A Comparative Study of Levofloxacin and
Ceftriaxone in the Treatment of Hospitalized
Patients with Pneumonia
S. RAGNAR NORRBY1, WOLFGANG PETERMANN2, PAUL ANTHONY WILLCOX3
NORBERT VETTER4 and ERICH SALEWSKI5
From the 1Department of Infectious Disease, University of Lund, Lund, Sweden, 2Innere Abteilung. Brüderkrankenhaus, St. Josef,
Paderborn, Germany, the 3Respiratory Clinic, Groote Schuur Hospital, Cape Town, South Africa, the 4Pulmologisches Zentrum
der Stadt Wien, Vienna, Austria and the 5Ήoechst Marion Roussel Deutschland, Clinical Development, Frankfurt, Germany
A multinational, multicentre, open, randomised study in hospitalised patients with pneumonia compared lerofloxacin 500 mg
twice daily with ceftríaxone 4 g i.v. once daily. Levofloxacin patients started on i.v. treatment and switched to oral on d 3-5
of therapy if signs and symptoms had improved. The minimum treatment duration was 5 d, except for treatment failure, and
the median 8 d. The primary efficacy analysis was based on the per-protocol assessment of the clinical cure rate determined
2-5 d after the end of treatment in the per-protocol (PP) population (levofloxacin 127, ceftríaxone 139). Of 625 patients
enrolled and randomized, 6 received no treatment, giving an intention-to-treat (ITT) population of 619 (levofloxacin 314,
ceftríaxone 305). At the clinical endpoint, 2-5 d after the end of treatment, the cure rates for levofloxacin and ceftríaxone
were similar in both the ITT (76% and 75%, respectively) and PP (87% and 86%, respectively) populations. Both drugs were
well tolerated. Twice-daily levofloxacin 500 mg, either i.v. or as sequential iv/oral therapy, was as effective as i.v. once-daily
ceftríaxone 4 g in the treatment of hospitalized patients with pneumonia and offers the advantage of sequential therapy.
S. R. Norrby, MD, University of Lund, Department of Infectious Diseases, University Hospital of Lund, S -221 85 Lund,
Sweden
INTRODUCTION
Levofloxacin, a new fluoroquinolone, is the L-isomer of the
racemate ofloxacin. The antibacterial action of ofloxacin
resides almost entirely in the L-isomer and levofloxacin
appears to be twice as active as the racemate ofloxacin in
vitro (1). Levofloxacin exerts antibacterial activity via antagonism of the interaction between bacterial DNA gyrase
and DNA.
Levofloxacin exhibits a broad spectrum of activity, including Gram-positive aerobic organisms such as Streptococcus pneumoniae and Staphylococcus aureus, Gramnegative bacteria such as Escherichia coli, Moraxella
catarrhalis, Haemophilus influenzae, Klebsiella pneumoniae
and Pseudomonas aeruginosa and intracellular pathogens
responsible for atypical pneumonia such as Chlamydia
pneumoniae, Legionella pneumophila and Mycoplasma
pneumoniae (1-3). Levofloxacin has been shown to be
more active against S. pneumoniae in vitro and in vivo in
an animal model than either ofloxacin or ciprofloxacin (4,
5).
Orally administered levofloxacin is rapidly absorbed,
with a bioavailability around 100%. Following oral or
intravenous administration it is eliminated relatively slowly,
with a terminal half-life of 6-8 h, primarily by renal
excretion (6). Levofloxacin is not metabolized to any clinically relevant extent.
Most cases of pneumonia are community-acquired (7).
Compared with younger patients, community-acquired
© 1998 Scandinavian University Press. ISSN 0036-5548
pneumonia (CAP) in the elderly has a greater incidence of
co-morbidity, and higher rates of hospitalisation and mortality.
Streptococcus pneumoniae, H. influenzae and M. pneumoniae are the most common bacterial pathogens in CAP
(8, 9). Staphylococcus aureus and Gram-negative bacteria
are recovered from sputum specimens in approximately 310% of cases although their causative role is debatable (7).
During the 1980s, M. catarrhalis, Legionella spp. and C.
pneumoniae emerged as significant pathogens and a
recent study by Márrie et al., in ambulatory patients with
CAP, showed that nearly half the cases were due to 'atypical' bacteria (10, 11). Up to 10% of CAP may have a
multiple aetiology with S. pneumoniae and H. influenzae
being the most common combination.
