506 CID 2001:32 (1 February) BRIEF REPORTS

B R I E F R E P O R T
Azithromycin and Gentamicin Therapy
for the Treatment of Humans
with Brucellosis
Javier Solera,
1
Jose L. Beato,
2
Elisa Martnez-Alfaro,
1
Juan C. Segura,
2
and Elena de Tomas,
2
for the Grupo de Estudio de Castilla la Mancha
de Enfermedades Infecciosas Group
1
Unit of Infectious Diseases, Service of Internal Medicine, Hospital General,
Albacete, and
2
Service of Internal Medicine, Hospital Comarcal, Hell n, Spain
Ten patients with brucellosis were treated with azithromycin
and gentamicin to assess the treatments safety and efcacy.
Seven patients had an excellent therapeutic response at the
end of therapy; however, relapse was noted in 3. When re-
lapse was considered in combination with an initial lack of
efcacy, 5 patients (50%; 95% condence interval, 18.7%
81.3%) did not respond to therapy; these results do not favor
the use of azithromycin to treat brucellosis in humans.
The combination of tetracyclines and either an aminoglycoside
or rifampin is effective in the treatment of brucellosis in hu-
mans; however, relapse remains a signicant problem [1].
Moreover, for patients for whom tetracycline is contraindicated
(i.e., pregnant women and children !8 years old), no other
antibiotic offers a consistent treatment option that is as effective
as doxycycline in the treatment of acute brucellosis [2]. There-
fore, better treatments for humans with brucellosis are needed.
In one early study [3], combination therapy with erythro-
mycin and streptomycin was as effective as therapy with tet-
racycline and streptomycin. Farid et al. [3] treated 94 patients
who had brucellosis. Relapse was seen in 7 (13.2%) of 53 pa-
tients who received combinations of erythromycin and strep-
tomycin for 21 days. The relapse rate was similar in 21 patients
(14.3%) treated with tetracycline and streptomycin and in 20
patients (10%) treated with tetracycline as monotherapy. How-
Received 10 March 2000; revised 21 June 2000; electronically published 24 January 2001.
Presented in part: 39th Interscience Conference on Antimicrobial Agents and Chemotherapy,
San Francisco, 2629 September 1999 (abstract 1072).
Informed consent was obtained from patients or their parents, and the guidelines of the
ethical review committees of the hospitals of Albacete and Hellin, Spain, were followed in
the conduct of this study.
Reprints or correspondence: Dr. Javier Solera, Dept. of Internal Medicine, Hospital General
de Albacete, 02006 Albacete, Spain (jsolera@chospab.es).
Clinical Infectious Diseases 2001; 32:5069
2001 by the Infectious Diseases Society of America. All rights reserved.
1058-4838/2001/3203-0025$03.00
ever, 16 patients (30%) treated with erythromycin and strep-
tomycin had moderate to severe adverse gastrointestinal effects,
compared with 10% of patients who received tetracyclines
( ). In this study, a high dosage of oral erythromycin (1 P p.03
g q6h) was given for 21 days. Other studies that used lower
dosages (2 g/day) for 12 weeks showed worse results (13 re-
lapses among 21 patients) [4]. Therefore, this combination
therapy was promptly abandoned because of severe and fre-
quent side effects.
Azithromycin is an azalide with in vitro activity against Bru-
cella melitensis (MIC
90
range, 0.52 mg/mL) [57]. Azithro-
mycin is signicantly more active than erythromycin (and other
macrolides) against Brucella species, showing an MIC
90
that is
8-fold lower than that of erythromycin [6]. Furthermore, azith-
romycin is concentrated and persists in phagocytic cells (neu-
trophils and macrophages), and it does not interfere with their
bactericidal activity [8, 9]. Such properties suggest that azith-
romycin might be useful in the treatment of Brucella species,
which are known to survive and multiply within host cells. In
a murine model of brucellosis, azithromycin that was admin-
istered orally for 10 days reduced the infection signicantly,
but it was not able to cure the animals as effectively as doxy-
cycline that was administered for a longer period [10]. In a
more recent study, azithromycin that was administered for 7
or 14 days achieved an effective primary cure and prevented
relapse in a murine model of brucellosis [11].
Azithromycin may be an effective therapy for treatment of
brucellosis, but, to our knowledge, its use has not been assessed
in humans. We report the ndings of a pilot investigation that
studied treatment with azithromycin for 21 days plus genta-
micin for the rst 7 days of therapy among 10 patients with
acute brucellosis.
