Forensic Science International 246 (2015) 86–91
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
Analysis of morphological characteristics and expression levels of extracellular matrix proteins in skin wounds to determine wound age in living subjects in forensic medicine Judith Fronczek a,b,1, Ronald Lulf c,1, H. Ibrahim Korkmaz a,d,*, Birgit I. Witte e, Franklin R.W. van de Goot b, Mark P.V. Begieneman a,d,f, Paul A.J. Krijnen a,d, Lawrence Rozendaal a, Hans W.M. Niessen a,d,g, Udo J.L. Reijnders c a
Department of Pathology, VU Medical Centre, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands Department of Pathology, Symbiant, Medisch Centrum Alkmaar, Wilhelminalaan 12, 1815 JD Alkmaar, The Netherlands c Public Health Service Amsterdam, Department of Forensic Medicine, Nieuwe Achtergracht 100, 1018 WT Amsterdam, The Netherlands d ICaR-VU, VU Medical Centre, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands e Department of Epidemiology and Biostatistics, VU Medical Centre, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands f Department of Pathology, Nederlands Forensisch Instituut (NFI), Laan van Ypenburg 6, 2490 GB The Hague, The Netherlands g Department of Cardiac Surgery, VU Medical Centre, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 5 September 2014 Received in revised form 29 October 2014 Accepted 13 November 2014 Available online 22 November 2014
Objective: Wound age determination in living subjects is important in routine diagnostics in forensic medicine. Macroscopical description of a wound to determine wound age however is inadequate. The aim of this study was to assess whether it would be feasible to determine wound age via analysis of morphological characteristics and extracellular matrix proteins in skin biopsies of living subjects referred to a forensic outpatient clinic. Methods: Skin biopsies (n = 101), representing the border area of the wound, were taken from skin injuries of known wound age (range: 4.5 h–25 days) in living subjects. All biopsies were analyzed for 3 morphological features (ulceration, parakeratosis and hemorrhage) and 3 extracellular matrix markers (collagen III, collagen IV and a-SMA). For quantification, biopsies were subdivided in 4 different timeframes: 0.2–2 days, 2–4 days, 4–10 days and 10–25 days old wounds. Subsequently, a probability scoring system was developed. Results: For hemorrhage, collagen III, collagen IV and a-SMA expression no relation with wound age was found. Ulceration was only found in wounds of 0.2–2, 2–4 and 4–10 days old, implying that the probability that a wound was more than 10 days old in case of ulceration is equal to 0%. Also parakeratosis was almost exclusively found in wounds of 0.2–2, 2–4 and 4–10 days old, except for one case with a wound age of 15 days old. The probability scoring system of all analyzed markers, as depicted above, however can be used to calculate individual wound age probabilities in biopsies of skin wounds of living subjects. Conclusions: We have developed a probability scoring system, analysing morphological characteristics and extracellular matrix proteins in superficial skin biopsies of wounds in living subjects that can be applied in forensic medicine for wound age determination. ß 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Immunohistochemistry Living subjects Forensic Immunology Skin wounds Wound age determination
1. Introduction Wound age estimation is important in forensic medicine. In the Netherlands living victims of violence or molestation with skin
* Corresponding author at: VU University Medical Center, Department of Pathology, Room Nr. 0E46, De Boelelaan 1117, 1007 HV Amsterdam, The Netherlands. Tel.: +31 20 444 0914; fax: +31 20 444 2964. E-mail address: [email protected] (H. Ibrahim Korkmaz). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.forsciint.2014.11.013 0379-0738/ß 2014 Elsevier Ireland Ltd. All rights reserved.
injuries (e.g., bruises, abrasions, etc.) are referred to a forensic physician for an assessment of the injury, including an estimation of wound age. As the reproducibility of macroscopic findings is low and not strong enough to hold in court, there is a need for additional methods to determine wound age in living subjects [1– 4]. In forensic autopsies several immunohistochemical markers are applied to determine wound age more precisely, in addition to more traditional histological characteristics of wound healing [5–15]. To this end skin excisions are made in which the actual damage (e.g. hemorrhage), mostly at the dermo-subcutaneous
J. Fronczek et al. / Forensic Science International 246 (2015) 86–91
junction, can be assessed [1,16]. However, in living subjects it is not ethical to excise skin wounds. We wondered whether superficial skin biopsies at the site of the injury, representing the border area of the wound, could also be used for (immuno)histochemical analysis to determine wound age. Herein we analyzed both morphological characteristics and extracellular matrix proteins that already have been applied in autopsy wounds [17–19]. Different histological markers have been described in wound healing. Ulceration of the epidermis can occur instantly at the time of infliction, dependent on the type of wound induction [20]. Parakeratosis can be found as early as 24 h after wounding of the skin in living people [21,22]. An almost consistent feature of a vital wound is hemorrhage, which leads to a hematoma (i.e. a bruise) that may appear almost instantaneously or it may take 24–48 h after wounding before extravasated blood rises to the surface to become macroscopically visible [23]. In an autopsy study it was shown that the extracellular matrix marker alpha smooth muscle actin (a-SMA) (myofibroblasts) was detected in human skin wounds as early as approximately 5 days post wound infliction [24]. Collagen III and IV have been studied in skin wounds before by Betz et al. [25,26], albeit they did not analyze collagen in blood vessels, as we did in the present study [27]. In this study we have analyzed the above-mentioned parameters in superficial skin biopsies of living subjects, to improve wound age determination.
87
phosphate buffered saline. These tissue slides were negative (not shown). 2.2. (Immuno)histochemical analysis Slides were evaluated by light microscopy. The biopsies reflected the superficial border zone area of a wound, containing the epidermis and superficial dermis. In total 6 variables were scored as follows: *Morphology: - ulceration: yes/no; - parakeratosis: yes/no; - hemorrhage: percentage of the biopsy area (0–100%).
