2 Literature Review
2.5 Wound healing research
2.5.2 Clinical
Initial wound evaluation relies heavily on the eye of the clinician. Undoubtedly, adequate clinical history and physical examination help in determining a diagnostic and treatment plan. Certain wound characteristics like its anatomic location assists a clinician in probable etiology. For example, spontaneous ulcers located in the dorsum of the foot usually relate to arterial deficiency. The features of a wound bed and edges, such as color, smell, edge eversion and rolling also aid in determining a cause (Grey et al. 2006, Robson and Barbul 2006). Nonetheless, once the diagnosis is made, it is essential to have an objective method to assess the wound during every visit to determine its evolution and response to treatment.
Determining the size and depth of the wound are key characteristics in registering their progression (Kantor and Margolis 1998, Flanagan 2003). Because most wounds are irregular in shape, their measurement can be difficult. Planimetry, is the estimation of wound surface area and its progress until complete closure. It can be measured manually using transparent acetate film or through
photo/videography by tracing the edges of the wound (Bohannon and Pfaller 1983, Gethin and Cowman 2006, Kim et al. 1987, Rogers et al. 2010, Wunderlich et al.
2000). The measurement is then made manually from a millimetric grid or analyzed by computer software (Gorin 1996, Wendland and Taylor 2017, Wendelken et al.
2011, Lagan et al. 2000), but this method is considered imprecise (Mayrovitz and Soontupe 2009, van Zuijlen et al. 2004, Grey et al. 2006, Robson and Barbul 2006, Bloemen et al. 2012). Even though there has been advancement in the area of digital imaging and software processing, ideal image analysis software is still lacking. There are software tools that aid in analyzing biomedical imaging; however, these are costly, inaccessible, and require extensive training to obtain adequate analyses (Eliceiri et al.
2012, Meijering et al. 2016). The Image J “project” is an open source piece of software that includes advanced tools allowing the user to process and analyze images in a simple manner (Rasband 1997-2018, Abramoff, et al. 2004, Schneider et al. 2012, Rueden et al. 2017). However, most image software products have their limitations, and do not replace the precision of an experienced clinician. A completely re-epithelialized wound is one that is dry (free from secretions), and the patient able to return to normal activities.
2.5.3 Split-thickness skin graft donor sites as a wound healing model
After direct wound opposition, skin graft closure is considered the simplest form of wound coverage in reconstructive surgery. Skin grafts are thin dermo-epidermal components that can be split or full thickness (STSG and FTSG, respectively) (Ratner 1998). STSGs are harvested with an electric or manual dermatome, and classified according to their thickness as ultrathin, thin, intermediate, and thick (Andreassi et al. 2005). These wounds are so superficial that they are left to re-epithelialize on their own, which normally occurs within 15 days under optimal conditions (Voineskos et al. 2009, Demirtas et al. 2010, Masella et al. 2014, Wiechula 2003). STSG donor sites are great models to evaluate wound healing, because wounds are controlled for size, depth, and location and can test different treatments and predict healing patterns (Nuutila et al. 2012, Nuutila et al. 2013, Serebrakian et al. 2018, Kazanavičius et al. 2017). In donor site wound healing studies, the primary endpoint is usually complete epithelialization (Still et al. 2003, Ottomann et al. 2010, Innes et al. 2001, Gao et al. 1992, Demirtas et al. 2010, Dornseifer et al. 2011). In spite of the fact that donor sites heal spontaneously, a frequent problem is the resultant discoloration and scar (Rakel et al. 1998). This is why areas normally
covered by clothing are chosen for STSG harvest i.e. inner thighs, abdomen and gluteal area. (Neligan 2013, Osman and Emara 2018, Park et al. 2012).