Materials and methods. The study was done from De-
cember 1997 through September 1999 in 2 general hospitals in
Spain. Eligible patients were at 8 years of age and had bru-
cellosis diagnosed, as dened below. The diagnostic criteria
included either (1) isolation of a Brucella species from samples
of blood or other uids or tissues, or (2) the nding of anti-
bodies to Brucella species at a titer of 11:160, by use of a
standard tube agglutination method, in association with at least
2 of the following compatible clinical ndings: fever, arthralgias,
weight loss, hepatosplenomegaly, or signs of focal disease. Pa-
tient exclusion criteria included a known or suspected hyper-
sensitivity (or other contraindication) to azithromycin or ami-
noglycosides, severe concomitant disease, and effective
antimicrobial therapy within 7 days before the patient entered
BRIEF REPORTS CID 2001:32 (1 February) 507
the study. We also excluded patients with CNS involvement,
spondylitis, or endocarditis. Eligible patients received azithro-
mycin in a single daily dose of 500 mg (or 10 mg/kg/day if the
body weight was !50 kg) for 21 days, and they received gen-
tamicin in a dose of 240 mg (or 5 mg/kg/day if the body weight
was !50 kg) given im once daily for the rst 7 days of treatment.
Azithromycin was taken orally at least 1 h before or 2 h after
a meal.
Patients were monitored for therapeutic efcacy and signs
of drug toxicity by analysis of clinical data; assessment of com-
plete blood cell counts (with differential and platelet counts)
and erythrocyte sedimentation rates; urinalysis; measurements
of levels of creatinine, aspartate aminotransferase, alkaline
phosphatase, albumin, total protein, bilirubin, and electrolytes;
Brucella serological analysis; and culture of blood samples. The
patients were evaluated at baseline, on days 8 and 15, and at
the end of therapy. During these visits, the subjects were asked
whether they had missed any dosing. After therapy ended, the
patients were reassessed at months 1, 2, 3, 6, 9, and 12, as well
as whenever clinical symptoms reappeared. The clinical re-
sponse to treatment was classied in the following manner: no
apparent response to the study drug or a worsening of the signs
and symptoms after 7 days of therapy was considered failure;
reappearance of symptoms or signs of the disease or a new
positive result of blood culture during the 12 months after
therapy was considered a relapse; and complete resolution of
the signs and symptoms of infection and no relapse at follow-
up was considered a cure. Safety was assessed on the basis of
all reported adverse events and the results of laboratory tests and
other investigations. Clinical adverse events were recorded and
evaluated for severity, outcome, and relation to the study drugs.
The standard tube agglutination test, the rose bengal test,
and the anti-Brucella Coombs test were done by use of standard
methods [12] with commercial reagents (Knickerbocker).
Blood samples were cultured as reported elsewhere [12], and
the cultures were incubated for 30 days by use of the BACTEC
NR-900 system (Becton DickinsonSpain). All isolates were
identied as recommended by Hausler et al. [13]. Three of the
isolated strains were sent to a reference center (Servicio de
Microbiologa, Universidad de Navarra, Pamplona, Spain) for
conrmation and biotyping. All Brucella isolates were identied
as B. melitensis: 2 were biovar 3 and 1 was biovar 1.
Results. Demographic variables and therapeutic responses
are summarized in table 1. Of the 10 patients with acute bru-
cellosis who were enrolled in this study, 7 completed the 21
days of azithromycin therapy and 3 discontinued treatment
prematurely. One patient was unable to continue treatment
after she broke out in a generalized rash, presumably in relation
to azithromycin therapy on day 2 of therapy. Two other patients
had objective evidence of treatment failure. A patient who pre-
sented with fever, sweats, weight loss, and lumbar pain con-
tinued to have fever and progressive lumbar pain and was
withdrawn from therapy on day 8. In a second patient, symp-
toms improved during the rst week, but fever and sterno-
clavicular arthritis developed during week 2 of treatment with
azithromycin.
Seven of 10 patients achieved excellent therapeutic responses
at the end of therapy; initially, 1 had shoulder arthritis and 6
had nonfocal brucellosis with fever, constitutional symptoms,
and arthralgias. However, during the month after they nished
therapy, 3 patients (30%; 95% CI, 6.7%65.2%) had clinical
relapse (table 1). Clinical suspicion of relapse was conrmed
in 2 of them by means of isolation of Brucella from blood
samples. The third patient who had relapse developed sacro-
iliitis that was conrmed by means of MRI. Relapses were
treated with doxycycline, 200 mg/day for 45 days, and strep-
tomycin, 1.0 g/day for 15 days; clinical response was excellent
for all patients. When relapse was considered together with
initial lack of efcacy, 5 patients (50%; 95% CI, 18.7%81.3%)
failed to respond to therapy.