*Extracellular matrix: - collagen III-positive vessels (collagen surrounding vessels): percentage of the total number of vessels (0–100%); - collagen IV-positive vessels (collagen surrounding vessels): percentage of the total number of vessels (0–100%); - a-SMA-positive myofibroblast(s) in the dermis (1): yes/no.
2.3. Statistical analysis 2. Materials and methods 101 human skin biopsies (1 biopsy per wound) were collected from patients being referred to a forensic physician of the Public Health Service in Amsterdam from May 2008 until November 2009. Patients with skin injuries of known age were asked permission to obtain a wound biopsy. The mean age of the population was 37 years (range: 17–80 years), consisting of 52 males (51%) and 49 females (49%). Mean wound age, as reported by the victim, was 5.4 days (range: 4.5 h–25 days). The wounds were identified macroscopically. The different types of injuries were bruises (69%), abrasions (19%), bites (3%), stabs (2%), scratches (2%), unknown (3%) and firework (1%). Permission for this study was given by the Medical Ethical Commission (METC) of the VUmc. 2.1. (Immuno)histochemistry Skin biopsies were overnight fixed in 4% formaldehyde solution and embedded in paraffin. For histochemical analysis, 4 mm-thick tissue sections were stained with standard Hematoxylin/Eosin (H&E) staining. For immunohistochemistry, tissues were dewaxed and rehydrated and incubated in methanol/hydrogen peroxide (H2O2) (0.3%) solution for 30 min to block endogenous peroxidases. Antigen retrieval was performed by either boiling slides in a citrate pH 6.0 buffer (collagen IV) or by incubation with pepsin/ hydrochloric acid (HCl) (0.1%) (collagen III) for 30 min at 37 8C. Sections were incubated with either rabbit anti-human collagen III (1:1000, Acris Antibodies, San Diego, USA), mouse anti-human collagen IV (1:50, Dako, Glostrup, Denmark), or mouse anti-human a-smooth muscle actin (1:200, Dako, Glostrup, Denmark) antibody for 1 h at room temperature (RT). Next, slides were incubated with anti-mouse/rabbit Envision (Dako, Glostrup, Denmark) for 30 min at RT. Staining was visualized using 3,30 diaminobenzidine (0.1 mg/ml, 0.02% H2O2). Sections were then counterstained with hematoxylin, dehydrated and covered. As a control, the same staining procedure was used, but then the primary monoclonal or polyclonal antibody was replaced by
Statistical analysis was performed with SPSS (Windows version 20, IBM Corp., Armonk, NY). Morphological features and extracellular matrix markers of the different wound age groups were compared with the chi-square test (dichotomous parameters) or the Kruskal–Wallis test (non-normal continuous parameters). Pvalues < 0.05 were considered significant. 2.4. Probability scores For quantification, wound age was divided into four groups: 0.2– 2 days old, 2–4 days old, 4–10 days old and 10–25 days. To compute the probability scores, the percentage of hemorrhage, collagen III and collagen IV first had to be binned into several groups of approximately equal size. Hemorrhage was binned in 3 groups (1%, 2–5% or 6%), collagen III and collagen IV in 4 groups (10%, 10–40%, 40–80% and >80% or 50%, 50–75%, 75–90% and >90%, respectively). Then, ordinal regression analyses were applied to compute the probability that the parameter has a certain value (e.g. no ulceration or a percentage of hemorrhage of 2–5%) with wound age as factor. Finally, by Bayes’ rule, these probabilities could be inverted to obtain the probability that a wound has a certain age given the value of the parameter, i.e. the probability score. 3. Results 3.1. Histological markers Ulceration (Fig. 1A/Fig. 2A) was only found in 11 out of 86 cases of skin injuries of 0.2–2, 2–4 and 4–10 days old, albeit not in wounds older than 5 days. No significant differences between the age groups were found (p = 0.247). Parakeratosis (Fig. 1B/Fig. 2B) was found in 23 out of 86 cases in these age groups, except for one patient with a wound age of 15 days old. No significant differences between age groups were found either (p = 0.276). Hemorrhage (Fig. 1C/Fig. 2C) was found in 51% of the wounds and in all wound age groups, albeit limited (mean was 80% collagen III positivity. The probabilities that a wound was 0.2–4 days old were 47, 56, 63 and 69%, respectively, and that a wound was 0.2–10 days old 81, 85, 88 and 90%. For collagen IV the lowest probabilities were found when blood vessel positivity was 50%: 25% that a wound was 0.2–2 days old, 48% that a wound was 0.2–4 days old and 81% that a wound was 0.2–10 days old. In case 50–75, 75–90 and >90% collagen IV-positive blood vessels were found, the probabilities were higher: 35/36/38% that a wound was 0.2–2 days old, 70/73/77% that a wound was 0.2–4 days old and 93/94/95% that a wound was 0.2–10 days old, respectively. When a-SMA-positive myofibroblasts were present the probability that a wound was 0.2–2 days old was 15%, that a wound was 0.2–4 days old was 50% and that a wound was 0.2–10 days old was
J. Fronczek et al. / Forensic Science International 246 (2015) 86–91
Ulceration
a
89
Collagen III+ blood vessels
a 100 80
no %
60 40
yes 20 0 0.2-2
0
25
0.2-22-4 25 28
4-10 28
2-4 27
4-10 26
10-25 13
10-25 13
Collagen IV+ blood vessels
b
Parakeratosis
100
b
80
no %
60 40 20
yes
0 0.2-2 23
2-4 24
4-10 22
10-25 12
0 0.2-2 2-4 25 28
4-10 28
10-25 13
SMA
c
no
Haemorrhage
c
100 yes
80
%
60 0
40
0.2-2 2-4 25 28
20
4-10 26
10-25 14
Fig. 3. Analyses of extracellular matrix proteins: the percentage of (a) collagen III and (b) collagen IV-positive blood vessels in the different groups. (c) The presence of a-SMA-positive cells in the dermis. Each dot represents 1 skin wound.