Brown et al. studied the epidermal regeneration of partial thickness donor site wounds employing daily photographic planimetry for surface area analysis (Brown et al.1989). Although they are a good model to evaluate wound healing, skin graft donor sites are difficult to evaluate. This is because, although most heal by day 10, it is difficult to get a precise daily measurement of epidermal closure, without interfering with its normal closure (Wiechula 2003). While authors have suggested wound dressing changes every 1-3 days (Brown et al. 1989, Guerid et al. 2013, Wiechula 2003), others suggest allowing the wound to close and evaluate it on day 7, 10 or 14 (Dornseifer et al. 2011). There are advantages and disadvantages to both methods. While evaluating every certain number of days lets us know the progress of wound closure, it may intervene and delay wound healing, and make it prone to infection. The opposite occurs with a single evaluation, where wound progress cannot be evaluated, and complete wound healing may have already occurred by the time one uncovers the dressing. There is definitely no one ideal procedure to date, which evaluates wound healing, and is cost-effective, painless, reliable and generates reproducible results.
2.5.4 Monitoring scar progression
Once the wound is closed, local scar characteristics such as color, malleability, positive pressure, tension, torsion, thickness, relief, pliability and surface area may be evaluated. (Hallam et al. 2013 Verhaegen et al. 2011). One of the many objective scar assessment tools is based on the principle of color reflectance (Draaijers et al.
2004). Reflectance colorimetry relies on three broad wavelength filters to quantify brightness, redness and pigmentation of the skin. Nonetheless, the lack of a clinical correlation and difficulty in performing the evaluation makes it less reliable.
Kaartinen et al. described the combined use of controlled lighting digital imaging and spectral modelling to measure the differences in light absorption of melanin and hemoglobin, the two skin chromofores. Through simple digital photography, one can interpret changes in concentration of these pigments in a scar and compare it with normal skin. While hemoglobin reflects red light (thus its red color), brown melanin has the ability to absorb all light wavelengths. This is the simple principle behind spectrophotometry and why it can be considered a reliable tool for measuring scar maturity. As a scar matures naturally, the hemoglobin and melanin concentrations become progressively similar to that of native skin. (Kaartinen et al.
2011). To measure scar thickness, a scar biopsy remains the gold standard.
Unfortunately, biopsies are invasive and may not precisely reflect what is happening in all parts of a scar. Employing innovative technology such as the Tissue Ultrasound Palpation system is highly practical, but not without costs. Using a small probe, one can determine an exact value of scar depth and stiffness (Zheng et al. 2000, Lau et al. 2005, Chao et al. 2010). Cutaneous scar depth and perfusion can be evaluated with high precision through a laser Doppler, (Fearmonti et al. 2010), laser perfusion imaging and 3D spectrophotometric intracutaneous analysis (Ud-Din et al. 2015, Hop et al. 2013). Concerning wound topography, high resolution imaging software such as the 3D optical profiling system (PRIMOS), generates high quality imaging of a scar, but is costly (Roques et al. 2003). Finally, surface area can be estimated though photogrammetry and sterophotogrammetry. These procedures employ one or two cameras placed at precise distances from the scar. The cameras come with built-in grids to obtain a calibrated image. It has been shown that photographic methods are more reliable for scar or wound planimetry than manual acetate methods because they reduce the inaccuracy caused by manual tracing (Stekelenburg et al. 2013, Verhaegen et al. 2011, Perry et al. 2010).
There are a number of scales available to monitor scar progression, many of which originated from burn scar evaluation (Durani et al. 2009, van de Kar et al.
2005, Tyack et al. 2012). They have been designed with the intention of measuring subjective parameters with an objective scoring system. Most scars are observer dependent and consider factors such as thickness, malleability, height, contour, texture, pigmentation and vascularization to evaluate scar progress. The most popular scales that assess scars in decreasing order are the Vancouver Scale (VSS), Manchester Scar scale, Patient and Observer Scar Assessment scale (POSAS) and Stony Brook scale (SBSES) (Draaijers et al. 2004, Perry et al. 2010 Brusselaers et al.
2010, Verhaegen et al. 2011, Rennekampff et al. 2006, Fearmonti et al. 2010). The VSS is most commonly applied in burn scar assessment and measures only a particular area of the scar, not taking into consideration the heterogenicity of wounds. The POSAS scale is unique in that it surveys both the physician and the patient, but it is inconsistent and has greater inter-observer variability compared with VSS. The Manchester scar scale is applicable for a variety of wounds, while the SBSES assesses the cosmetic outcome of wounds 5-10 days after the removal of a surgical suture or staple. These scales are particularly useful for small scars and get progressively inaccurate as the scar enlarges. Similarly, these scars only evaluate form and not functional outcomes (see Table 4).