Pneumonia is a serious hospital-acquired infection, particularly amongst critically ill patients. Crude mortality
rates range from 50% to 70%, although only one-third of
these deaths are directly attributed to the pneumonia (12).
Hospital-acquired pneumonia or pneumonia in immunocompromised patients is frequently caused by Gram-negative
bacilli such as P. aeruginosa. Anaerobic bacteria are most
likely to be found in patients with aspiration pneumonia, lung
abscess or empyema. In the hospital environment, the
likelihood of colonisation by coliforms may be increased
by serious underlying diseases, confinement in an intensive
care unit, tracheostomy, use of endotracheal tubes,
contaminated respiratory equipment, and antimicrobial
treatment which may select resistant organisms.
The present study was designed to compare levofloxacin,
either i.v. or sequential i.v. to oral, with i.v. ceftriaxone in
the treatment of hospitalised patients with pneumonia.
Ceftriaxone has been used extensively in patients with
pneumonia, either alone or in combination with other
antibiotics (7. 13-15).
PATIENTS AND METHODS
Study design
This was a multinational (13 countries), multicentre (64 centres),
open, randomised, comparative study in hospitalised patients
with pneumonia of levofloxacin (Hoechst AG, Frankfurt. Germany) 500 mg twice daily versus ceftriaxone (F. Hoffmann-La
Roche Ltd., Basel, Switzerland) 4 g i.v. once daily. Randomization was centralized using a computerized telephone system (G.
H. Besselaar Associates, Belgium) and a 1:1 allocation to
levofloxacin or ceftriaxone. Patients in the levofloxacin group
were started on i.v. treatment and switched to oral on d 3-5 of
therapy, after a minimum of 4 i.v. doses, if clinical signs and
symptoms of pneumonia had improved. The total duration of
treatment was at the discretion of the investigator although the
minimum treatment period was 5 d unless there was a treatment
failure. However, it was recommended that treatment be continued for 2 d after fever subsided. The primary efficacy analysis
was based on the protocol assessment of the clinical cure rate
determined 2-5 d after the end of treatment in the per-protocol
population. Three types of assessments were performed according
to the study protocol: protocol, conducted by computer (this was
the primary assessment); investigator, conducted by the investigator
and evaluator-blinded, conducted by an external eval-uator for
selected patients using blinded data.
The study was conducted in accordance with the Good Clinical Practice guidelines of the European Community and the
Declaration of Helsinki. The study protocol was approved by
the local ethics committee for each study centre and informed
consent was obtained from each patient.
Patients
Male or female hospitalized patients, aged > 18 y (21 y in parts of
Germany), were eligible for inclusion in the study if they had
clinical signs and symptoms of bacterial pneumonia, either nosocomial or CAP and chest X-ray findings consistent with the
clinical diagnosis of pneumonia. Clinical signs and symptoms
included fever, chills, tachypnoea, tachycardia, pleuritic chest
pain, cough productive of mucopurulent sputum, leucocytosis,
dullness over the involved lobe or segment, diminished breath
sounds, bronchial breath sounds, rales and pleural friction rubs.
Patients < 65 y of age had to have underlying disease(s) and/or
risk factor(s) which increased the risk of acquiring pneumonia or
which could complicate the clinical course. Nosocomial pneumonia
was defined according to the CDC definition (16). CAP was
defined as an infection that occurred outside the hospital environment. Exclusion criteria included: pneumonia strongly suspected to be caused by mycoplasma, chlamydia or legionella;
history of epilepsy: pregnancy or nursing mothers: severe hepatic
disease; renal impairment (creatinine clearance < 20 ml/min); antibiotic treatment for > 24 h in the 5 d prior to study entry;
azithromycin in the 7 d prior to study entry; ofloxacin or ceftriaxone for this infectious episode; any invcstigational drug in the 4
weeks prior to study entry or during the study period; hypersensitivity to fluoroquinolones or ceftriaxone.
Efficacy analysis
The intention-to-treat (ITT) analysis was performed on the clinical
response in all patients who received at least one dose of study
drug. The per-protocol analysis, which was the primary analysis,
was performed on the clinical and bacteriological response in all
patients with clinical signs and symptoms of pneumonia, consistent
X-ray findings and bacteriologically proven infection as treated,
excluding major protocol violators. The clinical response was
assessed at endpoint (2-5 d after the end of therapy), at change
from i.v. to oral levofloxacin and at follow-up (14-21 d after the
end of therapy).
