratories, Detroit, Mich.), and strains were
confirmed as H. influenzae by determination of the requirement for both the X and V factors. Only untypeable or non-type b
strains were included in the study.
Susceptibility testing.
The 10 oral antimicrobials
testedamoxicillin, amoxicillin-clavulanate (2:1 ratio), cefaclor,
cefixime, cefprozil, cefuroxime, loracarbef, azithromycin,
clarithromycin, and ciprofloxacinwere selected to reflect
representative current treatment options. Ampicillin was also tested to
characterize -lactamase-negative, ampicillin-resistant strains of
H. influenzae. Penicillin was also tested to characterize
the penicillin susceptibility of S. pneumoniae, and
ceftriaxone was used to characterize strains for which the MICs of this
agent are higher than the penicillin MICs. MICs were determined by
broth microdilution in accordance with the methods of the National
Committee for Clinical Laboratory Standards (NCCLS) (27).
Broth microdilution tests were performed in custom-dried 96-well
microdilution trays (Sensititre Division, Accumed International, Westlake, Ohio) in two configurations, one for testing of S. pneumoniae and the other for the testing of H. influenzae. Haemophilus inocula were prepared from
chocolate agar plates incubated for a full 24 h by the direct
colony suspension method. S. pneumoniae inocula were
prepared from blood agar plates incubated for 18 to 20 h, also by
direct colony suspension. Growth from these plates was then suspended
in tubes of Mueller-Hinton broth (Sensititre) to a density equivalent
to a 0.5 McFarland standard. Within 30 min of preparation, 20 µl of a
H. influenzae suspension or 200 µl of an S. pneumoniae suspension was added to a 10-ml tube of in-house fresh
Haemophilus test medium and Mueller-Hinton broth
supplemented with 5% lysed horse blood (Cleveland Scientific, Bath,
Ohio). Also within 30 min of preparation, doseheads were placed on the tubes and an autoinoculator (Sensititre) dispensed a 100-µl volume into each well of the Sensititre microdilution trays. The trays were
sealed and incubated for 22 to 24 h at 35°C in ambient air, and
the lowest drug concentration showing no growth was read as the MIC.
Inoculum checks were performed on all isolates by transfer of 10 µl
from the Haemophilus test medium or Mueller-Hinton
broth-lysed horse blood suspensions into tubes containing 6 ml of
saline; after mixing, 100 µl was transferred to a blood or chocolate
agar plate and spread over the surface of the plate. Colonies were counted after incubation for 20 to 24 h at 35°C in a 5%
CO2 atmosphere; 50 to 120 colonies represented the desired
range of 3 × 105 to 7 × 105 CFU/ml,
and strains with inocula beyond this range were retested until the
inocula were in the correct range. Quality control of MIC testing is
detailed below.
H. influenzae isolates were also tested for -lactamase
production by the nitrocefin disk method (Cefinase; Becton Dickinson Laboratories, Sparks, Md.), with positive and negative controls being
employed on each day of testing.
Quality control.
Initial quality control assessments
included evaluation of the performance characteristics of the Sensitire
panels and media used for S. pneumoniae and H. influenzae. Quality control strains specified by the NCCLS,
including S. pneumoniae ATCC 49619, H. influenzae
ATCC 49247 and 49766, Enterococcus faecalis ATCC 29212, and
Escherichia coli ATCC 25922 and 35218, were used
(27). Inocula of the nonfastidious strains were prepared as
described for S. pneumoniae, except that 50-µl suspension
volumes were added to 10-ml tubes of plain Mueller-Hinton broth. MICs
for quality control strains were required to fall within
NCCLS-specified ranges (28). In addition, a battery of
S. pneumoniae and H. influenzae strains for which
the MICs are known were tested. The performance characteristics of the
Sensititre trays were initially found to be inadequate (data not
shown), and this was traced to poor growth properties of the initial
Mueller-Hinton-lysed horse blood and Haemophilus test
medium broths obtained from Sensititre that were used to rehydrate the
trays. Strains were retested by using Mueller-Hinton-lysed horse blood
and Haemophilus test medium broths that were freshly prepared in house, and these results showed fully acceptable
performance characteristics (data not shown). Consequently, these media
were used throughout the study. The Mueller-Hinton-lysed horse blood broth was prepared on the day of testing by adding 1 ml of lysed horse
blood to 10 ml of Mueller-Hinton broth (Sensititre).
