burns 42 (2016) 1471–1476

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/burns

Analyzing contraction of full thickness skin grafts

in time: Choosing the donor site does matter

Carlijn M. Stekelenburg a,b,c,d,e

[3_TD$IF] , Janine M. Simons a,
Wim E. Tuinebreijer a,b,c,d,e, Paul P.M. van Zuijlen a,b,c,d,e,
[7_TD$IF] *
a
Burn Centre, Red Cross Hospital, Beverwijk, The Netherlands
b
Association of Dutch Burn Centres, Beverwijk, The Netherlands
c
Department of Plastic, Reconstructive and Hand Surgery, Red Cross Hospital, Beverwijk, The Netherlands
d
Department of Plastic, Reconstructive and Hand Surgery, VU Medical Centre, Amsterdam, The Netherlands
e
MOVE Research Institute, VU University, Amsterdam, The Netherlands

article info abstract

Article history: Background: In reconstructive burn surgery full thickness skin grafts (FTSGs) are frequently
Accepted 2 February 2016 preferred over split thickness skin grafts because they are known to provide superior
esthetic results and less contraction. However, the contraction rate of FTSGs on the long
Keywords: term has never been studied.
Skin grafting Methods: The surface area of FTSGs of consecutive patients was measured during surgery
Burn surgery and at their regular follow up (at approximately 1, 6,13 and 52 weeks postoperatively) by
Cicatrix, Burn scar contracture means of 3D-stereophotogrammetry. Linear regression analysis was conducted to assess
Full thickness skin graft the influence of age, recipient- and donor site and operation indication.
Results: 38 FTSGs in 26 patients, with a mean age of 37.4 (SD 21.9) were evaluated. A
significant reduction in remaining surface area to 79.1% was observed after approximately
6 weeks ( p = 0.002), to 85.9% after approximately 13 weeks ( p = 0.040) and to 91.5% after
approximately 52 weeks ( p = 0.033). Grafts excised from the trunk showed significantly less
contraction than grafts excised from the extremities (94.0% vs. 75.7% p = 0.036).
Conclusions: FTSGs showed a significant reduction in surface area, followed by a relaxation
phase, but remained significantly smaller. Furthermore, the trunk should be preferred as
donor site location over the extremities.
# 2016 Elsevier Ltd and ISBI. All rights reserved.

FTSGs are preferred over split thickness skin grafts (STSGs)

1. Introduction because they give a superior esthetic result and less contrac-
tion [2,3]. Remarkably, the extent of contraction of FTSGs in
Despite new developments in acute and reconstructive burn burn patients has never been objectified. Most often burn
surgery, such as dermal substitution and perforator based patients are treated with an FTSG to improve the range of
interposition flaps [1], full thickness skin grafts (FTSGs) are motion. Therefore, the extra tissue that is inserted should
regularly needed as first choice for reconstruction [2]. Usually retain its initial surface area to result in a successful

* Corresponding author at:. Department of Plastic, Reconstructive and Hand Surgery, Red Cross Hospital Beverwijk, Vondellaan 13, 1942 LE
Beverwijk, The Netherlands. Tel.: +0031 251265785; fax: +0031 251265342.
E-mail address: [email protected] (Paul P.M. van Zuijlen).
http://dx.doi.org/10.1016/j.burns.2016.02.001
0305-4179/# 2016 Elsevier Ltd and ISBI. All rights reserved.
1472 burns 42 (2016) 1471–1476

