ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Oct. 2008, p. 3633–3636
Copyright © 2008, American Society for Microbiology. All Rights Reserved.
Vol. 52, No. 10
Penetration of Chlorhexidine into Human Skin?
T. J. Karpanen,1* T. Worthington,1B. R. Conway,1A. C. Hilton,1T. S. J. Elliott,2and P. A. Lambert1
Life and Health Sciences, Aston University, Aston Triangle, Birmingham B4 7ET, United Kingdom,1and Selly Oak Hospital,
University Hospital Birmingham NHS Foundation Trust, Raddlebarn Road, Selly Oak, Birmingham B29 6JD, United Kingdom2
Received 15 May 2008/Returned for modification 4 July 2008/Accepted 25 July 2008
This study evaluated a model of skin permeation to determine the depth of delivery of chlorhexidine into
full-thickness excised human skin following topical application of 2% (wt/vol) aqueous chlorhexidine diglu-
conate. Skin permeation studies were performed on full-thickness human skin using Franz diffusion cells with
exposure to chlorhexidine for 2 min, 30 min, and 24 h. The concentration of chlorhexidine extracted from skin
sections was determined to a depth of 1,500 ?m following serial sectioning of the skin using a microtome and
analysis by high-performance liquid chromatography. Poor penetration of chlorhexidine into skin following
2-min and 30-min exposures to chlorhexidine was observed (0.157 ? 0.047 and 0.077 ? 0.015 ?g/mg tissue
within the top 100 ?m), and levels of chlorhexidine were minimal at deeper skin depths (less than 0.002 ?g/mg
tissue below 300 ?m). After 24 h of exposure, there was more chlorhexidine within the upper 100-?m sections
(7.88 ? 1.37 ?g/mg tissue); however, the levels remained low (less than 1 ?g/mg tissue) at depths below 300
?m. There was no detectable penetration through the full-thickness skin. The model presented in this study
can be used to assess the permeation of antiseptic agents through various layers of skin in vitro. Aqueous
chlorhexidine demonstrated poor permeation into the deeper layers of the skin, which may restrict the efficacy
of skin antisepsis with this agent. This study lays the foundation for further research in adopting alternative
strategies for enhanced skin antisepsis in clinical practice.
Effective skin antisepsis is essential in preventing infections
associated with invasive procedures, such as intravascular cath-
eter insertion or surgery. A range of skin antiseptic agents are
available in the clinical setting, such as povidone-iodine and
chlorhexidine compounds at various concentrations with alco-
holic or aqueous solutions. However, a 2% (wt/vol) chlorhexi-
dine solution is the recommended agent to be used prior to
invasive procedures according to the EPIC (evidence-based
practice in infection control) and CDC guidelines (18, 19).
Two percent chlorhexidine digluconate (CHG) has been
shown to significantly reduce intravascular catheter-related in-
fections (14), yet 2% (wt/vol) CHG in 70% (vol/vol) isopropyl
alcohol demonstrates activity superior to that of aqueous CHG
solution in a preoperative skin preparation (9) and in vitro
carrier tests (1). However, little is known about the kinetics of
chlorhexidine skin permeation from either of these solutions
(11, 25). Microorganisms colonizing the skin not only reside on
the skin surface but are also found to inhabit hair follicles and
lower skin depths (8). Many antimicrobial agents exhibit re-
stricted permeation of the skin (8) and fail to reach the deeper
layers, including the hair follicles, which harbor coagulase-
negative staphylococci (2, 7, 8, 13, 15) and propionibacteria
(13). Commensal microorganisms may therefore persist at the
site of incision following skin antisepsis (4, 22), and such res-
ident organisms may cause infection when the protective skin
barrier is breached during surgical procedures (12, 20, 26).
Therefore, effective and rapid permeation of the applied anti-
septic agent into the deeper layers of the skin is essential in
preventing infections associated with invasive procedures.