The clinical response was defined as: cure (all infection-related
signs and symptoms disappeared or returned to preinfection state
and chest X-ray findings at least improved; at least one infectionrelated sign or symptom had improved, follow-up chest X-ray had
at least improved or was unchanged and no subsequent antibiotic
treatment was started); failure (all infection-related signs and symptoms unchanged or worsened; patient developed new clinical findings consistent with active infection which did not improve or
disappear at clinical endpoint; patient died due to the infection;
study drug discontinued due to clinical and/or bacteriological
treatment failure; one or more antibiotics were added to the study
drug due to treatment failure) or indeterminate (circumstances
precluded classification as cure or failure e.g. missing follow-up
data, premature discontinuation for nonefficacy-related reasons or
major protocol violation (incorrect entry diagnosis; insufficient
treatment duration; non-compliance; antibiotic pre-treatment; additional antibiotic treatment; clinical post-treatment evaluation
missing or out of time window); patient received a subsequent
antibiotic for an infection other than the indication).
The bacteriological response was evaluated from the clinical
endpoint culture (2-5 d after the end of treatment) as: satisfactory
(eradication or presumed eradication [negative culture] of pathogen; colonization with a different pathogen without any clinical
evidence of infection); unsatisfactory (persistence of baseline pathogen; relapse; superinfection; eradication and reinfection; persistence
and reinfection; presumed persistence; resistance) or indeterminate
(lack of subsequent culture due to withdrawal, death etc.; treatment with a systemic antibiotic in addition to the study drug).
Sputum cultures were set up within 48 h before the start of
treatment, on d 2 or 3 of therapy and at each efficacy evaluation for
the isolation, identification and susceptibility testing of the
responsible pathogen(s). In addition, at least 2 sets of blood
cultures (aerobic and anaerobic) were set up prior to the start of
treatment. Blood cultures were repeated after 48 h treatment in
patients with persistent fever but were optional in patients whose
temperature was < 37.8°C.
Statistical analysis
The primary efficacy evaluation was based on the clinical cure rate
in the per-protocol population with bacteriologically proven infection at the clinical endpoint. Assuming a success rate of 80% in
both treatment arms and a δ of 15% (maximum difference between
treatments to be accepted as equivalent), 125 evaluable patients per
group were needed to provide a 84.6% chance of showing equivalence (power = 84.6%) by calculating a 2-sided 95% confidence
interval (CI) for the difference in cure rates (17. 18). If the upper
bound was greater than 0 and the lower bound greater than
— 0.15, levofloxacin was considered to be as effective as ceftriaxone. The Wilcoxon rank test was used to evaluate the time between
start of treatment and discharge from hospital. Patients who died
in hospital were assigned the longest duration of time observed for
any patient. It was assumed that 33% of the patients treated would
be evaluable for efficacy and, therefore, a total of 750 patients
would be required. However, the study was closed with < 750
enrolled patients because the evaluability during the trial was 46%.
RESULTS
Fig. 1 shows the distribution of patients over the different
populations for analysis. 625 patients were enrolled and
randomized (619 treated) at 64 study centres in 13 countries. No treatment was given in 6 cases, giving an ITT
population of 619 patients, 314 assigned to levofloxacin and
305 to ceftriaxone. Patient demographics were similar in
both treatment groups, with 65% of patients being male.
The median age was 65 y (range 18-94 y), 51% (315/619)
were elderly (65 and 80% (495/619) were Caucasian. The
median weight was 66 kg (range 36-129 kg) and the
median body mass index was 23.5 kg/m 2 (range 13-51
kg/m). Forty per cent (247/619) were ex-smokers and 34%
(209/619) were non-smokers. For 94% (582/619) of patients, the diagnosis was CAP and nosocomial pneumonia
was diagnosed in the remaining 6% (37/619).
A history of disease and concomitant illnesses was reported in 93% of patients (575/619), with respiratory (330
patients, 53%) and cardiovascular (276 patients, 45%) diseases the most common. Chronic obstructive airways disease (198 patients, 32%) and emphysema (87 patients, 14%)
were the most common respiratory diseases. Drug/alcohol
abuse was reported in 14% of patients. One hundred and
thirty-three patients (21%) had an APACHE II score > 15.
Bronchopneumonia was diagnosed in 40% (249/619), lobar
pneumonia in 54% (336/619) and in 6% of patients (35) the
type of pneumonia was not specified. At admission, disease
severity was assessed as moderately severe in 70% (436/619)
and severe in 20% (125/619) of patients. Pretreatment with
systemic antibiotics during the 5 d before study onset was
reported for 25% of patients (154/619) and 96% of patients
received at least one concomitant medication during the
study.