Visual analog scales are also commonly employed to evaluate pain and patient satisfaction, but this requires a high level of understanding and collaboration from the patient (Monstrey et al. 2008, Brusselaers et al. 2010). Additionally, the VAS scale has been modified to produce a photograph-based evaluation that includes traits such as pigmentation, vascularity, patient acceptability and observer comfort.
Unfortunately, this is highly observer-dependent (Fearmonti et al. 2010). Blinded expert evaluation of patient wounds or via photography has served as an excellent alternative to evaluate wound healing in an objective manner. Even if evaluators are blinded to treatment, there is considerable subjectivity involved in this type of evaluation (Eginton et al. 2003).
Table 4. Frequently employed scales for scar assessment.
Scale Indications Evaluation type
VSS Burn scars, scar in general Numeric scale for vascularity, pliability, pigmentation and height
Manchester Scars in general Visual analog scale for color, contour, distortion, texture, shine
POSAS Linear scar assessment Observer: vascularity, pigmentation, thickness, relief, pliability
Patient: pain, pruritus, color, stiffness, thickness, irregularity
Stoney Brook Surgical wounds Width, height, color, suture marks, overall appearance
3 AIMS OF THE STUDY
The goal of this thesis was to determine the efficacy of Adipose Tissue Extract (ATE) in wound healing. The specific aims were:
1. To transfer the original laboratory ATE method to a new, simple method of ATE extraction in an operating theatre (Study I)
2. To verify the growth factor (GF) and protein content of ATE (Studies I and II)
3. To study in vitro the effects of ATE and platelet-rich plasma (PRP) in the proliferation and migration of the main cells involved in wound healing:
keratinocytes, fibroblasts, endothelial cells and stem cells (Study II) 4. To evaluate the effect of ATE and PRP in human skin graft donor sites in
terms of re-epithelialization area over time, scar quality and pain scores to determine ATE’s clinical potential (Study III)
5. To correlate ATE’s effects on cell function in vitro and in clinical wound healing settings
6. To determine ATE’s progressive protein release among different wound dressings
4 SUBJECTS, MATERIALS AND METHODS
4.1 Ethical considerations
This investigation complies with the principles outlined in the Declaration of Helsinki. The human tissue used in this study conforms with the Ethics Committee of the Pirkanmaa Hospital District, Tampere, Finland, approval number R15034.
4.2 Patients and volunteers (I-III)
The adipose tissue lipoaspirate samples were obtained from otherwise healthy donors programmed for surgical procedures at the Department of Plastic Surgery of the Tampere University Hospital, with signed consent. For study I, 27 samples were obtained from patients with an indication of liposuction either as a complement to body contouring or fat grafting. In study I, all patients were females with a mean age of 52 years, mean weight of 77 kg and mean BMI of 28 kg/m2 (ranging from 21-38). For studies II-III, a total of 24 patients were included, 18 male and 6 females, with an indication for STSG. The indication for the skin graft reconstruction included trauma (10 patients), burns (7), ulcers or postoperative complicated wounds (5) and localized skin cancer (2). From these, 13 patients (10 males and 3 females) had ATE extracted, with a mean age of 56 years and mean BMI of 30 kg/m2. In contrast, 11 patients (8 males and 3 females) in this group had PRP extracted, with a mean age of 62 years and mean BMI of 25 kg/m2 (see Table 5).
Table 5. Patient demographics.
