Methods for removing bacterial biofilms: In vitro study
using clinical chronic rhinosinusitis specimens
Martin Desrosiers, M.D.,* Matthew Myntti, Ph.D.,# and Garth James, Ph.D.§
ABSTRACT
Background: Bacterial biofilms may be involved in refractory chronic rhinosinusitis (CRS). In vitro, we studied methods for removing
biofilms formed by Staphylococcus aureus and Pseudomonas aeruginosa.
Methods: Bacterial isolates were obtained from patients with refractory CRS and were plated and treated with either static administration
of citric acid/zwitterionic surfactant (CAZS), saline delivered with hydrodynamic force, or CAZS delivered hydrodynamically. Results were
assessed by counting colony-forming units (CFUs) and by confocal scanning laser microscopy (CSLM).
Results: All treatments produced significant reductions in CFU counts (p 
 0.002). Hydrodynamic CAZS provided the greatest
reduction, decreasing CFU counts from control values by 3.9 
 0.3 logs and 5.2 
 0.5 logs for S. aureus and P. aeruginosa, respectively
(99.9% reduction; p  0.001). CSLM showed decreases in biofilm coverage.
Conclusion: Hydrodynamic delivery of a soap-like surfactant and a calcium-ion sequestering agent may disrupt biofilms associated with
CRS. Our results may be relevant to a new approach to refractory CRS.
(Am J Rhinol 21, 527–532, 2007; doi: 10.2500/ajr.2007.21.3069)
Key words: Bacterial biofilm, chronic rhinosinusitis, citric acid, confocal scanning laser microscopy, hydrodynamic force,
Pseudomonas aeruginosa, Staphylococcus aureus, zwitterionic surfactant
A biofilm is a community of bacteria attached to a nonliv-ing surface or viable tissue, with the organisms in the
community contained within an extracellular polymeric sub-
stance (EPS) matrix produced by the bacteria.1 The EPS matrix
allows the biofilm to stick to the surface and also protects the
embedded organisms; thus, bacteria in biofilms are 100–
1000 times more resistant to the effects of antibiotics than
planktonic (suspended) bacteria.2 Biofilms have been found
on the surfaces of numerous medical devices and in a wide
range of human diseases, including dental caries, periodontal
disease, musculoskeletal infections, necrotizing fascitis, endo-
carditis, cystic fibrosis pneumonia, otitis media (OM), and
chronic rhinosinusitis (CRS).1,3–6 The most commonly ob-
served biofilms are those formed by Streptococcus species,
Staphylococcus epidermidis, Staphylococcus aureus, Pseudomonas
aeruginosa, and Haemophilus influenzae. All of these organisms
are important in otorhinolaryngologic disease.7 Recent re-
search has provided substantial evidence that many patients
with OM or CRS refractory to standard antibiotic therapy may
harbor a biofilm-related infection.3,6,8–10
A comprehensive consensus document produced in 2004
by representatives of five professional otorhinolaryngologic
societies (The American Academy of Allergy, Asthma and
Immunology; The American Academy of Otolaryngic Aller-
gy; The American Academy of Otolaryngology–Head and
Neck Surgery [AAO-HNS]; The American College of Allergy,
Asthma and Immunology; and the American Rhinologic So-
ciety) suggested that elimination of a biofilm in patients with
CRS may require mechanical removal of the biofilm.11 Me-
chanical disruption of biofilm has a long history of success in,
e.g., dentistry, in which it is used to remove dental plaque (a
known biofilm). Such disruption may occur during functional
endoscopic sinus surgery (FESS),5 which produces symptom-
atic relief in 69% to 90% of patients with CRS refractory to
medical therapy.12,13 However, in patients in whom neither
medical therapy nor FESS alone is effective, a different ap-
proach to mechanical removal of a biofilm may be required.
We hypothesized that an effective biofilm-removal method
should include two principal components: a nontoxic, water-
soluble, low-viscosity chemical solution (surfactant) capable
of breaking up the biofilm and allowing the debris to be
flushed from the sinuses; and a device for delivering this
solution with pressure (or shear force, similar to that provided
by a Waterpik dental water jet system [Waterpik Technolo-
gies, Fort Collins, CO]). We further posited that for this tech-
nique to be effective, the solution would have to interfere with
the calcium-ion bridges that produce gelling and “cross-links”
in the polymeric chains in the EPS structure and thereby
prevent it from dissolving. Thus, the solution should include
both a sequestering agent to address the calcium-ion bridge
and a surfactant molecule to bond with the unbound polymer
chains and dissolve them.
