Efficacy and safety of colistin (colistimethate sodium)
for therapy of infections caused by
multidrug-resistant Pseudomonas aeruginosa and
Acinetobacter baumannii in Siriraj Hospital,
Pornpan Koomanachaia, Surapee Tiengrima, Pattarachai Kiratisinb,
aDivision of Infectious Diseases and Tropical Medicine, Department of Medicine,
Siriraj Hospital, Mahidol University, Prannok Road, Bangkok, Thailand
bDepartment of Microbiology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand
Received 22 June 2006; received in revised form 31 August 2006; accepted 18 September 2006
Corresponding Editor: Andy I.M. Hoepelman, Utrecht, The Netherlands
International Journal of Infectious Diseases (2007) 11, 402—406
local pharmaceutical company in Thailand for the treatment of infections caused by multidrug-
resistant (MDR) Pseudomonas aeruginosa and Acinetobacter baumannii.
Methods: Patients hospitalized at Siriraj Hospital between January 2005 and April 2006, who had
infections caused by MDR P. aeruginosa or A. baumannii, were enrolled in the study. Colistin
(colistimethate sodium) atadosage of5 mg/kg/daywas givenintravenouslyintwo divideddoses.
Primary outcomes were the clinical response and 30-day mortality; secondary outcomes were
microbiological response and adverse events.
Results: Ninety-three patients infected with MDR P. aeruginosa and A. baumannii were enrolled.
Seventy-eight patients (71 with A. baumannii and seven with P. aeruginosa) received colistin,
whereas 15 patients (12 with A. baumannii and three with P. aeruginosa) received other
antibiotics. The mean age, gender, underlying conditions and severity of illness of the patients
in both groups were not significantly different. In the colistin group, 63 patients (80.8%) had a
favorable clinical response and 94.9% had a microbiological response. The overall mortality of the
patientsinthecolistingroupwas 46.2%andthatinthenon-colistin groupwas80%.Nephrotoxicity
was found in 24 patients (30.8%) in the colistin group and 17 of them had predisposing factors
contributing to their renal dysfunction. No neurotoxicity was observed among the 78 patients.
* Corresponding author. Tel.: +66 2 412 5594; fax: +66 2 412 5994.
E-mail address: email@example.com (V. Thamlikitkul).
1201-9712/$32.00 # 2006 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Nosocomial infections caused by multidrug-resistant (MDR)
organisms are emerging worldwide.1—3The incidence of MDR
pathogens, particularly Acinetobacter baumannii and Pseu-
increased.4A prospective cohort study of 208 clinical isolates
of A. baumannii recovered from the patients in Siriraj Hos-
pital from January to December 2002, revealed that 86
strains (41.3%) were isolated from the infected patients
and the remaining 58.7% were colonizers.5In this study,
57% of A. baumannii isolates were resistant to all antimicro-
bial agents available in Thailand including beta-lactams,
aminoglycosides and fluoroquinolones, and the overall mor-
tality rate of the patients infected with pandrug-resistant A.
baumannii was 79%.5The study of 104 clinical isolates of A.
baumannii from 100 hospitalized patients at Maharaj Nakorn
Chiang Mai hospital, Thailand also observed that 46% of the
isolates were pandrug-resistant and the overall mortality
Over the past few years there have been reports on
treating patients infected with MDR A. baumannii and P.
polymyxin B and colistin had modest efficacy and were safe.
In vitro activity of polymyxin B and colistin against 100
clinical isolates of MDR A. baumannii and 100 isolates of P.
aeruginosa collected from the patients hospitalized at Sir-
iraj Hospital from 2002 and 2003, revealed that all isolates
were susceptible to polymyxin B and colistin.10However
polymyxins are not available in Thailand and international
pharmaceutical companies do not have a policy to import
polymyxins to Thailand. Therefore we asked a local phar-
been approved by the Thai Food and Drug Administration
The objective of this study was to determine the efficacy
and safety of colistin produced by a local pharmaceutical
company in Thailand for the treatment of infections caused
by MDR P. aeruginosa and A. baumannii.
The study was approved by the ethics committee on human
research of the Faculty of Medicine Siriraj Hospital, and all
was a pragmatic clinical trial conducted at Siriraj Hospital,
Bangkok, Thailand, between January 2005 and April 2006.
