- Email: [email protected]
Retrospective analysis of tissue expansion in reconstructive burn surgery: Evaluation of complication rates
burns 34 (2008) 1113–1118
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/burns
Retrospective analysis of tissue expansion in reconstructive burn surgery: Evaluation of complication rates A. Bozkurt 1,*, A. Groger 1, D. O’Dey, F. Vogeler, A. Piatkowski, P.Ch. Fuchs, N. Pallua Department of Plastic, Reconstructive, and Hand Surgery, Burn Center, University Hospital RWTH Aachen, Germany
article info
abstract
Article history:
Tissue expansion is a widespread and accepted concept in reconstructive surgery, but is also
Accepted 11 May 2008
afflicted with a variety of complications. In burn patients, this technique allows large areas of burn scar to be replaced by tissue of similar texture and colour to the defect.
Keywords:
We retrospectively reviewed our results with tissue expanders in 57 burn patients over a
Thermal injury
period of 8 years including 102 expanders. Statistical analyses revealed a significant influence
Burn scarring
of the anatomical region ( p = 0.0156; Chi-square = 15.6811) and of the expander volume
Tissue expansion
( p = 0.0417; Chi-square = 18.8918) on the failure rate outcome. Factors such as age, gender, number of expanders per patient and shape of expander showed no statistical correlation in relation to the failure rate ( p-values >0.05). Furthermore, we present a short review of the recent literature of complications after tissue expansion. The presented study may help to draw attention on different aspects in tissue expansion and critically focus on each step of the tissue expansion procedure from implantation over inflation to explantation. # 2008 Elsevier Ltd and ISBI. All rights reserved.
1.
Introduction
The concept of tissue expansion in surgical practice was first reported by Neumann in 1957, while skin expansion was pioneered independently by Radovan, Austad, and Lapin [1,2]. Primarily established for breast reconstruction, skin expansion represents one of the major developments in reconstructive surgery in recent years, particularly as a valuable approach for many problems in reconstructive burn surgery [1,2]. In the past, reconstruction strategies like facial resurfacing procedures generally included the use of split thickness skin grafts. Incomplete graft take resulted in recurrent scarring and pigment imbalances with a reduced aesthetic outcome. Tissue expansion on the other hand, allows large areas of burn scar to be resurfaced and provides tissue of similar texture and colour to the defect to be covered and has the advantage of minimal donor site morbidity. Furthermore, the expanded tissue displays high vascularity, which is considered to be superior to surgically delayed flaps [3–6]. * Corresponding author. Tel.: +49 241 80800; fax: +49 241 82448. E-mail address: [email protected] (A. Bozkurt). 1 Joint first authors. 0305-4179/$34.00 # 2008 Elsevier Ltd and ISBI. All rights reserved. doi:10.1016/j.burns.2008.05.008
However, to fully harness these advantages and achieve success, the expansion process must also aim to minimise complications. Tissue expansion places a foreign body beneath the skin, which is then inflated over a period of time. Although the procedure is based on a simple concept, this technique is associated with complications [7–16]. The current study comprises a detailed evaluation of complications after tissue expansion in burn patients. The purpose is to present a critical analysis of the outcome over an 8 year period at a German university hospital burn centre. We retrospectively analysed the clinical outcome including complication rates and failure rates of our burn patient collective. Furthermore, we reviewed previous retrospective studies and compared our results to those reported in the literature. Tissue expansion is a popular and effective treatment strategy in reconstructive burn surgery. However, since this technique is afflicted with complications, a comprehensive retrospective analysis maybe of great interest to direct the attention on possible complications.
1114
burns 34 (2008) 1113–1118
2.
Patients and methods
2.2.
2.1.
Patient population and method of expansion
Significance analyses were studied by means of Chi-square (Epistat, T.F.Gustafson, Epistat services Richardson, TX). The significance of influence factors such as expander shape, expander volume, number of expanders per patient, gender and the anatomical site of implantation on the patient’s outcome (failure rate) were studied. p-Values less than 0.05 were regarded as significant.
The details of all patients having undergone tissue expansion between 1998 and 2006 were obtained by retrospective case note view. Age, gender, diagnosis, site of expander, expansion timings and method of closure were recorded. Major complications of expansion were defined as infection (clinical redness associated with pain, pyrexia, and solution of symptoms with antibiotic treatment), dislocation, leakage, exposure, wound dehiscence, extrusion, and necrosis. Minor complications included haematoma, seroma, delayed wound healing longer than 14 days (see also Pisarski et al. (1998) or http://www.emedicine.com). If active treatment of complications led to completion of expansion this was termed ‘‘salvage’’; if the programme had to be abandoned it was termed ‘‘failure’’. All burn patients were operated on either by the senior author or under his direct supervision. One-hundred and two expanders were placed in 57 patients between 1998 and 2006; this series comprised 28 men and 29 women with a mean age of 28 years (range: 5–58 years; median: 20 years). The mean of the total burn surface area (TBSA) was 27.4% (S.D.: 20.1; range: 2–70%; median: 25%). Twenty-three patients received a single expander, while 25 patients received two expanders; three expanders were implanted in seven patients and four expanders in two patients. Tissue expansion was not used to resurface acute burns. The protocol for expansion was standardised and was not changed since the start of the observation period (1998). Retrospectively, these standards were in agreement with the recommendations as summarized seven years later by Hudson and Grob (2005, [17]). After precise preoperative planning, tissue expanders were inserted under general anaesthesia with preoperative prophylactic antibiotics. The rationale in choosing the site for expansion was defined by the location of the scar tissue: the expander was implanted in a sufficiently large area of healthy skin adjacent to the burn scar. The incision was made in healthy skin next to the burn scar (not through the scar) with a short incision. The pocket had to be larger than the dimensions of the expander (approximately 1 cm larger in length and breadth) and was created by using non-traumatic blunt dissectiors. The pocket was flushed with a mixture of betadine diluted in normal saline to achieve local antisepsis. After inserting a suction drain, the wound was closed in two layers and the expander was filled to 10% of its capacity. At the same time care was taken to eliminate any sharp crease or knuckling which could cause rupture or exposure in the future. The surgical wound was left visible through a window in the cast to monitor the undermined skin and detect any early signs of damage. The sequential filling procedure was started after the wounds had healed soundly for a minimum of three weeks, using volumes dictated by patient comfort at weekly and fortnightly intervals in the outpatient department. Following complete expansion, a further time interval of 2–4 weeks was allowed to elapse prior to definitive surgery. When the prostheses were removed, a local flap with a suction drain was used to resurface the site and the wound was again placed in a plaster cast with an open window.
