570
Dosing of Amoxicillin/Clavulanate Given Every 12 Hours Is as Effective as
Dosing Every 8 Hours for Treatment of Lower Respiratory Tract Infection
Alistair D. Calver, Niall S. Walsh, Patrick F. Quinn,
Constantin Baran, Val Lonergan, Krishan P. Singh,
Wendy S. Orzolek, and the Lower Respiratory Tract
Infection Collaborative Study Group
From the West Vaal Hospital, Orkney, South Africa; Longford Centre,
County Longford, Ireland; Health Centre, Sligo, Ireland; private
practice, Dachau, Germany; Health Centre, Graignamanagh, County
Kilkenny, Ireland; and SmithKline Beecham Pharmaceuticals,
Collegeville, Pennsylvania, USA
In this double-blind study, 557 patients with lower respiratory tract infection were randomly
assigned to receive amoxicillin/clavulanate orally either every 12 hours (875/125 mg) or every 8
hours (500/125 mg) for 7 —15 days. For the 455 patients evaluable for clinical efficacy at the end of
therapy, clinical success was similar in the two groups: 93% and 94% in the 12-hour and 8-hour
groups, respectively (P = .42). Bacteriologic success at the end of therapy was also comparable:
97% and 91% in the 12-hour and 8-hour groups, respectively (P = .86). The occurrence of adverse
events related to treatment was similar for the two groups, but fewer patients in the 12-hour group
reported moderate or severe diarrhea. Amoxicillin/clavulanate (875/125 mg) given every 12 hours
is as effective and safe as every-8-hours administration of the combination (500/125 mg) for the
treatment of lower respiratory tract infection.
Bacterial pneumonia and acute exacerbation of chronic bronchitis remain among the most common infections treated in the
community. In the United States up to 3.3 million people develop community-acquired pneumonia each year [1], and 50%
to 90% are treated as outpatients [2]. Because of the relatively
high frequency of individuals whose host defenses are compromised by age or coexistent disease, along with increasing bacterial resistance, careful selection of antimicrobial therapy is warranted.
The microbiology of community-acquired lower respiratory
tract infection is changing with the emergence of resistant
pathogens. This is partially attributable to changing characteristics of the general population [3, 4]. There are now greater
numbers of people over age 65 years and of ambulatory patients
with compromised immune function. New lower respiratory
tract pathogens have emerged (such as Legionella species),
and f3-lactamase-producing organisms (such as Haemophilus
influenzae and Moraxella catarrhalis) are increasingly common. In addition, antibiotics tend to be used early in infectious
episodes, leading to a high proportion of patients with unidentifiable pathogens.
Received 26 April 1996; revised 23 September 1996.
This study was supported by a grant to each center by SmithKline Beecham
Pharmaceuticals.
The study was conducted following the guidelines of the Declaration of
Helsinki and was approved by an institutional review board/ethics committee
at each site. Informed consent was obtained from each patient prior to study
initiation.
Reprints or correspondence: Wendy S. Orzolek, SmithKline Beecham Pharmaceuticals, 1250 South Collegeville Road, P.O. Box 5089, UP4330, Collegeville, Pennsylvania 19426-0989.
Clinical Infectious Diseases 1997; 24:570-4
© 1997 by The University of Chicago. All rights reserved.
1058-4838/97/2404 — 0006$02.00
Amoxicillin/clavulanate (Augmentin; SmithKline Beecham
Pharmaceuticals, Philadelphia) is a combination product containing the semisynthetic penicillin amoxicillin and the 0-lactamase inhibitor clavulanate potassium. The established oraldose regimen for lower respiratory tract infections in adults is
500/125 mg (amoxicillin/clavulanate equivalent) given every
8 hours [5].
In recognition of the fact that compliance is a major factor
in the outcome of outpatient treatment of many infections, it
is necessary to determine the most convenient dose regimen
that is effective. The objective of this study was to compare
the safety and efficacy of amoxicillin/clavulanate given every
12 hours (875/125 mg) to that of the combination given every
8 hours (500/125 mg) for the treatment of lower respiratory
tract infections (community-acquired pneumonia and acute
bacterial exacerbation of chronic bronchitis).