Haemophilus test medium was prepared in accordance with the
methods of the NCCLS by using Mueller-Hinton broth base (Difco), 0.5%
yeast extract (Difco), 15 µg of NAD per ml, and 15 µg of hematin
per ml (Sigma). Haemophilus test medium was prepared in
batches and stored at 4°C for use within 2 weeks of preparation (its
performance was found to degrade after this time) or stored at 20°C
for up to 6 weeks.
After the adequacy of the susceptibility testing materials was
confirmed, the relevant quality control strains for each organism were
tested on each day of testing and results were accepted only if the
MICs for the quality control strains were within specified limits. In
addition, at the end of the study, all quality control values were
analyzed; this analysis confirmed that modal values for all agents were
the same as the modal NCCLS values and that the values from the two
testing laboratories were comparable.
Susceptibility interpretation criteria.
MICs were
interpreted as indicating susceptible, intermediate, or resistant
categories in accordance with NCCLS guidelines (28), where
available, and on the basis of PK/PD parameters as well (7, 8,
29). PK/PD breakpoints were based on standard dosing regimens and
criteria appropriate to each agent. For -lactams and clarithromycin,
these breakpoints were based on drug concentrations in serum present
for 40 to 50% of the dosing interval, while for azithromycin and
ciprofloxacin, they were based on 24-h AUC/MIC ratios exceeding 25 (8, 29).
Data collection and analysis.
All pertinent data, including
demographical and susceptibility data, were entered into a computerized
data base. The MIC ranges and distributions and the MICs that inhibited
50 and 90% of the organisms tested (MIC50 and
MIC90, respectively) were determined for each agent.
Haemophilus data were analyzed for all strains, -lactamase-positive strains, and -lactamase-negative strains. For
S. pneumoniae isolates, the analysis included all of the
strains, as well as the penicillin-susceptible, intermediate, and
penicillin-resistant strains. For all strains of both pathogens, data
were also analyzed by geographic region, isolation site, and patient
age. Statistical significance was determined by chi-square analysis,
and P values of 
0.05 were regarded as significant.
| |
RESULTS |
This comprehensive study evaluated 1,476 strains of S. pneumoniae and 1,676 untypeable strains of H. influenzae isolated from specimens submitted from patients in 31 states grouped into six regions (Fig. 1).
Tables 1 and
2 show the distribution of strains by
region and by specimen source, respectively. S. pneumoniae was isolated more frequently than H. influenzae from
children 10 years of age, while H. influenzae was more
common than S. pneumoniae in specimens from adults older
than 30 years (Table 3).
View larger version (56K):
[in this window]
[in a new window]
|
FIG. 1.
Isolates from the six regions shown were analyzed.
Abbreviations: NW, northwest; SW, southwest; NC, north central; SC,
south central; NE, northeast; and SE, southeast.
|
|
In children, isolates of S. pneumoniae and H. influenzae were cultured predominantly from specimens obtained
from the eye, blood (S. pneumoniae only), middle ear, and
nasopharynx, while in adults, strains were isolated predominantly from
specimens obtained from the lower respiratory tract (mainly H. influenzae) and paranasal sinuses. Of S. pneumoniae
isolates, 113 were isolated from blood (60% in the 2-year age
group), 440 were from the middle ear (66%, 2 years; 21%, 3 to 10 years), 279 were from the eye (50%, 2 years), 415 were from the
nasopharynx (76%, 10 years), 69 were from the paranasal sinus (28%,
31 to 40 years), and 131 were from sputum (40%, >70 years). Of
H. influenzae isolates, 295 were from the middle ear (78%,
2 years; 12%, 3 to 10 years), 404 were from the eye (74%, 10
years), 374 were from the nasopharynx (67%, <10 years), 67 were from
the paranasal sinus (43%, 31 to 50 years), and 374 were from sputum
(85%, >30 years); no strains were isolated from blood.
H. influenzae susceptibility.
Of the 1,676 H. influenzae strains tested, 41.6% (697 strains) were
-lactamase positive and 58.4% (979 strains) were -lactamase negative by the nitrocefin disk method (Tables 2 and 3). The MIC ranges
and the MIC50s and MIC90s for all strains and
for -lactamase-positive and -negative strains are shown in Table
4. Only one fluoroquinolone-resistant H. influenzae strain was detected. This strain was
additionally highly resistant to azithromycin and clarithromycin (MICs
of >64 µg/ml).