Side effects that were considered by the investigators to be
related to treatment were reported in 4 (40%) of 10 patients
(95% CI, 12.2%73.7%). The most common treatment-related
side effects involved the gastrointestinal tract (nausea in 1 pa-
tient, diarrhea in 1, and epigastric discomfort in 1). All side
effects were judged to be mild, with the exception of 1 case of
generalized skin rash. Abnormal laboratory test results that may
have been related to azithromycin treatment were recorded for
3 patients (30%). Small increases in liver enzyme levels were
observed in 3 patients, 1 of whom had a small transient decrease
in WBC count (nadir, cells/L). In this study, im ad-
9
3.5 10
ministration of gentamicin was well tolerated. Laboratory stud-
ies did not reveal any drug-related renal abnormalities. The
mean (SD) serum creatinine concentrations at baseline and
after 7 days of gentamicin treatment were mg/dL 0.90 0.06
and mg/dL, respectively. 0.88 0.06
Discussion. In patients who were treated with azithro-
mycin for 21 days and with gentamicin for the rst 7 days in
this study, the failure rate of 50% was higher than that in our
previous studies, in which doxycycline was given for 3045 days
and aminoglycosides were given for 12 weeks (with failure
rates ranging from 5.9% to 22.9%) [1416]. These ndings are
especially notable not only for the relapse of brucellosis within
only a few weeks of stopping azithromycin therapy in 3 of 7
patients who completed 21 days of therapy but, also, for the
apparent lack of a signicant treatment effect in 2 patients, 1
of whom developed sternoclavicular arthritis during the second
week while undergoing azithromycin therapy. This suggests that
azithromycin could not prevent the development of osteoar-
ticular complications.
The maximum level of azithromycin in serum is only 0.4
mg/L after an oral dose of 500-mg [17], and the MIC
90
for B.
508 CID 2001:32 (1 February) BRIEF REPORTS
Table 1. Characteristics of 10 patients with acute brucellosis treated with azithromycin and gentamicin.
Patient Age, y Sex
Symptom
duration, d Focal disease
Blood
culture
result STA titer
a
Outcome Comment
1 24 F 45 None 160 Cured
2 35 M 15 None 80 Cured
3 61 M 45 Shoulder arthritis
b

c
640 Cured
4 42 M None 1280 Cured
5 16 F 30 Sacroiliitis
b
2560 Adverse effect Withdrawn from study because
of rash after 2 doses of
azithromycin
6 39 M 20 None 320 Therapy failed Persistent fever, arthralgias, and
back pain 7 days after starting
therapy; spondylitis was diag-
nosed 12 weeks later
7 18 M 18 None 2560 Therapy failed Fever and sternoclavicular arthri-
tis during week 2 after starting
therapy; resolution after doxy-
cycline-gentamicin therapy
8 11 M 6 None 1280 Relapse (C) Clinical relapse with sacroiliitis
(MRI) 8 days after ending ther-
apy; resolution after doxycy-
cline-streptomycin therapy
9 57 M 9 None 640 Relapse (C/B) Relapse with wrist bursitis 1
month after ending therapy.
Resolution after doxycycline-
streptomycin therapy
10 39 F 9 None 1280 Relapse (C/B) Fever, malaise, and arthralgias 1
month after ending therapy;
resolution after doxycycline-
streptomycin therapy
NOTE. B, bacterial relapse; C, clinical relapse; STA, standard tube agglutination; , positive; , negative.
a
Reciprocal titer.
b
Diagnostic procedure, MRI
c
Brucella species were isolated from a synovial uid samples from the shoulder.
melitensis had a range of 0.52 mg/L [5, 6]. On the other hand,
although the pharmacokinetic prole of azithromycin is char-
acterized by sustained high concentrations in cells and tissues
[9, 10, 17], the activity of azithromycin against Brucella species
is 6- to 8-fold lower at pH 5.0 than at pH 7.0 [18]. Brucella
species grow and replicate in the phagolysosomes of macro-
phages, where the pH is 5.0 [19]. These data are supported by
the results of animal studies. Domingo and colleagues [10, 20]
reported that, for experimental brucellosis in mice, azithro-
mycin was less effective than doxycycline used either alone or
in combination with azrithromycin and streptomycin. These
authors think that Brucella species in intracellular conditions
would be resistant to azithromycin (MIC, 18 mg/mL) [21].
Other drugs used in the treatment of brucellosis seem to be
more effective. Rifampin (MIC range, 0.051 mg/L) also con-
centrates in neutrophils and reaches a higher serum concen-
tration than does azithromycin [1]. Moreover, the activity of
rifampin is increased 2- to 8-fold at pH 5.0, whereas the MIC
of azithromycin increases to well above its breakpoint [18]. In
our study, 6 patients who either failed to respond to therapy
(5 patients) or withdrew because of side effects (1 patient) were
eventually cured with combination regimens of rifampin plus
gentamicin or doxycycline plus aminoglycosides (table 1).
