The effect of a combination of 0.1% octenidine dihydrochloride and 2% 2‐phenoxyethanol (octenisept^®) on wound healing in pigs in vivo and its in vitro percutaneous permeation through intact and barrier disrupted porcine skin

Abstract

A combination of 0.1% octenidine dihydrochloride and 2% 2‐phenoxyethanol (octenisept^®) is a commonly used disinfectant in human medicine. As porcine skin represents an adequate model for human skin, the effect of octenidine dihydrochloride and phenoxyethanol on wound healing is studied in pigs. Furthermore, the in vitro percutaneous permeation of the test substances is studied. The impact of the test formulations on wound healing is examined (A) under non occlusive conditions and (B) in comparison to another disinfectant based on povidone‐iodine under occlusive conditions, while wounds are treated daily with the test substances. The percutaneous permeation of octenidine dihydrochloride and phenoxyethanol is studied in Franz‐type diffusion cells with intact skin as well as barrier disrupted after tape stripping. Compared with povidone‐iodine or vehicle treatment as well as untreated control wounds the treatment of wounds with the test formulation has no influence on the healing rate in pigs and does not induce retardation of wound healing. The in vitro diffusion experiment reveals that octenidine dihydrochloride is only detectable in the acceptor chamber of three‐barrier disrupted skin samples. Phenoxyethanol permeates through intact porcine skin in amounts of 11.3% and through barrier disrupted skin in amounts of 43.9%

INTRODUCTION

As antiseptics aim to kill or inactivate microorganisms by reducing their number or inhibiting their multiplication after topical administration, they are used locally on the surface of the body, in body cavities as well as on surgically exposed tissues (1). To achieve best conditions in killing or inhibition of microbes, many studies were conducted in the last decades to develop new antiseptic agents like chlorhexidine, iodine or iodophores 1, 2. Out of these studies octenidine dihydrochloride (OCT) was designed as an alkanediylbis[pyridine] germicidal agent with optimal antiseptic properties (3). Stability is given in the pH range 1·6–12·2, and under exposition to light and in contrast to chlorhexidine chloride OCT is not sensitive to hydrolysis and shows no reduction in the effect as a result of albumin or mucin (protein error) (1). It is levurocidal, effective against resident skin microflora and effects Gram positive and negative as well as plaque‐forming bacteria 1, 3, 4, 5, 6, 7, 8. Antibacterial effects of OCT on the oral microflora in vivo and in vitro are comparable to those of chlorhexidine gluconate and povidone‐iodine (9). The comparison of its biocompatibility with other disinfectants like PVP iodine or chlorhexidine digluconate showed that OCT represents an antiseptic with low cytotoxicity and high microbicidal effects 10, 11. Today OCT [in combination with 2‐phenoxyethanol (PH)] is a licensed product in several European countries (octenisept® with 0.1% OCT and 2% PH). It has been reported that the use of this combination of OCT and PH in premature newborn infants of 23–26 weeks gestation does not cause skin damage. Interestingly, PH permeates newborn's skin but seems to be inactivated effectively by oxidative metabolisation to 2‐phenoxyacetic acid (12).

Nevertheless, only few are known about the effects of the combination of OCT and PH on wound healing. Therefore, we investigated the impact of two antiseptic products on wound healing in a standardised porcine wound model simulating wound care in the clinical setting. To detect the influence of occlusive conditions, an additional study was conducted under occlusive conditions. Because intact and wounded skin gets in contact with the two components during skin disinfection, percutaneous permeation of OCT and PH through intact and wounded porcine skin was determined in vitro. Pigs were chosen for the in vivo and in vitro experiments, as porcine skin is comparable to human skin regarding skin morphology and biochemical composition 13, 14, 15, 16, 17, 18.

MATERIALS AND METHODS

Trial preparations

Octenisept^®, a combination of 0.1% OCT and 2% PH, was provided by Schülke & Mayr GmbH, Norderstedt, Germany. Vetsept^®, a combination of 5% poly(1‐vinyl‐2‐pyrrolidon)‐iodine and 9% nonoxinol, was used as brand medicinal product of AniMedica GmbH, Senden‐Bösensell, Germany.

The study design of the wound treatment of studies A and B is described in Table 1.

Table 1.