For -lactamase-positive strains of H. influenzae, the
MIC90 of ampicillin and amoxicillin was >16 µg/ml, that
of clarithromycin was 16 µg/ml, and that of cefaclor, cefprozil, and
loracarbef was 32 µg/ml (Table 4). Only one -lactamase-negative
strain demonstrated the features of a -lactamase-negative,
ampicillin-resistant strain, with an ampicillin MIC of 4 µg/ml, 1 dilution above the susceptibility breakpoint. It is of note, however,
that more than 97% of the H. influenzae strains were
susceptible to amoxicillin-clavulanate, cefixime, and ciprofloxacin by
both the NCCLS and PK/PD breakpoints (Table
5). For cefuroxime, 99% of the strains
were susceptible at the NCCLS breakpoint of 4 µg/ml, compared to 78%
at the PK/PD breakpoint of 1 µg/ml. The susceptibility values for
cefaclor, cefprozil, and loracarbef were 79, 86, and 91%,
respectively, at the NCCLS breakpoint of 8 µg/ml for these agents but
were 2, 14, and 9%, respectively, at the PK/PD breakpoints of 0.5, 1, and 0.5 µg/ml, respectively. These three agents were more active against -lactamase-negative strains than against
-lactamase-positive strains; nonetheless, only 21% of the
-lactamase-negative strains were susceptible to cefprozil, the most
active of these three agents. By NCCLS breakpoints, 99.7% of the
H. influenzae isolates were susceptible to azithromycin and
76.6% were susceptible to clarithromycin; using PK/PD breakpoints,
however, no strains were susceptible to these agents.
View this table:
[in this window]
[in a new window]
|
TABLE 5.
Percentages of H. influenzae isolates
susceptible to 11 antimicrobial agents on the basis of NCCLS and
PK/PD breakpoints
|
|
S. pneumoniae susceptibility.
The MIC
susceptibility data for the 1,476 strains of S. pneumoniae
tested are summarized in Tables 2, 3, 6,
and 7. Penicillin-susceptible strains
accounted for 49.6% of the isolates, while 17.2% were penicillin
intermediate and 32.5% were penicillin resistant (Table 2). Only one
strain with a ceftriaxone MIC higher than that of penicillin was
detected. No strain with a ciprofloxacin MIC of >4 µg/ml was found.
View this table:
[in this window]
[in a new window]
|
TABLE 7.
Percentages of S. pneumoniae isolates
susceptible to antimicrobial agents on the basis of NCCLS and
PK/PD breakpoints
|
|
It is of note that the MIC90s for all strains of cefaclor
(>64 µg/ml), cefixime (32 µg/ml), and loracarbef (>64 µg/ml)
were very high (Table 6), and susceptibility values at PK/PD
breakpoints (Table 7) were low for these antimicrobial agents (22.4%
for cefaclor, 52.1% for cefixime, and 10.7% for loracarbef); these agents do not currently have NCCLS breakpoints for S. pneumoniae. The overall susceptibility values by NCCLS breakpoints
for the other antimicrobial agents tested were 60.7% for cefuroxime,
63.5% for amoxicillin, 65.8% for amoxicillin-clavulanate, 68% for
ceftriaxone, and 69.8% for azithromycin and clarithromycin (Table 7).
The overall susceptibility values by PK/PD breakpoints were the highest for amoxicillin (93.5%) and amoxicillin-clavulanate (93.9%), followed by cefuroxime at 62.9% and cefprozil at 62.6%; the values for azithromycin and clarithromycin were unchanged.
Analysis of the macrolide-azalide and penicillin cross-resistance of
S. pneumoniae strains showed that 5.1% of the
penicillin-susceptible strains were resistant to macrolides-azalides
while 36.7% of the penicillin-intermediate strains and 65.6% of the
penicillin-resistant strains were macrolide-azalide resistant (Table
8). All differences between penicillin
and macrolide susceptibility categories were statistically
significant (P < 0.001).
Susceptibility variations by region, age, and specimen source.
Susceptibility patterns were determined for each of six regions of the
United States (Table 9). The highest
proportions of penicillin-intermediate and penicillin-resistant strains
of S. pneumoniae were from the south central region (62.4%)
(P < 0.001 compared to the overall proportion) and the
southeastern region (60.7%) (P < 0.001), while the
northwestern region (39.3%) (P = 0.003) and
northeastern region (35.8%) (P < 0.001) had the
lowest. In all regions, but particularly in the south central and
southeastern regions, more strains of S. pneumoniae were
resistant than intermediately susceptible (Table 9). Overall, the
proportion of penicillin-intermediate and penicillin-resistant strains
in the three southern regions (60.7%) was significantly higher than in
the three northern regions (39.2%) (P < 0.001).