We concluded that a regimen of azithromycin of 3 weeks
duration plus gentamicin given during the rst week results in
a high rate of therapeutic failures and relapses. These results
do not favor the use of azithromycin to treat brucellosis in
humans.
References
1. Young EJ. Brucella species. In: Yu VL, Merigan TC Jr, Barriere SL, eds.
Antimicrobial therapy and vaccines. Baltimore: Williams & Wilkins,
1999:7189.
2. Solera J, Martnez-Alfaro E, Espinosa E. Recognition and optimum
treatment of brucellosis. Drugs 1997; 53:24556.
3. Farid Z, Miale A Jr, Omar MS, Van Peenen PFD. Antibiotic treatment
of acute brucellosis caused by Brucella melitensis. J Trop Med Hyg
1961; 64:15763.
BRIEF REPORTS CID 2001:32 (1 February) 509
4. Hall HW. Modern chemotherapy for brucellosis in humans. Rev Infect
Dis 1990; 12:106099.
5. Landnez R, Linares J, Loza E, et al. In vitro activity of azithromycin
and tetracycline against 358 clinical isolates of Brucella melitensis. Eur
J Clin Microbiol Infect Dis 1992; 11:2657.
6. Garca-Rodrguez JA, Munoz-Bellido JL, Fresnadillo MJ, Trujillano I.
In vitro activities of new macrolides and rifapentine against Brucella
spp. Antimicrob Agents Chemother 1993; 37:9113.
7. Qadri HSM, Halim MA, Ueno Y, Abumustafa FM, Postle AG. Anti-
bacterial activity of azithromycin against Brucella melitensis. Chemo-
therapy 1995; 41:2536.
8. Gladue RP, Bright GM, Isaacson RE, Newborg MF. In vitro and in vivo
uptake of azithromycin (CP-62,993) by phagocytic cells: possible mech-
anism of delivery and release at sites of infection. Antimicrob Agents
Chemother 1989; 33:27782.
9. McDonald PJ, Pruul H. Phagocyte uptake and transport of azithro-
mycin. Eur J Clin Microbiol Infect Dis 1991; 10:82833.
10. Domingo S, Gastearena I, Vitas AI, et al. Comparative activity of azith-
romycin and doxycycline against Brucella spp. infection in mice. J
Antimicrob Chemother 1995; 36:64756.
11. Lang R, Torten D, Shasha B, Rubinstein E. Macrolides in the therapy
of acute experimental murine brucellosis [abstract B-92]. In: Program
and abstracts of the 37th Interscience Conference on Antimicrobial
Agents and Chemotherapy (Toronto). Washington, DC: American So-
ciety for Microbiology, 1997:43.
12. Solera J, Medrano F, Rodrguez M, et al. Ensayo terapeutico compar-
ativo y multicentrico de rifampicina y doxiciclina frente a estreptom-
icina y doxicilina en la brucelosis humana. Med Clin (Barc) 1991; 96:
64953.
13. Hausler WJ Jr, Moyer NP, Holcomb LA. Brucella. In: Lennette EH,
Balows A, Hausler WJ Jr, Shadomy HJ, eds. Manual of clinical mi-
crobiology. 4th ed. Washington, DC: American Society for Microbi-
ology, 1984:3826.
14. Solera J, Rodriguez-Zapata M, Geijo P, et al. Doxycycline-rifampin
versus doxycycline-streptomycin in treatment of human brucellosis due
to Brucella melitensis. Antimicrob Agents Chemother 1995; 39:20617.
15. Solera J, Espinosa A, Geijo P, et al. Treatment of human brucellosis
with netilmicin and doxycycline. Clin Infect Dis 1996; 22:4415.
16. Solera J, Espinosa A, Martnez-Alfaro E, et al. Treatment of human
brucellosis with doxycycline and gentamicin. Antimicrob Agents
Chemother 1997; 41:804.
17. Foulds G, Shepard RM, Johnson RB. The pharmacokinetics of azith-
romycin in human serum and tissues. J Antimicrob Chemother
1990; 25(Suppl A):7382.
18. Akova M, Gur D, Livermore DM, Kocagoz T, Akalin HE. In vitro
activities of antibiotics alone and in combination against Brucella mel-
itensis at neutral and acidic pHs. Antimicrob Agents Chemother
1999; 43:1298300.
19. Kornfeld S, Mellman I. The biogenesis of lysosomes. Annu Rev Cell
Biol 1989; 5:483525.
20. Domingo S, Gamazo C. Inadequate azithromycin activity against Bru-
cella melitensis in mice with acute or chronic infections. J Chemother
1996; 8:558.
21. Domingo S, Gastearena I, Diaz R, Gamazo C. Signicance of envi-
ronmental conditions (pH and serum) on the in vitro potency of
azithromycin against Brucella melitensis. J Chemother 1995; 7(Suppl 4):
2931.