3.
Statistical analysis
Results
A total of 102 expanders were placed for reconstruction in 57 burn patients. Time of treatment ranged between 4 and 18 weeks with a mean of 11.73 3.88 weeks. In most cases, rectangular expanders (n = 49) were applied, followed by croissant-shaped expanders (n = 31). Round (n = 10), tearshaped (n = 6) and oval expanders (n = 5) were used infrequently. Expanders with internal valves were implanted in 90.2% of the cases (n = 92), external valves were more rarely applied (n = 10; 9.8%). Anatomically, most expanders were implanted in the trunk (n = 46), followed by the upper limb (n = 24). Almost identical numbers were implanted in the lower limbs (n = 17) and the head region (n = 15). Further detailed, the back (n = 18), arms (n = 18), the thigh (n = 14) and the shoulder (n = 13) received the most expander implantations. Expanded volumes ranged from <100 ml to 1000 ml. Most frequently, volumes of 400 and 200 ml were used. The size of area for expansion was indirectly calculated from the area of the used expanders (for example: A = p r2 for round expanders [A = area; p = 3.14; r = radius] or A = a b for rectangular expanders [a = length; b = width]) and the mean size was 130.5 cm2 (20.5). In 73 out of 102 cases (71.6%) tissue expansion was achieved without complications. At the same time, in 19 cases (18.6%) minor complications and in 10 cases (9.8%) major complications occurred. These complications resulted in failure of tissue expansion of 8 cases (7.8%) (Table 1). Regarding age, there were no failures in those <10 years and >40 years (n = 41 cases) observed. Between 10 and 20 years, 12 out of 38 cases developed complications leading to 5 failures. Likewise, in the group between 20 and 30 years, 9 out of 23 cases had complications with 3 three cases resulting in unsuccessful tissue expansions (Table 1). Further detailed information is provided in Table 2 regarding the respective reasons leading to complications with salvage or failure. Statistical analyses revealed a significant influence of the region on the failure rate outcome ( p = 0.0156; Chi-square = 15.6811). Determination of the failure case ratio in respective anatomic locations (complications with failure/total number of cases) revealed highest failure rates in the head with 20% (3/15), followed by the trunk with 9% (4/46) and the upper limb with 4% (1/24): no failure was observed in the lower limb 0% (0/17) (see Table 2, Fig. 1). Regarding the general complication case ratio (complications with failure + complications with salvage/total number of cases), the sequence changed with the highest ratio in the lower limb 47% (8/17), followed by the head with 40% (6/15), upper limb 30% (7/24) and the trunk with 17% (8/46). A more detailed distribution for the
1115
burns 34 (2008) 1113–1118
Table 1 – Expander outcome related to the age of patients Age
<10 <20 <30 <40 <50 <60
Without complications years years years years years years
Total (S = 102)
With complications !Salvage
11 26 14 9 10 3
3 7 6 4 1 0
73 (71.6%)
!Failure 0 5 3 0 0 0
21 (20.6%)
8 (7.8%)
corresponding anatomical locations can be depicted from Table 2 and Fig. 1. Likewise, there was also a significant influence of the expander volume on the failure rate outcome ( p = 0.0417; Chisquare = 18.8918). The failure case ratio for each volume class was highest for 400 ml and 800 ml expanders with each 16.6% (3/18), followed by 100 ml with 12.5% (1/8) and 700 ml with 10% (1/10). In contrast, remaining factors such as age, gender, number of expander per patient and shape of expander had no statistical influence of the failure outcome ( p-values >0.05). A statistical significant correlation between co-morbidities of enrolled patients could not be found.
4.
Fig. 1 – Distribution of cases without and with complications (implantation site).
Fig. 2 – Distribution of cases without and with complications (volume of expander).