The aim of the current study was to use the Franz-cell skin
model (6) to determine the penetration profile for CHG
through excised human skin and to evaluate the skin perme-
ation of 2% (wt/vol) aqueous CHG into the skin using this
MATERIALS AND METHODS
Materials. CHG, diethylamine (high-performance-liquid-chromatography
[HPLC] grade), dimethyl sulfoxide, phosphate-buffered saline (PBS), sodium
heptane sulfonate (HPLC grade), and Tween 80 were purchased from Sigma-
Aldrich (Dorset, United Kingdom). Acetic acid and methanol (both HPLC
grade) were purchased from Fisher Scientific (Leicestershire, United Kingdom).
Skin samples. Full-thickness human skin samples were obtained from patients
undergoing breast reduction surgery, and full ethical committee approval was
obtained prior to this study (REC 2002/169). The full-thickness human skin was
frozen on the day of excision and stored at ?70°C until required.
Quantification of CHG. HPLC was used to measure the amounts of CHG in
the skin samples obtained during the permeation studies. The analyses were
performed using an Agilent 1200 series HPLC system (Agilent Technologies,
United Kingdom). The samples were run at a flow rate of 1.2 ml/min at room
temperature through a reverse-phase chromatography column (CPS-2 Hypersil
5-?m column; dimension, 150 by 4.6 mm [Thermo Electron Corporation, United
Kingdom]), with UV detection at 254 nm. The isocratic mobile phase consisted
of a methanol:water mixture (75:25) with 0.005 M sodium heptane sulfonate and
0.1% (vol/vol) diethylamine adjusted to pH 4 with glacial acetic acid. The HPLC
method was validated by repeating a series of standardized CHG concentrations
five times and plotting a graph of peak area versus CHG concentration. The level
of detection (LOD) and level of quantification (LOQ) were calculated from the
standard curve according to the following equations: LOD ? (3 ? standard
deviation)/slope; LOQ ? (10 ? standard deviation)/slope.
Skin permeation studies. Skin permeation studies were performed with ver-
tical Franz diffusion cells (Fig. 1). The receptor compartment was filled with 29
ml of PBS, maintained at 37°C by using a circulating water jacket, and agitated
by stirring with a magnetic bar. Skin samples were thawed in PBS at room
temperature, dried with an absorbent towel, and mounted on Franz diffusion
cells with the stratum corneum (SC) uppermost, facing the donor compartment.
The surface area exposed to the test compound was 3.14 cm2(2 cm in diameter).
All entrapped air between the skin and receptor fluid was removed, and the skin
was left to equilibrate for 30 min to reach the skin surface temperature of 32°C.
Twenty percent (wt/vol) aqueous CHG was diluted with distilled water and
* Corresponding author. Mailing address: Life and Health Sciences,
Aston University, Aston Triangle B4 7ET, United Kingdom. Phone: (44)
121 204 3951. Fax: (44) 121 204 4187. E-mail: firstname.lastname@example.org.
?Published ahead of print on 1 August 2008.
0.1% (vol/vol) Tween 80 to obtain the final test solution of 2% (wt/vol) CHG.