Similarly, the proportion of macrolide-azalide-resistant strains in the
southern regions (39.5%) was higher than in the northern regions
(22.7%) (P < 0.001).
The highest proportions of -lactamase-positive strains of
H. influenzae were from the northeastern (50.3%)
(P = 0.047 compared to the overall proportion) and
north central (48.3%) regions, while the lowest were from the
northwestern (36.7%) and southwestern (35.4%) regions, although the
differences for the last three regions did not reach statistical
significance. Overall, however, the proportion of
-lactamase-positive strains from the three northern regions (44.6%)
compared with the three southern regions (39.9%) just reached
statistical significance (P = 0.05).
For S. pneumoniae isolates, the overall intermediate
penicillin intermediate and penicillin resistance values were 17.9 and 32.5%, respectively. The prevalence of penicillin resistance in S. pneumoniae was the highest in isolates from the middle
ear (40.9%) (P < 0.001 versus non-ear isolates) and
sinus (44.9%) (P = 0.03 versus non-sinus isolates)
specimens (Table 2). The proportion of strains nonsusceptible
(intermediate and resistant) to penicillin was also the highest in
the < 2-year-old age group (58.3% versus 42.1%) (P < 0.001 compared with the >2-year-old age group) (Table 3). In
the youngest age groups, the prevalence of penicillin-resistant strains
from middle ear specimens (49.7%) was higher than that of
penicillin-intermediate strains (19.5%) in the <2-year-old age group
(P < 0.001). The prevalence of penicillin-intermediate strains (21.2%) was similar to that of penicillin-resistant strains (17.7%) (no statistically significant difference) in blood specimens. The prevalence of penicillin-intermediate and penicillin-resistant strains was also high (resistant > intermediate) in
nasopharyngeal specimens from all age groups.
Overall, 41.6% of the H. influenzae isolates were
-lactamase positive; the highest occurrence of
-lactamase-positive strains was in children younger than 2 years of
age (47.0 versus 37.4% for >2-year-olds; P < 0.001),
while the lowest was in nasopharyngeal (35.0%) and other (35.4%)
specimen sources (Tables 2 and 3). In an analysis of H. influenzae resistance by both age and specimen source, the
prevalence of -lactamase-positive strains was higher in the
following groups: (i) ear specimens from children <2 years of age
(45.0%), compared with those from children 2 years old (37.3%);
(ii) nasopharyngeal specimens from children <10 years of age (44.9%),
compared with those from children 10 years of age (37.6%); and (iii)
sinus specimens in adults 31 to 50 years of age (58.6%). However, none
of these differences were statistically significant.
| |
DISCUSSION |
The recent increase in the resistance of the major respiratory
pathogens H. influenzae and S. pneumoniae to oral
antimicrobial agents has produced a need to re-evaluate treatment
options for respiratory tract infections (16, 20). This is
particularly important for the established oral agents, many of which
have decreased activity against contemporary isolates, and also for newer agents like the fluoroquinolones, which currently have broader spectra of activity against these and other respiratory tract pathogens. Recent studies have shown that up to 33% of the strains of
S. pneumoniae are penicillin intermediate or penicillin
resistant in many parts of the country (13, 40).
Furthermore, over 30% of the strains of H. influenzae and
90% of the strains of Moraxella catarrhalis now produce
-lactamases (12, 40). This severely limits the activity
of many oral antimicrobial agents, particularly for pediatric use
(1-4, 20, 23, 30).
Applying PK/PD breakpoints to the results of this study has identified
amoxicillin-clavulanate and the fluoroquinolones as active against more
than 90% of the strains of both S. pneumoniae and H. influenzae. Although ciprofloxacin itself has marginal activity
against S. pneumoniae, no strains for which the
ciprofloxacin MIC was >4 µg/ml were found and the newer
fluoroquinolones, such as levofloxacin, sparfloxacin, grepafloxacin,
and trovafloxacin, are active against pneumococci (23).
