ORIGINAL ARTICLE

Biofilm Detection With Hematoxylin-Eosin Staining

Christian J. Hochstim, MD, PhD; Judy Yujin Choi, BA; Derek Lowe, MD; Rizwan Masood, PhD; Dale H. Rice, MD

Objective: To demonstrate that hematoxylin-eosin
staining can be used to detect the presence of bacterial
biofilm in patients with chronic rhinosinusitis (CRS).
Design: A prospective study.
Setting: The University of Southern California University
Hospital and the Department of Otolaryngology–Head
and Neck Surgery, University of Southern California, Keck
School of Medicine, Los Angeles.
Patients: A total of 34 patients: 24 undergoing endo-
scopic sinus surgery for CRS and 10 undergoing septo-
plasty with or without turbinate reduction with no his-
tory of sinusitis, were enrolled in the study.
Main Outcome Measures: Contiguous sections from
patient samples were examined by both hematoxylin-
eosin staining and fluorescence in situ hybridization
(FISH) with the bacterial-specific probe EUB338 for evi-
dence of bacterial biofilm.
Results: Biofilm was detected by hematoxylin-eosin stain-
ing in 15 of 24 patients with CRS and 1 of 10 control pa-
tients. In all cases, hematoxylin-eosin staining was found
to be an accurate predictor of the presence or absence of
biofilm using FISH as a control standard.
Conclusion: Hematoxylin-eosin staining of surgical speci-
mens is a reliable and available method for the detection
of bacterial biofilm in chronic infectious disease.
Arch Otolaryngol Head Neck Surg. 2010;136(5):453-456

Author Affiliations:
Department of
Otolaryngology–Head and Neck
Surgery and Keck School of
Medicine (Drs Hochstim, Lowe,
Masood, and Rice and Ms Choi)
and Department of Pathology
(Dr Masood), University
of Southern California,
Los Angeles.
W
E DESCRIBE FOR THE
first time, to our
knowledge, how to
detect biofilm on
routine hematoxy-
lin-eosin (H-E) preparations. The Centers
for Disease Control and Prevention esti-
mate that at least 65% of human bacterial
infectious processes involve biofilms.1 A bio-
film is a group of bacteria associated with a
surface, such as a mucous membrane, and
enclosed in a matrix of extracellular poly-
saccharide material. Bacterial biofilm has
been implicated in many human diseases.2
These include cystic fibrosis, otitis media,
periodontitis, chronic prostatitis, and arti-
ficial joint infections, among others. De-
spite this, biofilm has been difficult to iden-
tify and study. Multiple types of bacteria can
be living in a biofilm structure. They are fre-
quently difficult to culture.3 Therefore, di-
agnosis is often difficult. Bacteria in bio-
films have numerous defense mechanisms,
and therefore their response to antibiotics
is usually incomplete.4
Bacteria in the biofilm state live in a com-
munity protected by a 3-dimensional extra-
cellular polymeric substance that constitutes
90% of the volume and can account for more
than 50% of the total carbon content. Bio-
films may be composed of mixed species of
bacteria surrounded by this extracellular
polymeric substance, which is a physical,
noncellularbarrier.StaphylococcusandPseu-
domonas species are frequently implicated
in sinonasal disease biofilms.
Traditionally, biofilm has been de-
tected by scanning electron microscopy5 or
confocal scanning laser microscopy.6 More
recently, fluorescence in situ hybridiza-
tion (FISH) has been used.7 In the course
of our ongoing institutional review board–
approved study of biofilm in patients with
chronic rhinosinusitis (CRS), we discov-
ered that biofilm can be reliably detected
with ordinary H-E preparations. This can
be performed on fresh tissue as well as ar-
chival material. Histologically, biofilm ap-
pears as clusters of basophilic bacteria and
host cells entrapped in a layer of extracel-
lular polymeric substance. To our knowl-
edge, this has not been previously de-
scribed. To verify this, we performed the
following prospective study.
METHODS
Biopsy specimens were obtained for analysis
from patients undergoing either endoscopic si-

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 136 (NO. 5), MAY 2010 WWW.ARCHOTO.COM
453

©2010 American Medical Association. All rights reserved.