The median duration of treatment was 9 d for
levofloxacin and 8 d for ceftriaxone patients in the ITT
population (NS). A total of 89 patients [45/319 (14%)
levofloxacin, 44/306 (14%) ceftriaxone] did not complete
the study and were withdrawn for various reasons (Table
I). More patients in the ceftriaxone group (χ 2 = 3.9, p =
0.05) (22) withdrew due to clinical failure than in the
levofloxacin group (14). Sequential treatment, i.e. change
from i.v. to oral, was administered in 274/314 (87%)
levofloxacin patients with a median duration of 4 d on i.v.
and 5 d on oral treatment. The cure rate for levofloxacin
(76%) was similar to that for ceftriaxone (75%) (Table II).
The cure rates in both treatment groups were lower than in
the per-protocol population due to the higher number of
indeterminate cases (mainly due to protocol violations).
Levofloxacin was as effective as ceftriaxone (95% Cl 5.7%; + 7.8%).
The per-protocol population, patients with a proven
bacteriological infection at baseline (isolation of a responsible pathogen from a baseline culture) excluding major
protocol violators (40 levofloxacin, 26 ceftriaxone), consisted of 272 patients (130 levofloxacin, 142 ceftriaxone),
44% of the ITT population. The clinical response was
indeterminate in 6 patients and, therefore, the total
analysed was 266 (127 levofloxacin, 139 ceftriaxone). At the
clinical endpoint, the cure rate for levofloxacin (87%) was
similar to that for ceftriaxone (86%) and the 95% Cl
indicates that levofloxacin was as effective as ceftriaxone
(Table II).
Clinical response at endpoint was also assessed by the
investigator. Cure rates for the 2 drugs were similar for the
ITT population (levofloxacin, 87%; ceftriaxone, 85%) and
the per-protocol population (levofloxacin, 87%; ceftriaxone,
89%). The protocol and investigator assessments of clinical
response were in agreement for 456 of 502 patients (91%) in
the ITT population and for 225 of 240 patients (94%) in the
per-protocol population who had both assessments.
236 patients were selected for assessment by an external,
blinded evaluator, of which two-thirds were all the patients
who had differing outcomes in the protocol (i.e. by computer) and investigator assessments. The remaining third
were patients randomly selected from those classified as
cured in both of these assessments. The protocol and
evaluator-blinded assessments of clinical response were in
agreement in 75% (177/236) of ITT patients and 85%
(75/88) of per-protocol patients.
A total of 351 bacterial isolates, considered responsible
for infection by the investigator, were isolated at baseline.
The most common pathogens (> 5 isolates) are shown in
Table III. The majority of the pathogens isolated were
susceptible to both study drugs but 3/332 (1%) tested
isolates were resistant to levofloxacin (I S. pneumoniae and
2 P. aeruginosa) and 17/330 (5%) were resistant to ceftriaxone (10 P. aeruginosa, 1 P. fluorescens, 1 Pseudomonas
spp., 1 H. influenzae, I K. oxytoca, 1 Enterococcus spp., 1
E. faecalis and 1 Enterobacter cloacae). Similar results were
found for isolates of patients assigned to each treatment
group (levofloxacin, 1/157, l%; ceftriaxone, 8/174, 5%). In
the per-protocol population, 26 (20%) levofloxacin patients
and 30 (21%) ceftriaxone patients had bacteraemia.
In the per-protocol bacteriological efficacy analysis at
clinical endpoint, 61/130 levofloxacin patients and 57/142
ceftriaxone patients had an indeterminate bacteriological
response and, therefore, the total number of patients
analysed for bacteriological response was 69 and 85, respectively. Of the patients analysed, 57/69 levofloxacin (83%)
and 71/85 ceftriaxone (83%) patients had a satisfactory
bacteriological response. Persistent or presumed persistent
pathogens were the main reason for an unsatisfactory
response (levofloxacin 10, ceftriaxone 11). The percentage
difference in the comparison of the two treatments was 0.9% and the 95% Cl (- 12.8%; + 11.0%) indicates that
levofloxacin was as effective as ceftriaxone. The bacteriological and clinical responses were in agreement for 147/153
patients (96%) who had evaluable data for both assessments. Four patients had a satisfactory bacteriological response but a failed clinical response (levofloxacin 2,
ceftriaxone 2) and 2 patients, both ceftriaxone, had an
unsatisfactory bacteriological response but a satisfactory
clinical response (cure). The remaining patients had an
indeterminate bacteriological or clinical response, including
105 patients with a clinical response of cure but an indeterminate bacteriological response (levofloxacin 56, ceftriaxone 49).
