ORIGINAL REPORTS
Bacterial Adherence to Suture Materials
Brendan D. Masini, MD,* Daniel J. Stinner, MD,* Scott M. Waterman, MD,† and
Joseph C. Wenke, PhD‡
*Brooke Army Medical Center, Ft Sam Houston, Texas, †Walter Reed Army Medical Center, Washington,
DC, and ‡United States Army Institute of Surgical Research, Ft Sam Houston, Texas
BACKGROUND: Wound infections may be problematic for
physicians. Whether a practitioner is managing complex pene-
trating trauma or a skin biopsy, there may be a need for suture
closure. Suture material is an operator dependent variable and
while little objective data exist to guide the choice of suture, it
may play a role in wound infection. This study evaluates bac-
terial adherence to commonly used suture materials with a bio-
luminescent in vitro model.
METHODS: In all, 11 strands of size 2-0 poliglecaprone suture
(Monocryl; Ethicon, Inc, Somerville, New Jersey), polypropyl-
ene suture (Prolene; Ethicon, Inc), silk suture (Ethicon, Inc),
polyglycolic acid suture (Vicryl; Ethicon, Inc), and antimicro-
bial polyglycolic acid suture treated with triclosan (VicrylPlus;
Ethicon, Inc) were immersed in a broth of Staphylococcus aureus
engineered to emit photons. After biofilm formation, the suture
strands were irrigated and imaged with a photon-capturing
camera system yielding a total photon count that correlates with
residual bacteria.
RESULTS: The Vicryl suture had the highest counts and was
statistically significant in bacterial adherence versus all other
sutures. No other suture material was significantly different
from any other.
CONCLUSIONS: This study gives data to guide the selection
of suture materials. Absorbable braided suture should not be
used in closure of contaminated wounds or wounds at risk for
developing infection. The antibiotic impregnated absorbable
braided suture was similar to the other suture types; however, it
is at risk for reverting to the properties of its untreated coun-
terpart over time. The bacterial adherence of suture materi-
als should be taken into account by all practitioners when
closing wounds or debriding infected wounds. (J Surg 68:
101-104. © 2011 Association of Program Directors in Sur-
gery. Published by Elsevier Inc. All rights reserved.)
KEY WORDS: suture, infection, bacterial adherence, wound
complication
Correspondence: Inquiries to Brendan D. Masini, MD, Department of Orthopaedics and
Rehabilitation, Brooke Army Medical Center, 3851 Roger Brooke Drive, Ft Sam Hous-
ton, TX 78234; fax: (210) 916-7323; e-mail: Brendan.Masini@amedd.army.mil
Journal of Surgical Education • © 2011 Association of Program Directors in Surgery
Published by Elsevier Inc. All rights reserved.
COMPETENCIES: Patient Care, Medical Knowledge, Practice
Based Learning and lmprovement
INTRODUCTION
Wound infections may be problematic for physicians. Whether
a practitioner is managing complex penetrating trauma or a skin
biopsy, there may be a need for suture closure. Suture material
is an operator dependent variable, and although it may have a
significant role in wound infection management, little objective
data exist to guide material selection. Much of what we “know”
about suture material selection is passed down as dogma in
our surgical training programs without data-driven recom-
mendations. We know that suture material is a foreign body
that potentiates infection when implanted and that the abil-
ity of the sutured tissue to resist infection varies on the
material implanted.1 Suture functions to approximate tissue
edges, which promotes healing and at the same time limits
further deep contamination. Often, whether by the nature of a
wound itself or by a later inoculation event, suture is present in
a wound that has been exposed to a bacterial insult. The suture
may potentiate the infection by harboring adherent bacteria.
Previous studies have examined the bacterial adherence proper-
ties of sutures via methods, such as plate cultures, electron mi-
croscopy, and radiolabeling of bacteria, and have suggested that
not only does bacterial adherence vary by the physical proper-
ties of the suture material but also that increased adherence of
bacteria to suture directly correlates to the ability of that con-
taminated suture to cause a wound infection.2,3
The clinical problem addressed herein is one most basic to
the practice of surgical medicine. In the course of wound man-
agement, suture may be required for closure of wounds at risk
for contamination, or wounds that have been closed with the
expectation of sterility may become contaminated or infected.
Suture materials clearly cannot overcome sound surgical prin-
ciples, such as thorough debridement and not closing contam-
inated wounds; however, knowledge of the behavior of bacteria
on the surface of a suture material only enhances the surgeon’s
ability to make appropriate choices for the patient. The conflict
present is the lack of clear recommendations from previous
1931-7204/$30.00 101
doi:10.1016/j.jsurg.2010.09.015
literature to make a compelling argument in support of current
surgical training dogma.