The eligible subjects were hospitalized patients over the age
of 18 years who were infected with A. baumannii or P.
aeruginosa resistant to beta-lactams, fluoroquinolones and
aminoglycosides. We excluded patients with infections
caused by A. baumannii or P. aeruginosa with other bacteria
from our study because we felt that it was difficult to
determine the efficacy of colistin for treatment of infections
caused by MDR A. baumannii or P. aeruginosa. Colistin was
offered to all such patients and if the patients and their
responsible physicians agreed to have colistin treatment, the
sodium) of 5 mg/kg/day in two divided doses. The dosage
of colistin was adjusted according to the patients’ renal
function.11If the patients or their responsible physicians
did not wish to join the study, they received other antibiotics
according to their physicians’ decisions and these patients
were defined as the ‘non-colistin group’.
All isolates of A. baumannii and P. aeruginosa from the
eligible patients were tested for colistin susceptibility by E-
test according to the manufacturer’s guidelines (AB Biodisk,
Sweden). A suspension of each isolate in Mueller—Hinton
broth (BBL-Becton Dickinson, USA), adjusted to the density
of a 0.5 McFarland standard, was swabbed in three directions
to ensure uniform growth onto Mueller—Hinton agar (BBL-
Becton Dickinson, USA) plates. Once the agar surface was
completely dry, an E-test colistin strip (ranging from 0.06 to
1024 mg/ml) was applied to each plate and the plates were
incubated at 35 8C for 16—20 hours. The minimum inhibitory
concentration (MIC) was read where inhibition of growth
intersected the E-test strip. Quality control strains of Escher-
with the reference MIC range of 0.125—0.5 and 0.5—2 mg/l,
respectively. The susceptible isolate was defined as having a
MIC of ?2 mg/l. Quantitative colistin serum level was deter-
mined by microbiological assay.12
The primary outcomes were the clinical response and 30-
day mortality. A good clinical response referred to a combi-
nation of clinical cure and clinical improvement. Clinical
cure was defined as a disappearance of symptoms and signs
of infection and clinical improvement was defined as a
partial resolution of the symptoms and signs of infection.
The secondary outcomes were microbiological response and
adverse effects. Successful microbiological response was
defined as an eradication of the causative organisms at
the end of treatment. Nephrotoxicity was defined as an
increase in serum creatinine of at least two-fold of the
baseline value or a 30% decrease of creatinine clearance
from the baseline value.
Between January 2005 and April 2006, 93 patients met the
inclusion criteria. Seventy-eight patients were in the colistin
group and 15 patients in the non-colistin group. The baseline
characteristics of the patients are shown in Table 1. The
mean age, gender, underlying conditions, and severity of
illness of the patients in both groups were not significantly
Presenting infections in the colistin group were: pneumo-
nia (54), bacteremia and/or catheter related infection (9),
and soft tissue infection (5), and sinusitis (1). In the colistin
Colistin and multidrug-resistant P. aeruginosa and A. baumannii
Conclusion: Locally produced colistin appears to be safe and effective for the treatment of
infections caused by MDR P. aeruginosa and A. baumannii in Thai adult patients.
# 2006 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
seven (9%) were infected with P. aeruginosa, whereas 12
patients (80%) were infected with A. baumannii and three
(20%) were infected with P. aeruginosa in the non-colistin
group. In vitro susceptibility tests determined by E-test
revealed that all A. baumannii and P. aeruginosa isolates
had a MIC of colistin less than 2 mg/l and were considered
susceptible to colistin. In the colistin group, 33 patients
(42.3%) received colistin alone, whereas 45 patients
(57.7%) received colistin with other antibiotics including
vancomycin, aminoglycosides, metronidazole or carbape-
nems. In the non-colistin group, the patients received car-
bapenems (6), cefoperazone/sulbactam (3), cefoperazone/
sulbactam combined with netilmicin (4), and cefoperazone/
sulbactam combined with carbapenem (2).
The treatment outcomes are shown in Table 2. Sixty-four
patients (82.1%) in the colistin group had a good clinical
response. The clinical response in the patients who received
colistin alone was 84.8% and in those who received colistin
the non-colistin group responded.
All cause mortality within 30 days was 46.2% in the colistin
group and 80% in the non-colistin group (p = 0.03). The
relative risk of death in the colistin group was 0.58 of the
non-colistin group with a 95% confidence interval (CI) of 0.41
to 0.82. The difference in mortality was statistically signifi-
cant and the number needed to treat (NNT) was approxi-
mately three, which implies that only three patients infected
with MDR A. baumannii or P. aeruginosa needed to be treated
with colistin in order to prevent one additional death. The
overall mortality rates of the patients infected with A.
baumannii and P. aeruginosa in the colistin group were
46.5% and 42.9%, respectively.