One milliliter of test solution was spread over the skin surface in the donor
compartment, and the compartment was sealed with a moisture-resistant film
(Parafilm M, Alcan packaging) to prevent evaporation. One milliliter of receptor
fluid was removed every 30 min for 2 h, every hour between 2 to 6 h, and at 8 h,
12 h, and 24 h. Fluid removed from the receptor compartment was immediately
replaced with an equal volume of fresh PBS solution. All samples were filtered
through a 0.45-?m nylon filter (Kinesis, United Kingdom) and analyzed by
HPLC. The assay was performed in triplicate and on two different donor skin
CHG penetration profile studies. Excised full-thickness human skin samples
were mounted on the Franz diffusion cells as described above and exposed to 2%
(wt/vol) CHG for 2 min, 30 min, and 24 h. Following exposure, the skin samples
were removed, washed with PBS, and dried with an absorbent towel. The skin
samples were immediately sprayed with a cryospray (Bright Instruments) and
frozen at ?20°C. Punch biopsy samples (7 mm in diameter) were cut from each
frozen sample in triplicate and placed on a cork disc in embedding compound
(Bright Instruments, Cambs, United Kingdom). The frozen samples were sec-
tioned horizontally with a microtome (Bright Instruments) into 20-?m sections
(from the surface to a depth of 600 ?m) and 30-?m sections (from depths of 600
to 1,500 ?m). Each section was placed in an Eppendorf tube and the total weight
of each skin sample determined. Chlorhexidine was extracted from the skin by
placing 1 ml of HPLC mobile-phase solution in each tube, followed by incubation
of the sealed tubes at 60°C for 1 h. Following this, the samples were analyzed by
HPLC and the concentration of CHG (?g/mg of skin) determined. Control skin
(skin without treatment) was analyzed parallel to the test samples. Effective
elution and recovery of CHG from the skin by this method were confirmed prior
to the experiment by injecting a standardized quantity of CHG (128 ?g) into 10
skin samples, extracting the CHG, and determining the recovered amount (94.4 ?
1.82%; data not shown).
HPLC validation. The mean retention time for CHG was 3.6
min. There were no intervening peaks from endogenous con-
taminating compounds within skin samples. The HPLC
method gave a linear response (r2? 0.999) over the concen-
tration range of 0.0039 ?g/ml to 128 ?g/ml. The level of de-
tection and level of quantification were calculated at 0.016
?g/ml and 0.052 ?g/ml, respectively.
Skin permeation studies. No CHG was detected in the re-
ceptor compartment during the 24-h exposure of excised full-
thickness human skin to 2% (wt/vol) aqueous CHG.
CHG retention studies. After 2 min, 30 min, and 24 h,
concentrations of chlorhexidine within the skin were highest in
the surface 100-?m sections and reduced below depths of 300
?m (Fig. 2 and 3). The concentrations of CHG within the top
100-?m sections of skin were 0.157 (? 0.047) ?g/mg tissue and
0.077 (? 0.015) ?g/mg tissue after 2-min and 30-min exposures
to 2% (wt/vol) CHG, respectively (Fig. 2). The concentration
of CHG within deeper layers (below 300 ?m) fell to less than
0.002 ?g/mg tissue following both 2-min and 30-min exposures.
The difference between the amounts of chlorhexidine within
the top layers between 2 min and 30 min of exposure was not
significant (P ? 0.05) (Student’s t test, INSTAT2; Graphpad,
San Diego, CA). The concentration of CHG was significantly
higher within all skin sections following 24 h of exposure to
CHG than with the shorter exposure times. The concentration
of CHG was 7.88 (? 1.37) ?g/mg tissue within the upper
100-?m sections and less than 1 ?g/mg of tissue at depths of
300 ?m and below.
This study demonstrates that 2% (wt/vol) chlorhexidine, the
antiseptic agent recommended within EPIC and CDC guide-
lines for skin antisepsis prior to central venous catheter inser-
tion, poorly permeates into deeper layers of the skin after 2
FIG. 1. Diagram of Franz diffusion cell. The receptor compartment
was filled with PBS, which was kept at 37°C by circulating water jacket.
The skin was mounted between the receptor and donor compartments
and clamped. The test drug was aliquoted into the donor compart-
ment. The drug diffused through the skin was sampled by removing
receptor fluid via the sampling port.
FIG. 2. Penetration profile showing the concentration and location
of chlorhexidine (?g/mg tissue) in excised human skin after 2 min or 30
min of exposure to aqueous 2% (wt/vol) chlorhexidine digluconate
(mean ? standard error; n ? 15).
FIG. 3. Penetration profile showing the concentration and location
of chlorhexidine (?g/mg tissue) in excised human skin after 2 min or 30
min (n ? 15) or 24 h (n ? 30) of exposure to aqueous 2% (wt/vol)
CHG (mean ? standard error).