0 0.2-2 27
2-4 28
4-10 27
10-25 14
Fig. 2. Analyses of the histological markers: the occurrence of (a) ulceration or (b) parakeratosis. Each dot represents 1 skin wound. (c) The percentage of hemorrhage in the different groups.
90%. When a-SMA-positive myofibroblasts were absent these probabilities were 30%, 59% and 84%, respectively. 4. Discussion Additional methods are needed to determine wound age in living subjects more accurately. We have studied histological markers and extracellular matrix proteins in skin biopsies of living
subjects. We found ulceration in wounds of 0.2–2, 2–4 and 4–10 days old (i.e. maximal 5 days old), implying that the probability that a wound was more than 10 days old in case of ulceration is 0%. Except for one subject, the same was true for parakeratosis. Hemorrhage, collagen III, collagen IV and a-SMA however were not discriminating as such. Notwithstanding this, for all of the abovementioned markers a probability scoring system can be used to determine the likelihood of wound age more precisely. Our finding of ulceration in skin wounds up to maximal 5 days old, is consistent with the study of Betz et al., who found complete reepithelialisation in autopsy wounds as early as 5 days post wound infliction [28]. The almost exclusive finding of parakeratosis in wounds up to 10 days old is also compatible with this,
J. Fronczek et al. / Forensic Science International 246 (2015) 86–91
90
Table 1 Probability of the wound age for the scores of ulceration, parakeratosis, hemorrhage, collagen III, collagen IV and a-SMA. The probabilities in bold denote cumulative probabilities, i.e. the probability that the wound is 0.2–2, 0.2–4 and 0.2–10 days old. Marker
Score
0.2–2 days
2–4 days
4–10 days
10–25 days
Ulceration
Yes No
45% 24%
36% (81%) 29% (53%)
18% (100%) 31% (84%)
16% (100%)
Parakeratosis
Yes No
21% 29%
38% (59%) 27% (56%)
38% (97%) 27% (83%)
4% (100%) 17% (100%)
% Hemorrhage
1 2–5 >6
30% 24% 23%
26% (56%) 32% (56%) 34% (57%)
29% (85%) 29% (85%) 29% (86%)
15% (100%) 15% (100%) 14% (100%)
% Collagen III + blood vessels
10 10–40 40–80 >80
25% 29% 31% 33%
22% 27% 32% 36%
(47%) (56%) (63%) (69%)
34% 29% 25% 21%
(81%) (85%) (88%) (90%)
19% 15% 12% 10%
(100%) (100%) (100%) (100%)
% Collagen IV + blood vessels
50 50–75 75–90 >90
25% 35% 36% 38%
23% 35% 37% 39%
(48%) (70%) (73%) (77%)
33% 23% 21% 18%
(81%) (93%) (94%) (95%)
19% 8% 6% 5%
(100%) (100%) (100%) (100%)
a-SMA
Yes No
15% 30%
35% (50%) 29% (59%)
since parakeratotic corneocytes appear to secure a temporary barrier to cover a wound until complete reepithelialisation [22]. It has to be noticed that the skin biopsies obtained in the present study reflect the superficial border zone of a wound, which can explain the relatively low percentages (mean of 90%). The highest score was found when the positive ratio was >90%, giving a probability of 95% that an injury was inflicted 10 days ago or less. For a-SMA the probability scores were more or less comparable between wounds with and without a-SMA-positive myofibroblasts.
40% (90%) 25% (84%)
10% (100%) 16% (100%)
We were not able to control manipulations to the wounds, e.g. scraping by the wounded person. However, when sampling, macroscopic abnormalities were taken into account. In addition, the victims self are sort of ‘witnesses’ for happening of such manipulations to the wound, however we cannot exclude for sure that the wounds were not manipulated (limitation of the study). To the best of our knowledge, this is the first study assessing the reliability and reproducibility of wound age determination of human skin biopsies from living subjects by assessing morphological characteristics and extracellular matrix proteins. It was shown that they are of use to determine wound age in forensic medicine. More importantly, the probability scoring system can be used to calculate the likelihood of wound age based on a panel of markers. Acknowledgements This study was financed by the Public Health Service in Amsterdam and the Netherlands Forensic Institute (Nederlands Forensisch Instituut), project no. 34. References [1] B. Knight, The pathology of wounds, in: Forensic Pathology, 3rd ed, Edward Arnold, London, 2004, pp. 136–173. [2] T. Kanchan, R.G. Menezes, S. Manipady, Haemorrhoids leading to post-mortem bleeding artefact, J. Clin. Forensic Med. 13 (July (5)) (2006) 277–279. [3] R.B. Dettmeyer, Thrombosis and embolism; vitality, injury age, determination of skin wound age, and fracture age, in: Forensic Histopathology: Fundamentals and Perspectives, 1st ed., Springer, Berlin, 2012, pp. 173–210. [4] DiMaio, Time of death-decomposition, in: Handbook of Forensic Pathology, 2nd ed., CRC Press, London, 2007. [5] W. Grellner, Time-dependent immunohistochemical detection of proinflammatory cytokines (IL-1beta, IL-6, TNF-alpha) in human skin wounds, Forensic Sci. Int. 130 (December 4 (2–3)) (2002) 90–96. [6] J.A. Ortiz-Rey, J.M. Suarez-Penaranda, M.P. San, J.I. Munoz, M.S. Rodriguez-Calvo, L. Concheiro, Immunohistochemical analysis of P-selectin as a possible marker of vitality in human cutaneous wounds, J. Forensic Leg. Med. 15 (August (6)) (2008) 368–372. [7] P. Betz, Immunohistochemical parameters for the age estimation of human skin wounds. A review, Am. J. Forensic Med. Pathol. 16 (September (3)) (1995) 203–209. [8] J. Dressler, L. Bachmann, R. Koch, E. Muller, Estimation of wound age and VCAM-1 in human skin, Int. J. Legal Med. 112 (3) (1999) 159–162. [9] J. Dressler, L. Bachmann, R. Koch, E. Muller, Enhanced expression of selectins in human skin wounds, Int. J. Legal Med. 112 (1) (1999) 39–44. [10] A. Fieguth, H. Feldbrugge, T. Gerich, W.J. Kleemann, H.D. Troger, The time-dependent expression of fibronectin, MRP8, MRP14 and defensin in surgically treated human skin wounds, Forensic Sci. Int. 131 (January 28 (2–3)) (2003) 156–161.