Methods
Study design. This was a multinational, multicentered
study with 87 participating centers in the United States, Republic of Ireland, Canada, Australia, United Kingdom, Italy, Germany, Belgium, The Netherlands, Switzerland, and South Africa. Treatment was randomized and administered in a doubleblinded, double-dummy manner to two parallel treatment
groups.
Patients. Inpatients or outpatients at least 18 years of age
presenting with a community-acquired lower respiratory tract
infection who could be treated with an oral antimicrobial were
enrolled. We ensured that —50% of the patients had community-acquired pneumonia and —50% had acute exacerbation of
chronic bronchitis.
Patients with community-acquired pneumonia were required
to have clinical symptoms (e.g., chest pain, shortness of breath)
CID 1997;24 (April)
Amoxicillin/Clavulanate Every 12 vs. Every 8 Hours and features suggestive of community-acquired pneumonia
(e.g., fever, crackles on auscultation, bronchial breathing, dullness on percussion), cough productive of sputum, and infiltrates
evident on a chest roentgenogram. Patients with acute exacerbation of chronic bronchitis were required to have a history of
a cough productive of sputum on most days of the week for 3
months in each of the 2 preceding years, with an increase in
quantity and/or purulence of sputum in the 48 hours before
enrollment.
Patients included in the assessment of bacteriologic response
were required to have an organism identified from a sputum
sample collected within 2 days prior to the start of treatment.
The pretreatment gram stain must have contained >25 polymorphonuclear leukocytes and < 10 epithelial cells per lowpower field. Identification of organisms was confirmed by a
central study laboratory.
Antimicrobial treatment. Patients were randomly assigned
to receive either oral amoxicillin/clavulanate every 12 hours
(875/125 mg) plus an oral placebo (identical in appearance to
the active treatment) every 8 hours or amoxicillin/clavulanate
every 8 hours (500/125 mg) plus an oral placebo every 12
hours. Each regimen was administered for 7-15 days. Use of
additional antimicrobials or probenecid was not permitted.
At the end of treatment, compliance was assessed to ensure
that patients had received a minimum of 80% and a maximum
of 120% of the prescribed medication doses.
Evaluations. Patients were evaluated for clinical and bacteriologic efficacy of treatment 48-96 hours after its initiation
and then 2-4 days after its completion. In addition, assessment
was performed by telephone 5-9 days after the start of treatment. Patients returned for clinical and bacteriologic evaluation
2-4 weeks after the end of treatment.
Clinical response (cure, improvement, failure, or inevaluable
because of poor compliance or extenuating circumstances) was
determined 2-4 days after completion of treatment. Cure was
defined as complete resolution of signs and symptoms of infection, with no additional antimicrobial therapy required. Improvement was defined as incomplete resolution of the signs
and symptoms of infection, with no additional antimicrobial
therapy required. Failure was defined as no diminishment of the
signs and symptoms of infection and/or the need for additional
antimicrobial therapy. Cure and improvement were judged to
be evidence of clinical success.
Sputum samples were collected for gram stain, culture, and
susceptibility testing within 2 days prior to the start of treatment
and then during therapy, at the end of therapy, and 2-4 weeks
after treatment if the sputum was available. Serological tests
were not routinely performed to identify atypical pathogens.
Bacteriologic response was determined by presence or absence of an organism 2-4 days following completion of treatment and was assessed as proven eradication (elimination of
the pathogen documented by culture); presumed eradication
(symptomatic response was cure or improvement and a culture
was not clinically indicated); colonization (an organism ap-
571
peared after treatment that did not contribute to the symptoms
and signs of infection); superinfection (sputum culture showed
a significant number of colonies of an organism that was different from pretreatment pathogens, that contributed to the symptoms and signs of the infection, and that required additional
antimicrobial therapy); failure (noneradication of the initial
pathogen); or inevaluable. Proven eradication, presumed eradication, and colonization were judged to be evidence of bacteriologic success.