procedure. FTSGs that are used for scar contracture release are the use of reliable and valid measurement instruments is
positioned in scar tissue, which differs considerably from important. 3D stereophotogrammetry is one of the most
healthy tissue in terms of elasticity and contractile forces. The recent advances in the field of surface area measurement and
graft that is inserted in the defect is subject to these contractile has been proven to reliably and validly measure surface area
forces and this could affect the contraction rate of FTSGs. [9]. The aim of this study was to evaluate the surface area of
Several other factors may influence the contraction rate of FTSGs over time using a reliable and valid measurement tool
FTSGs such as the age of the patient and the location where and to identify potential predictive factors that influence the
the skin is harvested [3]. It has been reported that FTSGs on the surface area over time.
nose and peri-orbital area demonstrate more contraction than
other recipient areas [4]. As differences in contraction rates
according to the recipient location were found, likely 2. Methods
differences in donorsite location may as well be present,
which has only been observed in animal studies [5]. Further- 2.1. Patients
more patient characteristics like age of the patient at time of
surgery have been assumed to influence the contraction rate In this clinical observational study, we analyzed a cohort of
of grafts [3,6]. Skin laxity is thought to increase with age and consecutive patients that received FTSGs as a reconstructive
thereby older patients might show less contraction than procedure between April 2011 and November 2013 at the
younger patients [7]. department of plastic, reconstructive and hand surgery in the
Literature up to now, though providing some information Red Cross Hospital (Beverwijk, the Netherlands). Patients were
on potential influencing parameters, does not suffice in a seen at their regular follow up moments as part of the medical
clear understanding of the contraction rate of FTSGs on the treatment. Fig. 1 represents a flow chart. Also patients
long term and its potential predictive factors. Two studies participating in other clinical studies were included. All
describe contraction of FTSGs in reconstructive procedures patients of 12 years and older with scars that are treated in
over time [4,8]. One study found a significant reduction in our clinic, undergo a standard scar evaluation protocol at
surface area within the first month after surgery, but no follow up. This scar evaluation protocol was approved by the
significant difference was found beyond the first month [8]. local medical ethical committee and includes scar surface area
Another study stated that FTSGs undergo a significant measurements. From all patients informed consent was
amount of contraction; a mean remaining surface area of obtained. The following data were collected: age of the
62% was found [4]. As these studies use a relatively short patient, the donor site, the recipient site, the indication for
follow up period [4] and non validated surface area operation and the presence of risk factors such as diabetes
measurement techniques [8], results from these studies are mellitus or smoking was registered for each patient. These
less applicable for interpretation in clinical practice. To characteristics were registered to include in the analysis as
measure the outcome of a treatment technique objectively, potential risk factors.

[(Fig._1)TD$IG]

Fig. 1 – Flow chart representing the drop outs, due to lack of follow up.
 burns 42 (2016) 1471–1476 1473

[8_TD$IF]2.2. Operation technique considered clinically relevant and provided a substantial

number of measurements. The first group represented the
The FTSGs were applied using the following technique. The measurements performed directly after surgery, the second
skin together with a thin layer of subcutaneous fat was group the measurements within the first three weeks after
harvested. Defatting of the subcutaneous fat from the dermis surgery (C1w), the third group (C6w) the short term follow-up
was performed adequately to facilitate survival of the graft. measurements (around 6 weeks), the fourth group (C13w) the
The graft was fixated using synthetic absorbable polyglactin measurements from approximately 13 weeks, and the fifth
suture. A tie-over was made and[2_TD$IF] removed after approximately group (C52w) the long term (i.e. about 1 year) follow up
seven days. measurements.
In the linear mixed effects model absolute values of surface
[9_TD$IF]2.3. Procedure areas were used. However, because of the different compila-
tions of the groups, also the means of the differences in
Measurement of the surface area of the graft was performed percentages were calculated and presented per group. The
by means of 3D-stereophotogrammetry, for which the 3D main effects of age, reason for reconstruction, recipient and
LifeViz (DermaPix software; QuantifiCare S.A., Sophia Anti- donor site location, were included in a separate linear
polis, France) was used. This is a reliable and valid measure- regression model. The significance criterion was set to 0.05.
ment method to measure surface area of scars, which was
proven in a validation study, previously published by our
group [9]. Surface area measurements were performed during [1_TD$IF]3. Results
surgery and at follow up. At surgery, the FTSG was photo-
graphed and measured directly after being sutured on the [12_TD$IF]3.1. Demographics
recipient location (before applying the tie-over). This mea-
surement was the baseline measurement for comparison with 38 grafts in 26 patients (13 male and 13 female) with a mean
the follow up measurements and is referred to as the initial age of 37.4 (SD 21.9) were included. Five patients were
measurement. Follow up surface area measurements were smokers and none had a history of vascular disorders such as
performed at each planned visit of the patient to the diabetes mellitus, peripheral arterial disease or vascular
outpatient clinic. Care was taken that during each measure- malformations. The indication for grafting was burn scar
ment of the graft, the patient adopted the same position. Post- contracture release in 27 cases. The mean age of the scar at
operative complications were registered. Necrosis was sub- time of surgery was 11.9 years (SD 15.5). In 11 cases the
divided in complete and partial necrosis. Patients were indication of grafting was wound closure, of which 6 were
excluded from analysis in case no photographs were taken small burn wounds, 4 defects were non-healing wounds after
beyond 5 months follow up. regular surgery and 1 defect was due to excision of instable
scar tissue. Recipient locations were: the face and neck
[10_TD$IF]2.4. Statistical analysis (N = 20), the upper extremity (N = 11), the lower extremity
(N = 5) and the trunk (N = 2). Donor sites were located on: the
Statistical analysis was performed using SPSS 21.0. Normal trunk (N = 25 of which 2 from the thorax, 7 from the groin, 9
distribution was tested by calculating the skewness and from the abdomen and 7 from the back), the upper extremity
kurtosis, evaluating frequency histograms, and performing (N = 12), and the lower extremity (N = 1). Two grafts suffered
the Shapiro–Wilcoxon test. To compare repeated measure- from[13_TD$IF] partial necrosis consisting of respectively 23.6 and
ments over time, a linear mixed effects model with a random 39.4% of the surface area.
intercept and analysis of covariance was used. The covariance
structure was set to variance components. The surface area at [14_TD$IF]3.2. Surface area reduction
surgery and at different follow up moments was used as
dependent variable in the models. Time from surgery was Contraction was observed in 28 of the 38 FTSGs (74.0%). Two
divided into categories and set as fixed effect. Five follow up grafts suffered from partial necrosis, but no re-operation was
categories (C0w, C1w, C6w, C13w and C52w) were created needed to close the defect. The linear mixed effects model
(Table 1). The categories were chosen in a way they were showed that within the first three weeks after surgery the