Therefore, we performed preliminary studies with a wide
variety of surfactants, including acetone, isopropanol, tolu-
ene, and several other agents. These investigations allowed
identification of the optimal agent (a combination of citric acid
and zwitterionic surfactant [CAZS; caprylyl sulfobetaine]).
This agent was used subsquently for a controlled in vitro
study of its effectiveness in eradicating S. aureus and P. aerugi-
nosa biofilms formed by bacteria isolated from sinus samples
From the *Centre Hospitalier de l’Universite´ de Montre´al and McGill University
Health Center, Montreal, Quebec, Canada, #Medtronic ENT, Jacksonville, Florida, and
the §Center for Biofilm Engineering, Montana State University, Bozeman, Montana
Presented at the American Rhinologic Society meeting, San Diego, California, April
2007
Financial support for this research and the preparation of the article was provided by
Medtronic ENT, Jacksonville, Florida; Dr. Myntti is an employee of Medtronic ENT;
Dr. Derosiers is a paid consultant to Medtronic ENT
Address correspondence and reprint requests to Matthew Myntti, Ph.D., Materials
Science and Engineering, Medtronic ENT, 6743 Southpoint Drive North, Jacksonville,
FL 32216
E-mail address: matthew.f.myntti@medtronic.com
Copyright © 2007, OceanSide Publications, Inc., U.S.A.
American Journal of Rhinology 527
from patients with CRS. Here, we describe the results
achieved with CAZS delivered under static conditions and
with use of hydrodynamic shear force.
METHODS
This study was approved by the Ethical Review Board for
Human Subjects of the Centre Hospitalier de l’Universite´ de
Montre´al.
Sinus Sample Collection and Initial Processing
Bacterial isolates were recovered from the sinuses of a
consecutive series of patients with sinus disorders. Patients
with cystic fibrosis or an underlying immunosuppressive dis-
ease (human immunodeficiency virus [HIV] infection, insulin-
dependent diabetes mellitus, or renal disease) and patients
who had taken antibiotics or oral prednisone in the previous
month were excluded from the study. All patients had refrac-
tory sinusitis, i.e., persistent symptoms resistant to medical
therapy despite having undergone technically successful
FESS for CRS (with or without nasal polyposis) diagnosed in
accordance with the 2003 AAO-HNS guidelines14 and refrac-
tory to medical therapy for 12 months before sample collec-
tion. In all patients enrolled in the study, the failure of FESS
was judged not to be associated with technical factors such as
obstructive synechiae, frontal sinus obstruction, or a retained
uncinate process. Samples were collected consecutively until
10 specimens each of S. aureus and P. aeruginosa were ob-
tained.
All samples were obtained under direct endoscopic guid-
ance, using the procedure described by Nadel et al.15 Briefly,
a topical anesthetic agent was administered, the nasal ala was
retracted, and an endoscope was used to visualize the middle
meatus and sinus cavities. A thin, flexible calcium alginate
swab (Starwab Collection and Transport System; Starplex Sci-
entific, Etobicoke, Ontario, Canada) was inserted and directed
to the site with the most purulence. If no purulence was
observed, the surface of the maxillary sinus was swabbed for
15 seconds. Care was taken to avoid contact with the lateral
nasal wall or nasal vestibule.
Samples were plated and incubated using standard proce-
dures. Bacteria were identified with the Vitek 2 system (Bi-
ome´rieux, Durham, NC). Crystal violet staining to confirm the
presence of biofilms was performed according to the method
described by Stepanovic et al.16 This method was previously
used to assess biofilms in sinus isolates and provided consis-
tent results.17 For incubation and culture, previously frozen
strains were inoculated on trypticase soy agar (TSA) with
0.5% sheep blood. After 24 hours, one to four colonies per
strain were cultured on TSA. Cultures were incubated at 37°C
for 24 hours to condition them to the trypticase soy broth
(TSB)–TSA medium and ensure noncontamination. Colonies
grown on TSA solid medium were then amplified in 5 mL of
TSB medium with 0.5% glucose18 and incubated at 37°C for 24
hours. For the biofilm-removal experiments, previously fro-
zen strains of S. aureus and P. aeruginosa were inoculated on
TSA in glass tubes and transported by overnight courier to the
Medical Biofilm Laboratory, Center for Biofilm Engineering,
Montana State University (Bozeman, MT).
