JAC

Journal of Antimicrobial Chemotherapy (2000) 45, 315320

Evaluation of combined ceftriaxone and dexamethasone therapy in

experimental cephalosporin-resistant pneumococcal meningitis
C. Cabellosa*, J. Martnez-Lacasaa, F. Tubaub, A. Fernndeza, P. F. Viladricha, J. Liaresb
and F. Gudiola
a

Laboratory of Experimental Infection, Infectious Diseases Service; bMicrobiology Service, Ciutat Sanitria i
Universitria de Bellvitge, C/Feixa Llarga s/n, 08907 LHospitalet, Barcelona, Spain
The treatment of meningitis caused by strains of Streptococcus pneumoniae with decreased
susceptibility to third-generation cephalosporins is an increasingly frequent and difficult problem. In this study a rabbit model of meningitis was used to determine the efficacy of ceftriaxone
at different dosages, and to establish the effect of the addition of dexamethasone to the chemotherapeutic regimen. Groups of eight rabbits were inoculated with 106 cfu/mL of a cephalosporinresistant strain of S. pneumoniae (MIC of cefotaxime/ceftriaxone 2 mg/L). Eighteen hours after
inoculation, ceftriaxone (50 or 100 mg/kg/day) with or without dexamethasone (0.25 mg/kg/
day) was administered for a period of 48 h. The ceftriaxone dose of 50 mg/kg/day was not fully
effective in this model (therapeutic failure rate 28%). With a dose of 100 mg/kg/day there were
no therapeutic failures and all CSF cultures were below the level of detection at 48 h. CSF ceftriaxone concentrations, area under the timeconcentration curve and time above the MIC were
not significantly different with or without dexamethasone. However, concomitant use of dexamethasone resulted in higher CSF bacterial counts and a higher number of therapeutic failures
(57% with the 50 mg/kg/day dose and 28% with the 100 mg/kg/day dose). Increasing doses of
ceftriaxone might be an effective mode of therapy for meningitis caused by S. pneumoniae with
MIC
2 mg/L. However, in contrast to cephalosporin-sensitive cases, in cases caused by
ceftriaxone-resistant strains, concomitant use of dexamethasone was associated with a higher
failure rate even when a higher dosage of ceftriaxone was used.

Introduction
The increasing prevalence of penicillin- and cephalosporinresistant Streptococcus pneumoniae worldwide has changed
the empirical therapy of pneumococcal meningitis. Since
the MICs of cefotaxime and ceftriaxone for S. pneumoniae
were 24 times lower than that of penicillin G, these drugs
were considered to be the therapy of choice for meningitis
caused by organisms with intermediate penicillin resistance.14 Efficacy against pneumococcal strains with higher
levels of resistance is controversial. The use of ceftriaxone
combined with vancomycin or rifampicin in adults and children has been suggested as the best alternative,5 but no
clinical series has been published to date. Increasing the
dosage of cephalosporins in order to achieve higher CSF
antibiotic concentrations is an alternative strategy, and a
previous study described several cases that were cured with
high doses of cefotaxime.6 Dexamethasone has been shown
*Corresponding author. Tel:

to reduce inflammatory activity in experimental bacterial

meningitis and to reduce neurological sequelae in paediatric cases of Haemophilus influenzae and pneumococcal
meningitis,710 but its routine use as adjunctive therapy is
not yet widely accepted.10 In experimental models it has
been demonstrated that use of dexamethasone reduces
CSF concentrations of vancomycin significantly, and also
ceftriaxone (although not to a significant degree).11,12 The
aim of this study was to determine the efficacy of increased
doses of ceftriaxone in the therapy of meningitis caused by
S. pneumoniae with MICs
2 mg/L and to establish the
effect of dexamethasone on ceftriaxone therapy.