Discussion
Tissue expanders are one of the most frequently used implants in plastic and reconstructive surgery [15,17,18]. Here, we systematically examined the outcome of tissue expansion in our burn patient collective over the last 8 years (Figs. 1 and 2, Tables 1 and 2). In addition, for reasons of comparability, we used the Medline database (pubmed.org) for reviewing and comparing our data with previous retrospective studies (Table 3). In this study we observed that both expander volume and anatomical region had a significant influence on the failure rate ( p-values < 0.05) (Figs. 1 and 2; Table 2). In contrast, remaining factors like age, gender, number of expander per patient and shape of expander had no statistical influence on failure outcomes ( p-values >0.05). From methodological aspects, all expander cases (n = 102) were treated as independent observations. The number of expanders per patient (ratio of expander number per patient) was tested as not being significantly influential
( p = 0.9273) for the outcome (i.e. failure rate). Despite this pre-test, this assumption may cause some inflation of the statistical power. However, we believe that there is independence of outcomes in having multiple expanders per patient. Thus, all 102 expander cases were tested for statistical differences regarding risk factors like gender, age, volumes, side, and expander shape. Statistical differences were regarded as p < 0.05. In general, tissue expanders are considered as a great advance in reconstructive surgery. This method particularly shows beneficial effects for the reconstruction of burn patients with extensive scarring. Despite the disadvantage of being (at least) a two-stage procedure, the expander technique provides tissue of the same texture and colour with minimal donor site morbidity. The success of the tissue expansion procedure depends on the appropriate indication for its use, individual risk
Table 2 – Detailed presentation of the occurred complications and correlation with the outcome (S = salvage, F = failure) in the respective anatomical sites Head Haematoma Leakage Infection Exposure Necrosis Wound dehiscence
Face
Neck
Shoulder
Thorax
Abdomen
Back
2S 1F
2S 1F 1F
1S
Upper limb
1F
1S
1S
1F
2F
Lower limb 2S 2S
1S 1F 2S
1116
Table 3 – Comparison with other studies Year
Patients (n)
Cases
Age: mean
Age: range
Indication
Region
Complication (%)
Marks [25] Cohen [19] Zaki [27] Neligan [18] Governa [12] Chun [21] Lozano [22] Hudson [23] Pitanguy [26] Filho [16] Pisarski [14] Pisarski [14] Friedmann [10] Antonyshyn [8] Manders [2] Zoltie [28] Vo¨gelin [20] Gibstein [11] Hawary [24] Youm [15] Casanova [7] Cunha [9] Pandya [13]
1987 1988 1989 1989 1996 1998 2000 2001 2002 2007 1998 1998 1996 1988 1988 1990 1995 1997 1998 1999 2001 2002 2002
45 22 10 12 157 25 28 14 132 23 224 (1987–1995) 77 (1984–1987) 82 66 15 56 (burn = 22) 34 105 97 30 95 164 88
n/a n/a n/a 42 262 51 34 67 346 54 281 122 180 (burn:54) 76 16 (burn:2) 76 37 191 (burn:3) 128 34 (burn:8) 103 315 (burn:157) 113 (burn:9)
n/a n/a n/a n/a 26 24 6 7.8 10 ! 19 n/a 10.7 n/a Median: 5 34 n/a 22 34 6 21 25 26.8 21 n/a
n/a 3 ! 59 5 ! 25 n/a n/a <1 ! 46 3 ! 14 3 ! 11 1 ! 60 5 ! 48 n/a n/a 1.5 ! 15 4 ! 89 6 ! 56 5 ! 74 18 ! 78 <1 ! 28 3 ! 56 1 ! 65 3 ! 77 3 ! 52 <1 ! 58
Burn Burn Burn Burn Burn Burn Burn Burn Burn Burn Burn Burn General General General General General General General General General General General
General General Head General General Head & Neck Head & Trunk Head & Neck General General General (head:254) General General General Lower limb General Lower limb General General General Lower limb General General
18 27 1 45.23 28 12–32% 18 21–28 7.5 25.9 18.1 30.3 17.7 (burn:22.2) 39 80 24 76 13 18 65 (burn:87) 19.4 22 32.7
Note: if not directly stated, complication and failure rates were partially calculated and interpreted from original data.
Failure (%) 8.80% 0 0 4.76 6.4 0 0 14–25 n/a 7.4 11 18.8 n/a 17 n/a 12 32.3 n/a 5 38 7.7 19.3 15
burns 34 (2008) 1113–1118
First author
burns 34 (2008) 1113–1118
factors, detailed individualized planning and, most importantly, following procedural factors as outlined by Hudson and Grob (2005) [size and form of the expander, use of antibiotics, insertion place of the expander, size of the expander pocket, drainage, wound closure technique, expander filling, placement of the port and begin of tissue expansion] [17]. Nevertheless, similar to every surgical procedure, tissue expansion is associated with postoperative complications (see Tables 1 and 2). Thus, detailed analyses of complications occurring during tissue expansion are essential to optimize operative results. We used the Medline database (pubmed.org) for reviewing and comparing the present survey with previous retrospective studies. However, direct comparison is limited due to varying definitions of complications (e.g. major or minor complication vs. absolute or relative) and rates (either presented only general complication rates or failure rates). Furthermore, complication or failure rates of expander therapy for burn patients were not always explicitly stated, but had to be calculated from provided data (for example [10,15]). However, as outlined in Table 3, we collected data from previous studies and arranged them based on factors including (a) number of subjects, (b) number of cases, (c) age (mean & range), (d) indication, (e) region, (f) complication rate and g) failure rate for reasons of better comparability. Studies providing data of all anatomical regions (‘‘Region: General’’, see Table 3) failure rates ranged between 0% (n = 22 patients in [19]) and 18.8% (n = 77 in [14]). This is comparable with our overall failure rate of 7.8%. Interestingly, we had highest complication rates in the lower limb (8/17 = 47%) and lowest in the trunk (8/46 = 17%). This observed frequency of complications was comparable with findings of Manders et al. (80%) [2] and Vo¨gelin et al. (76%) [20], but not with Casanova et al. (19.4%) [7]. However, direct comparison is limited since these three studies examined not only the indication ‘‘Burn’’ but also other indications (e.g. naevi, tumour, etc.). Concerning our data we, expected that the expander volume has a statistical significant influence. From our clinical experience, this is most likely due to the larger tissue pocket with the required tissue mobilization. This, on the other hand, is associated with a higher risk for infections, haematoma, larger peri-prosthetic capsular formation and contracture, or, most importantly, severe impairment of the intra- and subdermal vascular system [1]. In contrast, when analyzing the statistical influence of the corresponding anatomical region, we were surprised that (major and minor) complications in the lower limbs could always be salvaged and did not lead to a failure of the expander procedure. We actually expected higher failure rates in the lower extremity due to factors like the thinner protective soft tissue covering the osseous skeleton, increased hydrostatic pressure, decreased venous backflow (resulting in oedema, low oxygen partial pressure, etc.) or cumulative retention of metabolites [1]. A possible reason may be intensified education as well as increased awareness of the medical staff (e.g. closer wound monitoring) and higher compliance of the patients, since it is widely known that the lower extremity is more prone for complications than other localizations. In the head region we had 3 failure cases. In contrast to Chun et al. and Lozano et al. [21,22] who
1117
presented failure rates in the head and neck of 0%, our data is more consistent with that of Hudson et al. [23] with failure rates between 14 and 25% in the head and neck region.