J. Fronczek et al. / Forensic Science International 246 (2015) 86–91 [11] J.A. Ortiz-Rey, J.M. Suarez-Penaranda, E.A. Da Silva, et al., Immunohistochemical detection of fibronectin and tenascin in incised human skin injuries, Forensic Sci. Int. 126 (April 18 (2)) (2002) 118–122. [12] P. Betz, A. Nerlich, J. Wilske, et al., Immunohistochemical localization of fibronectin as a tool for the age determination of human skin wounds, Int. J. Legal Med. 105 (1) (1992) 21–26. [13] M. Martinez-Ferrer, A.R. fshar-Sherif, C. Uwamariya, C.B. de, J.M. Davidson, N.A. Bhowmick, Dermal transforming growth factor-beta responsiveness mediates wound contraction and epithelial closure, Am. J. Pathol. 176 (January (1)) (2010) 98–107. [14] W. Grellner, T. Georg, J. Wilske, Quantitative analysis of proinflammatory cytokines (IL-1beta, IL-6, TNF-alpha) in human skin wounds, Forensic Sci. Int. 113 (September 11 (1–3)) (2000) 251–264. [15] M. Takamiya, S. Fujita, K. Saigusa, Y. Aoki, Simultaneous detection of eight cytokines in human dermal wounds with a multiplex bead-based immunoassay for wound age estimation, Int. J. Legal Med. 122 (March (2)) (2008) 143–148. [16] L.L. Randeberg, O.A. Haugen, R. Haaverstad, L.O. Svaasand, A novel approach to age determination of traumatic injuries by reflectance spectroscopy, Lasers Surg. Med. 38 (4) (2006) 277–289. [17] M.C. Robson, Wound infection: a failure of wound healing caused by an imbalance of bacteria, Surg. Clin. North Am. 77 (1997) 637–650. [18] G.R. Tobin, Wound repair: biologic foundations and clinical considerations, in: J.D. Richardson, H.C. Polk, Jr., L.M. Flint, Jr. (Eds.), Trauma: Clinical Care and Pathophysiology, Yearbook Medical Publishers, Chicago, IL, 1987, pp. 213–261. [19] S. Howes, S.C. Harrey, Healing of wounds as determined by their tensile strength, JAMA 92 (1929) 42.
91
[20] Miller-Keane, Encyclopedia and Dictionary of Medicine, Nursing, and Allied Health, 7th ed., Elsevier, Saunders, 2003. [21] J.N. Mansbridge, A.M. Knapp, Changes in keratinocyte maturation during wound healing, J. Invest. Dermatol. 89 (1987) 253–263. [22] M.L. Usui, R.A. Underwood, P. Fleckman, J.E. Olerud, Parakeratotic corneocytes play a unique role in human skin wound healing, J. Invest. Dermatol. 133 (March (3)) (2013) 856–858. [23] N.E.I. Langlois, G.A. Gresham, The aging of bruises; a review and study of the colour changes with time, Forensic Sci. Int. 50 (1991) 227–238. [24] P. Betz, A. Nerlich, J. Wilske, J. Tu¨bel, R. Penning, W. Eisenmenger, Time-dependent appearance of myofibroblasts in granulation tissue of human skin wounds, Int. J. Legal Med. 105 (2) (1992) 99–103. [25] P. Betz, et al., Analysis of the immunohistochemical localization of collagen type III and V for the time-estimation of human skin wounds, Int. J. Legal Med. 105 (6) (1993) 329–332. [26] P. Betz, et al., The time-dependent rearrangement of the epithelial basement membrane in human skin wounds – immunohistochemical localization of collagen IV and VII, Int. J. Legal Med. 105 (2) (1992) 93–97. [27] B.V. Shekhonin, V.F. Kondalenko, S.P. Domogatskiı˘, A.V. Rudin, V.S. Rukosuev, Distribution of collagen types I, III, IV and V in the walls of human arteries, Kardiologiia 24 (April (4)) (1984) 95–99. [28] P. Betz, Histological and enzyme histochemical parameters for the age estimation of human skin wounds, Int. J. Legal Med. 107 (1994) 60–68. [29] N. Vedrenne, B. Coulomb, A. Danigo, F. Bonte´, A. Desmoulie`re, The complex dialogue between (myo)fibroblasts and the extracellular matrix during skin repair processes and ageing, Pathol. Biol. (Paris) 60 (February (1)) (2012) 20–27.