Eradication rates for the most common pathogens are
shown in Table IV. The overall eradication rates were equal
for both treatments (87%). The eradication rate for
levofloxacin was greater than for ceftriaxone for Gram-negative pathogens (96% versus 88%) but lower for Gram-positive pathogens (76% versus 85%). However, the number of
pathogens per treatment group was low relative to the
number of patients treated and this must be taken into
account when comparing these rates.
There were 5 pathogens which persisted at clinical endpoint in 4 patients [levofloxacin: 3 (S. pneumoniae, Group
A beta-haemolytic streptococcus and S. aureus) in 2 patients; ceftriaxone: 2 (P. aeruginosa and Acinetobacter baumannii) in 2 patients]. Only the P. aeruginosa isolate in the
ceftriaxone group was resistant to the study drug in the
post-treatment culture. 95 new pathogens (levofloxacin 35,
ceftriaxone 60) were isolated in the ITT population during
the study. Gram-negative isolates were more prevalent in
the ceftriaxone group (30 vs 9). 39 of the 57 pathogens
tested (68%) were susceptible to the assigned study drug.
The clinical response of patients designated failure or
indeterminate at clinical endpoint was carried over to the
follow-up assessment. Patients who received subsequent
antibiotics for clinical or bacteriological failure from d 6
post-treatment onwards were classed as failures. For both
treatment groups and both populations considered, cure
rates at follow-up were approximately 3-5% lower than at
clinical endpoint (Table II).
In the ITT population, 274/314 patients (87%) switched
from i.v. to oral levofloxacin. Of these, 217 patients (79%)
changed on d 2-5 of i.v. treatment and the median time to
change was d 4 of therapy. The clinical response at time of
change was determined as cure in 66% of patients (164 of
250 analysed) and improved in 32% (80). Similar results,
64% and 32%, respectively, were obtained in the per-protocol population analysis. No significant difference in length
of hospital stay was detected between treatment groups,
with a median duration of 12 d for both groups in the ITT
population and 11 d for the levofloxacin group and 13 d for
the ceftriaxon group in the per-protocol population.
The safety profile of levofloxacin was similar to that of
ceftríaxone and both drugs were well tolerated. Adverse
events were reported in 334/619 patients (169/314 [54%]
levofloxacin, 165/305 [54%] ceftriaxone). The most frequently affected body systems were: body as a whole,
cardiovascular system, digestive system, metabolic and nutritional system, nervous system and respiratory system (see
Table V). The incidence of adverse events possibly associated with study medication was 68/314 (22%) for
levofloxacin and 79/305 (26%) for ceftriaxone (see Table V).
There were no significant differences between the treatment
groups. Serious adverse events (see Table VI), irrespective
of relationship to study drug, were reported in 55/314
levofloxacin patients (18%) and 52/305 ceftriaxone patients
(17%). Frequencies were generally similar apart from cardiovascular events, which occurred more frequently in the
levofloxacin group (28/314, 9% vs 14/305, 5%). Events
which contributed to this difference were heart failure (7
levofloxacin, 4 ceftriaxone); congestive heart failure (3, 0);
arrhythmia (2, 0) and tachycardia (2, 0). None of these
events were considered to be drug related and all were
thought to be due to pre-existing disease. Study treatment
was permanently discontinued due to adverse events in
15/314 levofloxacin patients (5%) and 20/305 ceftriaxone
patients (7%).
There were 43 deaths among the patients who were
randomised and treated: 21/314 (7%) levofloxacin and 22/
305 (7%) ceftriaxone patients. Fifteen deaths occurred during treatment (7 levofloxacin, 8 ceftriaxone), 14 posttreatment to 14-d follow-up (10 levofloxacin, 4 ceftriaxone)
and 14 after the 14-d follow-up (4 levofloxacin, 10
ceftriaxone). Cause of death was related to the patients'
severe pre-existing conditions and none were considered
related to the study drug. There was no significant difference between the death rates in both treatment arms.
DISCUSSION
The results indicate that levofloxacin, either i.v. or as
sequential iv/oral treatment, was as clinically effective as
i.v. ceftriaxone in the treatment of hospitalised patients
with pneumonia (cure rates of 87% and 86%, respectively,
in the per-protocol analysis at clinical endpoint). The study
population, pneumonia assessed as moderately severe in
70% and severe in 20% of patients at admission, and the
most common baseline pathogens isolated, S. pneumoniae
(36%), H. influenzae (21%) and M. catarrhalis (8%), was
representative of a hospital population with pneumonia.