Wound treatment of the two in vivo studies; Ringer solution (Braun, Melsungen, Germany) was used for dilution of the treatment of wound number 2–4 (study design A) and 2 and 3 (study design B)

Wound number	Study design A		Study design B
	Treatment	Occlusive bandage	Treatment	Occlusive bandage
1	Octenisept®	No	Octenisept®	yes
2	Octenisept®1:2	No	Octenisept®1:5	yes
3	Octenisept®1:5	No	Octenisept®1:10	yes
4	Octenisept®1:10	No	Vetsept®	yes
5	Ringer solution	No	Ringer solution	yes
6	Untreated	No	Untreated	no
7	—	—	Untreated	yes

Animals

For each study six male pigs with a weight of 10–15 kg (study A) respectively 20–25 kg (study design B) were used. They were fed restrictively, received water ad libitum and were kept on straw in single boxes (temperature in the stable: 22°C) according to legal requirements.

Study description

In vivo wound healing study

The study was conducted according to the requirements of good clinical practice (GCP) and the German animal welfare law. After 7 days of acclimatisation, each pig was put under general anaesthesia using a combination of azaperon (Stresnil, Janssen‐Cilag, Neuss, Germany) and ketamin (ketamin 10%, Wdt, Garbsen, Germany) and the wounds were set.

The following studies were carried out:

Study design A (non occlusive conditions): Six full thickness wounds with a diameter of 6 mm (= 28·3 mm²) each were set with biopsy punches (Stiefel laboratories, Offenbach, Germany) on the back skin of each pig on day 0. To avoid cross‐contamination of the wounds, at least 4‐cm distance were maintained between the wounds. Directly after wounding until day 3, 100 µl of the test formulations were applied once daily according to Table 1. Because there was no occlusive cover, the wounds got the possibility to get infected with environmental bacteria. On days 1–4, 7–10 and 14 wound areas were measured planimetrically (ScionImage 4.0.3.2; Scion Corporation, Maryland, USA) and inflammation parameters (rubor and tumour) were documented.

Study design B (occlusive conditions): According to study design A, seven full thickness wounds were inflicted on the back skin of each pig. Directly after wounding six wounds were covered with occlusive patches (Hydrosorb, Hartmann, Heidenheim, Germany), which were fixed with adhesive tape (Leukoplast, BSN medical GmbH, Hamburg, Germany). Wound number 7 served as non occlusive control. Wound area measurement was conducted on days 1, 4, 7, 10 and 11 and wound treatment according to Table 1 started on day 1 and was repeated once daily until day 11. Therefore, new occlusive patches were applied on the wounds daily after treatment.

According to Table 1 two puffs (210 µl) of the test solution were applied on the wounds. Inflammation parameters were documented daily (rubor and tumour). Pig number 6 was euthanised on day 3, as an advanced state of greasy pig disease was diagnosed (there was no correlation between the wound healing experiment and the disease).

In vitro percutaneous permeation studies

All studies were conducted in accordance to OECD‐guideline 428 (19). Skin samples were obtained from the lateral abdominal skin of sacrificed pigs of study design B and were stored at—20°C until use. After gentle defrosting of the skin hair was clipped off (hair clipper: Wahl GmbH, Unterkirnach, Germany). Split skin with a thickness of 500 µm was obtained by using a dermatom (Zimmer, Freiburg, Germany) and was incubated in phosphate buffered saline (pH 7·4, disodiumhydrogenphosphate 1·44 g, sodiumchloride 8·0 g, potassiumdihydrogenphosphate 0·24 g, potassiumchloride 0·2 g, aqua bidest. ad 1000 ml) for 30 min before starting the experiment. Visual examinations were used to study the integrity of the skin.

To induce barrier disruption in six skin samples, the skin surface was treated with tapes (Beiersdorf GmbH, Hamburg, Germany) up to 100 times. Permeation experiments were conducted in Franz‐type diffusion cells (Permgear, Bethlehem, USA). The skin samples were mounted dermal site down between donor (upper part) and acceptor (lower part) chamber (permeation area of 1·77 cm²). The acceptor compartment was filled with 12 ml Soerensen phosphate buffer (pH 7·4, potassiumdihydrogenphosphate 2·0, disodiumhydrogenphosphate 9·2 g, sodiumchloride 4·2 g, aqua bidest. ad 1000 ml) and was stirred continuously with a magnetic bar at 500 rpm. The donor compartment was filled with 2 ml octenisept^®(1·13 ml/cm²) and was covered with parafilm (American Can Company, Baltimore, USA).

A water bath provided a constant temperature of 32°C in each Franz‐type diffusion cell. Samples were taken from the acceptor compartment (2 × 400 µl) at each time point and were filled up with the same amount of Soerensen phosphate buffer (0, 1, 2, 3, 4, 5, 6, 22, 24, 26, 28 h).