Study I Study II-III
ATE
(n=27) PRP
(n=11) ATE
(n=13) Age (yrs) 51.9 (9.7) 62.5 (17.2) 56.1 (16.5) Gender
Male 0 8 (72.7%) 10 (76.9%)
Female 27 3 (29.2%) 3 (23.0%)
BMI (kg/m2) 28.1 25.2 (4.4) 30.3 (6.9)
Diabetes N/A 4 (36.4%) 4 (30.1%)
Hypertension N/A 5 (45.4%) 7 (53.8)
Cardiopathy N/A 3 (27.3%) 4 (30.8%)
Smoker N/A 1 (9.0%) 2 (15.4%)
Adipose tissue origin
Abdomen 22 - 13
Lateral thorax 4 - 0
Inner thigh 1 - 0
4.3 Cells and tissue samples (I-II)
Four different cell lines, 2 primary and 2 commercial, were employed in the studies.
Human adipose stem cells (hASC) were isolated from subcutaneous fat obtained from surgical samples of healthy volunteers (ethics approval no. R03058) from the Tampere University Hospital. Human umbilical vein endothelial cells (HUVECs) were isolated from the umbilical cords of healthy donors acquired from caesarean sections (ethics approval no. R08028). Finally, human keratinocytes (HaCaT; ATCC, Manassas, VA, USA) and human foreskin fibroblasts (BJ; ATCC) were employed (Study I).
4.4 Materials
Cell culture reagents included: DMEM (Lonza, Basel, Switzerland), 1% L-glutamine (Gibco, Grand Islands, NY, USA), 1% antibiotics (penicillin/streptomycin, Gibco, Invitrogen, Carlsbad, CA, USA), 10% FBS (Gibco), DMEM/F12 (Gibco), 10%
human serum AB (Biowest, Nuaille, France), EGMTM-2 (Endothelial Cell Growth Medium-2 Bullet Kit, Lonza Group Ltd), bovine serum albumin (BSA), 10% bovine serum albumin (Sigma). Also, 16, 48 and 96-well plate inserts were employed from Scaffdex Oy, Tampere.
4.4.1 Chemicals
Commercial analysis kits/reagents: Pierce™ BCA Protein Assay Kit (Thermo Scientific, Waltham, MA), Varioskan™ Flash Multimode Reader (ThermoScientific), PRP GLO Pro kit (Glofinn Oy, Salo, Finland), BacT/ALERT aerobic and anaerobic bacteria culture systems (bioMérieux SA, France), BacT/ALERT 3D (bioMérieux SA), colorimetric sandwich ELISA, Custom made ELISA strips (Signosis®, Santa Clara, CA) for VEGF, tumor necrosis factor alpha (TNFα), interferon gamma (IFNγ), granulocyte colony stimulating factor (G-CSF), granulocyte macrophage colony stimulating factor (GM-CSF), IL-6, IL-8, IGF-1, IL-1α, bFGF, resistin, macrophage inflammatory protein (MIP-1), adiponectin, leptin, and rantes, Protein standards for custom human cytokine ELISA Strip (Signosis), KGF Kit (Boster Biological Technology Co, Ltd, Pleasanton, CA, USA), assay wash buffer, Streptavidin-HRP, Presto blue (Life Technologies, Invitrogen Corporation, Carlsbad, CA, USA) cell viability reagent (Lopez et al. 2016).
4.4.2 Other materials
Ringer lactate (Baxter Healthcare Corporation, Helsinki, Finland), 0.2 µm pore size syringe filters (Acrodisc® filter, polyethersulfone PES membrane (PALL Life Sciences, New York); Minisart NML filter, cellulose acetate membrane (Sartorius AG, Germany); Filtropur S Plus filter, cellulose acetate membrane (Sarstedt & Co, Germany); and Millex GP filter, PES membrane (Merck, Millipore, Germany).
4.4.3 Liposuction materials
Water-assisted liposuction device (Bodyjet, Human Med AG, Germany), canister (LipoCollector, Human Med, Germany), Ringer lactate (Baxter Healthcare Corporation, Helsinki, Finland), adrenaline.