This study uses a novel imaging technique to quantify bac-
terial adherent to commonly used suture materials and we hy-
pothesize that there will be an increase in bacterial adherence to
polyglycolic acid sutures versus monofilament sutures.
MATERIALS AND METHODS
Suture Selection
The sutures selected for study were chosen to represent several
common categories of suture material and basic construction
properties. Poliglecaprone suture (Monocryl; Ethicon, Inc,
Somerville, New Jersey), which is a copolymer of glycolide and
epsilon-caprolactone, is a synthetic absorbable monofilament
suture. It has been found to be nonantigenic and nonpyogenic,
and it elicits slight tissue reaction during absorption. Polypro-
pylene suture (Prolene; Ethicon, Inc) is a synthetic nonabsorb-
able monofilament. It has a high tensile strength and has been
found to have minimal tissue reactivity. Silk suture (Ethicon,
Inc) is a nonabsorbable suture composed of an organic protein
called fibroin derived from the domesticated species Bombyx
mori. The silk for the suture material is processed to remove the
natural waxes and gums. Polyglycolic acid suture (Vicryl; Ethi-
con, Inc) is a synthetic absorbable braided suture composed of a
copolymer made of 90% glycolide and 10% l-lactide. This fam-
ily of sutures has been found to elicit only a slight tissue reaction
during absorption. The antimicrobial polyglycolic acid suture
(VicrylPlus; Ethicon, Inc) has a triclosan antimicrobial coating
to limit bacterial growth.
All sutures were size 2-0. They were obtained from commer-
cially available, unexpired, sterilized packets. All sutures were
taken directly from the package into the bacterial broth without
additional machinations. Interventions, such as knot tying,
abrasion with forceps or needle driver, or immersion in a serum
broth to simulate an in vivo environment could all reasonably
be expected to alter the bacterial adherence to the materials and
were not performed in effort to maintain an environment that
supported the ideal performance characteristics of each suture
material.
Bioluminescent Bacteria
The bacterial broth prepared for this investigation consisted of
108 c.f.u./ml Staphylococcus aureus (lux) (Xenogen 29, Caliper
Life Science, Hopkinton, MA). These bacteria are genetically
engineered to emit photons, allowing for quantification with a
photon-counting camera system. Bioluminescent bacteria emit
light in proportion to their number allowing correlation of
photon counts with bacterial counts.4 – 6 Postinoculation plat-
ing of the broths was performed to confirm the consistency of
concentrations in each batch prepared. The concentration se-
lected for this study has previously been demonstrated to cause
clinical infection in a large animal wound model in our labora-
102 Journal of Surgical Education • Volume 68/Number 2 • March/April 2011
tory and consistency in this area may facilitate future in vivo
study of this question. It is greater than concentrations of bac-
teria that are found to cause infection in human models and
thus represents a worst-case scenario for suture contamination.
Sutures that demonstrate low bacterial adherence in this model
can be expected to perform well in a wound where considerably
less contamination would likely be present.
Inoculation
In all, 11 strands each of the suture materials were taken from
the sterile, unopened, unexpired packages and cut into 10-cm
strands. This was the width of our custom suture frame and also
represented the image field for our camera system. The strands
were immersed in the broth of bioluminescent Staphylococcus
aureus (lux).
After 12 hours in the broth, sufficient time to allow biofilm
formation,7 the suture strands were removed individually from
the broth and irrigated with 10 ml of normal saline expressed at
low pressure from a syringe. In model development, broth im-
mersion of 24 and 48 hours showed no greater bacterial adher-
ence to the sutures. Also, this irrigation protocol was found in
model testing to reproducibly remove nonadherent bacteria.
Quantification
Once removed from the broth and irrigated, the suture was
suspended in a custom frame and placed within our dark-box
for imaging. The IVIS100 imaging system (Xenogen Corpo-
ration, Alameda, California), uses an optical charge couple
device camera to count photon emissions. Imaging software
(LIVINGIMAGE V. 2.12; Xenogen Corporation and IGOR
v.4.02A, WaveMetrics, Lake Oswego, Oregon) was used to
superimpose the photon count onto a gray-scale background
image yielding the location and photon intensity. A standard
size region of interest (ROI) was placed around the suture on
the image and from this ROI the total photon count was taken.
This photon count correlates directly with the bacteria adherent
to the suture material. This process was repeated for each suture
with a rotating selection of each suture type from the broth to
minimize any small difference that would exist between the
times of exposure to the broth. Of note, only living bacteria
continue to release photons. Nonviable bacteria even if present
on the imaged sutures are not counted by our camera system.