Cefuroxime is the next most active agent against both species, with
78% of H. influenzae and 63% of S. pneumoniae
strains being susceptible at PK/PD breakpoints. While 100% of H. influenzae strains are susceptible to cefixime, only 52% of
S. pneumoniae strains are susceptible. In contrast, while
94% of S. pneumoniae strains are susceptible to
amoxicillin, only 59% of H. influenzae strains are
susceptible. Cefprozil has activity similar to that of cefuroxime
against S. pneumoniae (63% susceptible) but poor activity
against H. influenzae (14% susceptible). Cefaclor and
loracarbef have poor activity against both species, with only 22 and
11% of S. pneumoniae and 2 and 9% of H. influenzae isolates, respectively, being susceptible. Although
69% of S. pneumoniae strains were susceptible to
azithromycin and clarithromycin, no H. influenzae strains
were susceptible based on serum PK/PD parameters. However, a human
volunteer study on the intrapulmonary distribution of clarithromycin
and azithromycin demonstrated much higher concentrations of these
agents in epithelial lining fluid (ELF) than in serum (35).
Drug concentrations present in ELF for 50% of the dosing interval
were 15 to 30 µg/ml (mean of five values, 26.1 µg/ml) for
clarithromycin and <0.1 to 1 µg/ml (mean of five values [excluding
one value of <0.1 µg/ml], 0.95 µg/ml). These concentrations
exceed the MICs of clarithromycin for most strains of H. influenzae and some strains of macrolide-resistant S. pneumoniae and the MICs of azithromycin for some strains of H. influenzae. However, the importance of these drug
concentrations in ELF for the clinical outcome of pulmonary infections
is unclear.
Analysis of the pathogen distribution obtained in this study by patient
age, specimen source, and geographic area shows some interesting
patterns. The number of strains isolated from middle ear specimens is
alarming, since such specimens can only be obtained by tympanocentesis
or after the tympanic membrane ruptures spontaneously. Thus, these
isolates likely represent treatment failures, which is supported by the
resistant nature of many of the S. pneumoniae isolates from
ear specimens compared to other sites. Similarly, paranasal sinus
specimens were presumably obtained by sinus puncture and may well also
represent treatment failures. The majority of S. pneumoniae
(68%) and H. influenzae (51%) strains were isolated from
children <10 years of age, although many of these strains were
recovered from eye and nasopharyngeal specimens. The significance of
these strains in eye specimens, predominantly from conjunctival exudates, is questionable. Their significance in nasopharyngeal specimens is unknown but presumably results from practitioners wanting
to know whether patients are carrying resistant strains to guide the
therapy of treatment failures or complications.
This study documents the dramatic and alarming increase in -lactam
and macrolide-azalide resistance in S. pneumoniae,
particularly the alarming increase in strains both fully penicillin
resistant and also resistant to macrolides-azalides. A few case reports have documented the clinical failure of macrolides-azalides (21, 32), and the clinical value of these agents remains controversial (4, 42). Amoxicillin and amoxicillin-clavulanate still
retain their activity, however, with 94% of the S. pneumoniae strains being susceptible in our study. In addition,
ceftriaxone-resistant strains with lower penicillin MICs remain rare
(only one strain being found in this study) and most strains causing
nonmeningeal infections should respond to this agent administered
intravenously or intramuscularly. An appropriate PK/PD breakpoint for
this agent in nonmeningeal infections is 2 µg/ml (8), with
97.7% of the strains being susceptible at this concentration in our
study. Treating meningitis with ceftriaxone may be a problem, however, since only 68% of the strains were susceptible at the meningitis breakpoint of 0.5 µg/ml and for a further 9.4% of the strains the
MIC was 1 µg/ml, the current intermediate value in meningitis. Although no fluoroquinolone-resistant strains were found in this study,
such strains have been identified (19, 40) and the indiscriminate use of these agents to treat respiratory tract infections could lead to the development and spread of resistance.
In contrast to S. pneumoniae, our results showed no major
changes in susceptibility in H. influenzae isolates,
although the prevalence of -lactamase positivity continued to
increase, with 42% of the strains being positive. The
MIC50s and MIC90s were generally unchanged from
those of previous studies, however, despite the fact that the
proportions of strains susceptible to many agents differ markedly from
those in other studies due to our use of PK/PD breakpoints (2, 12,
15, 25, 40). As with S. pneumoniae, fluoroquinolone
resistance has not yet emerged in H. influenzae, with only
one resistant strain detected. Additionally, no major shift in
non--lactamase-mediated -lactam resistance was found, with only
one possible -lactamase-negative, ampicillin-resistant strain
being found. No amoxicillin-clavulanate-resistant strains were found,
in contrast to a previous study, in which 3.6% of the strains from a
comparable patient population were found to be resistant to the NCCLS
breakpoint of 4 µg/ml, with MICs mainly 1 dilution above the
breakpoint (12). The MIC distributions of several agents in
that study, including amoxicillin-clavulanate and cefuroxime, were
wider than those most investigators have reported, and the
MIC50 and MIC90 of amoxicillin-clavulanate were 1 and 2 µg/ml, rather than the usual 0.5 and 1 µg/ml reported in
previous studies (22, 24, 36), as well as in the current study. Problems associated with the methods used to test
Haemophilus susceptibility appear to be responsible for
these differences (24).