Laboratory Model System
An in vitro system, the drip-flow reactor (DFR) containing
microscope slides (designated “coupon” in Fig. 1), was used
to determine the effectiveness of the test solutions, delivered
with and without hydrodynamic force, in removing S. aureus
and P. aeruginosa biofilms. The DFR models a low shear
environment. In this study, bacteria were inoculated on hy-
droxyapatite (HA)-coated glass slides because previous inves-
tigations showed that biofilms are better retained on HA
slides than on other surfaces during processing for microscop-
ical analysis.
The DFR experiments were conducted in an incubator at
37°C and under aerobic conditions. Approximately 20 min-
utes before bacterial inoculation, sterile medium (10% TSB for
S. aureus; 1% TSB for P. aeruginosa) was dripped on the slides
in the DFR and allowed to collect over the slides to form a
conditioning layer. The slides were then inoculated with 1 mL
of a culture of either S. aureus or P. aeruginosa. The DFR was
placed in a horizontal position for 4 hours to allow bacterial
attachment to the substrate. Subsequently, the device was
placed at a 10° angle, with sterile medium dripping on the
slides at a rate of 10 mL/hour. After 3 days, the biofilm-
removal experiments were performed.
Biofilm-Removal Experiments
Three methods were used to treat the biofilms formed by
each bacterial species. The first involved applying a solution
of 0.02 M of CAZS (constituent chemicals from Sigma Aldrich,
St. Louis, MO) on slides in the DFR (static treatment). The
other two treatments were delivery of 0.09% saline and deliv-
ery of 0.02 M of CAZS with use of hydrodynamic shearing
force. For all treatments, preliminary runs were done to en-
sure that variations among slides were within acceptable lim-
its. In addition, multiple plates of both bacterial species were
produced to determine the within-run and run-to-run varia-
tions. A control slide was made for each DFR run. Subse-
quently, the experiments included three runs for each treat-
ment of each type of bacteria.
For the static treatments, flow to the DFR was halted, the
device was placed in a horizontal position, and the cover was
removed. Either 25 mL of 0.02 M of CAZS (active treatment)
or 25 mL of 0.9% saline was applied to one slide. Control
Figure 1. DFR used in biofilm-removal experiments.
528 September–October 2007, Vol. 21, No. 5
slides were not treated with either saline or CAZS. After 10
minutes, the slides were rinsed with saline (25 mL). The DFR
was then disconnected from the inflow tube, and each slide
was removed under a laminar flow hood and placed in a
sterile 50-mL tube. After another saline rinse (2 mL), the
surface of the slide was scraped repeatedly, and the scrapings
and saline were collected in the tube. The tube was then
vortexed for 10 seconds, sonicated for 2 minutes, and vor-
texed again for 10 seconds to disperse the bacteria into sus-
pension.
Next, the suspensions were serially diluted and 100-L
aliquots were applied to three plates containing TSA. After
incubation at 37°C for 24 hours, colony-forming units (CFUs)
were counted manually, and the number of CFUs per square
centimeter was calculated. The resulting plate counts were log
(10) transformed and expressed as the mean (
SD) value
derived from plate counts from two DFR runs of three slides
each.
For the hydrodynamic treatments, the slides were removed
from the DFR and placed in a glove box. The slides were
placed in a holder and sprayed for 20 seconds with either
saline or 0.02 M of CAZS (150 mL of fluid) by using a device
that provided pressurized jet lavage. Applied pressure was
31.4 psi with a flow rate of 5.5 mL/s, delivered via a nozzle
with a 0.03-in. diameter. The resulting force on the tissue from
the spray was 10  g, applied over a 0.03-in. diameter (or
0.000706 in.2), remaining constant at distances from 0.1 inch to
1 inch. The spraying was done with both a side-to-side and an
up-and-down sweeping motion so that all areas were sprayed
twice, once in each axis. Then, the slides were placed in sterile
50-mL tubes and scraped as described previously. The scrap-
ings were processed and plate counts were obtained as de-
scribed previously.