The safety of the regimen was determined for all randomized
patients by interview at each visit. Clinical chemistry and hematology screening tests were performed at the start of treatment and then, if clinically indicated, at the end of treatment
and 2-4 weeks after completion of treatment.
Data analysis. The study was designed to enroll at least
150 evaluable patients per treatment regimen (300 patients total) to determine with 80% power (/3 = 0.20) that the lower
confidence limit of the two-sided 95% confidence interval
(a = 0.05) of the difference in the clinical success rates between the two treatment groups was not below —10%, with
the assumption that the clinical response rate in the two groups
was 90%.
Continuous data (age and duration of therapy) were analyzed
by Student's t-test, while categorical data were analyzed with
the x 2 test. The difference in the success rates between treatment groups was assessed by analyzing a linear model with
effects due to center and treatment (Statistical Analysis System,
version 6.07; SAS Institute, Cary, NC). The equivalence of the
two treatment groups was also assessed by determining the
two-tailed 95% confidence interval of the difference in the
proportions of patients with clinical and bacteriologic success.
The treatment groups were considered equally effective if the
lower 95% confidence limit was not below —10%.
Results
Patients. Five hundred and fifty-seven patients were randomized to receive study medication: 273 received amoxicillin/
clavulanate every 12 hours, and 284 received it every 8 hours.
A total of 472 patients (85%) were evaluated at all four visits,
and 455 (80%) were considered evaluable for clinical efficacy
at the end of therapy. No significant differences were detected
(P > .05) with respect to any of the baseline characteristics
of patients (table 1).
One hundred and two patients were excluded from the perprotocol evaluation of clinical efficacy at the end of therapy
because of protocol violations (17% of the 12-hour group and
20% of the 8-hour group). Two hundred and thirty-seven patients (52%) had community-acquired pneumonia and 218
(48%) had acute exacerbation of chronic bronchitis. Compliance with the antibiotic regimen (^, 80% and 120% of doses
taken) was similar for the 12-hour and 8-hour groups (96%
and 94%, respectively; P = .22).
572
Calver et al.
CID 1997; 24 (April)
Table 1. Characteristics of the 455 evaluable patients from among the 557 with lower respiratory tract infection who were enrolled in the
study.
No. of patients or other data per amoxicillin/
clavulanate dosing schedule
Variable
Every 12 h
Patients enrolled
Patients who completed therapy
and were clinically evaluable
Patients withdrawn from the study
Age (y)
Mean ± SD
Range
Gender (male/female)
Race
White
Black
Asian
Other
Duration of therapy (d)
Mean ± SD
Range
Patients with
Community-acquired
pneumonia
Acute exacerbation of chronic
bronchitis
273
284
227
41
228
44
56.2 ± 18.0
19-93
143/84
57.1 ± 17.1
18 —89
148/80
204
19
1
3
213
13
11.0 ± 3.0
2 —15
10.8 ± 2.9
4 —15
120 (53%)
117 (51%)
107 (47%)
111 (49%)
Every 8 h
P value
.38
.41
.92
.06
0
2
Clinical response. Clinical response at the end of therapy
was similar for the two treatment groups, with success rates
of 93% and 94% in the 12-hour and 8-hour groups, respectively
(P = .42; 95% CI for the difference was — 5.3% —3.5%) (table
2). Clinical response was similar between the groups when
community-acquired pneumonia and acute exacerbation of
chronic bronchitis were considered separately (table 2). The
95% confidence intervals for the differences between regimens
.19
were — 7.1% —4.0% for community-acquired pneumonia and
— 7.2%— 6.7% for acute exacerbation of chronic bronchitis.
Clinical success was maintained at the time of the follow-up
visit for 89% (193 of 216 patients) and 86% (185 of 215 patients)
of the clinical successes at the end of treatment (P = .30) in the
12-hour and 8-hour groups, respectively. Recurrence of infection
was observed in 8 patients (3.7%) and 17 patients (7.9%) in the
12-hour group and 8-hour group, respectively (P > 0.05).
Table 2. Summary of clinical responses to treatment among the 455 evaluable patients.