The results of this study should be applied to clinical practice based
on the clinical presentation of the patient, the probability of the
patient's having a bacterial rather than a viral infection, the
natural history of the disease, the potential of pathogens to be
susceptible to various oral antimicrobial agents, the potential for
cross-resistance between agents with S. pneumoniae, and the potential of pathogens to develop further resistance (37).
Antibiotics should be used judiciously to maintain any remaining
activity (16, 17) and chosen carefully based on activity
determined by PK/PD-based breakpoints (8).
In summary, this study has highlighted the continued dramatic rise in
-lactam and macrolide-azalide resistance in S. pneumoniae and a modest continued rise in -lactamase production in untypeable H. influenzae. In addition, pharmacodynamic parameters have
been used to interpret susceptibility data in a more clinically
meaningful way. Judicious antibiotic use is the key to avoiding the
development of further resistance to available agents in these bacteria.
Financial support for this work was provided by SmithKline
Beecham Pharmaceuticals, Philadelphia, Pa.
We thank Raymond Kaplan and Linda Miller for coordinating collection of
strains, Gwen Kendall for secretarial assistance, and PMSI Scott-Levin,
Inc., Newtown, Pa., for permission to use antibiotic prescription data.
| 1.
|
Barry, A. L., and P. C. Fuchs.
1995.
In vitro activities of a streptogramin (RP59500), three macrolides, and an azalide against four respiratory tract pathogens.
Antimicrob. Agents Chemother.
39:238-240[Abstract].
|
| 2.
|
Barry, A. L.,
M. A. Pfaller,
P. C. Fuchs, and R. R. Packer.
1994.
In vitro activities of 12 orally administered antimicrobial agents against four species of bacterial respiratory pathogens from U.S. medical centers in 1992 and 1993.
Antimicrob. Agents Chemother.
38:2419-2425[Abstract/Free Full Text].
|
| 3.
|
Barry, A. L.
1995.
Antimicrobial agents for community-acquired respiratory tract infections.
Infection
23(Suppl. 2):S59-S63. (Discussion, S64.)
|
| 4.
|
Bartlett, J. G.
1997.
Empirical therapy of community-acquired pneumonia: macrolides are not ideal choices.
Semin. Respir. Infect.
12:329-333[Medline].
|
| 5.
|
Black, S. B.,
H. R. Shinefield, and The Kaiser Permanente Vaccine Study Group.
1992.
Immunization with oligosaccharide conjugate Haemophilus influenzae type b (HbOC) vaccine on a large HMO population: extended follow-up and impact on Haemophilus influenzae disease epidemiology.
Pediatr. Infect. Dis. J.
11:610-613[Medline].
|
| 6.
|
Bradley, J. S., and J. D. Connor.
1991.
Ceftriaxone failure in meningitis caused by Streptococcus pneumoniae with reduced susceptibility to -lactam antibiotics.
Pediatr. Infect. Dis. J.
10:871-872[Medline].
|
| 7.
|
Craig, W. A., and D. Andes.
1996.
Pharmacokinetics and pharmacodynamics of antibiotics in otitis media.
Pediatr. Infect. Dis. J.
15:255-259[Medline].
|
| 8.
|
Craig, W. A.
1998.
Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial dosing of mice and men.
Clin. Infect. Dis.
26:1-10[Medline].
|
| 9.
|
Dagan, R.,
O. Abramson,
E. Leibovitz,
D. Greenberg,
R. Lang,
S. Goshen,
P. Yagupsky,
A. Leiberman, and D. M. Fliss.
1997.
Bacteriologic response to oral cephalosporins: are established susceptibility breakpoints appropriate in the case of acute otitis media?
J. Infect. Dis.
176:1253-1259[Medline].
|
| 10.
|
Dagan, R.,
O. Abramson,
E. Leibovitz,
R. Lang,
S. Goshen,
D. Greenberg,
P. Yagupsy,
A. Leiberman, and D. M. Fliss.
1996.
Impaired bacteriologic response to oral cephalosporins in acute otitis media caused by pneumococci with intermediate resistance to penicillin.
Pediatr. Infect. Dis. J.