Confocal Scanning Laser Microscopy (CSLM)
CSLM was performed on three slides (for each treatment
and bacteria species) not subjected to plate counts to allow
imaging of the biofilm architecture in control and treated
samples. The CSLM method has been found to be especially
effective in assessing bacterial biofilms, including those ob-
served in patients with otorhinolaryngologic diseases such as
OM and CRS.3,19 Staining of slides for CSLM was done with
the BacLight Live/Dead kit (Molecular Probes, Invitrogen,
Carlsbad, CA). This kit consists of two nucleic acid stains:
SYTO 9, which detects living cells by fluorescing green, and
propidium iodide, which detects dead cells by fluorescing
red.
After staining, the slides were examined by using CSLM
(magnification, 630; Leica acoustic-optical beam splitter SP2
with a 2-photon Mai Tai attachment; Leica Microsystems,
Bannockburn, IL) with fluorescence excitation and detection
in both the green and the red spectra. Each slide area was
divided into 10 equally sized segments. A microscopical field
was selected at random from each segment, and images were
obtained at 1-m intervals from the top of the biofilm to the
substrate, thereby creating an image stack for each location.
Statistical Analysis
The mean (
SD) percentage of reduction from control val-
ues in the quantity of S. aureus and P. aeruginosa bacteria (i.e.,
number of CFUs on plates) after each treatment was calcu-
lated. Results were assessed by using two-sample t-tests
(Minitab version 14; Minitab, State College, PA). A value of
p  0.05 was considered a significant difference from the
control value.
RESULTS
Plate Counts
Figures 2 and 3 and Table 1 show quantitative results from
the biofilm-removal experiments. Before treatment, the bio-
films formed in the DFR cultures of both S. aureus and P.
aeruginosa from patients with CRS were extensive, with CFU
counts for these controls ranging from 7.8 to 9.5 log/cm2. For
both bacterial species, all three treatment methods signifi-
cantly reduced the quantity of CFUs from control values.
Static administration of CAZS resulted in a 2.5-log reduction
(5.11  108 to 1.65  106; p  0.001) in the number of S. aureus
Figure 2. Log and CFUs per squared centimeter representations of
mean plate counts for samples of S. aureus for controls and after
treatment with static 0.02 M of CAZS, saline delivered with hy-
drodynamic force (hydrodynamic saline), and 0.02 M of CAZS
solution delivered with hydrodynamic force (hydrodynamic CAZS).
Figure 3. Log and CFUs per squared centimeter representations of
mean plate counts for samples of P. aeruginosa biofilms for controls
and after treatment with static 0.02 M of CAZS, saline delivered
with hydrodynamic force (hydrodynamic saline), and 0.02 M of
CAZS solution delivered with hydrodynamic force (hydrodynamic
CAZS).
American Journal of Rhinology 529
CFUs and a 2.9-log reduction (1.69  109 to 1.91  106; p 
0.002) in the number of P. aeruginosa CFUs. Mechanical dis-
ruption alone (hydrodynamic saline–only treatment) de-
creased the number of S. aureus CFUs by 2.3 logs (5.11  108
to 2.38 106; p 0.001) and the number of P. aeruginosa CFUs
by 2.4 logs (1.69  109 to 7.31  106; p  0.001). Clearly,
however, the delivery of CAZS with hydrodynamic force had
the greatest effect on both bacterial species, decreasing the S.
aureus CFU count by 3.9 logs (5.11  108 to 6.37  104; p 
0.001) and the P. aeruginosa CFU count by 5.2 logs (1.69  109
to 1.04  104; p  0.001)
CSLM Imaging
Figure 4 shows representative CSLM images of control and
treated S. aureus biofilms. The control (Fig. 4 A) shows a thick
biofilm carpeting the slides. Hydrodynamic treatment with
saline and static treatment with CAZS (Figs. 4, B and C)
decreased the amount of biofilm coverage markedly and re-
duced the organization of the remaining biofilm. Hydrody-
namic treatment with CAZS (Fig. 4 D) produced a greater
reduction both in biofilm coverage and in the amount of order
in the biofilm community. These results reflect those of the
plate count assessments with respect to the relative reductions
in the amount of biofilm achieved with each treatment.
DISCUSSION
This in vitro study assessed the effect of three biofilm-
removal treatments on S. aureus and P. aeruginosa biofilms that
formed in specimens of purulence obtained from patients
Table 1 Bacterial plate counts according to type of
treatment
Treatment Type of bacteria*
S. aureus P. aeruginosa
None (no saline) 8.7
 0.4 9.2
 0.2
Static delivery of CAZS 6.2
 0.3 6.3
 1.3
Hydrodynamic delivery of
saline
6.4
 0.2 6.9 
 0.1
Hydrodynamic delivery of
CAZS
4.8
 0.3 4.0 
 0.5
*Values are mean (
SD) numbers of CFUs per centimeter
(log) derived from three plates assessed twice.