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
Analysis of morphological characteristics and expression levels of extracellular matrix proteins in skin wounds to determine wound age in living subjects in forensic medicine Judith Fronczek a,b,1, Ronald Lulf c,1, H. Ibrahim Korkmaz a,d,*, Birgit I. Witte e, Franklin R.W. van de Goot b, Mark P.V. Begieneman a,d,f, Paul A.J. Krijnen a,d, Lawrence Rozendaal a, Hans W.M. Niessen a,d,g, Udo J.L. Reijnders c a
Department of Pathology, VU Medical Centre, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands Department of Pathology, Symbiant, Medisch Centrum Alkmaar, Wilhelminalaan 12, 1815 JD Alkmaar, The Netherlands c Public Health Service Amsterdam, Department of Forensic Medicine, Nieuwe Achtergracht 100, 1018 WT Amsterdam, The Netherlands d ICaR-VU, VU Medical Centre, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands e Department of Epidemiology and Biostatistics, VU Medical Centre, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands f Department of Pathology, Nederlands Forensisch Instituut (NFI), Laan van Ypenburg 6, 2490 GB The Hague, The Netherlands g Department of Cardiac Surgery, VU Medical Centre, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 5 September 2014 Received in revised form 29 October 2014 Accepted 13 November 2014 Available online 22 November 2014
Objective: Wound age determination in living subjects is important in routine diagnostics in forensic medicine. Macroscopical description of a wound to determine wound age however is inadequate. The aim of this study was to assess whether it would be feasible to determine wound age via analysis of morphological characteristics and extracellular matrix proteins in skin biopsies of living subjects referred to a forensic outpatient clinic. Methods: Skin biopsies (n = 101), representing the border area of the wound, were taken from skin injuries of known wound age (range: 4.5 h–25 days) in living subjects. All biopsies were analyzed for 3 morphological features (ulceration, parakeratosis and hemorrhage) and 3 extracellular matrix markers (collagen III, collagen IV and a-SMA). For quantification, biopsies were subdivided in 4 different timeframes: 0.2–2 days, 2–4 days, 4–10 days and 10–25 days old wounds. Subsequently, a probability scoring system was developed. Results: For hemorrhage, collagen III, collagen IV and a-SMA expression no relation with wound age was found. Ulceration was only found in wounds of 0.2–2, 2–4 and 4–10 days old, implying that the probability that a wound was more than 10 days old in case of ulceration is equal to 0%. Also parakeratosis was almost exclusively found in wounds of 0.2–2, 2–4 and 4–10 days old, except for one case with a wound age of 15 days old. The probability scoring system of all analyzed markers, as depicted above, however can be used to calculate individual wound age probabilities in biopsies of skin wounds of living subjects. Conclusions: We have developed a probability scoring system, analysing morphological characteristics and extracellular matrix proteins in superficial skin biopsies of wounds in living subjects that can be applied in forensic medicine for wound age determination. ß 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Immunohistochemistry Living subjects Forensic Immunology Skin wounds Wound age determination
1. Introduction Wound age estimation is important in forensic medicine. In the Netherlands living victims of violence or molestation with skin
* Corresponding author at: VU University Medical Center, Department of Pathology, Room Nr. 0E46, De Boelelaan 1117, 1007 HV Amsterdam, The Netherlands. Tel.: +31 20 444 0914; fax: +31 20 444 2964. E-mail address: [email protected] (H. Ibrahim Korkmaz). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.forsciint.2014.11.013 0379-0738/ß 2014 Elsevier Ireland Ltd. All rights reserved.
injuries (e.g., bruises, abrasions, etc.) are referred to a forensic physician for an assessment of the injury, including an estimation of wound age. As the reproducibility of macroscopic findings is low and not strong enough to hold in court, there is a need for additional methods to determine wound age in living subjects [1– 4]. In forensic autopsies several immunohistochemical markers are applied to determine wound age more precisely, in addition to more traditional histological characteristics of wound healing [5–15]. To this end skin excisions are made in which the actual damage (e.g. hemorrhage), mostly at the dermo-subcutaneous
J. Fronczek et al. / Forensic Science International 246 (2015) 86–91
junction, can be assessed [1,16]. However, in living subjects it is not ethical to excise skin wounds. We wondered whether superficial skin biopsies at the site of the injury, representing the border area of the wound, could also be used for (immuno)histochemical analysis to determine wound age. Herein we analyzed both morphological characteristics and extracellular matrix proteins that already have been applied in autopsy wounds [17–19]. Different histological markers have been described in wound healing. Ulceration of the epidermis can occur instantly at the time of infliction, dependent on the type of wound induction [20]. Parakeratosis can be found as early as 24 h after wounding of the skin in living people [21,22]. An almost consistent feature of a vital wound is hemorrhage, which leads to a hematoma (i.e. a bruise) that may appear almost instantaneously or it may take 24–48 h after wounding before extravasated blood rises to the surface to become macroscopically visible [23]. In an autopsy study it was shown that the extracellular matrix marker alpha smooth muscle actin (a-SMA) (myofibroblasts) was detected in human skin wounds as early as approximately 5 days post wound infliction [24]. Collagen III and IV have been studied in skin wounds before by Betz et al. [25,26], albeit they did not analyze collagen in blood vessels, as we did in the present study [27]. In this study we have analyzed the above-mentioned parameters in superficial skin biopsies of living subjects, to improve wound age determination.
87
phosphate buffered saline. These tissue slides were negative (not shown). 2.2. (Immuno)histochemical analysis Slides were evaluated by light microscopy. The biopsies reflected the superficial border zone area of a wound, containing the epidermis and superficial dermis. In total 6 variables were scored as follows: *Morphology: - ulceration: yes/no; - parakeratosis: yes/no; - hemorrhage: percentage of the biopsy area (0–100%).