Statistics
A 1-way analysis of variance was performed to compare means.
When the PHYLIP null hypothesis of equal means was rejected,
the Fisher least significant difference procedure was performed
for pairwise comparisons of the groups. SAS Statistical Software
(SAS, Inc, Cary, North Carolina) was used for all statistical
calculations. The alpha was set as 0.05.
 FIGURE 1. Mean photon counts ⫾ standard error of the mean.
FIGURE 2. Composite image of bioluminescent bacteria on sutures. a,
Vicryl (Ethicon, Inc). b, Silk (Ethicon, Inc). c, Monocryl (Ethicon, Inc). d,
RESULTS VicrylPlus (Ethicon, Inc). e, Prolene (Ethicon, Inc).

The mean photon counts are shown in Fig. 1. Table 1 shows the
p values for comparison between the groups. The polyglycolic than monofilament sutures.9 This study supports using caution
acid suture had the highest mean counts and had significantly with selecting braided sutures; however, two braided materials
greater bacterial adherence when compared with each of the currently studied showed statistically different bacterial adher-
other suture groups tested. No other group shows a statistical ence. This indicates that it is not necessarily the braid itself that
difference with any other group, including the other braided predisposes adherence but the specific material and likely the
suture tested (silk suture) or the antimicrobial polyglycolic acid proprietary weave of the suture material. Silk sutures represent
suture. Figure 2 is a composite photograph of the biolumines- a category of natural material in a braided suture that is not
cent imaging for each suture. commonly used in the wound closure applications, such as the
synthetic braided suture tested here. These results, however,
Discussion suggest that silk suture present in a wound that has become
contaminated might clear its bacterial burden with appropriate
The selection of surgical suture is a surgeon-dependent variable. irrigation and debridement, whereas the polyglycolic acid su-
Previous investigations have outlined criteria to use when se- ture likely would harbor bacteria after similar debridement and
lecting suture materials. These include resistance to traction, irrigation. This observation manifests clinically in the situation
knot stability, absence of traumatizing effects to tissue, elastic- of traumatic amputation where silk sutures may be used to
ity/plasticity, constant time of resorption, high tolerance, low achieve hemostasis and are typically left in place even if subse-
allergenic properties toward host, and low bacterial adherence.8 quent debridement and irrigation is required for decontamina-
Bacterial adherence of sutures has not historically been easy to tion of the wound before definitive closure.
evaluate and quantify, and this study has applied a novel Another method of evaluating bacteria on suture linked a
method to evaluate this source of mismatch between quality contaminated medium to sterile medium with a suture bridge
scientific literature and surgical training dogma. to assess the transmission of bacteria along the sutures.10 This
Previous studies have evaluated tissue inflammation to assess study concluded that polyglycolic acid sutures did not transmit
the effect of suture on the host environment; many studies bacteria as effectively as silk sutures; however, the sutures trans-
found that braided sutures had greater inflammatory response mitted more bacteria than catgut or steel. That finding would
support the polyglycolic acid suture as being less adherent to
bacteria. This finding is in contradiction to the findings of this
TABLE 1. p Values
study. The likely reason for this has to do with our study design.
Suture p Value The migration of the suture along bacteria is an interesting
Vicryl vs Prolene ⬍0.001 property but likely is related more to the absorbency of a suture
Vicryl vs VicrylPlus ⬍0.001 material rather than the true bacterial adherence. Our model
Vicryl vs Monocryl ⬍0.001 allows a static image of total bacteria on the suture to be evalu-
Vicryl vs Silk ⬍0.001
Silk vs Prolene 0.379
ated and shows significantly more bacteria on the polyglycolic
Silk vs VicrylPlus 0.500 suture compared with the silk suture. We feel that this method
Silk vs Monocryl 0.538 is more accurate and is a better guide for assisting in suture
Prolene vs Monocryl 0.790 selection.
Prolene vs VicrylPlus 0.836 In this study, we use a novel approach that allows the quan-
Monocryl vs VicrylPlus 0.953
tification of bacterial adherence. Optical imaging technology

Journal of Surgical Education • Volume 68/Number 2 • March/April 2011 103

allows the measurement of light produced by biological or
chemical moieties. Light-producing proteins, such as luciferase,
can be engineered into cells using plasmid vectors and can be
visualized and measured with specialized camera equipment
and software. These techniques have been well accepted in the
scientific and medical literature with numerous publications
using these techniques. The high sensitivity of these imaging
systems can potentially yield more consistent and accurate data
that, in turn, can yield significant biostatistic results with fewer
animals or materials required for testing.