A microbiological response was found in 94.9% of the
patients in the colistin group and none in the non-colistin
group. Nephrotoxicity was observed in 24 patients (30.8%) in
the colistin group. The incidence of nephrotoxicity of the
patients in the colistin group was significantly less than that
in the non-colistin group. Seventeen (70.8%) of 24 patients in
the colistin group who developed nephrotoxicity had other
predisposing factors contributing to a decline in renal func-
tion including nephrotoxic drugs, chronic kidney diseases,
and hypovolemia. Nephrotoxic effects were mild and rever-
sible without requiring renal replacement therapy. No neu-
rotoxicity or drug reaction was observed in the patients who
received colistin. The average dose of colistin was 179.6 mg/
day, the average duration of colistin treatment was 11.9
days, and the average total dose of colistin was 2.1 g/
404 P. Koomanachai et al.
Baseline characteristics of the patients
Characteristic Colistin group (N = 78)Non-colistin group (N = 15)
Mean age, years (range)
Mean APACHE II score
Pre-existing renal impairment (serum creatinine ?1.5 mg/dl)
Diabetes mellitus with or without other medical conditions
Chronic obstructive pulmonary disease
Other chronic medical conditions
Traumatic surgical patients
Recent cardiovascular surgery
Recent brain surgery
Recent abdominal surgery
Treatment outcomes of the patients
OutcomeColistin group (N = 78)Non-colistin group (N = 15)p Value
Good clinical response
All cause mortality within 30 days
This study used colistimethate sodium (also called colistin
methanesulfate, pentasodium colistimethane sulfate, or
colistin sulfonyl methate), which is less potent and less toxic
than colistin sulfate.13,14Colistin has a narrow spectrum of
antimicrobial activity and is active against most aerobic
Gram-negative bacilli including P. aeruginosa and Acineto-
bacter spp, even the organisms that are multidrug-resis-
tant.13Several reports published during the period 1999 to
2003 revealed that polymyxins were effective and safe for
treatment of patients infected with MDR Gram-negative
bacteria including A. baumannii and P. aeruginosa.7—9We
therefore attempted to study the efficacy and safety of
locally produced colistin.
We were unable to do a randomized controlled study to
compare colistin with other antibiotics since it would be
unethical to provide antibiotics likely to be ineffective to
patients, while the antibiotic active against the causative
pathogens, colistin, was available. Therefore we had to
offer colistin to all patients who had infections caused by
A. baumannii or P. aeruginosa resistant to beta-lactams,
fluoroquinolones and aminoglycosides. However, the base-
line characteristics of the patients including mean age,
gender, underlying conditions, severity of illness and the
sites of infections of the patients in both groups were
The results from our study also showed a good clinical
outcome and less overall mortality in patients who received
A good clinical outcome was found in 82.1% of patients
treated with colistin no matter how the patients received
it, alone or with other antibiotics. Overall mortality
decreased from 79% in a previous study of A. baumannii
infections in the same hospital to 46.5% of the patients
infected with A. baumannii treated with colistin in this
study.5The overall mortality in the non-colistin group in this
study was still up to 80%. Furthermore, NNT for mortality
from our study was only three, indicating that only three
patients infected with MDR A. baumannii or P. aeruginosa
needed to be treated with colistin in order to prevent one
additional death. Moreover the cost of colistin was approxi-
mately 10 to 20 times lower than that of other antibiotics
used to treat MDR A. baumannii and P. aeruginosa such as
carbapenems, cefoperazone/sulbactam, and cephalosporins
with or without aminoglycosides.
A microbiological response was observed in 74 patients
(94.9%) in the colistin group. Three patients who did not have
a microbiological response also had a good clinical outcome.
However, antibiotic susceptibility profiles of these persistent
isolates were different from those of the original isolates and
these isolates could be new colonizers. In four patients who
had no microbiological response after 72 hours of colistin
treatment, the serum levels of colistin were measured by
bioassay and the results showed that colistin levels were
adequate at 4—8 times above the MIC of the organism.
Therefore the same dose of colistin was continued for 7 days
and all patients eventually had a microbiological response.
We excluded patients with infections caused by A. baumannii
orP.aeruginosa withotherbacteria fromourstudy,therefore
the efficacy of colistin for treatment of mixed infections is
Nephrotoxicity is an important side effect of colistin. In
our study, nephrotoxicity was found in 30.8% of the patients
receiving colistin; this is comparable to the results found in
a previous report.15Some patients in the colistin group who
developed nephrotoxicity also had other contributing fac-
tors. Nephrotoxicity in these patients was mild and rever-
sible without requiring renal replacement therapy. Some
patients had improvement in their renal function after
colistin treatment, which implies that the worsening of
renal function was probably due to a severe infection or
other conditions. The incidence of nephrotoxicity of the
patients in the non-colistin group was significantly more
than that in the colistin group. This observation might be
due to uncontrolled infections and the side effects of
medications including antibiotics given to the patients.