High‐performance liquid‐chromatography (HPLC) with the following HPLC conditions was used to analyse the content of OCT and PH in each sample: autosampler (508, Beckmann, Muenchen, Germany), column (LichroCART 125‐4, 5 µm, 10 cm, Merck, Darmstadt, Germany), pre‐column (LichroCART 4‐A, 5 µm, Merck, Darmstadt, Germany), heater (SpH 99, Spark Holland, Emmen, Netherlands): 40°C, UV‐VIS‐detector (168, Beckmann, Muenchen, Germany).

The flow rate was set up at 1.5 ml/min, the pressure was approximately 20 mPa. Hundred microlitres of each sample were injected into the HPLC. OCT was detected at 280 nm with a mobile phase containing 80% methanol and 20% McIlvaine buffer (pH 2.2, citric acid (anhydrous) 20.8 g, disodiumhydrogenphosphate 0.4 g aqua bidest. ad 1000 ml). PH was detected at 270 nm with a mobile phase consisting of 80% McIlvaine buffer and 20% methanol.

Quantitative analysis was based on comparison with the calibration curve. The area under the curve were analysed using the method of external standards. The concentration in the acceptor chamber was used to calculate the drug amount permeated through cm² skin according to Niedorf et al. 2008 (20). The maximum flux (J _max) and the apparent permeability coefficient P _app were calculated from the maximum absorption after 28 h.

Analysis

Analysis was done using the computer software Scion Image (Scion Corporation, Frederick, Maryland, USA). Significance of the in vivo study was tested using a Friedman's test with a post hoc Dunn's test at an error level of 5% (P < 0·05).

Results

In vivo studies

Study A

Determination of the wound areas: All wounds were covered with eschar from day 2 to day 14 and independent of the wound treatment all wound areas showed the same healing rate, which did not differ significantly in the different treatments (Table 2).

Table 2.

Wound areas (mean and standard deviation: STD) of six pigs after topical treatment with different test formulations on different days (study design A: non occlusive)

Day	Wound area [%]
	octenisept®		octenisept®1:2		octenisept®1:5		octenisept®1:10		Ringer solution		untreated
	Mean	STD	Mean	STD	Mean	STD	Mean	STD	Mean	STD	Mean	STD
1	100·0	13·6	100·0	10·6	100·0	10·0	100·0	14·9	100·0	14·5	100·0	19·2
2	91·4	7·8	97·2	10·6	92·8	6·3	93·8	8·3	91·2	8·0	99·7	10·2
3	84·3	11·2	93·8	9·5	91·5	10·7	97·3	5·8	95·0	4·3	95·5	14·0
4	81·6	9·5	86·5	11·7	84·8	8·5	92·0	13·0	80·7	8·7	98·1	14·3
7	85·5	8·6	85·4	9·2	84·3	9·3	94·5	15·2	87·3	11·1	105·4	11·5
8	95·7	13·0	83·1	13·5	90·2	8·0	99·5	18·5	86·1	9·8	104·1	9·1
9	85·4	10·8	92·5	10·3	82·1	9·2	86·7	7·4	91·1	15·9	97·1	9·9
10	85·9	8·1	84·5	7·9	93·2	14·6	94·1	13·1	88·8	7·7	99·5	13·1
14	71·9	8·4	61·8	6·9	64·1	12·9	67·3	6·3	72·1	13·3	72·3	9·8

Inflammation parameters of the wounds: Redness (rubor) and swelling (tumour) were found in the wounds on day 1. While the untreated wound showed inflammation parameters until day 10, the wounds treated with Ringer solution showed inflammation parameters until day 8. Wounds treated with the combination of OCT and PH showed inflammation parameters until day 4, while higher dilutions of this combination of OCT and PH resulted in a longer appearance of inflammation parameters (until day 7 in the dilution 1:10).

Study B

Determination of the wound areas: Independent from the treatment all wounds showed the same rate of reepithelialisation, without significant differences between the wound areas (Table 3).

Table 3.

Wound areas (mean and standard deviation: STD) of six pigs after topical treatment with different aqueous solutions on day 1 and 14 (study design B: occlusive)

Day	Wound area [%]
	Ringer solution		octenisept®1:10		octenisept®1:5		octenisept®		Vetsept®		untreated		untreated + uncovered (eschar till day 11)
	Mean	STD	Mean	STD	Mean	STD	Mean	STD	Mean	STD	Mean	STD	Mean	STD
1	100·0	39·4	100·0	38·7	100·0	38·6	100·0	60·2	100·0	44·5	100·0	40·2	100·0	39·9
3	122·9	52·8	96·8	43·6	92·3	42·4	126·6	53.3	100·8	43·5	95·7	38·8	91·8	36·1
7	76·3	33·6	65·4	29·4	75·3	39·5	104·0	45.8	64·9	31·1	82·5	44·0	77·4	35.5
10	82·0	34·6	72·8	43.8	64·3	33·4	110·3	51.8	70·3	31·8	68·2	31·8	81·3	37·5
11	80·3	35·3	56·8	35·0	63·5	35·1	92·8	42.4	67·1	31·8	66·0	27·4	85·3	39·8
14	42·1	25·6	39·0	17·9	28·2	16·5	36·4	30.7	28·4	35·1	37·3	35·2	85·1	30·1