Table 1 – Time categories of the linear mixed effects model.

Category Phase Time Range Description
C0w At surgery 0 weeks – Surface area directly after surgery
C1w Graft survival Approximately 1 week 1–21 days The viability of the FTSG/presence of necrosis is
assessed after approximately 1 week
C6w Contraction—short Approximately 6 weeks 21–84 days After 6 weeks, the FTSG has definitely taken
term and the first signs of contraction may be visible
C13w Contraction phase Approximately 84–168 days End of the contraction phase. Entering the transition
three months phase to relaxation and maturation
C52w Maturation Approximately one year 168–1095 days End of the maturation phase
phase—end
1474 burns 42 (2016) 1471–1476
[(Fig._2)TD$IG]
( p = 0.036 95% CI 1.054–31.321), with a mean contraction rate
of 94.0% compared with 75.7%.

[16_TD$IF]4. Discussion

This is the first study that investigated the long-term outcome

of FTSGs with respect to contraction. Interestingly, our results
are inconsistent with the idea that FTSGs hardly contract. The
strength of this research is that measurements were carried
out at multiple occasions during follow up for each patient,
making it possible to visualize a trend in the contraction
pattern. Also, surface area measurements were carried out by
a reliable and valid 3D-stereophotogrammetry measurement
technique [9].
FTSGs are frequently used for the reconstruction of various
defects. They are often preferred over STSGs because STSGs
are known to contract considerably and moreover provide an
inferior cosmetic result. The purpose of reconstruction of a
Fig. 2 – Graph of the mean remaining graft surface area in defect or release of a scar contracture is to add tissue on a
mm2 of the grafts over time from surgery. Data points on location where tissue is too little or of poor condition. Ideally,
the x-axis represent the different ‘‘time from surgery’’ the added tissue remains its initial surface area or even
categories. The means and ranges for the different time increases in surface area, thereby anticipating the intended
categories are as follows: C0w: 1506.1 (48.6–10438.4), C1w: result. However, this study, with multiple follow up measure-
1642.7 (47.1–8472.5), C6w: 903.8 (29.1–2138.6), C12w: ments within a mean follow period of approximately 52
1382.7 (164.8–4524.0) and C52: 1414.6 (32.8–9491.8). weeks, showed that the surface area of full thickness grafts
decreased over time. Moreover, the vast majority of the FTSGs
showed contraction at final follow up.
Literature on the contraction rate of FTSGs for the
surface area of the grafts does not change substantially. A reconstruction of scar contractures is lacking up to now.
significant reduction in surface area to 79.1% was observed in Nevertheless there is one study that describes the contraction
category C6w (short term follow up) ( p = 0.002). In category rate in FTSGs used for reconstruction of defects caused by
C13w, a reduction to 85.9% ( p = 0.040) was seen and in category excision of skin tumors [4]. A remaining surface area of 62%
C52w a reduction to 91.5.0% ( p = 0.033). Fig. 2 shows the was observed after a mean follow up period of 111 days [4]. It is
change in mean remaining surface area (absolute values) at however thought that contraction of scar tissue develops over
the different follow up moments. Figs. 3 and 4 show two time, with a strong contraction phase in the first couple of
examples of contraction in FTSGs. months, followed by a relaxation phase. This pattern of
contraction has been seen clinically, but has not been proven
[15_TD$IF]3.3. Predictive factors yet by clinical studies in human participants. Therefore, to
approach the long-term results more adequately, the present
The linear regression model with the analysis of possible study included participants with a mean follow up period of
predictive factors showed that only the donor site location approximately 52 weeks. Our results showed that within the
significantly influenced the surface area at final follow up first three weeks (C1w) after surgery, the surface area of FTSGs
(Table 2). Grafts excised from the trunk showed less contrac- does not change substantially. After approximately 6 weeks
[(Fig._3)TD$IG]tion than grafts that were excised from the extremities (C6w) however, the surface area declined significantly (Fig. 2).