Notably, 7% of the patients died due to their infections
and/or underlying conditions. As expected, cure rates in the
ITT population were lower but equally so for both treatment groups so the conclusion for the ITT and per-protocol populations was the same. Equivalence of treatment
was further confirmed by comparison of the investigator
and protocol assessments and by an evaluator-blinded assessment. Clinical cure rates at follow-up were also similar
for both treatment groups in the per-protocol population
(83% for levofloxacin and 81% for ceftriaxone).
There have been relatively few comparative studies on
the efficacy of fluoroquinolone monotherapy in pneumonia,
particularly in patients with moderately severe or severe
pneumonia. This is probably due to the poor activity of
older fluoroquinolones, such as ofloxacin and ciprofloxacin,
against S. pneumoniae, the most common pathogen.
Plouffe et al. showed that ofloxacin was as effective as
standard therapy (ß-lactam alone or plus a macrolide) in
patients hospitalized for CAP (clinical success rate 92% and
87%, respectively) (19). A comparison of ciprofloxacin and
imipenem-cilastin, in 405 patients with severe pneumonia,
mainly of nosocomial origin, obtained a good clinical and
bacteriological response in approximately 70% of patients
(20). In a comparative study of oral sparfloxacin as
monotherapy versus oral amoxycillin plus ofloxacin as
combination therapy in 211 hospitalized patients with
CAP. the overall efficacy rates were 92% and 82%, respectively (21). Patients in this study were either elderly or had
failed a previous course of antibacterial therapy and the
main pathogens isolated were S. pneumoniae, H. influenzae, M. pneumoniae, S. aureus and Enterobacteriaceae.
In an open, non-comparative study, i.v. or oral
levofloxacin once daily was highly effective in mild/moderate
and severe CAP and all the pathogens isolated (S.
pneumoniae, H. influenzae, S. aureus and C. pneumoniae)
were eradicated (22). The results of our study are simitar to
those of File et al. who compared i.v. and/or oral
levofloxacin once daily with ceftriaxone and/or cefuroxime
axetil in an open study in 590 patients, treated in hospital
or as outpatients, with mild/moderate or severe CAP (23).
However, they showed that levofloxacin was superior to the
comparator(s) in terms of both clinical success rates (96%
vs 90%) and bacterial eradication rates (98% vs 85%).
In our study, the baseline pathogens were generally more
often susceptible to levofloxacin than to ceftriaxone. Only
0.9% (3/332) were resistant to levofloxacin compared to
5.2% (17/330) resistant to ceftriaxone, this difference being
mainly due to ceftriaxone-resistant Gram-negative pathogens such as P. aeruginosa. However, this difference did not
affect the overall bacteriological success rates which were
similar for both treatments (levofloxacin 84%, ceftriaxone
83%). Superinfections were rare, 8 in 7 patients
(levofloxacin 3, ceftriaxone 4), and there were no reinfections.
Both drugs were well tolerated and the majority of
adverse events were associated with the severe course of the
infectious disease itself. The difference in the incidence of
cardiovascular serious adverse events between the
levofloxacin and ceftriaxone groups was attributable to the
patients' underlying disease and was not related to the
study drug. The mortality rate was similar in both treatment groups. The incidence of possible drug-related events,
which is higher than in some studies, may in part be due to
the study protocol requiring a possible rather than a probable causal relationship to the study drug and/or the severity
of the patients' underlying disease (19, 23).
Levofloxacìn offers the advantage over ceftriaxone of
sequential or switch therapy from i.v. to oral without the
need to change the treatment drug. The results of this study
indicate that sequential i.v. to oral levofloxacin was as
effective as i.v. ceftriaxone. This provides an important
benefit in terms of treatment costs. There is a growing
pressure in most countries to limit the rise in healthcare
costs which, in the hospital environment, has resulted in a
trend towards reducing the length of time a patient spends in
hospital and the use of more cost-effective treatments. As part
of these changes, there is increasing use of switch, therapy
due to the advantages which oral administration offers over
i.v. therapy. These include reduced complications from i.v.
devices, reduced hospital stay and reduced treatment costs
(24-26).Several studies have reported the economic benefits
of switching from i.v. to oral treatment. Salewski et al.
showed a reduction in costs of approximately 50% when
switching from i.v. to oral treatment with ofloxacin or
ciprofloxacin in a multi-centre hospital study in patients
with a variety of infections (27). Garber et al.
reported an annual saving of nearly SC60000 in 1 hospital due
to switch therapy (28). In a review of the cost-effectiveness
of switch therapy, Jewesson reported savings of approximately SC85000 over a 4-y period at the Vancouver
General Hospital (29).