No. (%) of patients per amoxicillin/clavulanate dosing schedule and disease category
Every 12 h
Variable
Pneumonia
Evaluable patients
Clinical outcome at end of therapy
Cure
Improvement
Overall clinical success (cure or improvement)
Failure
Clinical outcome at follow-up
Recurrence
Failure
120
78
35
113
7
AECB
107
(65)
(29)
(94)
(6)
4 (4)
7 (6)
NOTE. AECB = acute exacerbation of chronic bronchitis.
Every 8 h
81
18
99
8
Pneumonia
117
(76)
(17)
(93)
(7)
4 (4)
8 (8)
80
32
112
5
AECB
111
(68)
(27)
(96)
(4)
4 (4)
5 (5)
74
29
103
8
Every
12h
Every
8h
Total
Total
227
(67)
(26)
(93)
(8)
13 (12)
8 (8)
159
53
212
15
228
(70)
(23)
(93)
(7)
8 (4)
15 (7)
154
61
215
13
(68)
(27)
(94)
(6)
17 (8)
13 (6)
CID 1997;24 (April)
Amoxicillin/Clavulanate Every 12 vs. Every 8 Hours The reason for the difference in the recurrence rates could
not be ascertained. Recurrence was more frequent in the 8-hour
group with acute exacerbation of chronic bronchitis (12.3% of
patients), but differences between subgroups were not tested
since these subgroup comparisons were not part of the study
design.
Bacteriologic response. One hundred and thirty-nine organisms were reported for the 122 patients (59 in the 12-hour
group and 63 in the 8-hour group) who were evaluable for
bacteriologic efficacy. Most patients (108 of 122) had one organism isolated at presentation. The pattern of organisms isolated was similar for the two treatment groups (table 3) and
for the two indications for treatment, with the exceptions that
M catarrhalis was isolated more frequently with acute exacerbation of chronic bronchitis (17 of 58 patients [29%], vs. 9 of
64 [14%] with pneumonia) and Streptococcus pneumoniae was
isolated more frequently with pneumonia (28 of 64 patients
[44%], vs. 13 of 58 [22%] with acute exacerbation of chronic
bronchitis).
There were no significant differences between treatment
groups in the frequency of isolation of S. pneumoniae, Staphylococcus aureus, or gram-negative bacilli (P > .05). All isolates
of the four most common pretherapy organisms (S. pneumoniae, H. influenzae, M catarrhalis, and S. aureus) were susceptible to amoxicillin/clavulanate. Three of 31 H. influenzae isolates (10%) and 19 of 26 M catarrhalis isolates (73%) were
ampicillin-resistant and 0-lactamase-positive. All S. pneumoniae isolates were susceptible to ampicillin.
Bacteriologic success (proven eradication, presumed
eradication, or colonization) was similar in the two groups:
97% with the every-12-hours regimen and 91% with the
Table 3. Bacteria isolated from the 122 patients who were evaluable
for bacteriologic efficacy of the regimen.
No. (%) of isolates per amoxicillin/
clavulanate dosing schedule
Isolate(s)
Streptococcus pneumoniae
Staphylococcus aureus
Streptococcus group B
Other streptococci
Haemophilus influenzae
Moraxella catarrhalis
Citrobacter species
Escherichia coli
Enterobacter species
Haemophilus
parainfluenzae
Proteus mirabilis
Pseudomonas aeruginosa
Serratia marcescens
Pseudomonas fluorescens
Other gram-negative bacilli
Every 12 hours
19
8
0
2
16
13
0
2
3
(3)
(4)
2
1
1
1
1
2
(3)
(1.4)
(1.4)
(1.4)
(1.4)
(3)
(27)
(11)
(2.8)
(23)
(18)
Every 8 hours
22
10
1
2
15
13
2
0
0
(32)
(15)
(1.5)
(3)
(22)
(19)
(3)
1 (1.5)
0
1 (1.5)
0
0
1 (1.5)
573
every-8-hours regimen (P = .86; 95% CI for the difference,
—2.5%-14.7%). Colonization was observed in five patients
in the every-12-hours group and no patients in the
every-8-hours group (this difference was not tested since it
was a subgroup analysis that was not included in the protocol design). The responses were similar when patients with
pneumonia and those with acute exacerbation of chronic
bronchitis were considered separately. At the follow-up
visit, bacteriologic success was maintained in 95% and 87%
of patients in the 12-hour and 8-hour groups, respectively
(P = .15).
