Human allograft skin is generally used to effect a temporary (it will mostly be rejected) wound closure on full thickness (burn) wounds. It acts as a mechanical and biological barrier and thus decreases the loss of water, protein, and heat through the (burn) wound (Mackie 2002; Kearney 2005; Leon-Villapalos et al. 2010). Human allograft skin can be preserved by numerous methods: cool storage, cryopreservation, lyophilisation and dehydration (e.g. glycerolisation). These methods amount to different ranges of allograft skin viability, integrity and immunogenicity. In some cases a certain level of allograft skin viability, integrity and permeability is required, but often skin allografts do not need to be viable at any expense.

Glycerol or propylene glycol preserved allograft skin, for example, is non-viable and has been used successfully in burn surgery for decades (Kreis et al. 1989; de Backere 1994; Huang et al. 2004; Khoo et al. 2010). Glycerolisation was shown to attenuate allograft antigenicity (Hettich et al. 1994; Richters et al. 1997), and thus extend the period of graft take, and to inhibit bacteria and viruses (Marshall et al. 1995; Saegeman et al. 2008). On the down side, glycerolized skin tends to be more rigid than fresh or cryopreserved skin. Hence, glycerolised donor skin is usually not the first choice of temporary biological dressing of the burn surgeons of the Burn Wound Centre of the Queen Astrid Military Hospital (BWC-QAMH).

Cryopreservation has been shown to maintain a degree of viability and the basic physiochemical and permeability properties of fresh human skin (Aggarwal et al. 1985). First reports of frozen skin storage date from before the Second World War (Mider and Morton 1939). In 1952, Billingham and Medawar described the cryopreservation of skin using glycerol as cryoprotectant. Since then many protocols for skin cryopreservation have been developed, based on the use of a variety of cryoprotectants and constant-rate cooling using commercially available programmable controlled-rate freezers (Biagini et al. 1979; Kets et al.1985; Kearney et al. 1990; Ingham et al. 1993; Bravo et al.2000; Alotto et al. 2002; Franchini et al. 2009; Pianigiani et al. 2010). The physicians in our centre have the experience that cryopreserved and partly viable donor skin allografts generate extra beneficial effects, mainly a better up granulation of the wound bed in comparison with glycerolised skin. They also observed that partly viable cryopreserved allograft skin stimulates wound healing in superficial burn wounds in children, thus preventing autografting. This could be due to the fact that living skin cells can deliver organisational signals or growth factors to the wound bed, thus mediating the formation of granulation tissue or the stimulation of wound healing (Galkowska et al. 2006) in the same way as do cultured allogenic keratinocytes and fibroblasts (Roseeuw et al. 1990; Duinslaeger et al. 1996; Goedkoop et al. 2010). Down side of the cryopreservation methods, however, is that they tend to support bacterial and fungal survival, which can be problematic in the case of skin allograft preservation. The only variables significantly affecting microbiological contamination of donor skin allografts are the type of donor (live or dead) and the type of processing (cryo- or glycerol preservation), with highest levels of contamination found in cryopreserved cadaveric donor skin (Pianigiani et al. 2010). In contrast to some other harvested donor tissues (e.g. bone, tendons and heart valves), skin tissue is inherently colonised by skin associated micro-organisms and thus non-sterile at the time of harvesting. Superficial decontamination of the donor’s skin before harvesting, using antiseptics, is not indefectible. Therefore, in the BWC-QAMH, freshly harvested (post mortem) donor skin is collected in an antibiotics containing transport solution. This transport solution is composed of 800 ml custom made Medium for Culturing of Epithelial Cells (MCEC, Gibco Invitrogen Corporation) supplemented with 200 ml of custom made Cambridge Antibiotic Solution (CAS, Inverclyde Biologicals). MCEC consists of 3:1 ratio of DMEM/Ham’s F12 Nutrient mix and 10.6 g/l sodium bicarbonate, while CAS consists of Gentamicin sulphate (4 g/l), Primaxin (0.2 g/l), Polymyxin B (0.2 g/l), Vancomycin (0.05 g/l) and Nystatin (2,500,000 U/l) in Medium 199 with 25 mM HEPES. The harvested donor skin allografts are kept in this transport solution (static container) for minimum 24 h (max. 72 h) at 2–8°C, before further processing.