Materials and methods

The design of the rabbit meningitis model followed an
established protocol.13 Four groups of eight rabbits were

34-93-335-7011 ext. 2487; Fax: 34-93-260-7637; E-mail: ccabellos@csub.scs.es

315
2000 The British Society for Antimicrobial Chemotherapy

C. Cabellos et al.
used, as well as a control group which was inoculated but
not treated. Female New Zealand white rabbits weighing
2 kg were anaesthetized intramuscularly with 35 mg/kg
ketamine (Ketolar, Parke-Davis, El Prat de Llobregat,
Spain) and 5 mg/kg xylazine (Rompun, Bayer AG, Leverkusen, Germany), and a dental acrylic helmet was affixed
to each rabbits calvaria. Twenty-four hours later the
animals were anaesthetized again and placed in a stereotaxic frame. A spinal needle was introduced into the
cisterna magna, 200 L of CSF was withdrawn and 200 L
of 106 cfu/mL of a strain of S. pneumoniae belonging
to serotype 23 F (MIC of penicillin 4 mg/L; MIC of ceftriaxone 2 mg/L) was instilled into the subarachnoid space.
Eighteen hours later the rabbits were anaesthetized again
with urethane (Sigma Chemical Company, St Louis, MO,
USA) 1.75 g/kg subcutaneously and thiopental (Penthotal
sodico, Abbott Laboratories, Madrid, Spain) 5 mg/kg iv
and a baseline CSF sample was taken. Then an iv dose of
0.25 mg of dexamethasone (Fortecortin, Merck, Mollet
del Valls, Barcelona, Spain) or saline (Suero fisiolgico
Braun, Braun SA Rub, Barcelona, Spain) was administered, and 10 min later an iv dose of either 50 or 100 mg/kg
of ceftriaxone (Roche, Madrid, Spain) was administered
(serum ceftriaxone concentrations at 30 min after a 75
mg/kg dose are c. 200 mg/L 12 and serum ceftriaxone concentrations at 6090 min after a 125 mg/kg dose are also c.
200 mg/L 5). Dexamethasone was administered every 12 h
over a 48 h period (four doses) and ceftriaxone was administered every 24 h (two doses). Serial CSF samples were
taken at 2 (peak), 6, 24 (trough), 26 (peak) and 48 h
(trough). WBC counts, lactic acid concentration, direct and
quantitative bacterial cultures, ceftriaxone concentrations
in CSF and CSF bactericidal activities at trough and peak
time points were determined for each sample of CSF. WBC
counts were performed by optical microscopy with a
Neubauer chamber after RBC had been lysed with Turk
solution (made in house; 0.2% acetic acid and methylene
blue). CSF lactic acid concentrations were determined by
Lactate PAP (bioMrieux SA, Marcy lEtoile, France) and
read by spectrophotometer.
Serial ten-fold dilutions were performed to determine
bacterial counts at each time point (the detection limit with
this method was 102 cfu/mL, and a value of 0 (log 10 1)
was assigned to the first and subsequent sterile cultures).
MICs of penicillin and ceftriaxone were determined
by a microdilution method with cation-adjusted Mueller
Hinton broth (Difco Laboratories, Detroit, MI, USA)
supplemented with 3% lysed horse blood, with an inoculum of approximately 5 105 cfu/mL as recommended
by the NCCLS. 14 The MIC was defined as the lowest concentration of antibiotic for which no visible turbidity was
apparent.
CSF ceftriaxone concentrations were determined by an
agar disc diffusion method15 in antibiotic medium 2 (Difco
Laboratories), using Bacillus subtillis ATCC 12432 as the
indicator strain. Standard antibiotic solutions were pre-

pared in saline. To avoid inter-day variability the concentrations of drugs from which a standard curve was
determined and all CSF samples were assayed in duplicate
in a single day. The assay variability for individual samples
was 10%. The level of detection was 0.06 mg/L.
CSF bactericidal activities were performed by a microdilution method16 with cation-adjusted MuellerHinton
broth supplemented with 25% lysed horse blood. Serial
two-fold dilutions (range 1:2 to 1:4096) of CSF were performed, and a concentration of 5 105 cfu/mL of the same
strain as the meningitis model was used for the inoculum.
Bacteriostatic activity was defined as the highest dilution
without visible turbidity and bactericidal activity was
defined as the highest dilution capable of killing 99.9% of
the initial inoculum.
To minimize the effect of carry over of antibiotic agent,
an entire agar plate was used for each sample. The sample
was placed on the plate in a single streak down the centre,
allowed to be absorbed into the agar until the plate surface
appeared dry, and then the inoculum was spread over the
plate.17
Therapeutic failure was defined as an increase in bacterial concentration compared with a previous count, and
the death of the rabbit.
Since slight but not statistically significant differences in
CSF ceftriaxone concentrations were found at certain time
points, it was decided to perform a further set of experiments in order to determine the area under the time
concentration curve (AUC; by the linear trapezoidal rule),
which might better demonstrate such differences. The
rabbit model was set up as before and 13 rabbits were
treated with ceftriaxone 100 mg/kg/day and 13 rabbits with
ceftriaxone 100 mg/kg/day plus dexamethasone. Several
samples of 50 L of CSF were withdrawn at 1, 2, 4, 6, 8, 10,
12, 24, 26, 28, 30, 32, 34, 36 and 48 h. Only bacterial concentrations and CSF ceftriaxone concentrations were
determined for these samples.