Downloaded From: http://archotol.jamanetwork.com/ on 05/27/2015

Table. Amount of Bacterial Biofilm Present
on Hematoxylin-Eosin–Stained Sections a
Patients, No. (%)
Extensive Biofilm Biofilm
Group Amount Present Absent
Patients with CRS (n=24) 7 (29) 8 (33) 9 (38)
Controls (n=10) 0 1 (10) 9 (90)
Abbreviation: CRS, chronic rhinosinusitis.
a Biofilms were classified as extensive (ⱖ50% of mucosal surface involved),
present (⬍50% involved), or absent. The percentage of patients with CRS with
detectable biofilm was significantly greater than that of control patients (62% vs
10%; P=.008, Fisher exact test).
nus surgery for CRS (n=24 study patients) or septoplasty with
or without turbinate reduction with no history of sinusitis (n=10
controls). All patients gave informed consent, and the study
was approved by the institutional review board.
TISSUES
Samples of nasal mucosa were taken from the ethmoid sinus
in study patients and from the middle turbinate in controls.
These mucosal specimens were fixed in 4% formalin for 48 hours
and processed and embedded into paraffin blocks according to
routine procedures. Fresh samples were collected in phosphate-
buffered saline (PBS) on ice for analysis by FISH or Baclight
staining (DEAD Baclight kit; Invitrogen, Molecular Probes, Carls-
bad, California), and confocal microscopy.
HISTOLOGIC EXAMINATION
Contiguous sections were taken for H-E preparation and for
FISH. Sections were stained with H-E using standard patho-
logic procedures. Briefly, sections were deparaffinized in
xylene (2 ⫻ 5 minutes) and rehydrated with successive
1-minute washes in 100%, 96%, 80%, and 70% ethanol. They
were then stained with hematoxylin (2 minutes), rinsed with
distilled water, rinsed with 0.1% hydrochloric acid in 50%
ethanol, rinsed with tap water for 15 minutes, stained with
eosin for 1 minute, and rinsed again with distilled water. The
slides were then dehydrated with 95% and 100% ethanol suc-
cessively followed by xylene (2 ⫻ 5 minutes) and mounted
with coverslips.
All paraffin-embedded samples analyzed by H-E staining in
this study were also subjected to FISH using the bacterial-
specific probe EUB338 (5⬘-Cy3-GCTGCCTCCCGTAGGAGT-
3⬘) 4,6-diamidino-2-phenylindole (DAPI; Vector Laborato-
ries, Burlingame, California), which is hybridized to the
conserved 16S ribosomal RNA sequence found in nearly all bac-
terial species.8 Thus, FISH with the EUB338 probe will specifi-
cally label bacteria but not eukaryotic cells within a tissue sample.
The protocol was performed as previously described.9 Briefly,
sections were deparaffinized, rehydrated, and postfixed in 4%
paraformaldehyde for 5 minutes. Sections were washed in PBS
and then treated with 10 mg/mL of lysozyme in Tris EDTA buffer
for 10 minutes at room temperature. Slides were preincubated
in hybridization buffer (0.9M of sodium chloride; 20mM of Tris
hydrochloric acid, pH 8; 0.01% sodium dodecyl sulfate [SDS];
30% formamide) for 5 minutes at 35° C, and then hybridized
overnight (12-18 hours) at 35° C with a 5 µg/mL final concen-
tration of 5⬘Cy3-labeled EUB338 probe in hybridization buffer.
Next, 2⫻15-minute stringent washes were performed at 37°C
in washing buffer (65mM sodium chloride, 20mM Tris hydro-
chloride [pH 8], 5mM EDTA, and 0.01% SDS). Slides were
(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 136 (NO. 5), MAY 2010
454
©2010 American Medical Association. All rights reserved.
Downloaded From: http://archotol.jamanetwork.com/ on 05/27/2015
washed in PBS and mounted with Vectashield (Vector Labo-
ratories) mounting medium with DAPI.
LIVE/DEAD BACLIGHT STAINING
Fresh tissue samples were obtained in parallel from selected
cases stained with SYTO9 (Invitrogen) and propidium
iodide using the LIVE/DEAD Baclight kit, as previously
described. 5 These samples were then prepared as whole
mounts with Vectashield and coverslips. The SYTO9 green
stains only live cells in a fresh sample, whereas the prop-
idium iodide red stains only dead or dying cells. The SYTO9
was used to verify live bacteria in the biofilm, which are
smaller in size than host cells.
MICROSCOPY
Hematoxylin-eosin–stained sections were analyzed by light
microscopy using a Leica DM LB2 epifluorescence micro-
scope (Wetzlar, Germany). The FISH and stained samples
were analyzed using the same microscope with UV and
Cy3-, DAPI-, and fluorescein isothiocyanate–specific filters.
Images that were original magnification ⫻20 and ⫻40 of
H-E staining and FISH were acquired with a CCD digital
camera (model 7.2; Diagnostic Instruments, Sterling
Heights, Michigan). A Zeiss 510 laser scanning confocal
microscope (Göttingen, Germany) was used to acquire
images that were original magnification ⫻40 and ⫻63 of
LIVE/DEAD Baclight–stained samples.
RESULTS
Of the 24 patients with CRS, 15 had biofilm detected by
H-E (62%). In the 10 control patients, only 1 had bio-
film (10%). This difference was statistically significant
(P=.008, Fisher exact test). The amount of biofilm present
in each sample was categorized as extensive, present, or
absent for both groups (Table). Patient samples classi-
fied as having extensive biofilm had involvement of 50%
or more of the mucosal surface analyzed, whereas any
amount of biofilm less than this was classified as present.
All biofilms detected by H-E were also detected by FISH.
Furthermore, all patient samples classified as negative for
biofilm by H-E were also classified as negative for bio-
film by FISH, which was considered positive for biofilm
when a significant quantity of staining was observed either
covering the epithelial surface or in clusters along the sur-
face. Therefore, in this study, H-E correctly identified the
presence or absence of biofilm in all cases, using FISH
as a control standard. In selected cases, fresh samples of
patient tissue were collected in parallel for Baclight LIVE/
DEAD staining and examination by confocal laser scan-
ning microscopy (CLSM). The detection of biofilm by
CLSM was consistent with that of H-E staining and FISH
(Figure 1).
Of interest, 3 of the patients with CRS had samples
with detectable fungal elements on H-E (Figure 2),
whereas no fungal elements were present in any control
patient samples. Two of the 3 patients with CRS with fun-
gal elements had no bacterial biofilms. In the 1 patient
with both bacterial biofilm and fungal elements, there was
no apparent overlap of these entities, which were iden-
tified in separate regions of the sample.
WWW.ARCHOTO.COM
 H-E EUB338 DAPI SYTO9 PL