5.
Conclusion
It is not always possible to measure the absolute efficacy of a surgical technique or determine a general guideline its indication of the tissue expansion procedure. However, although afflicted with a broad range of possible complications, the tissue expansion procedure remains a valuable and reliable technique for the reconstruction of burn patients suffering from extensive scarring. Knowledge of the frequency of complications, precise instruction of the medical stuff as well as detailed and continuous education of the patients may help to further increase the efficacy of the tissue expansion process.
references
[1] Pallua N, O’Dey D. II-7.1 Gewebeexpansion— Grundprinzipien und Funktion. In: Krupp S, Rennekampff S, Pallua N, editors. Plastische Chirurgie. Ecomed: Landsberg; 2006. p. 1–14. [2] Manders EK, Oaks TE, Au VK, Wong RK, Furrey JA, Davis TS, et al. Soft-tissue expansion in the lower extremities. Plast Reconstr Surg 1988;81(2):208–19. [3] Heitland AS, Pallua N. The single and double-folded supraclavicular island flap as a new therapy option in the treatment of large facial defects in noma patients. Plast Reconstr Surg 2005;115(6):1591–6. [4] Pallua N, von Heimburg D. Pre-expanded ultra-thin supraclavicular flaps for (full-) face reconstruction with reduced donor-site morbidity and without the need for microsurgery. Plast Reconstr Surg 2005;115(7):1837–44 [discussion 1837-1845]. [5] Pallua N, Magnus Noah E. The tunneled supraclavicular island flap: an optimized technique for head and neck reconstruction. Plast Reconstr Surg 2000;105(3):842–51 [discussion 852-844]. [6] Pallua N, Machens HG, Rennekampff O, Becker M, Berger A. The fasciocutaneous supraclavicular artery island flap for releasing postburn mentosternal contractures. Plast Reconstr Surg 1997;99(7):1878–84 [discussion 1876-1885]. [7] Casanova D, Bali D, Bardot J, Legre R, Magalon G. Tissue expansion of the lower limb: complications in a cohort of 103 cases. Br J Plast Surg 2001;54(4):310–6. [8] Antonyshyn O, Gruss JS, Zuker R, Mackinnon SE. Tissue expansion in head and neck reconstruction. Plast Reconstr Surg 1988;82(1):58–68. [9] Cunha MS, Nakamoto HA, Herson MR, Faes JC, Gemperli R, Ferreira MC. Tissue expander complications in plastic surgery: a 10-year experience. Rev Hosp Clin Fac Med Sao Paulo 2002;57(3):93–7. [10] Friedman RMIAJ, Rohrich RJ, Byrd HS, Hodges PL, Burns AJ, Hobar PC. Risk factors for complications in pediatric tissue expansion. Plast Reconstr Surg 1996;98(7):1242–6. [11] Gibstein LA, Abramson DL, Bartlett RA, Orgill DP, Upton J, Mulliken JB. Tissue expansion in children: a retrospective study of complications. Ann Plast Surg 1997;38(4):358–64. [12] Governa M, Bonolani A, Beghini D, Barisoni D. Skin expansion in burn sequelae: results and complications. Acta Chir Plast 1996;38(4):147–53.
1118
burns 34 (2008) 1113–1118
[13] Pandya AN, Vadodaria S, Coleman DJ. Tissue expansion in the limbs: a comparative analysis of limb and non-limb sites. Br J Plast Surg 2002;55(4):302–6. [14] Pisarski GP, Mertens D, Warden GD, Neale HW. Tissue expander complications in the pediatric burn patient. Plast Reconstr Surg 1998;102(4):1008–12. [15] Youm T, Margiotta M, Kasabian A, Karp N. Complications of tissue expansion in a public hospital. Ann Plast Surg 1999;42(4):396–401 [discussion 392-401]. [16] Tavares Filho JM, Belerique M, Franco D, Porchat CA, Franco T. Tissue expansion in burn sequelae repair. Burns 2007;33(2):246–51. [17] Hudson DA, Grob M. Optimising results with tissue expansion: 10 simple rules for successful tissue expander insertion. Burns 2005;31(1):1–4. [18] Neligan PC, Peters WJ. Advances in burn scar reconstruction: the use of tissue expansion. Ann Plast Surg 1989;22(3):203–10. [19] Cohen M, Marschall MA, Schafer ME. Tissue expansion for the reconstruction of burn defects. J Trauma 1988;28(2):158–63. [20] Vogelin E, de Roche R, Luscher NJ. Is soft tissue expansion in lower limb reconstruction a legitimate option? Br J Plast Surg 1995;48(8):579–82.
[21] Chun JT, Rohrich RJ. Versatility of tissue expansion in head and neck burn reconstruction. Ann Plast Surg 1998;41(1): 11–6. [22] Lozano S, Drucker M. Use of tissue expanders with external ports. Ann Plast Surg 2000;44(1):14–7. [23] Hudson DA, Arasteh E. Serial tissue expansion for reconstruction of burns of the head and neck. Burns 2001;27(5):481–7. [24] Hawary MB. Tissue expansion King Khalid University experience. Ann Saudi Med 1998;18(5):454–6. [25] Marks MW, Argenta LC, Thornton JW. Burn management: the role of tissue expansion. Clin Plast Surg 1987;14(3): 543–8. [26] Pitanguy I, Gontijo de Amorim NF, Radwanski HN, Lintz JE. Repeated expansion in burn sequela. Burns 2002;28(5): 494–9. [27] Zaki MS. The use of tissue expansion for reconstruction of post-burn alopecia. Annals MBC 1989:2. Published online. http://www.medbc.com/annals/review/vol_2/num_2/text/ vol2n2p95.htm. [28] Zoltie N, Chapman P, Joss G. Tissue expansion: a unit review of non-scalp, non-breast expansion. Br J Plast Surg 1990;43(3):325–7.