Inflammation parameters of the wounds: Only redness (rubor) was found in some wound areas from day 1 to day 10, while no relationship was observed between wound treatment and inflammation rate. The untreated and non occlusive wound number 7 showed formation of eschar from day 1 till day 11.

In vitro percutaneous permeation studies

OCT was only detectable in the acceptor medium of three‐barrier disrupted skin samples with amounts of 6.28 µg/cm² (0.6%) of the applied 2000 µg OCT. PH permeated through physiological skin samples as well as barrier disrupted skin after tape stripping with a 10.2‐fold higher flux J _max in skin after tape stripping. After 28 h PH permeated in amounts of 9.87 mg/cm² (43.7%) through barrier disrupted skin and of 2.54 mg/cm² (11.3%) through physiological intact porcine skin. All data obtained in the in vitro diffusion experiment are shown in Table 4 and Fig. 1.

Table 4.

Flux J _max and apparent permeability coefficient P _app of permeated 2‐phenoxyethanol (PH) and octenidine dihydrochloride (OCT) per cm² skin obtained in diffusion experiments with porcine split skin after 28 h (n = 6) calculated from the maximum absorption according to Niedorf et al. 2008 (20)

Parameter	PH				OCT
	Intact skin		Skin after tape stripping		Skin after tape stripping
	Mean	STD	Mean	STD	Mean	STD
Jmax [µg/cm2/h]	1·31E+02	3·39E+01	1·34E+03	2·06E+02	2·90E+00	2·28E+00
P app [cm/s]	1·82E−06	4·70E−07	1·86E−05	2·86E−06	8·05E−07	6·34E−07
Amount of substance permeated per cm2 skin after 28 h [µg/ml]	2544·30	685·12	9865·41	993·25	6·28	10·44
Amount of substance permeated per cm2 skin after 28 h [%]	11·3	3·0	43·7	4·4	0·6	0·9

Figure 1.

Permeation rate curve of PH obtained in diffusion experiments over 28 h with physiological intact and barrier disrupted porcine split skin (mean and standard deviation).

DISCUSSION

OCT has been developed as a potential broad spectrum topical antiseptic and is described to be effective against skin flora (3) with good effects in the treatment of mild to moderate inflammatory acne lesions, Pseudomonas nail infections as well as the antibacterial treatment of intravascular catheter insertion sites 21, 22, 23, 24. Although the combination of OCT and PH is licensed in several European countries as a skin antiseptic for humans (octenisept^®), only few is known about the impact on wound healing and skin permeability after topical application.

Thus, the present studies are made in pigs, which represent an adequate model for human skin, to examine: (a) the influence of a combination of OCT and PH on wound healing under non occlusive conditions, (b) the influence of two commonly used topical antiseptics (a combination of OCT and PH versus povidone‐iodine) on wound healing under occlusive conditions, and (c) the in vitro percutaneous permeation of a combination of OCT and PH through intact as well as barrier disrupted skin.

The summary of the effects observed under different study designs is given in the following sections.

In vivo experiment under non occlusive conditions

Wound treatment with a combination of OCT and PH results in no significant increase of the reepithelialisation and no wound healing retardation compared with control under non occlusive conditions in a porcine wound model. The most remarkable result concerns the different occurrences of wound inflammation parameters. Although redness and swelling are observed in all wounds, wound treatment with OCT and PH slightly accelerates the reduction of inflammation parameters compared with untreated and vehicle treated controls.

In vivo experiment under occlusive conditions

Under occlusive conditions no differences are found in the reepithelialisation of porcine wounds treated with a combination of OCT and PH compared with povidone‐iodine and control. Bennet et al. (25) described povidone‐iodine to decrease wound reepithelialisation in rodent back wounds compared with other solutions. The present study reveals neither octenisept^®nor Vetsept^®(povidone‐iodine based) to retard wound healing under occlusive conditions. Interestingly, compared with septic wounds in study A, no alteration was observed in the appearance of wounds between the different wound treatments in study B.