15:980-985[Medline].
|
| 11.
|
Dagan, R.,
L. Piglansky,
P. Yagupsky,
D. M. Fliss,
A. Leiberman, and E. Leibovitz.
1997.
Bacteriologic response in acute otitis media (AOM): comparison between azithromycin (AOM), (AZ), cefaclor (CEC) and amoxicillin (AMOX), abstr. K-103, p. 346.
In
Abstracts of the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy. American Society for Microbiology, Washington, D.C.
|
| 12.
|
Doern, G. V.,
A. B. Brueggemann,
G. Pierce,
H. P. Holley, Jr., and A. Rauch.
1997.
Antibiotic resistance among clinical isolates of Haemophilus influenzae in the United States in 1994 and 1995 and detection of -lactamase-positive strains resistant to amoxicillin-clavulanate: results of a national multicenter surveillance study.
Antimicrob. Agents Chemother.
41:292-297[Abstract].
|
| 13.
|
Doern, G. V.,
A. Brueggemann,
H. P. Holley, Jr., and A. M. Rauch.
1996.
Antimicrobial resistance of Streptococcus pneumoniae recovered from outpatients in the United States during the winter months of 1994 to 1995: results of a 30-center national surveillance study.
Antimicrob. Agents Chemother.
40:1208-1213[Abstract].
|
| 14.
|
Doern, G. V.,
J. H. Jorgensen,
C. Thornsberry,
D. A. Preston,
T. Tubert,
J. S. Redding, and L. A. Maher.
1988.
National collaborative study of the prevalence of antimicrobial resistance among clinical isolates of Haemophilus influenzae.
Antimicrob. Agents Chemother.
32:180-185[Abstract/Free Full Text].
|
| 15.
|
Doern, G. V.
1996.
Antimicrobial resistance among lower respiratory tract isolates of Haemophilus influenzae: results of a 1992-93 Western Europe and USA collaborative study.
J. Antimicrob. Chemother.
38(Suppl. A):59-69[Abstract/Free Full Text].
|
| 16.
|
Dowell, S. F.,
S. M. Marcy,
W. R. Phillips,
M. A. Gerber, and B. Schwartz.
1998.
Principles of judicious use of antimicrobial agents for pediatric upper respiratory tract infections.
Pediatrics
101:163-165[Abstract/Free Full Text].
|
| 17.
|
Dowell, S. F.,
J. C. Butler,
G. S. Giebink,
M. R. Jacobs,
D. Jernigan,
D. M. Musher,
A. Rakowsky, and B. Schwartz.
1999.
Acute otitis media: management and surveillance in an era of pneumococcal resistancea report from the Drug-Resistant Streptococcus pneumoniae Therapeutic Working Group.
Pediatr. Infect. Dis. J.
18:1-9[Medline].
|
| 18.
|
Drusano, G. L., and W. A. Craig.
1997.
Relevance of pharmacokinetics and pharmacodynamics in the selection of antibiotics for respiratory tract infections.
J. Chemother.
9(Suppl. 3):38-44.
|
| 19.
|
Drusano, G. L., and F. W. Goldstein.
1996.
Relevance of the Alexander Project: pharmacodynamic considerations.
J. Antimicrob. Chemother.
38(Suppl. A):141-154.
|
| 20.
|
Green, M., and E. R. Wald.
1996.
Emerging resistance to antibiotics: impact on respiratory infections in the out-patient setting.
Ann. Allergy Asthma Immunol.
77:167-173[Medline].
|
| 21.
|
Jackson, M. A.,
V. F. Burry,
L. C. Olson,
S. E. Duthie, and G. L. Kearns.
1996.
Breakthrough sepsis in macrolide-resistant pneumococcal infection.
Pediatr. Infect. Dis. J.
15:1049-1051[Medline].
|
| 22.
|
Jacobs, M. R.
1999.
Assessing the quality of the Alexander Project.
J. Chemother.
11:26-34.
|
| 23.
|
Jacobs, M. R., and P. C. Appelbaum.
1995.
Antibiotic-resistant pneumococci.
Rev. Med. Microbiol.
6:77-93.
|
| 24.
|
Jacobs, M. R., and S. Bajaksouzian.
1997.
Microbiologic evaluation of contemporary Haemophilus influenzae isolates having elevated MICs to amoxicillin-clavulanic acid.
Diagn. Microbiol. Infect. Dis.
28:105-112[Medline].
|
| 25.
|
Jones, R. N.,
M. R. Jacobs,
J. A. Washington, and M. A. Pfaller.
1997.