Statistical analysis
Fishers exact test was used, when appropriate, to determine categorical variables. Students t test was used to determine continuous variables. The N-par MannWhitney U
Wilcoxon rank sum test was used to compare median
bactericidal titres.

Results
Bacterial concentrations
Ceftriaxone at 50 mg/kg/day reduced bacterial counts by
3 log cfu/mL at 24 h, with a mean log cfu/mL of 1.69 at
24 h versus 5.91 at 0 h (Figure). However, in two animals
bacterial counts were above the level of detection at 24 h,
and in one animal bacteria were detected at 48 h.
Concomitant use of dexamethasone resulted in statistic-

316

Ceftriaxone and dexamethasone for pneumococcal meningitis

Figure. Mean bacterial concentrations in CSF during treatment

of experimental S. pneumoniae meningitis. Symbols: ,
control; , ceftriaxone 50 mg/kg/day; , ceftriaxone 50 mg/kg/day
plus dexamethasone; , ceftriaxone 100 mg/kg/day; , ceftriaxone 100 mg/kg/day plus dexamethasone.

ally significant higher bacterial counts at 24 h, with a mean

log cfu/mL of 1.69 in ceftriaxone-treated animals versus
4.61 in animals treated with ceftriaxone plus dexamethasone (P 0.05).
Ceftriaxone at 100 mg/kg/day decreased bacterial counts
by 3 log cfu/mL at 6 h, with a mean log cfu/mL 1.68 versus
5.6 at 0 h. With this regimen, at 24 h 5 of 8 animals had bacterial counts under the level of detection and at 48 h bacterial counts in all animals were under the level of detection.
A higher bacterial count was found with concomitant use of
dexamethasone, although the difference was not statistically significant: the mean log cfu/mL was 1.18 in ceftriaxonetreated animals versus 2.18 in ceftriaxone plus dexamethasone-treated animals at 24 h, and 0 in ceftriaxone-treated
animals versus 1.2 mean log cfu/mL in ceftriaxone plus
dexamethasone-treated animals at 48 h (Figure).
Comparisons between bacterial counts with the different regimens at a range of time points are shown in the
Figure. Bacterial counts at 24 h in animals treated with
ceftriaxone 100 mg/kg/day versus animals treated with ceftriaxone 50 mg/kg/day plus dexamethasone also showed
statistically significant differences (P 0.05).

CSF ceftriaxone concentrations

Mean CSF ceftriaxone concentrations at peak and trough
on the first and second days of therapy are shown in Table I.
For the 50 mg/kg/day dose these were lower when concomitant dexamethasone was used, but this difference was
not statistically significant. Differences were more evident
at 2 h (peak of the first day): 3.57 mg/L in the ceftriaxone
group versus 1.75 mg/L in the ceftriaxone plus dexamethasone group (P 0.06). Mean trough CSF ceftriaxone concentrations were below the MIC for the infecting strain at
24 and 48 h; the variations between groups were unremarkable.
The effects of increasing the dose to 100 mg/kg/day were
variable (Table I). The most noteworthy finding was a
higher peak concentration with the 100 mg/kg/day dose
317

C. Cabellos et al.
Table II. Mean CSF ceftriaxone concentrations (mg/L) in rabbits with pneumococcal meningitis included in the AUC
study
1h
(n 12)
Ceftriaxone (100 mg/kg/day) 2.54
Ceftriaxone
2.95
dexamethasone

2h
(n 12)

4h
(n 12)

6h
(n 12)

8h
(n 11)

10 h
(n 11)

12 h
(n 11)

24 h
(n 11)

3.65
3.61

2.31
2.73

1.72
1.69

1.03
1.18

0.77
0.64

0.57
0.63

0.35
0.30

than with the 50 mg/kg/day dose, although the difference

was disproportionate. Some differences were observed
with the use of dexamethasone, but these were not statistically significant: at 26 h (peak of the second day) concentrations were 6.24 mg/L in the ceftriaxone group versus
4.43 mg/L in the ceftriaxone plus dexamethasone group.