We feel that this is the most accurate assessment of bacterial
adherence present in the literature. Although using bacterial
counts related to photons is a nonstandard format compared
with the more commonly used colony forming units, previous
work demonstrating the correlation of these photon counts
with bacterial colony-forming units obviates the need for addi-
tional plating. Plating, in fact, might only introduce potential
sampling error, which is avoided completely by the current
technique. Additional assessment modalities, such as light or
electron microscopy, could add additional understanding of the
nature of bacterial adherence to these materials and could aid
future suture designs. However, the imaging of bacteria with
microscopy cannot differentiate between live and dead bacteria.
Our bioluminescent imaging only measures the live bacteria
present on the suture as an active metabolic cycle is required to
release photons. This technique is perhaps most relevant in the
evaluation of suture treated with antimicrobials. Antimicrobial
suture under microscopy might manifest bacterial adherence;
however, the bacteria may no longer be viable. These bacteria
are not counted in the current study, which eliminates this
source of error.
Antimicrobial forms of many of the monofilament materials
tested in this study are commercially available. With the excel-
lent performance of the monofilaments tested, it is questionable
whether additional costs associated with antimicrobial forms of
these sutures are justifiable. Any further reduction in bacterial
adherence is unlikely to have any clinical significance. In terms
of the antimicrobial braided polyglycolic acid suture; despite its
excellent performance in this study we do not recommend
that it be used in wounds that are known to be contami-
nated, as persistence of bacteria beyond the effective period
of the antimicrobial elution leave a nidus of braided suture in
the wound that this study has shown to have significant
bacterial adherence.
The evaluation of the antimicrobial suture in this study bears
out the findings of previous investigations that show effective-
ness against both gram-positive and gram-negative bacteria.11
Our study design only included gram-positive bacteria. Addi-
tional study could be undertaken to assess the effect of gram-
negative bacteria as well in this bioluminescent model.
With the findings of this study, we present data acquired
with a novel experimental technique. The findings support our
hypothesis that the polyglycolic acid suture has the greatest
104 Journal of Surgical Education • Volume 68/Number 2 • March/April 2011
bacterial adherence, and in addition, we confirmed the anec-
dotal teaching that monofilament suture be used in the closure
of wounds that are contaminated or at risk for contamination.
Furthermore, in the treatment of wounds that have become
infected and require debridement, it is important to be diligent
about the removal of all residual braided suture within the
wound to prevent persistence or recurrence of the infection.
REFERENCES
1. Alexander JW, Kaplan JZ, Altemeir WA. Role of suture
materials in the development of wound infection. Ann
Surg. 1967;165:192-199.
2. Banche G, Roana J, Mandras N, et al. Microbial adher-
ence on various intraoral suture materials in patients un-
dergoing dental surgery. J Oral Maxillofac Surg. 2007;65:
1502-1507.
3. Katz S, Izhar M, Mirelman D. Bacterial adherence to sur-
gical sutures: a possible factor in suture induced infection.
Ann Surg. 1981;194:35-41.
4. Svoboda SJ, Bice TG, Gooden HA, et al. Comparison of
bulb syringe and pulsed lavage irrigation with use of a
bioluminescent musculoskeletal wound model. Bone Joint
Surg. 2006;88:2167-2174.
5. Rocchetta HL, Boylan CJ, Foley JW, et al. Validation of a
noninvasive, real-time imaging technology using biolumi-
nescent Escherichia coli in the neutropenic mouse thigh
model of infection. Antimicrob Agents Chemother. 2001;
45:129-137.
6. Kadurugamuwa JL, Sin L, Albert E, et al. Direct contin-
uous method for monitoring biofilm infection in a mouse
model. Infect Immun. 2003;71:882-890.
7. Charles CA, Ricotti CA, Davis SC, Mertz PM, Kirsner
RS. Use of tissue-engineered skin to study in vitro biofilm
development. Dermatol Surg. 2009;35:1334-1341.
8. Greenwald D, Shumway S, Albear P, Gottlieb L. Mechan-
ical comparison of 10 suture materials before and after in
vivo incubation. J Surg Res. 1994;56:372-377.
9. Lilly GE, Cutcher JL, Jones JC, et al. Reaction of oral
tissues to suture materials. IV. J Oral Surg. 1972;33:152-
157.
10. Lilly GE, Osbon DB, Hutchinson RA, Heflich RH. Clin-
ical and bacteriologic aspects of polyglycolic acid sutures.
J Oral Surg. 1973;31:103-105.
11. Edmiston CE, Seabrook GR, Goheen MP, et al. Bacterial
adherence to surgical sutures: Can anti-bacterial-coated
sutures reduce the risk of microbial contamination? J Am
Coll Surg. 2006;203:481-489.