No neurotoxicity or drug reaction was observed in the
patients in our series.
Although the ability of Gram-negative bacteria to
develop resistance to colistin is rare, such Gram-negative
or adaptation mechanisms.13,16We therefore recommend
that colistin, as the only currently available drug for the
treatment of MDR Gram-negative bacteria in Thailand,
should be reserved for treatment of infections caused by
multidrug-resistant Gram-negative bacteria that are only
susceptible to colistin.
In summary, we found that colistin appears to be safe and
effective for treatment of infections caused by multidrug-
resistant P. aeruginosa and A. baumannii in Thai adult
patients. Treatment with colistin decreases patient mortality
and is cost-effective.
The authors thank the Thailand Research Fund for supporting
the study, Atlantic Pharmaceutical Co. Ltd for supplying
colistin and colistin E-test strips, Ms Pornsiri Chinsawangwa-
tanakul and Ms Sunee Thanakhumtorn for coordinating the
study, and Dr Methee Chayakulkeeree for reviewing the
Conflict of interest: No conflict of interest to declare.
1. Bergogne-Berezin E, Towner KJ. Acinetobacter spp. as nosoco-
mial pathogens: microbiological, clinical, and epidemiological
features. Clin Microbiol Rev 1996;9:148—65.
2. Livermore D. Multiple mechanisms of antimicrobial resistance in
Pseudomonas aeruginosa: our worst nightmare? Clin Infect Dis
3. Hamberger H, Diekema D, Fluit A. Surveillance of antibiotic
resistance in European ICUs. J Hosp Infect 2001;48:161—76.
4. Thamlikitkul V, Jintanothaitavorn D, Sathimethakul R, Vaiyhaya-
phichet S, Trakulsomboon S, Danchaivijitr S. Bacterial infections
5. Keerasuntonpong A, Samakeepanich C, Tribuddharat C. Epide-
miology of Acinetobacter baumannii infections in Siriraj Hospi-
tal. Siriraj Med J 2006;58:951—4.
6. Chaiwarith R, Mahatthanaphak S, Boonchoo M, Supparatpinyo K,
Sirisanthana T. Pandrug-resistant Acinetobacter baumannii at
Maharaj Nakorn Chiang Mai Hospital. J Infect Dis Antimicrob
Colistin and multidrug-resistant P. aeruginosa and A. baumannii
7. Markou N, Apostolakos H, Koumoudiou C, Athanasiou M, Kout-
soukouA,AlamanosI, et al.Intravenouscolistinin the treatment
patients. Crit Care 2003;7:78—83.
8. Ouderkirk JP, Nord JA, Turett GS, Kislak JW. Polymyxin B nephro-
toxicity and efficacy against nosocomial infections caused by
multiresistant Gram-negative bacteria. Antimicrob Agents Che-
9. Levin AS, Barone AA, Penco J, Santos MI, Marinho IS, Arruda EAG,
et al. Intravenous colistin as therapy for nosocomial infections
caused by multidrug-resistant Pseudomonas aeruginosa and Aci-
netobacter baumannii. Clin Infect Dis 1999;28:2008—11.
10. Tribuddharat C, Tiensasiton C, Techachaiwiwat W, Rugdeekha S,
Dhiraputtra C, Thamlikitkul V. In vitro activity of polymyxin E
against multi-drug resistant Pseudomonas aeruginosa and Aci-
netobacter baumannii. J Infect Dis Antimicrob Agents 2003;20:
antibiotics for emerging multiresistant Gram-negative bacteria.
Ann Pharmacother 1999;33:960—7.
method for colistin sulphomethate. Clin Microbiol Infect
13. Falagas ME, Kasiakou SK. Colistin: the revival of polymyxins for
infections. Clin Infect Dis 2005;40:1333—41.
14. Kaye D. Current use for old antibacterial agent. Inf Dis Clin N Am
15. Falagas ME, Kasiakou SK. Toxicity of polymyxins: a systematic
review of the evidence from old and recent studies. Crit Care
16. Moore RA, Chan L, Hancock RE. Evidence for two distinct
mechanisms of resistance to polymyxin B in Pseudomonas aer-
uginosa. Antimicrob Agents Chemother 1984;26:539—45.
406P. Koomanachai et al.