Area under the timeconcentration curve

No statistically significant differences were observed in
CSF ceftriaxone concentrations between the time points
during the first 24 h (Table II). Nor were statistically significant differences observed in the AUC of the two therapy
groups: 25.81 mgh/L in the ceftriaxone group versus 23.61
mgh/L in the ceftriaxone plus dexamethasone group. After
24 h there was a higher mortality in the ceftriaxone plus
dexamethasone group, and so comparisons of CSF levels
and AUC calculations were not possible.

peutic efficacy between the ceftriaxone 100 mg/kg/day and

ceftriaxone 50 mg/kg/day plus dexamethasone groups was
statistically significant (P 0.05).

Inflammatory activity
Concomitant use of dexamethasone resulted in differences
in CSF inflammatory activity. With the dose of 50 mg/kg/
day the mean CSF WBC at 24 h was 11,900 cells/ L in the
ceftriaxone group versus 4780 cells/ L in the ceftriaxone
plus dexamethasone group (P
0.05). With ceftriaxone
100 mg/kg/day no differences were observed in the mean
WBC between the groups treated with or without dexamethasone, but there were statistically significant differences
in the mean CSF lactic acid concentration at 48 h: 6.71 mg/L
in the ceftriaxone group versus 3.18 mg/L in the ceftriaxone
plus dexamethasone group (P 0.05).

Percentage of time above MIC

Discussion

This set of experiments also allowed the time above the

MIC to be determined: on the first day this was 46 h (20%)
in the ceftriaxone (with or without dexamethasone) groups
and after the second dose it ranged 68 h (29%) in both
groups.

In this study, doses of ceftriaxone of 50 mg/kg/day failed to

eradicate a cephalosporin-resistant strain of S. pneumoniae
(MIC of ceftriaxone 2 mg/L). Increasing the dose to 100 mg/
kg/day improved bacterial killing, with sterilization of CSF
at 48 h, and no therapeutic failures occurred. In addition
CSF ceftriaxone levels were higher and bactericidal activities better when a dose of 100 mg/kg/day was used, showing
that increased dosing concentrations may be effective in
the therapy of cephalosporin-resistant pneumococcal meningitis. Overall, bactericidal activities were low, as predicted by the MIC for the infecting strain. In a recent
report, Klugman et al.1 showed that the CSF of children
receiving ceftriaxone at a range of doses, with dexamethasone, was unable to kill cephalosporin-intermediate or fully
-resistant strains when the CSF ceftriaxone concentration
was 5 mg/L, but at higher concentrations bactericidal
activity was present. In another study, the percentage of
time above the MBC was the variable that independently
predicted bacterial killing rate.18
Adult patients with pneumococcal meningitis cannot
usually be treated with 100 mg/kg/day of ceftriaxone, because of the limitation in total daily dose (4 g) recommended by the manufacturers. It is possible to increase the

CSF bactericidal activity

Median CSF bactericidal activities are shown in Table I.
There were no statistically significant differences between
groups, and all bacterial activities were low ( 1:2). With
the higher dose of ceftriaxone, bactericidal activities were
also low, but were one dilution higher.