4.4.4 Wound dressings and gel materials
Hydrocolloid dressing (Duoderm Extrathin, E.R. Squibb & Sons, LLC), semi-occlusive methylcellulose film (Opsite Flexigrid, T.J. Smith and Nephew, UK), Aquacel hydrofiberâ Aquacelâ extra, Aquacelâ Ag (ConvaTec, Desidee, UK), Acticoatâ Absorbent (Smith&Nephew, Inc., Watford, UK), Exufiberâ (Mölnlycke, Gothenburg, Sweden), Cutimedâ Siltec (BSN Medical GmbH,
Hamburg, Germany), Evercareâ adhesive surgical dressing (OneMed, Helsinki, Finland), Curea P1 superabsorbent wound dressing (Apodan, Hoersholm, Denmark), Mepilexâ Border (Mölnlycke, Gothenburg, Sweden), CarboFLEXâ dressing (ConvaTec, Desidee, United Kingdom), Intrasite gel (Smith&Nephew, Inc., Watford, UK), Restylane (Galderma, Lausanne, Switzerland) and Microdacynâ hydrogel (Microcyn Technology, Oculus Innovative Sciences, CA, USA).
4.4.5 Other equipment
Air-driven Zimmer dermatome and mesher (Zimmer Inc., Freiburg, Germany), Fluorometric analysis Varioskan Flash Multimode Reader (Thermo Scientific), digital camera (Nikon TS-100, Tokyo, Japan) connected to an inverted microscope (Nikon), fluorescence Nikon Eclipse Ti-S microscope (Nikon, Tokyo, Japan), Fuji IS Pro (Fujifilm Corporation, Tokyo, Japan) connected to a specialized lighting chamber operated by a computer, which was built at the University of Vaasa.
4.4.6 Computer programs
GraphPad Prism 5.0 (GraphPad Software, Inc., San Diego, CA, USA), SPSS® (IBM SPSS Statistics for Macintosh, Version 22.0, IBM Corp, Armonk, NY), Image J (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, https://imagej.nih.gov/ij/, 1997-2018). Microscopy images were processed with Adobe Photoshop software (Adobe Systems, San Jose, CA, USA).
4.5 Methods
4.5.1 Adipose tissue extract preparation
Lipoaspirate samples were obtained with patients under general or spinal anesthesia employing a water-assisted technique (Bodyjet, Human Med AG, Germany). The infiltrating tumescent solution was prepared by adding 1 mg of adrenaline and 250 mg of lidocaine to 1000 mL of saline or ringer lactate solution for study I. For studies II-III the lidocaine was omitted from the solution, on the basis of a series of studies that had demonstrated that local anesthesia, in particular lidocaine, interrupts preadipocyte differentiation and adipocyte cell growth and metabolism
(Keck et al. 2010, Moore et al. 1995, Girard et al. 2013). The settings of the jet infiltration pressure ranged from 2 (110 mL/min) to 3 (130/mL/min), which is equal to 50 and 70 bar, respectively. The suction pressure was set at −400 mbar.
Lipoaspirate samples were then collected into a sterile canister (LipoCollector, Human Med, Germany) and transferred to syringes, as can be seen on Figure 4.
Following sample collection, lidocaine was placed locally and evenly over the lipoaspirated region for analgesic purposes. The total amount of sample collection ranged from 40-200 mL per patient.
Figure 4 Lipoaspirate collection. Water-assisted liposuction device employed in the studies (left), 500ml cannister used to collect adipose tissue (center), fat collection into a sterile syringe (right).
4.5.2 ATE preparation in the laboratory (I)
Immediately after the lipoaspirate was obtained, it was transported to the laboratory and ringer lactate was added at a ratio of 1:1. The sample was incubated either in a 37°C water bath or at room temperature (RT). Finally, the sample was filtered through 0.2 µm syringe filters. In study I, three different incubation timepoints (15, 30, or 45 min) and four different syringe filters were tested: Acrodisc® filter polyethersulfone PES membrane (PALL Life Sciences, New York); Minisart NML filter, cellulose acetate membrane filter (Sartorius AG, Germany); Filtropur S Plus filter, cellulose acetate membrane filter (Sarstedt & Co, Germany); and Millex GP filter, PES membrane (Merck, Millipore, Germany). After filtration, the resulting extract, ATE, was stored at −20°C.