Fig. 3 – An example of a 22-year[4_TD$IF] old patient that suffered from a contracture of the elbow. The contracture was surgically
released with an FTSG (left). The photograph on the right was taken 4 weeks postoperatively.
 burns 42 (2016) 1471–1476 1475
[(Fig._4)TD$IG]

Fig. 4 – An example of a patient that suffered from a functional disabling contracture in the neck. The contracture was
released with an FTSG, which is shown on the left. The photograph on the right was taken 12 months after the surgical
release.

Table 2 – Linear regression analysis, with different potential predictive factors: reason for reconstruction (wound covering
compared to contracture release), donor site location (extremity compared to trunk), recipient site[1_TD$IF] (trunk compared to face/
neck) and age.
Parameter Regression coefficient 95% Confidence interval Significance

Lower Upper
Reason for operation 10.15 26.22 5.92 0.22
Donorsite 16.19 1.05 31.32 0.04
Recipient site 11.65 20.77 44.08 0.48
Age in years 0.30 0.04 0.64 0.08

This contraction phase was then followed by a relaxation In the present study, we found a significant difference in
phase after approximately 13 weeks (C13w). Up to around 52 contraction rates between donor site locations. This finding is
weeks (C52w) the grafts persisted to expand. They remained of clinical importance because it implicates that when
however significantly smaller at final follow up compared to possible, the trunk should be preferred over the extremities
the initial measurements ( p = 0.033, Fig. 2). A similar pattern of to serve as donor site location whenever a FTSG is to be
contraction followed by relaxation has been seen in STSGs; the harvested. Grafts that were excised from the trunk endured
grafts showed even higher rates of contraction with a mean less contraction than grafts excised from the extremities.
remaining surface area of 75.4% (SD 36.8 and p = 0.004) at a final This could be explained by the fact that skin of the trunk is
follow up of one year [10]. Assumptions have been made about generally thicker than skin of the extremities [16]. It may be
the factors that play a role in the etiology of contraction. suggested that a graft with a thicker dermis more effectively
Several studies implied an important role of the myofibroblast; protects the graft against contraction of the underlying
a differentiated fibroblast that is a key cell for connective tissue wound bed than a graft with a thinner dermis. Also, a thicker
remodeling and that has been identified in both normal and dermis likely contains a more extensive collagen network,
pathological tissues [11–13]. Because of its contractile structure which is more capable of stretching [17,18]. In the light of this
and its strong retractile activity compared with e.g. a hypothesis it is important to question whether contraction of
protomyofibroblast, the myofibroblast is thought to play an the graft is caused by the graft itself or by the underlying
important role in wound contraction [11]. Besides this, wound bed. In the present study a significant influence of the
myofibroblasts are over expressed in hypertrophic scars and recipient site could not be confirmed, however this does not
these scars often exist together with contractures [11,14,15]. exclude its influence. It is our belief that the contraction of
However, the exact role of the myofibroblast in the etiology of the underlying wound bed contributes to the final graft
wound contraction and its possibility as a target for treatment contraction. More in depth research should be performed to
remains poorly understood. Because of an increased tension in confirm this hypothesis. Finally, age was not found to be a
contractures, we expected grafts that were used to treat predictive factor for the contraction rates of grafts, which
contractures to react differently in terms of contraction could be caused by the relatively small inclusion number of
compared to grafts used to cover wound defects. Though our this study.
data showed that grafts used for contracture release endured In conclusion, this study was the first long term evaluation
more contraction on the long term (data not shown), no of the contraction patterns of FTSGs: a strong contraction
significant difference was found to support this assumption. phase was seen in the first 13 weeks, followed by a relaxation
1476 burns 42 (2016) 1471–1476