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/burns
Retrospective analysis of tissue expansion in reconstructive burn surgery: Evaluation of complication rates A. Bozkurt 1,*, A. Groger 1, D. O’Dey, F. Vogeler, A. Piatkowski, P.Ch. Fuchs, N. Pallua Department of Plastic, Reconstructive, and Hand Surgery, Burn Center, University Hospital RWTH Aachen, Germany
article info
abstract
Article history:
Tissue expansion is a widespread and accepted concept in reconstructive surgery, but is also
Accepted 11 May 2008
afflicted with a variety of complications. In burn patients, this technique allows large areas of burn scar to be replaced by tissue of similar texture and colour to the defect.
Keywords:
We retrospectively reviewed our results with tissue expanders in 57 burn patients over a
Thermal injury
period of 8 years including 102 expanders. Statistical analyses revealed a significant influence
Burn scarring
of the anatomical region ( p = 0.0156; Chi-square = 15.6811) and of the expander volume
Tissue expansion
( p = 0.0417; Chi-square = 18.8918) on the failure rate outcome. Factors such as age, gender, number of expanders per patient and shape of expander showed no statistical correlation in relation to the failure rate ( p-values >0.05). Furthermore, we present a short review of the recent literature of complications after tissue expansion. The presented study may help to draw attention on different aspects in tissue expansion and critically focus on each step of the tissue expansion procedure from implantation over inflation to explantation. # 2008 Elsevier Ltd and ISBI. All rights reserved.
1.
Introduction
The concept of tissue expansion in surgical practice was first reported by Neumann in 1957, while skin expansion was pioneered independently by Radovan, Austad, and Lapin [1,2]. Primarily established for breast reconstruction, skin expansion represents one of the major developments in reconstructive surgery in recent years, particularly as a valuable approach for many problems in reconstructive burn surgery [1,2]. In the past, reconstruction strategies like facial resurfacing procedures generally included the use of split thickness skin grafts. Incomplete graft take resulted in recurrent scarring and pigment imbalances with a reduced aesthetic outcome. Tissue expansion on the other hand, allows large areas of burn scar to be resurfaced and provides tissue of similar texture and colour to the defect to be covered and has the advantage of minimal donor site morbidity. Furthermore, the expanded tissue displays high vascularity, which is considered to be superior to surgically delayed flaps [3–6]. * Corresponding author. Tel.: +49 241 80800; fax: +49 241 82448. E-mail address: [email protected] (A. Bozkurt). 1 Joint first authors. 0305-4179/$34.00 # 2008 Elsevier Ltd and ISBI. All rights reserved. doi:10.1016/j.burns.2008.05.008
However, to fully harness these advantages and achieve success, the expansion process must also aim to minimise complications. Tissue expansion places a foreign body beneath the skin, which is then inflated over a period of time. Although the procedure is based on a simple concept, this technique is associated with complications [7–16]. The current study comprises a detailed evaluation of complications after tissue expansion in burn patients. The purpose is to present a critical analysis of the outcome over an 8 year period at a German university hospital burn centre. We retrospectively analysed the clinical outcome including complication rates and failure rates of our burn patient collective. Furthermore, we reviewed previous retrospective studies and compared our results to those reported in the literature. Tissue expansion is a popular and effective treatment strategy in reconstructive burn surgery. However, since this technique is afflicted with complications, a comprehensive retrospective analysis maybe of great interest to direct the attention on possible complications.
1114
burns 34 (2008) 1113–1118
2.
Patients and methods
2.2.
2.1.
Patient population and method of expansion
Significance analyses were studied by means of Chi-square (Epistat, T.F.Gustafson, Epistat services Richardson, TX). The significance of influence factors such as expander shape, expander volume, number of expanders per patient, gender and the anatomical site of implantation on the patient’s outcome (failure rate) were studied. p-Values less than 0.05 were regarded as significant.
The details of all patients having undergone tissue expansion between 1998 and 2006 were obtained by retrospective case note view. Age, gender, diagnosis, site of expander, expansion timings and method of closure were recorded. Major complications of expansion were defined as infection (clinical redness associated with pain, pyrexia, and solution of symptoms with antibiotic treatment), dislocation, leakage, exposure, wound dehiscence, extrusion, and necrosis. Minor complications included haematoma, seroma, delayed wound healing longer than 14 days (see also Pisarski et al. (1998) or http://www.emedicine.com). If active treatment of complications led to completion of expansion this was termed ‘‘salvage’’; if the programme had to be abandoned it was termed ‘‘failure’’. All burn patients were operated on either by the senior author or under his direct supervision. One-hundred and two expanders were placed in 57 patients between 1998 and 2006; this series comprised 28 men and 29 women with a mean age of 28 years (range: 5–58 years; median: 20 years). The mean of the total burn surface area (TBSA) was 27.4% (S.D.: 20.1; range: 2–70%; median: 25%). Twenty-three patients received a single expander, while 25 patients received two expanders; three expanders were implanted in seven patients and four expanders in two patients. Tissue expansion was not used to resurface acute burns. The protocol for expansion was standardised and was not changed since the start of the observation period (1998). Retrospectively, these standards were in agreement with the recommendations as summarized seven years later by Hudson and Grob (2005, [17]). After precise preoperative planning, tissue expanders were inserted under general anaesthesia with preoperative prophylactic antibiotics. The rationale in choosing the site for expansion was defined by the location of the scar tissue: the expander was implanted in a sufficiently large area of healthy skin adjacent to the burn scar. The incision was made in healthy skin next to the burn scar (not through the scar) with a short incision. The pocket had to be larger than the dimensions of the expander (approximately 1 cm larger in length and breadth) and was created by using non-traumatic blunt dissectiors. The pocket was flushed with a mixture of betadine diluted in normal saline to achieve local antisepsis. After inserting a suction drain, the wound was closed in two layers and the expander was filled to 10% of its capacity. At the same time care was taken to eliminate any sharp crease or knuckling which could cause rupture or exposure in the future. The surgical wound was left visible through a window in the cast to monitor the undermined skin and detect any early signs of damage. The sequential filling procedure was started after the wounds had healed soundly for a minimum of three weeks, using volumes dictated by patient comfort at weekly and fortnightly intervals in the outpatient department. Following complete expansion, a further time interval of 2–4 weeks was allowed to elapse prior to definitive surgery. When the prostheses were removed, a local flap with a suction drain was used to resurface the site and the wound was again placed in a plaster cast with an open window.