Therapeutic failures
The dose of 50 mg/kg/day resulted in therapeutic failures in
2/7 subjects (28%). Concomitant use of dexamethasone
resulted in a higher failure rate: 4/7 (57%). The only group
in which there were no therapeutic failures (0/8) was for
those given ceftriaxone 100 mg/kg/day alone, but in the
group given 100 mg/kg/day plus dexamethasone there were
2/7 (28%) therapeutic failures. The difference in thera-

318

Ceftriaxone and dexamethasone for pneumococcal meningitis

dose of other similar broad-spectrum cephalosporins, such
as cefotaxime, for which a total daily dose limitation is not
given. Our group has reported the successful use of cefotaxime 300 mg/kg/day concomitantly with dexamethasone
in ten cases of pneumococcal meningitis caused by strains
with decreased susceptibility to third-generation cephalosporins.6 Increased dosage of third-generation cephalosporins
might be an effective therapy for cephalosporin-resistant
pneumococcal meningitis, at least for MICs
2 mg/L.
However, in our study, with the use of dexamethasone the
results were unacceptable even at the higher recommended
dose. We have previously shown11 that dexamethasone can
be used safely with ceftriaxone to treat penicillin-sensitive
strains. In this study, although small differences were detected, there were no statistically significant differences in
ceftriaxone levels in the CSF, bactericidal activities or
therapeutic failures. Paris et al.12 found that ceftriaxone
concentrations were lower at each time point in animals
given dexamethasone in a meningitis model using two
cephalosporin-resistant pneumococcal strains, although
again the differences were not statistically significant. These
small, though not statistically significant differences might
be of biological relevance in cases owing to cephalosporinresistant pneumococci. In the present study, AUC data
showed that dexamethasone did not decrease ceftriaxone
levels in the CSF. However, concomitant use of dexamethasone worsened results in terms of bacterial killing and
bactericidal activity, with a higher number of therapeutic
failures for both dose regimens. The mechanism of these
failures remains unclear. Although possible differences in
CSF ceftriaxone concentrations may appear the best explanation, this is not supported by our data. Nevertheless,
percentage of time above MICs of the dosing intervals was
lower than accepted for achieving therapeutic success in
most cases (therapeutic failures with percentage of time
above the MIC values in the range of 30% might be
expected).19 The concomitant use of dexamethasone may
accentuate this therapeutic limitation. Our data indicate
that increasing doses of ceftriaxone may be an effective
mode of therapy for cephalosporin-resistant cases, but
concomitant use of dexamethasone makes this approach
unreliable.

Acknowledgements
This work was supported in part by grants from the
Fondo de Investigaciones Sanitarias de la Seguridad Social
(FISss) 92/277 and 95/385 of Spains National Health
Service. F. Tubau, J. Martnez-Lacasa and A. Fernndez
were supported by grants from the Fundaci August Pi i
Sunyer. This paper was presented in part at the Thirty-Fifth
Interscience Conference on Antimicrobial Agents and
Chemotherapy, American Society for Microbiology, in San
Francisco, California, September 1995 (abstract no. B25).
The study was approved by the Ethical Committee for

Animal Experiments at the University of Barcelona

(Campus de Bellvitge).