phase. The majority of grafts however, remained contracted [5] Corps BV. The effect of graft thickness, donor site and graft
after a mean follow up period of one year or more. bed on graft shrinkage in the hooded rat. Br J Plast Surg
1969;22:125–33.
Understanding of the contraction patterns of FTSGs is of
[6] Ashcroft GS, Horan MA, Ferguson MW. The effects of ageing
great help for the reconstructive surgeon. We believe that,
on cutaneous wound healing in mammals. J Anat
despite our findings, FTSGs should retain an important role in 1995;187(Pt 1):1–26.
the reconstruction of skin defects and contractures. However, [7] Hermanns-Le T, Jonlet F, Scheen A, Pierard GE. Age- and
an increased knowledge on the contraction pattern allows to body mass index-related changes in cutaneous shear wave
anticipate on the expected contraction and/or to use another velocity. Exp Gerontol 2001;36:363–72.
reconstruction technique in cases where a re-contraction is [8] Tuncali D, Ates L, Aslan G. Upper eyelid full-thickness skin
graft in facial reconstruction. Dermatol Surg 2005;31:65–70.
absolutely unwanted, such as scar contracture release. Based
[9] Stekelenburg CM, van der Wal MB, Knol DL, de Vet HC, van
on the results that were presented in this study, it should be Zuijlen PP. Three-dimensional digital
recommended to harvest a greater FTSG than is needed on the stereophotogrammetry: a reliable and valid technique for
recipient location, to anticipate the following contraction. measuring scar surface area. Plast Reconstr Surg
Furthermore, considering our finding that grafts excised from 2013;132:204–11.
the trunk endure significantly less contraction on the long [10] van Zuijlen PP. Perspectives on burn scar evaluation and
artificial skin. Amsterdam: Plastic Recontronstructive and
term than grafts excised from the extremities, we advise to,
Hand Surgery; 2002.
whenever possible, use the trunk as the donorsite location of
[11] Desmouliere A, Guyot C, Gabbiani G. The stroma reaction
preference when an FTSG is required. myofibroblast: a key player in the control of tumor cell
behavior. Int J Dev Biol 2004;48:509–17.
[12] Gabbiani G. The myofibroblast: a key cell for wound healing
Conflict of interest statement and fibrocontractive diseases. Prog Clin Biol Res
1981;54:183–94.
[13] Gomez DS, Mariani U, Pinto WS, Gemperli R, Ferreira MC.
The authors state no conflict of interest.
Contraction and myofibroblasts in restored skin. Scand J
Plast Reconstr Surg Hand Surg 1998;32:147–55.
[14] Micallef L, Vedrenne N, Billet F, Coulomb B, Darby IA,
Acknowledgements Desmouliere A. The myofibroblast, multiple origins for
major roles in normal and pathological tissue repair.
This research is supported by the Dutch Burns Foundation. We Fibrogenesis Tissue Repair 2012;5(1 Suppl):S5.
gratefully thank the patients for their participation in this [15] van der Veer WM, Bloemen MC, Ulrich MM, Molema G, van
Zuijlen PP, Middelkoop E, et al. Potential cellular and
study.
molecular causes of hypertrophic scar formation. Burns
2009;35:15–29.
[16] Nedelec B, Forget NJ, Hurtubise T, Cimino S, de Muszka F,
references
Legault A, et al. Skin characteristics: normative data for
elasticity, erythema, melanin and thickness at 16 different
anatomical locations. Skin Res Technol 2015;0:1–13. http://
[1] Verhaegen PD, Stekelenburg CM, van Trier AJ, Schade FB, dx.doi.org/10.1111/srt.12256.
van Zuijlen PP. Perforator-based interposition flaps for [17] Verhaegen PD, Schouten HJ, Tigchelaar-Gutter W, van
sustainable scar contracture release: a versatile, practical, Marle J, van Noorden CJ, Middelkoop E, et al. Adaptation of
and safe technique. Plast Reconstr Surg 2011;127:1524–32. the dermal collagen structure of human skin and scar
[2] Mathes SJ. Plastic surgery. 2nd ed. Philadelphia, PA: tissue in response to stretch: an experimental study.
Saunders; 2006. Wound Repair Regen 2012;20:658–66.
[3] Harrison CA, MacNeil S. The mechanism of skin graft [18] Verhaegen PD, van Zuijlen PP, Pennings NM, van Marle J,
contraction: an update on current research and potential Niessen FB, van der Horst CM, et al. Differences in collagen
future therapies. Burns 2008;34:153–63. architecture between keloid, hypertrophic scar,
[4] Stephenson AJ, Griffiths RW, La Hausse-Brown TP. Patterns normotrophic scar, and normal skin: an objective
of contraction in human full thickness skin grafts. Br J Plast histopathological analysis. Wound Repair Regen
Surg 2000;53:397–402. 2009;17:649–56.