3634KARPANEN ET AL.ANTIMICROB. AGENTS CHEMOTHER.
min and 30 min of exposure to the antiseptic. The concentra-
tions of CHG within the upper 100-?m sections of skin were
0.157 (? 0.047) ?g/mg tissue and 0.077 (? 0.015) ?g/mg tissue
after 2 min and 30 min, respectively. If 1 g of tissue is estimated
to equal 1 ml, these levels are higher than the concentrations
required to kill many common skin microorganisms, such as
Staphylococcus epidermidis, under in vitro conditions (10). Be-
low 300 ?m, the CHG concentration remained less than 0.002
?g/mg tissue, which may not be effective for eradicating mi-
croorganisms on the skin (17), especially microorganisms re-
siding deep in the hair follicles. Furthermore, chlorhexidine
activity is reduced in the presence of organic compounds, such
as fatty acids, and at lower pHs (16) and therefore may reduce
the efficacy of skin antisepsis with CHG. An exposure time of
2 min was used to reflect the clinical conditions used prior to
surgery (5). Although the 2-min study appears to show a larger
amount of bound chlorhexidine than the 30-min study, there is
variability in concentrations measured in the top layers, as is
expected with the shorter exposure period (24), and the dif-
ference between 2 min and 30 min of exposure is not significant
(P ? 0.05). It is likely that a steady state has not yet been
reached at 2 min. A similar phenomenon was reported by
Wagner et al. (23). Skin was also exposed to 2% (wt/vol) CHG
for 24 h, and the concentration of CHG in the deeper sections,
i.e., beyond 300 ?m, was less than 1 ?g/mg tissue. These levels
of CHG are more than the minimum bactericidal concentra-
tions for many skin commensals (10); however, this level of
CHG was obtained only after a prolonged time of contact of
the skin with CHG. In this study, no detectable levels of CHG
were recovered from the receptor compartment, suggesting
that aqueous CHG does not permeate through the full thick-
ness of excised skin and is retained within the tissue. These
results support previous research on another CHG-based com-
pound, chlorhexidine phosphanilate, which was also shown not
to permeate through full-thickness skin samples (25).
In this study, a model for studying the delivery of CHG into
excised full-thickness human skin was evaluated. Skin perme-
ation studies are commonly performed in vitro with vertical or
horizontal diffusion cells using skin or artificial membranes.
This study was performed using vertical diffusion cells (Franz-
type diffusion cells) to evaluate the delivery of CHG through
excised full-thickness human skin. Such conditions mimic the
in vivo environment by maintaining the physiological receptor
fluid at body temperature and the skin surface temperature at
32°C (6, 23). Skin permeation studies generally evaluate drug
delivery through the skin by measuring drug diffusion into the
receptor fluid through the SC or epidermis, which are the main
barriers for skin permeation. However, the use of stripped skin
layers, such as isolated SC or epidermal layers, for drug per-
meation studies may influence the results, with possible reten-
tion of the drug in the dermal layers of the skin. Full-thickness
skin was used in this study to determine the location of CHG
throughout the skin, rather than studying the flux of the drug
through the barrier layers. Following exposure to CHG, the
full-thickness human skin was sectioned to a depth of 1,500 ?m
by sequential sectioning with a microtome, producing a total of
60 sections per skin sample. Skin sectioning has been used in
many previous studies (21); however, the SC is often removed
by tape stripping prior to sectioning of the skin. In this study,
the full-thickness skin samples were sectioned throughout the
sample without prior removal of the surface layers. This study
demonstrates that the CHG permeation through the full-thick-
ness skin was not linear, which was expected due to the vari-
ation in structure at various layers. The top 100-?m layer of the
skin, which contains SC (average of 10 to 20 ?m thick) and
other epidermal layers (50 to 100 ?m thick), contained the
largest amount of CHG following exposure to 2% (wt/vol)
CHG over all time points studied. Previous research has shown
that the main permeation barrier for skin absorption is the SC
(3, 11, 25), which is thought to be due to its high-lipid matrix
and packed layers of keratinized epithelial cells. Furthermore,
this study found that below 300 ?m, at the dermal layer, the
level of CHG remained constantly low. Depending on the body
site, dermis contains hair follicles and other skin appendages,
including sebaceous glands and sudoriferous glands (sweat-
producing glands), which are of interest in skin antisepsis since
they may be niches for microbial colonization of the skin fol-
lowing skin antisepsis (7, 8). It is generally recognized that skin
antisepsis does not sterilize the skin; our study confirms this
and demonstrates that it may be due to poor permeation of
chlorhexidine into the deeper layers of the skin.