*Extracellular matrix: - collagen III-positive vessels (collagen surrounding vessels): percentage of the total number of vessels (0–100%); - collagen IV-positive vessels (collagen surrounding vessels): percentage of the total number of vessels (0–100%); - a-SMA-positive myofibroblast(s) in the dermis (1): yes/no.
2.3. Statistical analysis 2. Materials and methods 101 human skin biopsies (1 biopsy per wound) were collected from patients being referred to a forensic physician of the Public Health Service in Amsterdam from May 2008 until November 2009. Patients with skin injuries of known age were asked permission to obtain a wound biopsy. The mean age of the population was 37 years (range: 17–80 years), consisting of 52 males (51%) and 49 females (49%). Mean wound age, as reported by the victim, was 5.4 days (range: 4.5 h–25 days). The wounds were identified macroscopically. The different types of injuries were bruises (69%), abrasions (19%), bites (3%), stabs (2%), scratches (2%), unknown (3%) and firework (1%). Permission for this study was given by the Medical Ethical Commission (METC) of the VUmc. 2.1. (Immuno)histochemistry Skin biopsies were overnight fixed in 4% formaldehyde solution and embedded in paraffin. For histochemical analysis, 4 mm-thick tissue sections were stained with standard Hematoxylin/Eosin (H&E) staining. For immunohistochemistry, tissues were dewaxed and rehydrated and incubated in methanol/hydrogen peroxide (H2O2) (0.3%) solution for 30 min to block endogenous peroxidases. Antigen retrieval was performed by either boiling slides in a citrate pH 6.0 buffer (collagen IV) or by incubation with pepsin/ hydrochloric acid (HCl) (0.1%) (collagen III) for 30 min at 37 8C. Sections were incubated with either rabbit anti-human collagen III (1:1000, Acris Antibodies, San Diego, USA), mouse anti-human collagen IV (1:50, Dako, Glostrup, Denmark), or mouse anti-human a-smooth muscle actin (1:200, Dako, Glostrup, Denmark) antibody for 1 h at room temperature (RT). Next, slides were incubated with anti-mouse/rabbit Envision (Dako, Glostrup, Denmark) for 30 min at RT. Staining was visualized using 3,30 diaminobenzidine (0.1 mg/ml, 0.02% H2O2). Sections were then counterstained with hematoxylin, dehydrated and covered. As a control, the same staining procedure was used, but then the primary monoclonal or polyclonal antibody was replaced by
Statistical analysis was performed with SPSS (Windows version 20, IBM Corp., Armonk, NY). Morphological features and extracellular matrix markers of the different wound age groups were compared with the chi-square test (dichotomous parameters) or the Kruskal–Wallis test (non-normal continuous parameters). Pvalues < 0.05 were considered significant. 2.4. Probability scores For quantification, wound age was divided into four groups: 0.2– 2 days old, 2–4 days old, 4–10 days old and 10–25 days. To compute the probability scores, the percentage of hemorrhage, collagen III and collagen IV first had to be binned into several groups of approximately equal size. Hemorrhage was binned in 3 groups (1%, 2–5% or 6%), collagen III and collagen IV in 4 groups (10%, 10–40%, 40–80% and >80% or 50%, 50–75%, 75–90% and >90%, respectively). Then, ordinal regression analyses were applied to compute the probability that the parameter has a certain value (e.g. no ulceration or a percentage of hemorrhage of 2–5%) with wound age as factor. Finally, by Bayes’ rule, these probabilities could be inverted to obtain the probability that a wound has a certain age given the value of the parameter, i.e. the probability score. 3. Results 3.1. Histological markers Ulceration (Fig. 1A/Fig. 2A) was only found in 11 out of 86 cases of skin injuries of 0.2–2, 2–4 and 4–10 days old, albeit not in wounds older than 5 days. No significant differences between the age groups were found (p = 0.247). Parakeratosis (Fig. 1B/Fig. 2B) was found in 23 out of 86 cases in these age groups, except for one patient with a wound age of 15 days old. No significant differences between age groups were found either (p = 0.276). Hemorrhage (Fig. 1C/Fig. 2C) was found in 51% of the wounds and in all wound age groups, albeit limited (mean was 80% collagen III positivity. The probabilities that a wound was 0.2–4 days old were 47, 56, 63 and 69%, respectively, and that a wound was 0.2–10 days old 81, 85, 88 and 90%. For collagen IV the lowest probabilities were found when blood vessel positivity was 50%: 25% that a wound was 0.2–2 days old, 48% that a wound was 0.2–4 days old and 81% that a wound was 0.2–10 days old. In case 50–75, 75–90 and >90% collagen IV-positive blood vessels were found, the probabilities were higher: 35/36/38% that a wound was 0.2–2 days old, 70/73/77% that a wound was 0.2–4 days old and 93/94/95% that a wound was 0.2–10 days old, respectively. When a-SMA-positive myofibroblasts were present the probability that a wound was 0.2–2 days old was 15%, that a wound was 0.2–4 days old was 50% and that a wound was 0.2–10 days old was
J. Fronczek et al. / Forensic Science International 246 (2015) 86–91
Ulceration
a
89
Collagen III+ blood vessels
a 100 80
no %
60 40
yes 20 0 0.2-2
0
25
0.2-22-4 25 28
4-10 28
2-4 27
4-10 26
10-25 13
10-25 13
Collagen IV+ blood vessels
b
Parakeratosis
100
b
80
no %
60 40 20
yes
0 0.2-2 23
2-4 24
4-10 22
10-25 12
0 0.2-2 2-4 25 28
4-10 28
10-25 13
SMA
c
no
Haemorrhage
c
100 yes
80
%
60 0
40
0.2-2 2-4 25 28
20
4-10 26
10-25 14
Fig. 3. Analyses of extracellular matrix proteins: the percentage of (a) collagen III and (b) collagen IV-positive blood vessels in the different groups. (c) The presence of a-SMA-positive cells in the dermis. Each dot represents 1 skin wound.