Adverse events. One hundred and thirty-two of 557 patients
(24%) reported 179 adverse events related or possibly related
to the study medication (P = .39 between groups). The most
common adverse event was diarrhea (11% of patients). Most
of the diarrhea was mild (defined as easily tolerated by the
patient, causing minimal discomfort, and not interfering with
everyday activities).
Of all patients who reported moderate or severe diarrhea
(moderate diarrhea was defined as sufficiently discomforting
to interfere with normal everyday activities, while severe diarrhea was defined as incapacitating enough to prevent everyday
activities), 50% fewer were in the 12-hour group (2.9%) than
in the 8-hour group (4.9%) (P = .28). Thirty-eight patients
reported 58 adverse events that led to withdrawal from the
study (5.9% of the 12-hour group and 7.7% of the 8-hour
group; P = .38). Gastrointestinal disturbances were the most
common adverse events leading to withdrawal.
Intent-to-treat analysis. The two regimens were similar
when assessed by intent-to-treat analysis of 557 patients. Clinical success at the end of therapy was achieved in 230 of 273
patients (84%) who received treatment every 12 hours and in
236 of 284 (83%) who received it every 8 hours (P = .71;
95% CI for the difference, —5.0-7.3).
Discussion
A major determinant of the success of an antimicrobial regimen in an outpatient setting is compliance, which is encouraged
when a regimen is simplified to less-frequent daily dosing.
Amoxicillin/clavulanate has normally been administered as a
regimen of 500/125 mg (amoxicillin/clavulanate) every 8 hours
for adults. In this regimen, the combination is as effective for
treatment of lower respiratory tract infections as other agents
such as loracarbef [6, 7], cefuroxime [8], and azithromycin
[9, 10].
The pharmacodynamic properties of amoxicillin/clavulanate
supported consideration of every-12-hours dosing. Bacterial
killing by 0-lactam antimicrobials is related to the duration of
time the plasma concentration of the 0-lactam antimicrobial
remains above the MIC [11] . A study of healthy volunteers
demonstrated that amoxicillin/clavulanate given at a dosage of
875/125 mg every 12 hours resulted in a time above the MIC
574
Calver et al.
comparable to that of a dosage of 500/125 mg given every 8
hours [12].
Within the anticipated range of MICs for organisms causing
lower respiratory tract infection, the time above MIC between
the two regimens would be comparable. Although compliance
may be influenced by every-12-hours dosing, it is unlikely that
the every-12-hours regimen will increase antimicrobial activity
or the time above the MIC.
This trial documents the effectiveness of an every-12-hours
regimen of amoxicillin/clavulanate (875/125 mg) in direct comparison with the standard every-8-hours regimen (500/125 mg)
in patients with community-acquired pneumonia or acute exacerbation of chronic bronchitis, including patients infected with
fi-lactamase-producing pathogens. The numbers of enrolled patients in each of the two disease categories may not have been
sufficient to detect differences between treatment groups in
each category.
As with most recent reports of community-acquired lower
respiratory tract infections, the majority of pathogens isolated
were organisms other than S. pneumoniae. M catarrhalis accounted for 19% of isolates, and 81% of these were /3-lactamase-positive, while H. influenzae accounted for 22% of isolates, and 13% of those were /3-lactamase-positive.
A major difficulty in treating community-acquired pneumonia is that pathogens in -50% of patients cannot be identified
[3, 13], and therapy must be initiated prior to identification
even when this is possible. Antimicrobials effective against
potentially resistant pathogens must therefore be used for empirical treatment.
The most common pathogens identified in studies of outpatients with community-acquired pneumonia were S. pneumoniae and H. influenzae; S. aureus, gram-negative bacilli,
M. pneumoniae, Legionella species, and Chlamydia species
are identified with varying frequency [14]. In studies of acute
exacerbation of chronic bronchitis, M catarrhalis has gained
prominence, particularly in the elderly and in patients with
chronic lung diseases [15]; this pathogen is usually (3-lactamase-producing, as are up to 40% of strains of H. influenzae [16].