In conclusion, this study showed poor permeation of chlor-
hexidine through excised full-thickness human skin after 2 min
and 30 min of exposure to aqueous 2% (wt/vol) CHG. The
levels of CHG were highest within the top 100-?m sections of
skin and remained consistently low within the deeper layers.
Furthermore, the model presented in this study is a valuable
tool in determining a permeation profile for chlorhexidine
through human skin in vitro. This study lays the foundation for
further research within this area with a view to potentially
adopting alternative strategies for enhanced skin antisepsis in
This work was supported by EPSRC CASE grant CNA/05/09 with
funding from Insight Health Ltd., United Kingdom.
1. Adams, D., M. Quayum, T. Worthington, P. Lambert, and T. Elliott. 2005.
Evaluation of a 2% chlorhexidine gluconate in 70% isopropyl alcohol skin
disinfectant. J. Hosp. Infect. 61:287–290.
2. Brown, E., R. P. Wenzel, and J. O. Hendley. 1989. Exploration of the mi-
crobial anatomy of normal human skin by using plasmid profiles of coagu-
lase-negative staphylococci: search for the reservoir of resident skin flora.
J. Infect. Dis. 160:644–650.
3. Cal, K., S. Janicki, and M. Sznitowska. 2001. In vitro studies on permeation
of terpenes from matrix-type transdermal systems through human skin. Int.
J. Pharm. 224:81–88.
4. Elliott, T. S., H. A. Moss, S. E. Tebbs, I. C. Wilson, R. S. Bonser, T. R.
Graham, L. P. Burke, and M. H. Faroqui. 1997. Novel approach to investi-
gate a source of microbial contamination of central venous catheters. Eur.
J. Clin. Microbiol. Infect. Dis. 16:210–213.
5. Elliott, T. S., M. H. Faroqui, R. F. Amstrong, and G. C. Hanson. 1994.
Guidelines for good practice in central venous catheterization. J. Hosp.
6. Franz, T. J. 1975. Percutaneous absorption—on the relevance of in vitro
data. J. Investig. Dermatol. 64:190–195.
7. Hendley, J. O., and K. M. Ashe. 2003. Eradication of resident bacteria of
normal human skin by antimicrobial ointment. Antimicrob. Agents Che-
8. Hendley, J. O., and K. M. Ashe. 1991. Effect of topical antimicrobial treat-
ment on aerobic bacteria in the stratum corneum of human skin. Antimicrob.
Agents Chemother. 35:627–631.
9. Hibbard, J. S., G. K. Mulberry, and A. R. Brady. 2002. A clinical study
comparing the skin antisepsis and safety of ChloraPrep, 70% isopropyl
alcohol, and 2% aqueous chlorhexidine. J. Infus. Nurs. 25:244–249.
10. Karpanen, T. J., T. Worthington, D. Rathbone, and P. A. Lambert. 2006.
Activity of thiosemicarbazone and carboxamidrazone compounds and essen-
VOL. 52, 2008PENETRATION OF CHLORHEXIDINE INTO SKIN 3635
tial oils against microorganisms associated with intravascular device related
infections. J. Hosp. Infect. 64(Suppl. 1):S37.
11. Lafforgue, C., L. Carret, F. Falson, M. E. Reverdy, and J. Freney. 1997.
Percutaneous absorption of a chlorhexidine digluconate solution. Int.