3.
Statistical analysis
Results
A total of 102 expanders were placed for reconstruction in 57 burn patients. Time of treatment ranged between 4 and 18 weeks with a mean of 11.73 3.88 weeks. In most cases, rectangular expanders (n = 49) were applied, followed by croissant-shaped expanders (n = 31). Round (n = 10), tearshaped (n = 6) and oval expanders (n = 5) were used infrequently. Expanders with internal valves were implanted in 90.2% of the cases (n = 92), external valves were more rarely applied (n = 10; 9.8%). Anatomically, most expanders were implanted in the trunk (n = 46), followed by the upper limb (n = 24). Almost identical numbers were implanted in the lower limbs (n = 17) and the head region (n = 15). Further detailed, the back (n = 18), arms (n = 18), the thigh (n = 14) and the shoulder (n = 13) received the most expander implantations. Expanded volumes ranged from <100 ml to 1000 ml. Most frequently, volumes of 400 and 200 ml were used. The size of area for expansion was indirectly calculated from the area of the used expanders (for example: A = p r2 for round expanders [A = area; p = 3.14; r = radius] or A = a b for rectangular expanders [a = length; b = width]) and the mean size was 130.5 cm2 (20.5). In 73 out of 102 cases (71.6%) tissue expansion was achieved without complications. At the same time, in 19 cases (18.6%) minor complications and in 10 cases (9.8%) major complications occurred. These complications resulted in failure of tissue expansion of 8 cases (7.8%) (Table 1). Regarding age, there were no failures in those <10 years and >40 years (n = 41 cases) observed. Between 10 and 20 years, 12 out of 38 cases developed complications leading to 5 failures. Likewise, in the group between 20 and 30 years, 9 out of 23 cases had complications with 3 three cases resulting in unsuccessful tissue expansions (Table 1). Further detailed information is provided in Table 2 regarding the respective reasons leading to complications with salvage or failure. Statistical analyses revealed a significant influence of the region on the failure rate outcome ( p = 0.0156; Chi-square = 15.6811). Determination of the failure case ratio in respective anatomic locations (complications with failure/total number of cases) revealed highest failure rates in the head with 20% (3/15), followed by the trunk with 9% (4/46) and the upper limb with 4% (1/24): no failure was observed in the lower limb 0% (0/17) (see Table 2, Fig. 1). Regarding the general complication case ratio (complications with failure + complications with salvage/total number of cases), the sequence changed with the highest ratio in the lower limb 47% (8/17), followed by the head with 40% (6/15), upper limb 30% (7/24) and the trunk with 17% (8/46). A more detailed distribution for the
1115
burns 34 (2008) 1113–1118
Table 1 – Expander outcome related to the age of patients Age
<10 <20 <30 <40 <50 <60
Without complications years years years years years years
Total (S = 102)
With complications !Salvage
11 26 14 9 10 3
3 7 6 4 1 0
73 (71.6%)
!Failure 0 5 3 0 0 0
21 (20.6%)
8 (7.8%)
corresponding anatomical locations can be depicted from Table 2 and Fig. 1. Likewise, there was also a significant influence of the expander volume on the failure rate outcome ( p = 0.0417; Chisquare = 18.8918). The failure case ratio for each volume class was highest for 400 ml and 800 ml expanders with each 16.6% (3/18), followed by 100 ml with 12.5% (1/8) and 700 ml with 10% (1/10). In contrast, remaining factors such as age, gender, number of expander per patient and shape of expander had no statistical influence of the failure outcome ( p-values >0.05). A statistical significant correlation between co-morbidities of enrolled patients could not be found.
4.
Fig. 1 – Distribution of cases without and with complications (implantation site).
Fig. 2 – Distribution of cases without and with complications (volume of expander).