The European Human Cell and Tissue Directives (2004/23/CE, 2006/17/CE and 2006/86/CE) were recently (December 19, 2008) transposed to Belgian Law. To be accredited, Belgian banks for human body materials have to comply with quality and safety criteria defined in Royal Decrees (September 28, 2009). In addition, the Belgian Superior Health Council published (October 1, 2008) quality- and safety criteria for human skin allografts. The skin bank of the Queen Astrid Military Hospital is EN ISO 9001: 2008 certified for the full scope of its activities. Implementing this kind of relevant Quality Management Systems is important, also for tissue bankers (von Versen et al. 2000).

In Belgium, human donor skin allografts are tested for bacterial and mycological contamination using a protocol based on 14-day microbiological cultures. Culture-positive skin allografts can only be released for clinical use if growth of non-pathogens appeared post 7-days of culture. In our skin bank this results in a donor rejection rate (due to positive cultures) of approximately 15% of cryopreserved skin donations, which is in line with the expectations (Rooney et al. 2008). The CAS containing transport solution was (since it’s introduction in 2009) not capable of decontaminating the harvested fresh donor skin in 5 of 34 (14.7%) skin donations. The drug sensitivity patterns revealed that these bacteria were sensitive (in vitro) to at least one of the antibiotics present in the CAS-cocktail. As to why the cocktail was not efficient in some cases, we can think of two likely causes. First of all, bacteria can be hidden deep into the harvested skin (e.g. in the dept of the hair-follicles) where the antibiotics can’t reach them in due time and, secondly, the optimal operating temperature of these antibiotics is much closer to 36°C than to 2–8°C.

Recently, we decided to develop and implement a recovery procedure for the cryopreserved skin allografts that tested positive for bacteria and fungi.

Multiple publications describe the antimicrobial effect of glycerol and the use of glycerol as an inactivation agent for culture positive skin allografts (de Backere 1994; Hoekstra et al.1994; van Baare et al. 1994; Marshall et al. 1995; Richters et al. 1996; Mackie 1997; Saegeman et al. 2008). The hitherto published glycerol preservation and decontamination methods generally use high glycerol concentrations and elevated incubation temperatures and periods. We used the most recently published glycerol inactivation protocol (Saegeman et al. 2008) as a starting point for our own development. Saegeman and co-workers investigated the inhibiting effect of glycerol on bacterial suspensions (not on contaminated skin samples) in relation to the glycerol concentration and incubation temperature and time. They observed that no bacterium, with the exception of one spore-former (Bacillus subtilis) survived more than 14 days in the test-tubes with 85% glycerol at 36°C.

We transposed these findings to an inactivation method for the bacteria and fungi in our culture positive cryopreserved skin allografts. Twenty-one thawed (water bath at 37°C) culture positive skin donations were incubated in 50% glycerol (Pharma Belgium) in Hartmann’s solution (Baxter) for 2–24 h at 36°C, followed by long-term storage in 85% glycerol in Hartmann’s solution at 36°C (Fig. 1). Skin allografts had been previously routinely cryopreserved in 30% glycerol in Hartmann’s solution at <−135°C (vapour of liquid nitrogen) (Kets et al. 1985).

Fig. 1
figure 1

Incubation of skin allografts in 85% glycerol at 36°C

Full size image

After 5–6 weeks of incubation, several representative samples of 10–20 cm² each were rinsed in 0.9% NaCl and submitted to 14-day bacteriological and mycological cultures in thioglycolate (with resazurine, bioMérieux) and Sabouraud (bioMérieux) broth, respectively. In case of a positive culture, the 85% glycerol solution was renewed and incubation at 36°C was resumed. After 5–6 months new samples were taken and tested. A skin donation was released and transfered to final sterile (closed and locked) storage recipients (Fig. 2) upon negative 14-day cultures. Since the glycerolisation procedure had significantly reduced the initial donor skin surface (up to 15%), skin allografts were remeasured before final packaging.