References
1. Klugman, K. P., Friedland, I. R. & Bradley, J. S. (1995). Bactericidal
activity against cephalosporin-resistant Streptococcus pneumoniae
in cerebrospinal fluid of children with acute bacterial meningitis.
Antimicrobial Agents and Chemotherapy 39, 198892.
2. Liares, J., Alonso, T., Perez, J. L., Ayats, J., Dominguez, M. A.,
Pallares, R. et al. (1992). Decreased susceptibility of penicillinresistant pneumococci to twenty-four -lactam antibiotics. Journal of
Antimicrobial Chemotherapy 30, 27988.
3. Spangler, S. K., Jacobs, M. R. & Appelbaum, P. C. (1994). Susceptibilities of 177 penicillin-susceptible and -resistant pneumococci
to FK 037, cefpirome, cefepime, ceftriaxone, cefotaxime, ceftazidime, imipenem, biapenem, meropenem and vancomycin. Antimicrobial Agents and Chemotherapy 38, 898900.
4. Viladrich, P. F., Gudiol, F., Liares, J., Ruf, G., Ariza, J. &
Pallars, R. (1988). Characteristics and antibiotic therapy of adult
meningitis due to penicillin-resistant pneumococci. American
Journal of Medicine 84, 83946.
5. Friedland, I. R., Paris, M., Ehrett, S., Hickey, S., Olsen, K. D. &
McCracken, G. H., Jr (1993). Evaluation of antimicrobial regimens
for treatment of experimental penicillin and cephalosporin-resistant
pneumococcal meningitis. Antimicrobial Agents and Chemotherapy
37, 16306.
6. Viladrich, P. F., Cabellos, C., Pallares, R., Tubau, F., MartnezLacasa, J., Liares, J. et al. (1996). High dose of cefotaxime in treatment of adult meningitis due to Streptococcus pneumoniae with
decreased susceptibility to broad spectrum cephalosporins. Antimicrobial Agents and Chemotherapy 40, 21820.
7. Girgis, N. I., Farid, Z., Mikhail, I. A., Farrag, I., Sulstan, Y. &
Kilpatrick, M. E. (1989). Dexamethasone treatment for bacterial
meningitis in children and adults. Pediatric Infectious Diseases
Journal 8, 84851.
8. Lebel, M. H., Freij, B. J., Syrogianopoulos, G. A., Chrane, D. F.,
Hoyt, M. J., Stewart, S. M. et al. (1988). Dexamethasone therapy for
bacterial meningitis: results of two double blind, placebo controlled
trials. New England Journal of Medicine 319, 96471.
9. Odio, C. M., Faingezicht, I., Paris, M. I., Nassar, M., Baltodano,
A., Rogers, J. et al. (1991). The beneficial effects of early dexamethasone administration in infants and children with bacterial
meningitis. New England Journal of Medicine 324, 152531.
10. Schaad, U. B., Kaplan, S. L. & McCracken, G. H., Jr. (1995).
Steroid therapy for bacterial meningitis. Clinical Infectious Diseases
20, 68590.
11. Cabellos, C., Martnez-Lacasa, J., Martos, A., Tubau, F.,
Fernndez, A., Viladrich, P. F. et al. (1995). Influence of dexamethasone on efficacy of ceftriaxone and vancomycin therapy in
experimental pneumococcal meningitis. Antimicrobial Agents and
Chemotherapy 39, 215860.
12. Paris, M. M., Hickey, S. M., Uscher, M. I., Shelton, S., Olsen, K.
D., McCracken, G. H., Jr (1994). Effect of dexamethasone on therapy of experimental penicillin- and cephalosporin-resistant pneumococcal meningitis. Antimicrobial Agents and Chemotherapy 38,
13204.

319

C. Cabellos et al.
13. Dacey, R. G. & Sande, M. A. (1974). Effect of probenecid on
cerebrospinal fluid concentrations of penicillin and cephalosporin
derivates. Antimicrobial Agents and Chemotherapy 6, 43741.

activity. In Clinical Microbiology Procedures Handbook, (Isenberg,

H. D., Ed.), p. 12. American Society for Microbiology, Washington,
DC.

14. National Committee for Clinical Laboratory Standards. (1999).

Performance Standards for Antimicrobial Susceptibility Testing
Ninth Informational Supplement: M100-S9. NCCLS, Wayne, PA.

18. Lutsar, I., Ahmed, A., Friedland, I. R., Trujillo, M., Wubbel, L.,
Olsen, K. et al. (1997). Pharmacodynamics and bactericidal activity
of ceftriaxone therapy in experimental cephalosporin-resistant
pneumococcal meningitis. Antimicrobial Agents and Chemotherapy
41, 24147.

15. Chapin-Robertson, K. & Edberg, S. C. (1991). Measurements

of antibiotics in human body fluids: techniques and significance.
In Antibiotics in Laboratory Medicine, 3rd edn, (Lorian, V., Ed.),
pp. 295366. Williams & Wilkins, Baltimore.
16. Griffin, J. (1992). Serum inhibitory and bactericidal titers. In
Clinical Microbiology Procedures Handbook, (Isenberg, H. D., Ed.),
pp. 118. American Society for Microbiology, Washington, DC.
17. Knapp, C. & Moody, J. A. (1992). Test to assess bactericidal

19. Vogelman, B., Gudmundsson, S., Legett, J., Turnidge, J., Ebert,
S. & Craig, W. A. (1988). Correlation of antimicrobial pharmacokinetic parameters with therapeutic efficacy in an animal model.
Journal of Infectious Diseases 158, 83147.
Received 25 May 1999; returned 9 August 1999; revised 24
September 1999; accepted 12 November 1999

320