0 0.2-2 27
2-4 28
4-10 27
10-25 14
Fig. 2. Analyses of the histological markers: the occurrence of (a) ulceration or (b) parakeratosis. Each dot represents 1 skin wound. (c) The percentage of hemorrhage in the different groups.
90%. When a-SMA-positive myofibroblasts were absent these probabilities were 30%, 59% and 84%, respectively. 4. Discussion Additional methods are needed to determine wound age in living subjects more accurately. We have studied histological markers and extracellular matrix proteins in skin biopsies of living
subjects. We found ulceration in wounds of 0.2–2, 2–4 and 4–10 days old (i.e. maximal 5 days old), implying that the probability that a wound was more than 10 days old in case of ulceration is 0%. Except for one subject, the same was true for parakeratosis. Hemorrhage, collagen III, collagen IV and a-SMA however were not discriminating as such. Notwithstanding this, for all of the abovementioned markers a probability scoring system can be used to determine the likelihood of wound age more precisely. Our finding of ulceration in skin wounds up to maximal 5 days old, is consistent with the study of Betz et al., who found complete reepithelialisation in autopsy wounds as early as 5 days post wound infliction [28]. The almost exclusive finding of parakeratosis in wounds up to 10 days old is also compatible with this,
J. Fronczek et al. / Forensic Science International 246 (2015) 86–91
90
Table 1 Probability of the wound age for the scores of ulceration, parakeratosis, hemorrhage, collagen III, collagen IV and a-SMA. The probabilities in bold denote cumulative probabilities, i.e. the probability that the wound is 0.2–2, 0.2–4 and 0.2–10 days old. Marker
Score
0.2–2 days
2–4 days
4–10 days
10–25 days
Ulceration
Yes No
45% 24%
36% (81%) 29% (53%)
18% (100%) 31% (84%)
16% (100%)
Parakeratosis
Yes No
21% 29%
38% (59%) 27% (56%)
38% (97%) 27% (83%)
4% (100%) 17% (100%)
% Hemorrhage
1 2–5 >6
30% 24% 23%
26% (56%) 32% (56%) 34% (57%)
29% (85%) 29% (85%) 29% (86%)
15% (100%) 15% (100%) 14% (100%)
% Collagen III + blood vessels
10 10–40 40–80 >80
25% 29% 31% 33%
22% 27% 32% 36%
(47%) (56%) (63%) (69%)
34% 29% 25% 21%
(81%) (85%) (88%) (90%)
19% 15% 12% 10%
(100%) (100%) (100%) (100%)
% Collagen IV + blood vessels
50 50–75 75–90 >90
25% 35% 36% 38%
23% 35% 37% 39%
(48%) (70%) (73%) (77%)
33% 23% 21% 18%
(81%) (93%) (94%) (95%)
19% 8% 6% 5%
(100%) (100%) (100%) (100%)
a-SMA
Yes No
15% 30%
35% (50%) 29% (59%)
since parakeratotic corneocytes appear to secure a temporary barrier to cover a wound until complete reepithelialisation [22]. It has to be noticed that the skin biopsies obtained in the present study reflect the superficial border zone of a wound, which can explain the relatively low percentages (mean of 90%). The highest score was found when the positive ratio was >90%, giving a probability of 95% that an injury was inflicted 10 days ago or less. For a-SMA the probability scores were more or less comparable between wounds with and without a-SMA-positive myofibroblasts.
40% (90%) 25% (84%)
10% (100%) 16% (100%)
We were not able to control manipulations to the wounds, e.g. scraping by the wounded person. However, when sampling, macroscopic abnormalities were taken into account. In addition, the victims self are sort of ‘witnesses’ for happening of such manipulations to the wound, however we cannot exclude for sure that the wounds were not manipulated (limitation of the study). To the best of our knowledge, this is the first study assessing the reliability and reproducibility of wound age determination of human skin biopsies from living subjects by assessing morphological characteristics and extracellular matrix proteins. It was shown that they are of use to determine wound age in forensic medicine. More importantly, the probability scoring system can be used to calculate the likelihood of wound age based on a panel of markers. Acknowledgements This study was financed by the Public Health Service in Amsterdam and the Netherlands Forensic Institute (Nederlands Forensisch Instituut), project no. 34. References [1] B. Knight, The pathology of wounds, in: Forensic Pathology, 3rd ed, Edward Arnold, London, 2004, pp. 136–173. [2] T. Kanchan, R.G. Menezes, S. Manipady, Haemorrhoids leading to post-mortem bleeding artefact, J. Clin. Forensic Med. 13 (July (5)) (2006) 277–279. [3] R.B. Dettmeyer, Thrombosis and embolism; vitality, injury age, determination of skin wound age, and fracture age, in: Forensic Histopathology: Fundamentals and Perspectives, 1st ed., Springer, Berlin, 2012, pp. 173–210. [4] DiMaio, Time of death-decomposition, in: Handbook of Forensic Pathology, 2nd ed., CRC Press, London, 2007. [5] W. Grellner, Time-dependent immunohistochemical detection of proinflammatory cytokines (IL-1beta, IL-6, TNF-alpha) in human skin wounds, Forensic Sci. Int. 130 (December 4 (2–3)) (2002) 90–96. [6] J.A. Ortiz-Rey, J.M. Suarez-Penaranda, M.P. San, J.I. Munoz, M.S. Rodriguez-Calvo, L. Concheiro, Immunohistochemical analysis of P-selectin as a possible marker of vitality in human cutaneous wounds, J. Forensic Leg. Med. 15 (August (6)) (2008) 368–372. [7] P. Betz, Immunohistochemical parameters for the age estimation of human skin wounds. A review, Am. J. Forensic Med. Pathol. 16 (September (3)) (1995) 203–209. [8] J. Dressler, L. Bachmann, R. Koch, E. Muller, Estimation of wound age and VCAM-1 in human skin, Int. J. Legal Med. 112 (3) (1999) 159–162. [9] J. Dressler, L. Bachmann, R. Koch, E. Muller, Enhanced expression of selectins in human skin wounds, Int. J. Legal Med. 112 (1) (1999) 39–44. [10] A. Fieguth, H. Feldbrugge, T. Gerich, W.J. Kleemann, H.D. Troger, The time-dependent expression of fibronectin, MRP8, MRP14 and defensin in surgically treated human skin wounds, Forensic Sci. Int. 131 (January 28 (2–3)) (2003) 156–161.