Figure 4. Representative CSLM images showing S. aureus biofilm (A) in a control sample from the DFR and (B) in samples obtained from
the DFR after hydrodynamic administration of saline alone, (C) static CAZS treatment, and (D) hydrodynamic CAZS treatment.
530 September–October 2007, Vol. 21, No. 5
with CRS refractory to both medical therapy and FESS. The
DFR used in these experiments allowed growth of the bacteria
of interest and administration of the biofilm-removal treat-
ments in the same device. Of the three treatments investi-
gated, the one using delivery of CAZS by means of pressur-
ized jet lavage was the most effective in breaking up biofilms.
Although some bacteria remained after this treatment, large,
statistically significant reductions occurred, with the mean
decreases in bacterial plate counts being 3.9 and 5.2 log (a
reduction of 10,000–100,000 times), respectively, for the S.
aureus and P. aeruginosa biofilms. We suggest that a decrease
of this magnitude in vitro may translate into a similar reduc-
tion in vivo, with a potentially important clinical impact on
biofilms in patients with CRS.
Our findings are in accordance with those of Anglen et
al.,20 who investigated methods for removing S. epidermidis
biofilms from orthopedic stainless steel screws in vitro.
Their study, which also reported CFU counts as a means of
assessing bacterial reduction, showed that compared with
bulb-syringe irrigation, power irrigation increased the re-
moval of bacteria by a factor of at least 100, regardless of
the type of solution used. The effect of mechanical irriga-
tion on the bacteria was 100-fold that was achieved with
neomycin alone and 285-fold that was obtained with poly-
myxin alone. Adding liquid Castile soap to the irrigation
solution as a surfactant dramatically increased the quantity
of bacteria removed, whereas adding antibiotics produced
results not significantly different from those achieved with
saline alone.
In the study reported here, we also found that power
irrigation had considerable biofilm-reducing effects. How-
ever, we additionally showed that the presence of a surfactant
and citric acid in the irrigation solution significantly enhanced
the reduction in CFU counts in both P. aeruginosa and Staph-
ylococcus species biofilms.
The observation in both of these studies that a substance
other than an antibiotic was effective in removing biofilms
is especially intriguing with respect to possible clinical
implications. Currently, orally and topically administered
antibiotics are used extensively and, in many cases, unsuc-
cessfully, to treat CRS. This may be because of a reduced
effectiveness of antibiotics on bacterial biofilms. In support
of this idea, in vitro studies have shown that topical appli-
cation of even high concentration of antibiotics have only a
limited effect on bacterial viability in bacterial biofilms
from sinus isolates.21 However, the use of even a topical
antibiotic is not without risk because of the possibility that
a patient will have an allergic reaction to the agent or a
resistant antibiotic strain will develop. Antibiotics also are
relatively expensive. In contrast, Castile soap and CAZS are
inexpensive and nontoxic. The CAZS solution used in our
study was a blend of deionized water and citric acid buff-
ered to 5.4 pH with sodium citrate and containing a low
percentage of zwitterionic surfactant. The role of the citric
acid is to break the calcium-ion bridges that serve as chem-
ical binding sites connecting the EPS polymeric chains. The
surfactant brings the disconnected chains into solution.
Because of the relatively low pH of the CAZS solution and
the known relatively low toxicity of its constituent chemi-
cals, we suggest that it is unlikely that the solution would
adversely affect living tissue exposed to it.
Our study was an in vitro investigation, but its results
support the idea that delivery of a simple, soap-like sur-
factant with hydrodynamic force can disrupt biofilms
formed by bacteria commonly observed in otorhinolaryn-
gologic disorders, including CRS. In the light of the grow-
ing evidence of the presence of bacterial biofilms in patients
with refractory CRS,4–6,8–10 we believe that studies of hy-
drodynamic treatment of biofilms in living tissue and ani-
mal models of CRS are warranted. Such research may pro-
vide data that could aid development of a new treatment
for CRS that would benefit the many patients in whom the
best available medical and surgical therapy has failed to
produce a cure.
ACKNOWLEDGMENTS
The authors thank Rene´e J. Robillard for writing and editorial
assistance.