Fig. 2
figure 2

A vial containing glycerolised skin allografts and labelled with donor references, lot number, sizes and expiry date

Full size image

After 5–6 months all donations scored negative in 14-day bacterial and mycological testing, except one (containing Bacillus sp.), which was definitively rejected and destroyed (Table 1). The structural integrity of the rehydrated glycerolisation-recovered cryopreserved skin allografts was evaluated. Sections (Bouin fixation followed by haematoxylin eosin staining and PAS reaction) of “direct glycerolisation only”, “direct cryopreservation only” and “glycerolisation after cryopreservation” skin samples were evaluated under a microscope and compared (blind) by an experienced dermato-pathologist. This evaluation was based on the pigmentation, presence and integrity of hair follicles and their associated muscles, sweat and sebaceous glands, and the dermal-epidermal junction and the presence of apoptotic and/or necrotic cells, sentinels of local tissue stress and inflammation. No differences were observed between the structural features of the allografts that were immediately glycerolised or only cryopreserved and those that were cryopreserved prior to glycerolisation. Hair follicles and sweat and sebaceous glands were clearly present. The samples harboured neither apoptotic nor necrotic cells and showed a normal pigmentation and an intact dermal-epidermal junction. A more elaborate description and discussion on the functional and structural integrity of those retain samples was published earlier in this journal by Verbeken et al. (2010).

Table 1 Glycerol decontamination of skin allografts positive for bacteria and fungi
Full size table

Upon this satisfactory structural and histological evaluation, our burn surgeons started to use the recovered allografts in routine. Surprisingly, they observed that the grafts did not resist well to shearing forces when meshed (skin is fenestrated to allow drainage and expansion); the epithelium detached from the dermis (epidermolysis) once rinsed in physiological solution (0.9% NaCl) prior to use. Because the epidermis is not an absolute requirement for the proper functioning of meshed allograft skin, surgeons reported this as a minor event without clinical impact for the grafted patients. We preferred, however, to further optimise our glycerolisation-recovery protocol and to solve the epidermolysis problem. We suspected that epidermolysis was mainly due to the long storage at relatively high temperature (36°C). Therefore we reduced the 50% glycerol incubation temperature to 2–8°C and reduced the 85% glycerol incubation period at 36°C to 2.5 h, followed by long time incubation in 85% glycerol at 2–8°C. Eight culture positive skin donations were treated according to this modified protocol. With the exception of one (again containing Bacillus sp.), all skin donations were successfully decontaminated (Table 1) and, as could be expected, all samples exhibited normal structural aspects. Microscopical pictures show a piece of hydrated skin that was only cryopreserved (Fig. 3), one that was only glycerolised (Fig. 4) and one that was glycerolised (“decontaminating—glycerolisation”) after thawing (Fig. 5). This time the surgeons who used the skin grafts and the technicians and nurses who processed and prepared the grafts did not observe any mechanical or structural abnormalities, such as epidermolysis, when recovered skin allografts were meshed and rehydrated. Preliminary clinical observations (7555 cm² of glycerolisation-recovered skin was grafted) indicate that there are no effects on the functional characteristics of the donor skin.

Fig. 3
figure 3

Micrograph (200×) of a haematoxylin eosin and PAS reaction stained section of a rehydrated skin sample that was thawed after 10 years of cryopreservation at < −135°C

Full size image
Fig. 4
figure 4

Micrograph (200×) of a haematoxylin eosin and PAS reaction stained section of a rehydrated skin sample that was glycerolised, without previous cryopreservation

Full size image
Fig. 5
figure 5

Micrograph (200×) of a haematoxylin eosin and PAS reaction stained section of a rehydrated skin sample that was thawed after 14 weeks of cryopreservation at < −135°C and subsequently glycerolised (optimised protocol)

Full size image

Although the physicians of the BWC-QAMH usually prefer to use the more elastic, viable and presumably wound healing-stimulating cryopreserved skin, for some indications they opt for the less immunogenic glycerolised skin. Today, the skin bank can provide them with both types of allograft skin whilst recycling most of the culture positive cryopreserved skin allografts. Contamination with spore-formers remains however problematic. This results in an optimal use of the scarce and precious harvested donor skin (Fig. 6). The described protocol could also be of interest to tissue bankers using glycerolisation as main or sole method of preservation or to researchers involved in the development of human skin equivalents (Richters et al. 2008; Böttcher-Haberzeth et al. 2010; van der Veen et al. 2010).

Fig. 6
figure 6

Flow chart of the glycerolisation-based recovery procedure for cryopreserved human skin allografts that tested positive in bacterial and mycological culture

Full size image