In conclusion, amoxicillin/clavulanate (875/125 mg)
given every 12 hours is at least as effective and as safe as
that given every 8 hours (500/125 mg) for the treatment of
community-acquired lower respiratory tract infection. Reduced frequency of administration with the every-12-hours
regimen should improve compliance with treatment as well
as decrease the incidence of moderate to severe diarrhea.
The per-day cost of the 12-hour regimen is comparable to
that of the 8-hour regimen.
CID 1997; 24 (April)
Acknowledgment
The authors thank Lorna Piora for writing the programs used
to analyze the data for the manuscript.
References
1. Garibaldi RA. Epidemiology of community-acquired respiratory tract infections in adults. Incidence, etiology and impact. Am J Med 1985;
78(suppl 6B):32-7.
2. Pomilla PV, Brown RB. Outpatient treatment of community-acquired
pneumonia in adults. Arch Intern Med 1994;154:1793-802.
3. Marrie TJ. Community-acquired pneumonia. Clin Infect Dis 1994;18:
501-13.
4. Sue DY. Community-acquired pneumonia in adults. West J Med 1994;
161:383-9.
5. Augmentin prescribing information. Philadelphia: SmithKline Beecham
Pharmaceuticals, 1996.
6. Dere WH, Farlow D, Therasse DG, Jacobson KD, Guerra FJ. Loracarbef
(LY163892) versus amoxicillin/clavulanate in the treatment of acute
purulent bacterial bronchitis. Clin Ther 1992;14:166-77.
7. Zeckel ML. Loracarbef (LY163892) in the treatment of acute exacerbations of chronic bronchitis: results of U.S. and European comparative
clinical trials. Am J Med 1992; 92(suppl 6a):58S-64S.
8. Brambilla C, Kastanakis S, Knight S, Cunningham K. Cefuroxime and
cefuroxime axetil versus amoxicillin plus clavulanic acid in the treatment of lower respiratory tract infections. Eur J Clin Microbiol Infect
Dis 1992;11:118-24.
9. Hoepelman AIM, Sips AP, van Helmond JLM, et al. A single-blind comparison of three-day azithromycin and ten-day co-amoxiclav treatment
of acute lower respiratory tract infections. J Antimicrob Chemother
1993; 31(suppl E):147 -52.
10. Balmes P, Clerc G, Dupont B, Labram C, Pariente R, Poirier R. Comparative study of azithromycin and amoxicillin/clavulanic acid in the treatment of lower respiratory tract infections. Eur J Clin Microbiol Infect
Dis 1991;10:437-9.
11. Drusano GL. Role of pharmacokinetics in the outcome of infections. Antimicrob Agents Chemother 1988; 32:289-97.
12. Hust MR. Comparison of the 24-hour pharmacokinetic profile of oral
Augmentin administered with food to healthy male volunteers as 1 g 12
hourly versus 625 mg 8 hourly, and as 625 mg 12 hourly versus 375
mg 8 hourly [report HH-1001/BRL-025000/2/CPMS-360]. Philadelphia: SmithKline Beecham Pharmaceuticals, 24 May 1994.
13. Marrie TJ. New aspects of old pathogens of pneumonia. Med Clin North
Am 1994;78:987-95.
14. Niederman MS, Bass JB Jr, Campbell GD, et al. Guidelines for the initial
management of adults with community-acquired pneumonia: diagnosis,
assessment of severity, and initial antimicrobial therapy. Am Rev Respir
Dis 1993;148:1418-26.
15. Wright PW, Wallace RJ Jr, Shepherd JR. A descriptive study of 42 cases
of Branhamella catarrhalis pneumonia. Am J Med 1990; 88(suppl 5A):
2S-8S.
16. Pichichero ME. Assessing the treatment alternatives for acute otitis media.
Pediatr Infect Dis J 1994; 13(suppl 1):S27-34.