J. Pharm. 147:243–246.
12. Langgartner, J., H. J. Linde, N. Lehn, M. Reng, J. Scho ¨lmerich, and T.
Glu ¨ck. 2004. Combined skin disinfection with chlorhexidine/propanol and
aqueous povidone-iodine reduces bacterial colonisation of central venous
catheters. Intensive Care Med. 30:1081–1088.
13. Leeming, J. P., K. T. Holland, and W. J. Cunliffe. 1984. The microbial
ecology of pilosebaceous units isolated from human skin. J. Gen. Microb.
14. Maki, D. G., M. Ringer, and C. J. Alvarado. 1991. Prospective randomised
trial of povidone-iodine, alcohol, and chlorhexidine for prevention of infec-
tion associated with central venous and arterial catheters. Lancet 338:339–
15. Malcolm, S. A., and T. C. Hughes. 1980. The demonstration of bacteria on
and within the stratum corneum using scanning electron microscopy. Br. J.
16. McDonnell, G., and A. D. Russell. 1999. Antiseptics and disinfectants: ac-
tivity, action, and resistance. Clin. Microbiol. Rev. 12:147–179.
17. Messager, S., P. A. Goddard, P. W. Dettmar, and J.-Y. Maillard. 2001.
Determination of the antibacterial efficacy of several antiseptics tested on
skin by an ‘ex-vivo’ test. J. Med. Microbiol. 50:284–292.
18. O’Grady, N. P., M. Alexander, E. P. Dellinger, J. L. Gerberding, S. O. Heard,
D. G. Maki, H. Masur, R. D. McCormick, L. A. Mermel, M. L. Pearson, I. I.
Raad, A. Randolph, and R. A. Weinstein. 2002. Guidelines for the prevention
of intravascular catheter-related infections. MMWR Recommend. Rep.
19. Pratt, R. J., C. M. Pellowe, J. A. Wilson, H. P. Loveday, P. J. Harper,
S. R. L. J. Jones, C. McDougall, and M. H. Wilcox. 2007. epic2: national
evidence-based guidelines for preventing healthcare-associated infections in
NHS hospitals in England. J. Hosp. Infect. 65(Suppl. 1):S1–S64.
20. Safdar, N., and D. G. Maki. 2004. The pathogenesis of catheter-related
bloodstream infection with noncuffed short-term central venous catheters.
Intensive Care Med. 30:62–67.
21. Touitou, E., V. M. Meidan, and E. Horwitz. 1998. Methods for quantitative
determination of drug localized in the skin. J. Control. Release 56:7–21.
22. Traore ´, O., F. A. Allaert, S. Fournet-Fayard, J. L. Verrie `re, and H. Laveran.
2000. Comparison of in-vivo antibacterial activity of two skin disinfection
procedures for insertion of peripheral catheters: povidone iodine versus
chlorhexidine. J. Hosp. Infect. 44:147–150.
23. Wagner, H., K.-H. Kostka, C.-M. Lehr, and U. F. Schaefer. 2002. Human
skin penetration of flufenamic acid: in vivo/in vitro correlation (deeper skin
layers) for skin samples from the same subject. J. Investig. Dermatol. 118:
24. Wagner, H., K.-H. Kostka, C.-M. Lehr, and U. F. Schaefer. 2000. Drug
distribution in human skin using two different in vitro test systems: compar-
ison with in vivo data. Pharm. Res. 17:1475–1481.
25. Wang, J. C. T., R. R. Williams, L. Wang, and J. Loder. 1990. In vitro skin
permeation and bioassay of chlorhexidine phosphanilate, a new antimicro-
bial agent. Pharm. Res. 7:995–1002.
26. Worthington, T., and T. S. J. Elliott. 2005. Diagnosis of central venous
catheter related infection in adult patients. J. Infect. 51:267–280.
3636 KARPANEN ET AL.ANTIMICROB. AGENTS CHEMOTHER.