A 1994-95 survey of Haemophilus influenzae susceptibility to ten orally administered agents. A 187 clinical laboratory center sample in the United States.
Diagn. Microbiol. Infect. Dis.
27:75-83[Medline].
|
| 26.
|
Khan, W.,
S. Ross,
W. Rodriguez,
G. Controni, and A. K. Saz.
1974.
Haemophilus influenzae type b resistant to ampicillin: a report of two cases.
JAMA
229:298-301[Medline].
|
| 27.
|
National Committee for Clinical Laboratory Standards.
1997.
Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. Approved standard M7-A4.
National Committee for Clinical Laboratory Standards, Villanova, Pa.
|
| 28.
|
National Committee for Clinical Laboratory Standards.
1997.
Performance standards for antimicrobial susceptibility testing. Sixth information supplement M100-S7.
National Committee for Clinical Laboratory Standards, Villanova, Pa.
|
| 29.
| Nightingale, C. H. Pharmacokinetics and
pharmacodynamics of newer macrolides. Pediatr. Infect. Dis. J. 16:438-443.
|
| 30.
|
Pelton, S. I.
1996.
Defining resistance: breakpoints and beyond implications for pediatric respiratory infection.
Diagn. Microbiol. Infect. Dis.
25:195-199[Medline].
|
| 31.
|
PMSI Scott-Levin, Inc..
1997.
Physician drug and diagnosis audit (PDDA).
PMSI Scott-Levin, Inc., Newtown, Pa.
|
| 32.
|
Reid, R., Jr.,
J. S. Bradley, and J. Hindler.
1995.
Pneumococcal meningitis during therapy of otitis media with clarithromycin.
Pediatr. Infect. Dis. J.
14:1104-1105[Medline].
|
| 33.
|
Rittenhouse, S. F.,
L. A. Miller,
R. L. Kaplan,
G. H. Mosely, and J. A. Poupard.
1995.
A survey of -lactamase-producing Haemophilus influenzae.
Diagn. Microbiol. Infect. Dis.
21:223-225[Medline].
|
| 34.
|
Roberts, D. E.,
A. Ingold,
S. V. Want, and J. R. May.
1974.
Osmotically stable L-forms of Haemophilus influenzae and their significance in testing sensitivity to penicillins.
J. Clin. Pathol.
27:560-564[Abstract/Free Full Text].
|
| 35.
|
Rodvold, K. A.,
M. H. Gotfried,
L. H. Danziger, and R. J. Servi.
1997.
Intrapulmonary steady-state concentrations of clarithromycin and azithromycin in healthy adult volunteers.
Antimicrob. Agents Chemother.
41:1399-1402[Abstract].
|
| 36.
|
Rolinson, G. N.
1994.
A review of the microbiology of amoxicillin-clavulanic acid over the 15-year period 1978-1993.
J. Chemother.
6:283-318[Medline].
|
| 37.
|
Rosenfeld, R. M.
1996.
An evidence-based approach to treating otitis media. 1996.
Pediatr. Clin. N. Am.
430:1165-1181.
|
| 38.
|
Sloas, M. M.,
F. F. Barrett,
P. J. Chesney,
B. K. English,
B. C. Hill,
F. C. Tenover, and R. J. Leggiadro.
1992.
Cephalosporin treatment failure in penicillin- and cephalosporin-resistant Streptococcus pneumoniae meningitis.
Pediatr. Infect. Dis. J.
11:662-666[Medline].
|
| 39.
|
Sykes, R. B.,
A. Griffiths, and D. M. Ryan.
1977.
Comparative activity of ampicillin and cefuroxime against three types of Haemophilus influenzae.
Antimicrob. Agents Chemother.
11:599-604[Abstract/Free Full Text].
|
| 40.
|
Thornsberry, C.,
P. Oglivie,
J. Kahn, and Y. Mauritz.
1997.
Surveillance of antimicrobial resistance in Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis in the United States in 1996-1997 respiratory season.
Diagn. Microbiol. Infect. Dis.
29:249-257[Medline].
|
| 41.
|
Tomeh, M. O.,
S. E. Starr,
J. E. McGowan, Jr.,
P. M. Terry, and A. J. Nahamias.
1974.
Ampicillin-resistant Haemophilus influenzae type b infection.
JAMA
229:295-297[Medline].
|
| 42.
|
Vergis, E. N., and V. L. Yu.
1997.
Macrolides are ideal for empiric therapy of community-acquired pneumonia in the immunocompetent host.
Semin. Respir. Infect.
12:322-328[Medline].
|
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