Discussion
Tissue expanders are one of the most frequently used implants in plastic and reconstructive surgery [15,17,18]. Here, we systematically examined the outcome of tissue expansion in our burn patient collective over the last 8 years (Figs. 1 and 2, Tables 1 and 2). In addition, for reasons of comparability, we used the Medline database (pubmed.org) for reviewing and comparing our data with previous retrospective studies (Table 3). In this study we observed that both expander volume and anatomical region had a significant influence on the failure rate ( p-values < 0.05) (Figs. 1 and 2; Table 2). In contrast, remaining factors like age, gender, number of expander per patient and shape of expander had no statistical influence on failure outcomes ( p-values >0.05). From methodological aspects, all expander cases (n = 102) were treated as independent observations. The number of expanders per patient (ratio of expander number per patient) was tested as not being significantly influential
( p = 0.9273) for the outcome (i.e. failure rate). Despite this pre-test, this assumption may cause some inflation of the statistical power. However, we believe that there is independence of outcomes in having multiple expanders per patient. Thus, all 102 expander cases were tested for statistical differences regarding risk factors like gender, age, volumes, side, and expander shape. Statistical differences were regarded as p < 0.05. In general, tissue expanders are considered as a great advance in reconstructive surgery. This method particularly shows beneficial effects for the reconstruction of burn patients with extensive scarring. Despite the disadvantage of being (at least) a two-stage procedure, the expander technique provides tissue of the same texture and colour with minimal donor site morbidity. The success of the tissue expansion procedure depends on the appropriate indication for its use, individual risk
Table 2 – Detailed presentation of the occurred complications and correlation with the outcome (S = salvage, F = failure) in the respective anatomical sites Head Haematoma Leakage Infection Exposure Necrosis Wound dehiscence
Face
Neck
Shoulder
Thorax
Abdomen
Back
2S 1F
2S 1F 1F
1S
Upper limb
1F
1S
1S
1F
2F
Lower limb 2S 2S
1S 1F 2S
1116
Table 3 – Comparison with other studies Year
Patients (n)
Cases
Age: mean
Age: range
Indication
Region
Complication (%)
Marks [25] Cohen [19] Zaki [27] Neligan [18] Governa [12] Chun [21] Lozano [22] Hudson [23] Pitanguy [26] Filho [16] Pisarski [14] Pisarski [14] Friedmann [10] Antonyshyn [8] Manders [2] Zoltie [28] Vo¨gelin [20] Gibstein [11] Hawary [24] Youm [15] Casanova [7] Cunha [9] Pandya [13]
1987 1988 1989 1989 1996 1998 2000 2001 2002 2007 1998 1998 1996 1988 1988 1990 1995 1997 1998 1999 2001 2002 2002
45 22 10 12 157 25 28 14 132 23 224 (1987–1995) 77 (1984–1987) 82 66 15 56 (burn = 22) 34 105 97 30 95 164 88
n/a n/a n/a 42 262 51 34 67 346 54 281 122 180 (burn:54) 76 16 (burn:2) 76 37 191 (burn:3) 128 34 (burn:8) 103 315 (burn:157) 113 (burn:9)
n/a n/a n/a n/a 26 24 6 7.8 10 ! 19 n/a 10.7 n/a Median: 5 34 n/a 22 34 6 21 25 26.8 21 n/a
n/a 3 ! 59 5 ! 25 n/a n/a <1 ! 46 3 ! 14 3 ! 11 1 ! 60 5 ! 48 n/a n/a 1.5 ! 15 4 ! 89 6 ! 56 5 ! 74 18 ! 78 <1 ! 28 3 ! 56 1 ! 65 3 ! 77 3 ! 52 <1 ! 58
Burn Burn Burn Burn Burn Burn Burn Burn Burn Burn Burn Burn General General General General General General General General General General General
General General Head General General Head & Neck Head & Trunk Head & Neck General General General (head:254) General General General Lower limb General Lower limb General General General Lower limb General General
18 27 1 45.23 28 12–32% 18 21–28 7.5 25.9 18.1 30.3 17.7 (burn:22.2) 39 80 24 76 13 18 65 (burn:87) 19.4 22 32.7
Note: if not directly stated, complication and failure rates were partially calculated and interpreted from original data.
Failure (%) 8.80% 0 0 4.76 6.4 0 0 14–25 n/a 7.4 11 18.8 n/a 17 n/a 12 32.3 n/a 5 38 7.7 19.3 15
burns 34 (2008) 1113–1118
First author
burns 34 (2008) 1113–1118
factors, detailed individualized planning and, most importantly, following procedural factors as outlined by Hudson and Grob (2005) [size and form of the expander, use of antibiotics, insertion place of the expander, size of the expander pocket, drainage, wound closure technique, expander filling, placement of the port and begin of tissue expansion] [17]. Nevertheless, similar to every surgical procedure, tissue expansion is associated with postoperative complications (see Tables 1 and 2). Thus, detailed analyses of complications occurring during tissue expansion are essential to optimize operative results. We used the Medline database (pubmed.org) for reviewing and comparing the present survey with previous retrospective studies. However, direct comparison is limited due to varying definitions of complications (e.g. major or minor complication vs. absolute or relative) and rates (either presented only general complication rates or failure rates). Furthermore, complication or failure rates of expander therapy for burn patients were not always explicitly stated, but had to be calculated from provided data (for example [10,15]). However, as outlined in Table 3, we collected data from previous studies and arranged them based on factors including (a) number of subjects, (b) number of cases, (c) age (mean & range), (d) indication, (e) region, (f) complication rate and g) failure rate for reasons of better comparability. Studies providing data of all anatomical regions (‘‘Region: General’’, see Table 3) failure rates ranged between 0% (n = 22 patients in [19]) and 18.8% (n = 77 in [14]). This is comparable with our overall failure rate of 7.8%. Interestingly, we had highest complication rates in the lower limb (8/17 = 47%) and lowest in the trunk (8/46 = 17%). This observed frequency of complications was comparable with findings of Manders et al. (80%) [2] and Vo¨gelin et al. (76%) [20], but not with Casanova et al. (19.4%) [7]. However, direct comparison is limited since these three studies examined not only the indication ‘‘Burn’’ but also other indications (e.g. naevi, tumour, etc.). Concerning our data we, expected that the expander volume has a statistical significant influence. From our clinical experience, this is most likely due to the larger tissue pocket with the required tissue mobilization. This, on the other hand, is associated with a higher risk for infections, haematoma, larger peri-prosthetic capsular formation and contracture, or, most importantly, severe impairment of the intra- and subdermal vascular system [1]. In contrast, when analyzing the statistical influence of the corresponding anatomical region, we were surprised that (major and minor) complications in the lower limbs could always be salvaged and did not lead to a failure of the expander procedure. We actually expected higher failure rates in the lower extremity due to factors like the thinner protective soft tissue covering the osseous skeleton, increased hydrostatic pressure, decreased venous backflow (resulting in oedema, low oxygen partial pressure, etc.) or cumulative retention of metabolites [1]. A possible reason may be intensified education as well as increased awareness of the medical staff (e.g. closer wound monitoring) and higher compliance of the patients, since it is widely known that the lower extremity is more prone for complications than other localizations. In the head region we had 3 failure cases. In contrast to Chun et al. and Lozano et al. [21,22] who
1117
presented failure rates in the head and neck of 0%, our data is more consistent with that of Hudson et al. [23] with failure rates between 14 and 25% in the head and neck region.