In conclusion, the results of this study show that, in
hospitalized patients with pneumonia, twice-daily
levofloxacin 500 mg, either i.v. or as sequential iv/oral
therapy, was as effective as i.v. once-daily ceftriaxone 4 g.
ACKNOWLEDGEMENTS
This study was supported by a grant from Hoechst Marion Roussel. The authors would like to acknowledge the contribution of the
following members of the International Study Group—Argentina:
Ambasch G, Ariza HA, Jasovich A. Mingrone HO, Montero JM,
Piovano CF, Salvarezza CR; Australia: Looke D; Belgium:
Claessens J-J, Gepts L, Jordens P; France: Bernard JP, Boutin C.
Brouqui PL, Laaban J-P, Muir JF, Weitzenblum E; Germany:
Altrogge G, Berger K, Beyer JH, Bodmann K, Ferünz R, Freytag
F, Geier FP. Greiner A, Henrich K, Herms W. Janzik U. Kirsch
WD, Matthys H, Mõller A, Sell G. Simon B, Thai W, Vallee D,
Weiss R, Wiesner B; Ireland: Clancy LJ. O'Neill S; Italy: Clini V.
Vagliasindi M, Velluti G; The Netherlands: Maesen F; South
Africa: Heynecke ML, Irusen EM, Mpe MJ, Naude GE, Theron
W, van Aswegen GL, van Zyl L, Wood R; Spain: Garcia Bragado
F, Pigrau Serrallach C, Segura F; Switzerland: Bernasconi E,
Braendli O, Frei A. Vìlliger B; UK: Britton MG, Murphy PG.
OΉickey SP.
REFERENCES
1. Davis R, Bryson HM. Levofloxacin. A review of its antibacte
rial activity, pharmacokinetics and therapeutic efficacy. Drugs
1994; 47: 677-700.
2. Barry AL, Fuchs PC, Allen SD. Brown SD, Jorgensen JH,
Tenover FC. In-vitro susceptibility of Streptococcus pneumo
niae to the D- and L-isomers of ofloxacin: interpretive criteria
and quality control limits. J Antimicrob Chemother 1996; 37:
365-9.
3. Fu KP, Lafredo SC, Foleno B, Isaacson DM, Barrett JF,
Tobia AJ, Rosenthale ME. In vitro and in vivo antibacterial
activities of levofloxacin (L-ofloxacin), an optically active
ofloxacin. Antimicrob Agents Chemother 1992; 36: 860-6.
4. Klugman KP, Capper T. Levofloxacin in-vitro activity and
synergistic activity in combination with other antibacterials
against antibiotic-resistant S. pneumoniae and selection of
resistant mutants. Proceedings of 35th ICAAC, San Francisco,
USA 1995; Abstract E9.
5. Markus A, Isert D, Klesel N, Seibert G. Killing activity of
levofloxacin and ciprofloxacin against Streptococcus pneumo
niae in vitro and in vivo. Proceedings of 35th ICAAC, San
Francisco. USA 1995; Abstract E30.
6. Fish DN, Chow AT. The clinical pharmacokinetics of
levofloxacin. Clin Pharmacokinet 1997; 32: 101-19.
7. Bartlett JG, Mundy LM. Community-acquired pneumonia. N
Engl J Med 1995; 333: 1618-24.
8. Pass RJ. Aetiology and treatment of community-acquired
pneumonia in adults: an historical perspective. J Antimicrob
Chemother 1993; 32 Suppl A: 17-27.
9. Finch RG. The role of new quinolones in the treatment of
respiratory tract infections. Drugs 1995; 49, Suppl 2: 144-51.
10. Fang G-D, Fine M, Orloff J, Arisumi D, Yu VL, Kapoor W.
et al. New and emerging etiologies for community-acquired
pneumonia with implications for therapy: a prospective multicenter study of 359 cases. Medicine 1990; 69: 307-16.
11. Marrie TJ, Peeling RW, Fine MJ. Ambulatory patients with
community-acquired pneumonia: the frequency of atypical
agents and clinical course. Am J Med 1996: 101: 508-15.