REFERENCES
1. Costerton JW, Stewart PS, and Greenberg EP. Bacterial bio-
films: A common cause of persistent infections. Science 284:
1318–1322, 1999.
2. Morris DP, and Hagr A. Biofilm: Why the sudden interest?
J Otolaryngol 34(suppl 2):S56–S59, 2005.
3. Hall-Stoodley L, Hu FZ, Gieseke A, et al. Direct detection of
bacterial biofilms on the middle-ear mucosa of children with
chronic otitis media. JAMA 296:202–211, 2006.
4. Sanderson AR, Leid JG, and Hunsaker D. Bacterial biofilms
on the sinus mucosa of human subjects with chronic rhino-
sinusitis. Laryngoscope 116:1121–1126, 2006.
5. Sanclement JA, Webster P, Thomas J, and Ramadan HH.
Bacterial biofilms in surgical specimens of patients with
chronic rhinosinusitis. Laryngoscope 115:578–582, 2005.
6. Bendouah Z, Barbeau J, Hamad WA, and Desrosiers M.
Biofilm formation by Staphylococcus aureus and Pseudomonas
aeruginosa is associated with an unfavorable evolution after
surgery for chronic sinusitis and nasal polyposis. Otolaryn-
gol Head Neck Surg 134:991–996, 2006.
7. Bhattacharyya N, and Kepnes LJ. The microbiology of recur-
rent rhinosinusitis after endoscopic sinus surgery. Arch Oto-
laryngol Head Neck Surg 125:1117–1120, 1999.
8. Ferguson BJ, and Stolz DB. Demonstration of biofilm in
human bacterial chronic rhinosinusitis. Am J Rhinol 19:452–
457, 2005.
9. Cryer J, Schipor I, Perloff JR, and Palmer JN. Evidence of
bacterial biofilms in human chronic sinusitis. ORL J Otorhi-
nolaryngol Relat Spec 66:155–158, 2004.
10. Ramadan HH, Sanclement JA, and Thomas JG. Chronic
rhinosinusitis and biofilms. Otolaryngol Head Neck Surg
132:414–417, 2005.
11. Meltzer EO, Hamilos DL, Hadley JA, et al. Rhinosinusitis:
Establishing definitions for clinical research and patient care.
J Allergy Clin Immunol 114(suppl):S155–S212, 2004.
12. Chiu AG, and Kennedy DW. Surgical management of
chronic rhinosinusitis and nasal polyposis: A review of the
evidence. Curr Allergy Asthma Rep 4:486–489, 2004.
13. Bhattacharyya N. Clinical outcomes after endoscopic si-
nus surgery. Curr Opin Allergy Clin Immunol 6:167–171,
2006.
14. Benninger MS, Ferguson BJ, Hadley JA, et al. Adult chronic
rhinosinusitis: Definitions, diagnosis, epidemiology, and
pathophysiology. Otolaryngol Head Neck Surg 129(suppl
3):S1–S32, 2003.
15. Nadel DM, Lanza DC, andKennedyDW. Endoscopically guided
cultures in chronic sinusitis. Am J Rhinol 12:233–241, 1998.
American Journal of Rhinology 531
16. Stepanovic S, Vukovic D, Dakic I, et al. A modified
microtiter-plate test for quantification of staphylococcal
biofilm formation. J Microbiol Methods 40:175–179,
2000.
17. Bendouah Z, Barbeau J, Hamad WA, et al. Use of an in
vitro assay for determination of biofilm-forming capacity
of bacteria in chronic rhinosinusitis. Am J Rhinol 20:434–
438, 2006.
18. Gotz F. Staphylococcus and biofilms. Mol Microbiol 43:1367–
1378, 2002.
19. Ha KR, Psaltis AJ, Tan L, and Wormald PJ. A sheep model
for the study of biofilms in rhinosinusitis. Am J Rhinol
21:339–345, 2007.
20. Anglen JO, Apostoles S, Christensen G, and Gainor B. The
efficacy of various irrigation solutions in removing slime-
producing Staphylococcus. J Orthop Trauma 8:390–396,
1994.
21. Desrosiers M, Bendouah Z, and Barbeau J. Effectiveness of
topical antibiotics on Staphylococcus aureus biofilm in vitro.
Am J Rhinol 21:149–154, 2007. e
532 September–October 2007, Vol. 21, No. 5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.