5.
Conclusion
It is not always possible to measure the absolute efficacy of a surgical technique or determine a general guideline its indication of the tissue expansion procedure. However, although afflicted with a broad range of possible complications, the tissue expansion procedure remains a valuable and reliable technique for the reconstruction of burn patients suffering from extensive scarring. Knowledge of the frequency of complications, precise instruction of the medical stuff as well as detailed and continuous education of the patients may help to further increase the efficacy of the tissue expansion process.
references
[1] Pallua N, O’Dey D. II-7.1 Gewebeexpansion— Grundprinzipien und Funktion. In: Krupp S, Rennekampff S, Pallua N, editors. Plastische Chirurgie. Ecomed: Landsberg; 2006. p. 1–14. [2] Manders EK, Oaks TE, Au VK, Wong RK, Furrey JA, Davis TS, et al. Soft-tissue expansion in the lower extremities. Plast Reconstr Surg 1988;81(2):208–19. [3] Heitland AS, Pallua N. The single and double-folded supraclavicular island flap as a new therapy option in the treatment of large facial defects in noma patients. Plast Reconstr Surg 2005;115(6):1591–6. [4] Pallua N, von Heimburg D. Pre-expanded ultra-thin supraclavicular flaps for (full-) face reconstruction with reduced donor-site morbidity and without the need for microsurgery. Plast Reconstr Surg 2005;115(7):1837–44 [discussion 1837-1845]. [5] Pallua N, Magnus Noah E. The tunneled supraclavicular island flap: an optimized technique for head and neck reconstruction. Plast Reconstr Surg 2000;105(3):842–51 [discussion 852-844]. [6] Pallua N, Machens HG, Rennekampff O, Becker M, Berger A. The fasciocutaneous supraclavicular artery island flap for releasing postburn mentosternal contractures. Plast Reconstr Surg 1997;99(7):1878–84 [discussion 1876-1885]. [7] Casanova D, Bali D, Bardot J, Legre R, Magalon G. Tissue expansion of the lower limb: complications in a cohort of 103 cases. Br J Plast Surg 2001;54(4):310–6. [8] Antonyshyn O, Gruss JS, Zuker R, Mackinnon SE. Tissue expansion in head and neck reconstruction. Plast Reconstr Surg 1988;82(1):58–68. [9] Cunha MS, Nakamoto HA, Herson MR, Faes JC, Gemperli R, Ferreira MC. Tissue expander complications in plastic surgery: a 10-year experience. Rev Hosp Clin Fac Med Sao Paulo 2002;57(3):93–7. [10] Friedman RMIAJ, Rohrich RJ, Byrd HS, Hodges PL, Burns AJ, Hobar PC. Risk factors for complications in pediatric tissue expansion. Plast Reconstr Surg 1996;98(7):1242–6. [11] Gibstein LA, Abramson DL, Bartlett RA, Orgill DP, Upton J, Mulliken JB. Tissue expansion in children: a retrospective study of complications. Ann Plast Surg 1997;38(4):358–64. [12] Governa M, Bonolani A, Beghini D, Barisoni D. Skin expansion in burn sequelae: results and complications. Acta Chir Plast 1996;38(4):147–53.
1118
burns 34 (2008) 1113–1118
[13] Pandya AN, Vadodaria S, Coleman DJ. Tissue expansion in the limbs: a comparative analysis of limb and non-limb sites. Br J Plast Surg 2002;55(4):302–6. [14] Pisarski GP, Mertens D, Warden GD, Neale HW. Tissue expander complications in the pediatric burn patient. Plast Reconstr Surg 1998;102(4):1008–12. [15] Youm T, Margiotta M, Kasabian A, Karp N. Complications of tissue expansion in a public hospital. Ann Plast Surg 1999;42(4):396–401 [discussion 392-401]. [16] Tavares Filho JM, Belerique M, Franco D, Porchat CA, Franco T. Tissue expansion in burn sequelae repair. Burns 2007;33(2):246–51. [17] Hudson DA, Grob M. Optimising results with tissue expansion: 10 simple rules for successful tissue expander insertion. Burns 2005;31(1):1–4. [18] Neligan PC, Peters WJ. Advances in burn scar reconstruction: the use of tissue expansion. Ann Plast Surg 1989;22(3):203–10. [19] Cohen M, Marschall MA, Schafer ME. Tissue expansion for the reconstruction of burn defects. J Trauma 1988;28(2):158–63. [20] Vogelin E, de Roche R, Luscher NJ. Is soft tissue expansion in lower limb reconstruction a legitimate option? Br J Plast Surg 1995;48(8):579–82.
[21] Chun JT, Rohrich RJ. Versatility of tissue expansion in head and neck burn reconstruction. Ann Plast Surg 1998;41(1): 11–6. [22] Lozano S, Drucker M. Use of tissue expanders with external ports. Ann Plast Surg 2000;44(1):14–7. [23] Hudson DA, Arasteh E. Serial tissue expansion for reconstruction of burns of the head and neck. Burns 2001;27(5):481–7. [24] Hawary MB. Tissue expansion King Khalid University experience. Ann Saudi Med 1998;18(5):454–6. [25] Marks MW, Argenta LC, Thornton JW. Burn management: the role of tissue expansion. Clin Plast Surg 1987;14(3): 543–8. [26] Pitanguy I, Gontijo de Amorim NF, Radwanski HN, Lintz JE. Repeated expansion in burn sequela. Burns 2002;28(5): 494–9. [27] Zaki MS. The use of tissue expansion for reconstruction of post-burn alopecia. Annals MBC 1989:2. Published online. http://www.medbc.com/annals/review/vol_2/num_2/text/ vol2n2p95.htm. [28] Zoltie N, Chapman P, Joss G. Tissue expansion: a unit review of non-scalp, non-breast expansion. Br J Plast Surg 1990;43(3):325–7.