4.5.3 ATE preparation in the operation theatre (I-III)
In the operating theatre, the ATE was produced in sterile conditions on a separate Mayo table. The lipoaspirate was combined with warm (37°C) Ringer lactate at a 1:1 ratio. The mixture was gently mixed and incubated for 30 min at RT. The decanted lower layer was passed through a sterile syringe filter (Acrodisc®, Pall Corporation, Port Washington, NY, USA) and either immediately placed on the donor sites or frozen at −20°C for later use (Fig 5).
Figure 5. ATE preparation.
4.5.4 ATE sterility testing (I)
Six ATE samples prepared in the OR were sent to the microbiology laboratory at the Tampere University Hospital and tested for gram-positive, gram-negative bacteria and yeast to ensure a quality product. Two milliliters of ATE was added to each BacT/ALERT SA and SN system (bioMérieux SA, France) bottle. The samples were cultured for 10 days and results were reported as positive (bacterial growth detected) or negative (no bacterial growth detected).
4.6 PRP preparation (II-III)
PRP preparation began with the extraction of 9ml of peripheral blood. Following the manufacturer’s instructions of the GLO Pro kit, one ml of sodium citrate was combined with blood and centrifuged at 1200 x g for 5 min. The bottom layer of red blood cells was removed from the syringe, followed by a second centrifugation at 1200 x g for 10 min. The PRP layer was drawn from the syringe and activated with 10% calcium gluconate. After the PRP gel slowly formed, it was immediately placed over the donor site wound and the rest stored at - 20 °C for further use.
4.7 ATE and PRP characterization: protein and GF measurements (I-II)
4.7.1 Protein measurements
The ATE total protein yield was quantified employing the PierceTM BCA Protein Assay Kit (Thermo Scientific, Waltham, MA, USA) using BSA as a standard, following the manufacturer’s instructions. Readings were subsequently performed at 562 nm with a VarioskanTM Flash Multimode Reader (Thermo Scientific).
Unfortunately, due to the physical characteristics of the PRP gel, PRP protein readings could not be discerned.
4.7.2 Growth factor measurements
Sandwich ELISA Custom human cytokine and Angiogenesis strips (Signosisâ, Santa Clara, CA, USA) were employed to obtain the concentration of the following growth factors/cytokines: EGF, IGF-1, IL- 6, PDGFb, TNFα, TGFb, bFGF and VEGF. To measure KGF and other individual GFs the Growth Factor Kit (Boster Biological Technology Co, Ltd, Pleasanton, CA, USA) was employed. Readings were performed at 450 nm with Varioskan Flash multimode reader (Thermo Scientific).
4.8 Cell isolation and culture (II)
Four cell lines were individually investigated: keratinocytes, fibroblasts, endothelial and adipose stem cells. To perform keratinocyte cultures, human keratinocytes were combined with DMEM (Lonza, Basel, Switzerland), supplemented with 1%
L-glutamine (Gibco, Grand Islands, NY, USA), 1% antibiotics and 10% FBS (Gibco).
Fibroblasts were cultured in MEM along with 10% FBS and 1% L-glutamine.
Human adipose stem cells were isolated from adipose tissue and cultured in medium containing DMEM/F12, 10% human serum AB (Biowest, Nuaille, France) and 1%
L-glutamine. Finally, endothelial cells derived from human umbilical cords were cultured in EGM-2 (Endothelial Medium Bullet Kit, Lonza).
4.8.1 Cell proliferation assay (II)
As preparation for the cell proliferation assays, an optimal cell confluency was obtained within 6 days. This corresponded to the following values, depending on the cell type: keratinocytes 7500 cells/ cm2, fibroblasts 5000 cells/cm2, adipose stem cells 5000 cells/cm2 and endothelial cells 7500 cells/cm2. The cells were plated as monocultures in 96-well plates.
In previous studies, ATE samples demonstrated bioactivity from concentrations of 200 µg/mL and upwards in cell culture (Sarkanen 2012). For this
In previous studies, ATE samples demonstrated bioactivity from concentrations of 200 µg/mL and upwards in cell culture (Sarkanen 2012). For this