12. Craven DE. Prevention of hospital-acquired pneumonia: mea
suring effect in ounces, pounds, and tons. Ann Intern Med
1995; 122: 229-31.
13. Mangi RJ. Peccerillo K.M, Ryan J, Berenson C, Greco T,
Thornton G. Andriole VT. Cefoperazone versus ceftriaxone
monotherapy of nosocomial pneumonia. Diagn Microbiol In
fect Dis 1992; 15: 441-7.
14. Hirata-Dulas CA. Stein DJ, Guay DR. Gruninger RP, Peter
son PK. A randomized study of ciprofloxacin versus ceftriax
one in the treatment of nursing home-acquired lower
respiratory tract infections. J Am Geriatr Soc 1991; 39: 97985.
15. The British Thoracic Society, London. Guidelines for the
management of community-acquired pneumonia in adults ad
mitted to hospital. Br J Hosp Med 1993; 49: 346-50.
16. Garner JS, Jarvis WR, Emori TG, Horan TC. Hughes JM.
CDC definitions for nosocomial infections 1988. Am J Infect
Control 1988; 16: 128-40.
17. Harkins RD. Albrecht R. Design and analysis of clinical trials
for anti-infective drug products. Drug Infect J 1990; 24: 21324.
18. Makuch R, Simon R. Sample size requirements for evaluating
a conservative therapy. Cancer Treat Rep 1978; 62: 1037-40.
19. Plouffe JF, Herbert MT, File TM Jr, Baird I. Parsons JN,
Kahn JB, Rielly-Gauvin KT, and The Pneumonia Study
Group. Ofloxacin versus standard therapy in treatment of
community-acquired pneumonia requiring hospitalization. Antimicrob Agents Chemother 1996; 40: 1175-9.
20. Fink MP, Snydman DR, Niederman MS, Leeper KV, Johnson
RJ, Heard SO, Wunderink RG, Caldwell JW, Schentag JJ,
Siami GA, Zameck RL, Haverstock DC, Reinhart HH, Echols
RM, and The Severe Pneumonia Study Group. Treatment of
severe pneumonia in hospitalized patients: Results of a multicenter, randomized double-blind trial comparing intravenous
ciprofloxacin with imipenem-cilastin. Antimicrob Agents
Chemother 1994; 38: 547-57.
21. Portier H. May TH, Proust A. and The French Study Group.
Comparative efficacy of sparfloxacin in comparison with
amoxycillin plus ofloxacin in the treatment of community-ac
quired pneumonia. J Antimicrob Chemother 1996; 37, Suppl
A: 83-91.
22. Fogarty CM. A noncomparative study to evaluate the safety
and efficacy of levoñoxacin in the treatment of community-ac
quired pneumonia in adults. Proceedings of 19th ICC, Mon
treal, Canada 1995.
23. File TM. Segreti J, Dunbar L, Player R, Kohler R, Williams
RR, et al. A multicenter, randomized study comparing the
efficacy and safety of intravenous and/or oral levofloxacin
versus ceftriaxone and/or cefuroxime axetil in treatment of
adults with community-acquired- pneumonia. Antimicrob
Agents Chemother 1997; 41: 1965-72.
24. Conly JM, Low DE. Intravenous to oral antimicrobial stepdown therapy: improving the quality of care and reducing
costs. Can J Infect Dis 1995; 6 Suppl A: A3-5.
25. Louie TJ. Intravenous to oral stepdown antibiotic therapy:
another cost-effective strategy in an era of shrinking health
care dollars. Can J Infect Dis 1994; 5 Suppl C: C45-50.
26. Quintiliani R, Crowe HM, Nightingale C. Transitional antibi
otic therapy. Can J Infect Dis 1995; 6 Suppl A: A6-10.
27. Salewski E, Bassaris HP, Calangu S. Kitzes R;Kosmidis J,
Raz R, Kompan L. Cost-minimisation analysis of sequential
treatment with ofioxacin or ciprofloxacin in hospitalised pa
tients. Pharmacoecon 1997; 11: 359-66.
28. Garber GE, McLean W, Ticrney M, Page D. Reducing antibi
otic costs; the Ottawa General Hospital experience of cost
cutting, including intravenous to oral stepdown. Can J Infect
Dis 1995; 6, Suppl A: 17A.
29. Jewesson P. Cost-effectiveness and value of an i.v. switch.
Pharmacoecon 1994; 5 Suppl 2: 20-6.