Summary of the in vivo results under different study designs

The wound healing rate after 14 days differs considerably between the two study designs, which can be explained by the occlusive bandage as well as the different wound treatments. Study A is conducted without occlusive cover, so the wounds can get infected. Furthermore, wounds are treated with the different solutions from day 1 to 3 while wounds in study B are occlusive covered and wound treatment is performed daily. As compared with all other wounds the untreated uncovered wound in study B shows retardation of wound healing, it can be concluded, that occlusive cover accelerates wound reepithelialisation. This result is in accordance with former reports, who described wounds to heal faster with occlusive dressings 26, 27, 28, 29.

Kramer et al. 2004 described OCT to have a higher cytotoxicity compared with polyhexanide and control in vitro which leads to wound retardation without histological alterations compared with polyhexanide and control in pigs in vivo (30), while another examination revealed that OCT represents an antimicrobial agent with the highest biocompatibility index compared to other disinfectants like polyhexanide, povidone‐iodine or chlorhexidine digluconate (11). Compared with povidone‐iodine OCT, chlorhexidine digluconate and polyhexamethylene biguanide show a 20‐fold lower tolerability in murine fibroblasts L929 in vitro (31). As in our studies on a porcine wound model neither retardation of reepithelialisation nor increased duration of inflammation parameters, which argue for high cytotoxicity, were observed in octenisept^®as well as Vetsept^®treated wounds, the present study shows a high in vivo biocompatibility of the combination of 0.1% OCT and 2% PH as well as povidone‐iodine. Under experimental conditions five epidemic MRSA clones did not develop stable resistance after exposure to low concentrations of OCT and a strong inhibition of fungal phospholipase B1 was shown 3, 32, 33. Based on our data and because of the high antimicrobial effects, OCT seems to be a promising agent for topical antisepsis.

In vitro percutaneous permeation

OCT is described to be neither absorbed during the usage on the vagina nor by topical treatment of the skin for 24 h under occlusive conditions in humans (1). The present study illustrates, that after application of octenisept^®on physiological intact porcine skin no OCT can be found in the acceptor compartment, while 11.3% of the topical applied PH permeate per cm². To get further information of percutaneous permeation of wounded skin, skin samples were treated with adhesive tapes 100 times before application of the test solution. Tape stripping is an effective method to remove stratum corneum layers from the skin surface (19), therefore it can be used to weaken the skin barrier, maintained by the horny layer and it is also used to detect drug amounts in the stratum corneum after topical drug application (34). After removing the horny layer OCT could be found in three of six acceptor chambers (0.6% of the applied OCT), while PH permeated 10.2‐fold higher than through intact porcine skin. It has to be considered that the study was conducted under extreme conditions regarding the amount of solution per skin area (1.14 ml/cm²) for 28 h and the results obtained in the in vitro diffusion study may over‐estimate percutaneous permeation under physiological conditions in practice. Therefore, it can be assumed that under clinical conditions the absorption of OCT even after application on wounds is neglectible. Nevertheless, the permeation of PH observed in the present study confirms examinations conducted in newborn infants, in which after topical treatment of human infants with octenisept^®, both small amounts of PH and high amounts of its metabolite 2‐phenoxyacetic acid but no OCT were found in spot urine (12). And in vitro studies in rat skin showed that PH permeates up to 99%, while first‐pass metabolism in the skin does not have any influence on systemic availability of topical applied PH (35). Although few is known about systemic toxicity of PH, Scortichini et al. 1987 could neither show embryotoxic, fetotoxic, nor teratogenic effects of PH in rabbits after topical treatment (36).

In conclusion, the present in vivo study in pigs shows that topical wound treatment with a combination of OCT and PH leads to no retardation in wound healing no matter if under occlusive or non occlusive conditions, but to faster decay of wound inflammation without occlusive cover. The best reepithelialisation rate is observed in wounds with occlusive cover. In vitro diffusion of OCT and PH under occlusive conditions with high amounts of topically applied octenisept^®onto porcine split skin, leads to a low permeation rate of PH in intact and a 3.8‐fold higher permeation of PH in barrier disrupted skin. OCT only permeates three of six barrier disrupted skin samples in low amounts.

In summary, concerning its high antimicrobial effect as well as its good biocompatibility 3, 10, 21, 22, 23, 24, the combination of 0.1% OCT with 2% PH represents a promising antiseptic for topical skin and wound treatment.

ACKNOWLEDGEMENTS

The present study was sponsored by Schülke & Mayr GmbH, Norderstedt, Germany. Braun, M. and Siebert J. are employees of Schülke & Mayr GmbH, Norderstedt, Germany.