A B C

CRS

50 µm 20 µm

D E F

Control

Figure 1. Detection of bacterial biofilm by hematoxylin-eosin stain (H-E), fluorescent in situ hybridization (FISH), and confocal laser scanning microscopy (CLSM).
Representative images of a patient with chronic rhinosinusitis (CRS) with bacterial biofilm (A, B, C) and a control patient without bacterial biofilm (D, E, F).
A and D, Hematoxylin-eosin–stained sections reveal small basophilic bacterial clusters (arrow) on the surface epithelium of the sample of patients with CRS (A),
whereas the control patient sample (D) shows normal eosinophilic ciliated respiratory epithelium. B and E, FISH with the bacterial-specific probe EUB338
4,6-diamidino-2-phenylindole (DAPI) (Vector Laboratories, Burlingame, California) reveals positive staining of bacterial clusters (arrow) at the epithelial surface in
the CRS sample (B), whereas there is no staining in the control (E). Panels B and E are contiguous with A and D, respectively. C and F, CSLM reveals SYTO9
(Invitrogen, Carlsbad, California) positive live (PL) bacterial elements (arrows) in the CRS sample (C), whereas only much larger epithelial cellular staining is
observed in the control sample (D). Panels A, B, D, and E are original magnification ⫻20, whereas panels C and F are original magnification ⫻63.

COMMENT A B

Biofilm can be detected on routine H-E. It can be seen as

irregularly shaped groupings of small basophilic bacte-
ria, exopolymeric substance, and entrapped erythro-
cytes and leukocytes resting on the surface epithelium
(Figure 1 and Figure 2). Depending on fixation and pro- 50 µm
cessing, it may be tightly adherent to the surface epithe-
lium or pulled away slightly. Precise identification of the
C D
bacterial species involved in biofilms still requires cul-
turing, following physical disruption of the biofilm, or
molecular methods, such as FISH, with bacterial species–
specific probes. Because there is increasing evidence that
biofilm plays an important role in many chronic dis-
eases, it is important to identify easier and cheaper meth-
ods to study biofilm in clinical samples. In particular, the
wide availability of H-E staining of surgical specimens Figure 2. Appearance of bacterial biofilms on hematoxylin-eosin
through clinical pathology laboratories makes this a highly (H-E)–stained sections. Images of surface epithelium from H-E–stained
practical method for detecting biofilm in clinical prac- sections of patients with chronic rhinosinusitis with varying appearance of
tice. Gram staining is another simple histological tech- biofilm. A, Basophilic bacterial clusters (arrows) line the epithelial surface.
B, Extracellular polysaccharide substance (EPS) material with embedded
nique that has been previously used in conjunction with bacteria and entrapped erythrocytes and leukocytes, partially sheared from
other methods to describe biofilms.10 We are currently the epithelial surface. C, Dense EPS material with embedded basophilic
investigating the efficacy of Gram staining in detecting bacteria coating the epithelial surface. D, Image from a sample from a patient
without bacterial biofilm but in which fungal elements were identified
biofilms in CRS compared with H-E staining and FISH. (arrowhead). All images are original magnification ⫻20.
Although specific treatments are not currently available
to target biofilm, awareness of the presence of biofilm by
physicians may affect overall patient treatment and follow-
up, given the association between biofilm and potential outcomes will be needed to determine whether the ac-
for treatment failure and persistent symptoms.11 There- tual amount of biofilm detected is also an important fac-
fore, the presence of biofilm in tissue samples from pa- tor correlating with risk of persistent symptoms or whether
tients with CRS should be reported by pathology labo- merely reporting the presence or absence of biofilm is
ratories when noted. Additional studies of patient sufficient.