J. Fronczek et al. / Forensic Science International 246 (2015) 86–91 [11] J.A. Ortiz-Rey, J.M. Suarez-Penaranda, E.A. Da Silva, et al., Immunohistochemical detection of fibronectin and tenascin in incised human skin injuries, Forensic Sci. Int. 126 (April 18 (2)) (2002) 118–122. [12] P. Betz, A. Nerlich, J. Wilske, et al., Immunohistochemical localization of fibronectin as a tool for the age determination of human skin wounds, Int. J. Legal Med. 105 (1) (1992) 21–26. [13] M. Martinez-Ferrer, A.R. fshar-Sherif, C. Uwamariya, C.B. de, J.M. Davidson, N.A. Bhowmick, Dermal transforming growth factor-beta responsiveness mediates wound contraction and epithelial closure, Am. J. Pathol. 176 (January (1)) (2010) 98–107. [14] W. Grellner, T. Georg, J. Wilske, Quantitative analysis of proinflammatory cytokines (IL-1beta, IL-6, TNF-alpha) in human skin wounds, Forensic Sci. Int. 113 (September 11 (1–3)) (2000) 251–264. [15] M. Takamiya, S. Fujita, K. Saigusa, Y. Aoki, Simultaneous detection of eight cytokines in human dermal wounds with a multiplex bead-based immunoassay for wound age estimation, Int. J. Legal Med. 122 (March (2)) (2008) 143–148. [16] L.L. Randeberg, O.A. Haugen, R. Haaverstad, L.O. Svaasand, A novel approach to age determination of traumatic injuries by reflectance spectroscopy, Lasers Surg. Med. 38 (4) (2006) 277–289. [17] M.C. Robson, Wound infection: a failure of wound healing caused by an imbalance of bacteria, Surg. Clin. North Am. 77 (1997) 637–650. [18] G.R. Tobin, Wound repair: biologic foundations and clinical considerations, in: J.D. Richardson, H.C. Polk, Jr., L.M. Flint, Jr. (Eds.), Trauma: Clinical Care and Pathophysiology, Yearbook Medical Publishers, Chicago, IL, 1987, pp. 213–261. [19] S. Howes, S.C. Harrey, Healing of wounds as determined by their tensile strength, JAMA 92 (1929) 42.
91
[20] Miller-Keane, Encyclopedia and Dictionary of Medicine, Nursing, and Allied Health, 7th ed., Elsevier, Saunders, 2003. [21] J.N. Mansbridge, A.M. Knapp, Changes in keratinocyte maturation during wound healing, J. Invest. Dermatol. 89 (1987) 253–263. [22] M.L. Usui, R.A. Underwood, P. Fleckman, J.E. Olerud, Parakeratotic corneocytes play a unique role in human skin wound healing, J. Invest. Dermatol. 133 (March (3)) (2013) 856–858. [23] N.E.I. Langlois, G.A. Gresham, The aging of bruises; a review and study of the colour changes with time, Forensic Sci. Int. 50 (1991) 227–238. [24] P. Betz, A. Nerlich, J. Wilske, J. Tu¨bel, R. Penning, W. Eisenmenger, Time-dependent appearance of myofibroblasts in granulation tissue of human skin wounds, Int. J. Legal Med. 105 (2) (1992) 99–103. [25] P. Betz, et al., Analysis of the immunohistochemical localization of collagen type III and V for the time-estimation of human skin wounds, Int. J. Legal Med. 105 (6) (1993) 329–332. [26] P. Betz, et al., The time-dependent rearrangement of the epithelial basement membrane in human skin wounds – immunohistochemical localization of collagen IV and VII, Int. J. Legal Med. 105 (2) (1992) 93–97. [27] B.V. Shekhonin, V.F. Kondalenko, S.P. Domogatskiı˘, A.V. Rudin, V.S. Rukosuev, Distribution of collagen types I, III, IV and V in the walls of human arteries, Kardiologiia 24 (April (4)) (1984) 95–99. [28] P. Betz, Histological and enzyme histochemical parameters for the age estimation of human skin wounds, Int. J. Legal Med. 107 (1994) 60–68. [29] N. Vedrenne, B. Coulomb, A. Danigo, F. Bonte´, A. Desmoulie`re, The complex dialogue between (myo)fibroblasts and the extracellular matrix during skin repair processes and ageing, Pathol. Biol. (Paris) 60 (February (1)) (2012) 20–27.