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 136 (NO. 5), MAY 2010 WWW.ARCHOTO.COM
455

©2010 American Medical Association. All rights reserved.

Downloaded From: http://archotol.jamanetwork.com/ on 05/27/2015
 Submitted for Publication: May 18, 2009; final revi-
sion received October 30, 2009; accepted November
18, 2009.
Correspondence: Dale H. Rice, MD, Department of Oto-
laryngology–Head and Neck Surgery, Keck School of
Medicine, Room 4136, 1200 N State St, Los Angeles, CA
90033 (dhrice@usc.edu).
Author Contributions: All authors had full access to
all the data in the study and take responsibility for the
integrity of the data and the accuracy of the data
analysis. Study concept and design: Hochstim, Choi,
and Rice. Acquisition of data: Hochstim, Choi, Lowe,
Masood, and Rice. Analysis and interpretation of data:
Hochstim, Choi, and Masood. Drafting of the manu-
script: Hochstim, Choi, and Rice. Critical revision of
the manuscript for important intellectual content: Hoch-
stim, Lowe, Masood, and Rice. Statistical analysis:
Hochstim. Administrative, technical, and material sup-
port: Choi, Lowe, and Rice. Study supervision: Choi,
Masood, and Rice.
Financial Disclosure: None reported.
Funding/Support: This study was supported by funds
from the Department of Otolaryngology–Head and
Neck Surgery of the University of Southern California
(USC).
Additional Contributions: We received invaluable
assistance from Parish P. Sedghizadeh, DDS, Amita
Gorur, MS, and Christoph Schaudinn, PhD, of the
USC Center for Biofilms. Alexander Markarian, MD, of
the Department of Otolaryngology, assisted by provid-
ing surgical specimens. Christi Gordon and the
(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 136 (NO. 5), MAY 2010
456
©2010 American Medical Association. All rights reserved.
Downloaded From: http://archotol.jamanetwork.com/ on 05/27/2015
Pathology Department staff at USC University Hospital
assisted with patient samples.
REFERENCES
1. Costerton W, Veeh R, Shirtliff M, Pasmore M, Post C, Ehrlich G. The application
of biofilm science to the study and control of chronic bacterial infections. J Clin
Invest. 2003;112(10):1466-1477.
2. Potera C. Forging a link between biofilms and disease. Science. 1999;283(5409):
1837.
3. Lewis K. Persister cells, dormancy and infectious disease. Nat Rev Microbiol.
2007;5(1):48-56.
4. Stewart PS, Costerton JW. Antibiotic resistance of bacteria in biofilms. Lancet.
2001;358(9276):135-138.
5. Ferguson BJ, Stolz DB. Demonstration of biofilm in human bacterial chronic
rhinosinusitis. Am J Rhinol. 2005;19(5):452-457.
6. Psaltis AJ, Ha KR, Beule AG, Tan LW, Wormald PJ. Confocal scanning laser mi-
croscopy evidence of biofilms in patients with chronic rhinosinusitis. Laryngoscope.
2007;117(7):1302-1306.
7. Sanderson AR, Leid JG, Hunsaker D. Bacterial biofilms on the sinus mucosa of
human subjects with chronic rhinosinusitis. Laryngoscope. 2006;116(7):1121-
1126.
8. Daims H, Brühl A, Amann R, Schleifer KH, Wagner M. The domain-specific probe
EUB338 is insufficient for the detection of all bacteria: development and evalu-
ation of a more comprehensive probe set. Syst Appl Microbiol. 1999;22(3):
434-444.
9. Vasquez N, Mangin I, Lepage P, et al. Patchy distribution of mucosal lesions in
ileal Crohn’s disease is not linked to differences in the dominant mucosa-
associated bacteria: a study using fluorescence in situ hybridization and tempo-
ral temperature gradient gel electrophoresis. Inflamm Bowel Dis. 2007;13(6):
684-692.
10. Chole RA, Faddis BT. Anatomical evidence of microbial biofilms in tonsillar tis-
sues: a possible mechanism to explain chronicity. Arch Otolaryngol Head Neck
Surg. 2003;129(6):634-636.
11. Psaltis AJ, Weitzel EK, Ha KR, Wormald PJ. The effect of bacterial biofilms on
post-sinus surgical outcomes. Am J Rhinol. 2008;22(1):1-6.
WWW.ARCHOTO.COM